Columbus State University

CSU ePress

Theses and Dissertations Student Publications

6-2006

Online Survey System: A Web-Based Tool for Creating and

Administering Student Evaluations Online
Fatima Marie Idowu
Columbus State University

Follow this and additional works at: https://csuepress.columbusstate.edu/theses_dissertations

Part of the Computer Sciences Commons

Recommended Citation
Idowu, Fatima Marie, "Online Survey System: A Web-Based Tool for Creating and Administering Student
Evaluations Online" (2006). Theses and Dissertations. 71.
https://csuepress.columbusstate.edu/theses_dissertations/71

This Thesis is brought to you for free and open access by the Student Publications at CSU ePress. It has been
accepted for inclusion in Theses and Dissertations by an authorized administrator of CSU ePress.
 Digitized by the Internet Archive
in 2012 with funding from
LYRASIS Members and Sloan Foundation

http://archive.org/details/onlinesurveysystOOidow
 Columbus State University

The College of Science

The Graduate Program in Applied Computer Science

Online Survey System

A Web-Based Tool for Creating and Administering Student Evaluations Online

A Thesis in

Applied Computer Science

by

Fatima Marie Idowu

Submitted in Partial Fulfillment

of the Requirements
for the Degree of

Masters of Science

June 2006

K © 2006 by Fatima Marie Idowu

 11

I have submitted this thesis in partial fulfillment of the requirements for the degree of
Master of Science.

$ iW? ILooQ hsu^/naS^^fc

Date Fatima Marie Idowu

We approve the thesis of Fatima Marie Idowu as presented here.

3qrv£ g g00£?
Date" Christopher Whitehead, Assistant
Professor of Computer Science, Thesis
Advisor

Date Bhagyavati, Assistant Professor of

Computer Science

dK
J^rueL * £s00-£
Date Eugen Ionascu, Associate Professor of
Mathematics

Date Paulina Kuforiji, Associate Professor of

Education
 Ill

Abstract

With the advancement in technology over the years, the administering of online surveys has

expanded. In particular, universities are using the online medium to administer surveys to

students, in order to evaluate faculty performances. The move of surveys to the online realm

has meant a reduction in cost, time and efforts, of survey administrators, and the increase in

use of technology within universities. With the use of online surveys the challenges of

confidentiality, anonymity and response rates are as prominent as they are with paper-based

surveys.

This study researched the use of online surveys in education; detailing systems currently used

by many universities to facilitate the creation of an Online Survey System that would solve

the latter challenges. The principal idea was to create a system that would provide the

Columbus State University Computer Science department with a web-based tool for creating

surveys and administering them online. The surveys will be created by faculty members;

Survey Administrators, and administered to students; Survey lakers. With the system,

faculty members are able to create surveys for classes taught, providing questions and

responses deemed suitable. Once a survey is administered, students can access the survey by

logging onto the system. Upon authentication students are able to complete surveys online.

Surveys are generated dynamically, depending on the survey criteria supplied by the survey

administrator. The system stores survey criterion in a database, which is retrieved when

generating a survey for a survey taker to complete.

 IV

The system developed focused on providing functionality that would increase the

convenience of completing evaluation surveys and protect anonymity of students.

The created system was successful in its aim to provide these features, also providing ease

for faculty to create online surveys without having the extensive technical expertise that is

required to do so.
 1

Table of Contents

Abstract ill

Table of Contents v
List of Figures viii

List of Tables ix

List of Tables ix

Acknowledgements x
Acknowledgements x
1. Introduction 1

2. Purpose of the System/Motivation/Related Work 3

2.1 Online versus Paper Surveys 6
2.2 Advantages and Disadvantages of Online Surveys 1

2.3 Constructing Surveys 12

2.4 Issues with Response Rates 15
2.5 Overview of related work / Comparable Survey Systems in Current Use 17
3. Requirements Analysis 22
3.1 Introduction 22
3.1.1 Scope of the system 22
3.1.2 Objectives and success criteria of the project 23
3.2 Proposed System 24
3.2.1 Overview 24
3.2.2 Functional Requirements 25
3.2.2.1 User Authentication 25
3.2.2.2 Querying Database for previous survey questions 25
3.2.2.3 Dynamic Survey Creation 25
3.2.2.4 Survey Start and Cut-Off Dates 26
3.2.2.5 Email Notification 26
3.2.2.6 Generation of statistical survey result report 27
3.2.2.7 Allocation of Survey Administrators and user groups 27
3.2.2.8 Viewing completed surveys 27
3.2.2.9 Other aspects of the system 27
3.2.3 Justification of functional requirements 28
3.2.3.1 User Authentication 28
3.2.3.2 Querying Database for previous survey questions 28
3.2.3.3 Dynamic Survey Creation 28
3.2.3.4 Email Notification 28
3.2.3.5 Generation of statistical survey result report 29
3.2.3.6 Protection of anonymity 29
3.2.4 Non-functional requirements 29
3.2.4.1 User interface and human factors 29
3.2.4.2 Hardware and software considerations 30
3.2.4.3 Performance characteristics 30
3.2.4.4 Error Handling and Extreme Conditions 30
 VI

3.2.4.5 System Modifications 31

3.2.4.6 Operating Environment 31
3.2.4.7 Security Requirements 31
3.3 System models 31
3.3.1 Scenarios 31
3.3.2 Use Case Model 34
3.3.3 Use Case model Descriptions 36
3.3.3.1 The 'Log On' Use Case 36
3.3.3.2 The 'Create Survey' Use Case 37
3.3.3.3 The 'View Survey' Use Case 39
3.3.3.4 The 'Notify Survey Takers' Use Case 40
3.3.3.5 The 'Complete Survey' Use Case 41
3.3.3.6 The 'View Completed Survey' Use Case 42
3.3.4 Object Models 43
3.3.4.1 Class Diagram 44
3.5. Functional Requirements Cross-referenced and prioritized 46
Summary 48
Design and Implementation 49
4.1 Introduction 49
4.2 Design decisions 49
4.3 Proposed software architecture 52
Overview 52
4.4 Subsystem decomposition 53
4.5 Subsystem services 55
4.6 Survey Application Subsystem 56
4.6.1 User Interface 58
4.6.1.1 Login 58
4.6.1.2 Default 58
4.6.1.3 AdminMain 59
4.6.1.4 Create 61
4.6.1.5 Answer2 63
4.6.1.6 Survey 66
4.6.1.7 SendMail 68
4.6.1.8 Message Sent 70
4.6.1.9 Results 71
4.6.1.10 Main 72
4.6.1.11 Page 74
4.6.1.12 Survey Error 76
4.7 The Survey Application Object Subsystem 77
4.8 The Database Subsystem 79
Summary 81
5. System Evaluation 82
5.1. Functionality Testing 82
5.1.1. Choosing Test Cases 82
5.1.1.1. The 'Log On' Test Case 82
 Vll

5.1.1.2. The 'Create Survey' Test Case 83

5.1.1.3. The 'View Survey' Test Case 83
5.1.1.4. The 'Complete Survey' Test Case 84
5.1.2. Test cases and Results of Testing 84
Log_On_Test (1) 84
LogOn_Test (2) 84
Create_Survey_Test (1) 85
Create_Survey_Test (2) 85
Create_Survey_Test (3) 85
Create_Survey_Test (4) 86
View_Survey_Test (1) 86
View_Survey_Test (2) 86
View_Survey_Test (3) 87
View_Survey_Test (4) 87
Complete_Survey_Test 87
Summary 88
6. Conclusion 89
6.1. Future Enhancements 89
6.2. Project Summary 91
References 94
Appendix A 96
Ten Usability Heuristics 96
Appendix B 98
Survey Application Class Descriptions 98
Class _default 98
Class AdminMain 99
Class answers2 99
Class create 102
Class Main 104
Class messagesent 105
Class page 106
Class SendMail 108
Class survey 110
Class SurveyError 1 12
Appendix C 113
Class Descriptions 113
Class Answer 1 13
Class Pager 1 14
Class Question 115
Class QuestionCategory 116
Class Survey 117
Class SurveyConnection 119
Class SurveyUser 120
Class UserAuth 121
Class UserGroup 121
 Vlll

List of Figures

Figure 1: The use case model for the propose system 36

Figure 2: Abstract object model for proposed system 44
Figure 3: Login Interface of Proposed system 45
Figure 4: Create Survey Interface of Proposed system 46
Figure 5: Three Tier architecture for web applications 53
Figure 6: The subsystems for the Online Survey System 54
Figure 7: Object Model (class diagram) of the Survey Application subsystem 57
Figure 8: System default page 59
Figure 9: Survey Administrator Main page 61
Figure 10: Add New Question panel on the create page 63
Figure 11: Configuration of survey quesdon responses 65
Figure 12: Creating a new answer for a survey question 66
Figure 13: Survey created by survey administrator on survey page 67
Figure 14: Email notification on sendmail page 69
Figure 15: Reminders notificadon on sendmail page 70
Figure 16: Email confirmation 71
Figure 17: Survey response rate graph 72
Figure 18: Email nodfication received by survey taker 73
Figure 19: Survey Taker Main page 74
Figure 20: Survey compledon page 75
Figure 21: Previously completed survey 76
Figure 22: Page access error page 77
Figure 23: Object Model (class diagram) of the Survey Application Object subsystem 78
Figure 24: Entity Relationship diagram of the system database 80
 IX

List of Tables

Table 1: Actors for the system and their roles 35

Table 2: Initial objects Descriptions for the Online Survey System 44
Table 3: Functional requirements cross-referenced with use cases 48
Ac kno wle dge m e nts

Firstly I would like to give glory and praise to my Heavenly Father for giving me the peace,

wisdom and strength I needed to complete this thesis. Thank you Ford.

I want to express my deepest gratitude to Professor Whitehead for being so patient with me

and always being available when I needed help throughout the course of this project. For

your guidance and support, I am truly grateful.

I also want to thank the committee members, Bhagyavati, Eugen Ionascu, and Paulina

Kuforiji, for their comments and suggestion throughout the process of documenting this

thesis and for their approval of my work.

Finally I want to thank my parents and sisters for their love, support and prayers for me

during the course of this project. Thank you to Andrew Smith for guidance, suggestions and

constructive criticism of my work and Amir Fynton for encouraging and constantly assuring

me that I was able to complete this project.

God Bless You All and Thank You.

1. Introduction

For many decades, surveys have been used for research and as a means of obtaining

feedback from consumers. In fact, surveys are probably the most commonly used research

method world-wide [Pfleeger & Kitchenham, 2001]. Surveys can be administered in two

different forms: supervised and unsupervised. The means by which a survey is administered

is dependent upon the administrator's objectives and the resources available. Supervised

surveys are those where a survey researcher is assigned a respondent, or survey taker; for

example, telephone surveys require a researcher to ask the respondent a series of questions

and then to record the answers. Unsupervised surveys take the form of automated voice

telephone calls or mailed questionnaires.

In light of the advancement in technology over the years, the medium used to administer

surveys has expanded to the online realm. Companies have been able to administer their

surveys online in order to obtain feedback on their products and services. Due to the nature

of the Internet, respondents' answers are recorded, totaled, and ready for analysis

immediately, thus eliminating the need for the tallying of surveys' results by hand; thereby

saving time and money. In the past few years, this trend has moved into the education

system.

It is common practice among universities to give end-of-unit surveys in order to gain

feedback on the units offered at the institute. Over the years, the use of communication has

grown extensively in university teaching and learning. Many universities have expanded to

online curriculum and with this expansion have adopted online surveys as a method of
collecting student feedback. However, the use of online surveys for feedback on the

university's curriculum and performance of its professors is not limited to those taking

online classes. Universities are adopting online surveys as a mechanism to replace "pen and

paper" surveys across the board; online and face-to-face classes alike. Research done at

Murdoch University found that the growth in the use of online surveys developed from the

desire to align the surveys with the use of technology for classes, "to increase access to

external students, and to improve the efficiency of the survey process." [Cummings &

Ballantyne, 2000]

While online surveys are used in numerous universides all around the world, there is little

research literature invesdgating their use in education [Cummings & Ballantyne, 2000]. Few

researchers have investigated the use and creation of online surveys for the educational field

[Cummings & Ballantyne, 2000; Pfleeger & Kitchenham, 2002b; Ha & Mars, 1998].

The University of Leicester has a research project entitled 'Exploring Online Research

Methods.' The research resulted in the development of a website that aims to provide an

"online resource which provides training for researchers who are interested in using online

research methods such as online questionnaires and online interviews." The website targets

researchers and postgraduates in the higher education, and "researchers working for other

organizations such as those involved with public policy and market research," and is a great

resource providing online methodologies to those wishing to explore the online research

realm.
2. Purpose of the System/Motivation/ Related Work

The common practice at Columbus State University (CSU) is to administer paper-based end

of unit surveys. This allows departments to offer students the opportunity to evaluate their

classes and the faculty members who conducted these classes on the basis of their teaching

skills and the way the class or module was delivered. The main objectives of these

evaluations are to assess the instructor's performance, gain insight into student attitudes

about course content and assignments, and student satisfaction levels with quizzes, exams,

and the course in general. Evaluations are also used to assess changes in instructional

practices or new courses; specifically when this practice is used or the course is offered as

part of an experiment or a trial. The results of these surveys are generally used to determine

whether a class will be offered again, if the class was offered on a trial basis, or whether the

professor's teaching style needs to be adapted in anyway.

The Department of Computer Science at CSU currently uses both paper-based and online

evaluations. Some faculty members prefer the use of paper-based surveys while others prefer

their online counterparts. However, these ultimately contain the same information. The

Department of Computer Science has a designated individual who creates the online surveys

for the department and faculty are then responsible for pointing students to the appropriate

Internet address to take the survey.

The online surveys currently administered by the Department of Computer Science are static

in nature, meaning that the questions are predefined and each survey has the same questions

on them, these questions are the same as those used in the paper-based surveys; there is no

opportunity provided for faculty to pose their own questions regarding classes that they
teach. This is somewhat unfortunate as a faculty member teaching a class will more than

likely have questions specific to the class that he/she would like to ask the students.

The intent of this project, therefore, is to produce a system that is aimed at providing CSU

Computer Science faculty members with a tool which can be used to develop online

evaluation surveys for the courses taught and to administer these surveys to students when

they deem fit; most likely near the end of the semester or half semester, for those classes

offered for only half a semester. The system will allow faculty to supply questions that they

feel are relevant to their course rather than administering a static survey which is used college

wide. The system will be web based allowing faculty to create surveys when they wish and to

make them available to students. This system will be beneficial for both students and faculty,

as students will have the opportunity to take surveys in the convenience of their own time. It

is anticipated that this convenience will allow students to think about their responses; and

thus increase the quality of their responses. Feedback from students revealed that

completing paper-based surveys in a class room setting leads to the selection responses that

would not normally be chosen. The reason for this is that students feel rushed to complete

surveys in the time given. Whilst this notion is plausible, it has not been validated by

research.

The system currently in place has no way of restricting students from completing a class

survey more than once or even restricting students who did not take a particular class to

have access to complete the survey. The proposed system will utilize user authentication in

order to restrict users from completing a survey more that once and makes surveys available

only to those students specified by faculty.

There are three groups of individuals who are involved in the process of student evaluations

[Gaide, 2005]. These are system administrators, survey administrators and students. A survey

administrator is responsible for the creation of online surveys. In the proposed system the

faculty members will take on the role of survey administrators and students will take the role

of survey takers. The system will have a system administrator who will be responsible for

maintaining the system.

The success of student evaluations online is wholly dependent upon the faculty's support. It

is the faculty's responsibility to make students aware of the surveys they need to complete.

Faculty are, however, in a position where they can either undermine the survey by lack of

attention and despondent comments or promote the survey through supportive comments,

reminders and providing feedback on how past survey results were used to improve the unit

[Cummings & Ballantyne, 2000]. It is, therefore, anticipated that the system will encourage

faculty to promote surveys as it will be surveys that they created themselves and feedback

from those surveys will be more directed towards information they require about the

particular class rather than simply generic feedback about the teaching process.

Response rate is a major issue when administering surveys. Any reliable survey system

should measure and report its response rate and responses. Using the system presented in

this thesis, faculty will be able to view statistical results of completed surveys online, allowing

them to graphically see responses to the surveys instantly instead of having to wait for all

surveys to be taken and then tallied; as is the case with paper-based surveys.
The proposed system therefore aims to improve the process of administering surveys by

using the Internet as a medium and making the process more convenient for faculty and

students alike.

2.1 Online versus Paper Surveys

Two common types of surveys are paper-based and telephone surveys. However, with the

recent move made by higher education institutions to make their surveys available online, the

question arises as to the effectiveness of online surveys versus paper-based surveys. It has

been argued that comparison of responses between online and paper surveys shows no

difference [Kelly & Marsh, 1999].

There are various advantages when using an online surveying system for those involved in

the process. The common advantage to all involved is the flexibility and the "potendal to

improve the efficiency of the survey process." Alternatively, low response rates and

"increased reliance on access to technology" are idendfied as significant disadvantages

[Cummings & Ballantyne, 2000].

The creadon of surveys and the surveying process highlights many challenges. These

challenges range from confidentiality and anonymity to response rates. These challenges are

common, however, to both paper-based and online surveys.

The observed challenges that face an online surveying system are those of computer access

and literacy, security and confidentiality and response abuse, such as multiple submissions.
In asking the questions whether student feedback should be moved completely online, four

essential criteria were found to be influential [Cummings & Ballantyne, 2000].

These are "support of staff members involved, level of student access to information

technology, the lowest level of computer literacy in the student target group and the level of

student acceptance of undertaking tasks online." If surveys are to be fully administered

online, they must be promoted by staff or students will be unaware of their availability. Staff

members may be reluctant to promote evaluation surveys because they do not want to

change their teaching style or be critiqued by students. This also holds true with paper-based

surveys.

The access to hardware is a limitation to consider when thinking about online surveys. The

level of access that students across a campus or class have may vary considerably. The

physical access to appropriate hardware and software for students is not the only factor to be

considered when talking about access to information technology, but also access where and

when it is convenient for the student. With regard to access to hardware, class time could be

provided where students have access to the appropriate hardware and software needed to

complete a survey. Paper-based surveys are usually administered during class time.

Nevertheless, with both online and paper surveys, the issue of convenience is still at hand.

Are a couple of minutes before the end of class necessarily a convenient time for a student

to complete a survey? As a student myself, and this is true for many students that this issue

has been discussed with, when asked to complete a course evaluation in class, one would

prefer to complete it in his or her own time, as the student tends to feel rushed to complete

the survey within the few minutes at the end of a class or within the time frame given. The
 8

given responses therefore may not necessarily be the same as those students would have

given if given the opportunity to complete surveys in their own time. This is especially true

when one is required to give comments, as a student will not have time to ponder the

response they want to give.

The level of computer literacy among students and also among staff is another factor when

considering transferring surveys online. Unless their particular area of expertise affords

them to deal with information technology regularly, staff members, especially the ones less

familiar or comfortable with using technology, will have more problems administering online

surveys. The level of competence with computers is not necessarily evenly distributed among

the student population and also the staff population. However, many high school graduates

are fairly familiar with computer technology, as nowadays it is a requirement for high school

students to take computer literacy classes. There are however, some parts of the world where

it may not be necessary depending on the field the student wants to embark upon, .i.e.,

mature students also may lack experience or confidence with computers

If CSU is to administer all its surveys online, faculty need to be comfortable with using an

online surveying system, appropriate access to technology needs to be available to all

students asked to complete surveys. Students need a reasonable level of competency to

complete the survey and a willingness to do so. With regards to staff comfort in

administering online surveys, it is not necessary that all staff members become survey

administrators. Those comfortable with the task can be given the responsibility of

administering surveys for themselves and other staff members. With reference to student
access to technology, most university campuses are equipped with computers available for

student use; the only issue here is the convenience to the student.

As the primary focus for this system is the Department of Computer Science, the issue of

computer literacy among staff members should not be an issue. With students there is a

chance that computer literacy may be low, but as they are completing a computer science

degree this level will increase as they progress with their degree and there are tutors who are

available to help with any computer literacy issues.

When considering taking surveys online, there must be a means of informing the

respondents, students in this case, that they are required to take the survey. If a student has

the liberty to complete a survey in their own time, a mechanism should be in place to inform

the student of the requirement to complete the survey; this should be something other than

the instruction from the teacher. With existing systems, students are usually notified via

school email.

An effective survey system for student feedback should be password protected; there should

be some way of authenticating users to ensure that the intended survey taker is the one

actually taking the survey. With paper surveys, particularly those that are mailed and in the

case of students, those which students are given the opportunity to take away with them and

return at a later date, it is somewhat difficult to ensure that the correct person has completed

the survey. A student might have a friend or a sibling complete a survey for them. With the

means of authentication, a majority of students are aware of security issues and will be more
 10

reluctant to share their login credentials with others. A password system will therefore allow

the intended survey audience to respond and no one else.

With paper-based surveys administered in class, instructors usually tell students to drop the

completed surveys off when the students are leaving or return the completed surveys to the

faculty member's mailbox at a later date. These surveys are somewhat anonymous, so there

is no way to remind those who have yet to submit them to do so. With an online system

such as the one proposed, there will be a means of identifying those students who have yet

to complete survey for the purpose of reminding them to do so. The issue of confidentiality

and anonymity' will arise however if survey administrators have access to the identities of

students who have yet to or did not complete their surveys. For this reason the proposed

system will protect anonymity' by having a mechanism to track those who have not

completed surveys solely for the purpose of sending reminders. This information will be

encapsulated so that the instructor or persons administering the survey have no information

as to who reminders are sent to. The system will also have a mechanism of ensuring that

surveys are only responded to once by a particular respondent.

 11

2.2 Advantages and Disadvantages of Online Surveys

Advantages of online surveys include access to individuals in distant locations; this is

beneficial for those taking online classes. Automated data collection is another advantage

which as a result reduces researcher time and effort. Online surveys save time by allowing

large volumes of data to be collected for the given survey continuously and imported into

statistical tools and databases, increasing the speed and accuracy of analysis. In the case of

the Department of Computer Science, the resource which is normally used to collect and

analyze the data for surveys will no longer be required, as survey responses will be

automatically stored on a database system. As a result, there will be a reduction in the cost of

administering survey, "from less staff time required to handle forms and enter data"

[Cummings & Ballantyne, 2000] to saving cost of printed forms. Data can also be

automatically validated for online surveys; that is, the system can return error messages

requesting the correct format of data entry, resulting in low data entry errors. In addition,

online surveys eliminate handwriting interpretation problems.

Disadvantages of online surveys include uncertainties over the validity of the data and

sampling issues. Here validity refers to the accuracy of the specific conclusions and

inferences drawn from non-experimental data [Gunn]. For online researchers, sampling is an

issue as there is no access to a central registry or database where an accurate sampling frame

can be gathered, neither is there any way of discerning how many users are logging on from

a particular machine. For the proposed system, this is not an issue of concern, as the sample

will be the students registered for a particular class or those that meet the criteria set out by

the survey administrator. Another disadvantage is a concern surrounding the design,

implementation and evaluation of an online survey [Wright, 2005]. Administering an online

 12

survey requires a certain level of technical expertise. In addition, the time taken to prepare an

online questionnaire can be substantial and may outweigh some of the time savings noted in

the advantages. With the proposed system however, a substantial part of the designing will

be automated. The only concern for the survey administrator will be to provide the survey

questions and administer it to students by sending them email notifications and reminders,

which will be done through the system.

Another disadvantage to consider is that online surveys may need to be shorter than paper-

based surveys. Response rates for online surveys drop after 10-15 questions and are directly

and negatively correlated with questionnaire length [Madge, 2006].

Technical issues can also occur with online surveys. A server or computer can crash. There

are technical variances in computers, monitors, browsers and internet connections which

may have design implications.

2.3 Constructing Surveys

The construction of a survey goes far beyond the development of a questionnaire and asking

the intended audience to complete the survey. A survey should be seen as more than just an

instrument for gathering information. Surveys should be viewed as comprehensive systems

for "collecting information to describe, compare or explain knowledge, attitudes and

behavior" [Pfleeger & Kitchenham, 2001].

The "survey instrument" is part of a large survey process defined by the following ten

activities [Pfleeger & Kitchenham, 2001]:

1. Setting specific, measurable objectives
 .

13

2. Planning and scheduling the survey

3. Ensuring that appropriate resources are available

4. Designing the survey

5. Preparing the data collection

6. Validating the instrument

7. Selecting participants

8. Administering and scoring the survey instrument

9. Analyzing the data retrieved

10. Reporting the results, statistical analysis and inference of survey results

It is necessary that a survey be designed to provide the most effective means of obtaining

information needed to address the objectives of the survey [Kitchenham and Pfleeger,

2002b]. For a survey to provide the most effective means, it should be designed in a way that

it will not be swayed by a particular faction, aspect or opinion. The survey should make

sense in the context of the population, and the administration and analysis should be within

the resources allocated to the survey.

Those conducting surveys often have some idea of what they are seeking. As a result, the

way they build the survey instrument can inadvertently reveal their biases [Kitchenham and

Pfleeger, 2002a] Replies to survey questions can be influenced by

.

1 The way the question is asked

2. The number of questions asked

3. The range and type of response categories

4. The instructions to respondents

 14

5. The order in which the questions are asked.

6. The language used in the questions.

To avoid bias, survey construction must be done in a way that questions are neutral, the use

of words should not influence the respondent's thoughts, enough questions should be asked

to adequately cover the survey topic, attention should be paid to the order of questions (so

that answers to one do not influence responses to the next), provision should be made for

exhaustive, unbiased, mutually exclusive response categories, and instructions should be clear

and unbiased.

When constructing a survey, care must be taken as to how the questions are formulated and

structured. Questions must be formulated in a way that respondents can answer easily and

accurately. Questions should, therefore, be worded so that a respondent can see

relationships between the intention of the question and survey objectives [Kitchenham &
Pfleeger, 2002b]. That is, the purpose of the question should be clear, or the question is

likely to go unanswered or 'thought-less' responses provided. It is important to keep in mind

that the number of questions that can be realistically asked in a survey depends on the

amount of time respondents are willing to spend completing it [Kitchenham & Pfleeger,
2002c].

The time it takes to complete a survey can be dramatically reduced by having standardized

answer formats, for example "strongly agree, agree, disagree, strongly disagree" [Kitchenham

& Pfleeger, 2002c]. Standardized answer formats save time, as the respondent can anticipate

that the same choices are available for each question and do not have to take the time to
 15

read new answer choices for every question within the survey. Questions in a survey can be

either open or closed. Open questions avoid imposing restrictions on the respondent,

ultimately leaving room for misinterpretadon and provision of irrelevant or confusing

answers. Open quesdons are difficult to code and analyze [Kitchenham & Pfleeger, 2002c].
Closed questions restrict responses but are easier to analyze. However, on the subject of

standardized responses for closed questions, each question may require responses that may

not be necessarily aligned with the standard answer set, whatever it may be.

Quesdons can usually be grouped into topics where each topic addresses specific objecdves

for the survey. It is important not to have too many quesdons on a survey as this can make

the respondent lose interest [Kitchenham & Pfleeger, 2002c]. Kitchenham and Pfleeger

further suggest identifying a topic that is addressed by many quesdons and removing some

of the less vital ones as a way of reducing question size. However, a balance should be

maintained between what one wants to achieve with the survey and the willingness of the

respondent to provide the required information.

2.4 Issues with Response Rates

Some researchers oppose the view that there is little difference between the response rates of

online and paper-based surveys [Nulty, 2000]. Response rate is a major issue when

administering surveys. Any reliable survey should measure and report its response rate. The

validity of a survey is severely compromised if there is a significant level of non-response

[Kitchenham & Pfleeger, 2002b].

 16

[Cummings and Ballantyne, 2000) found that when they established an online system at

Murdoch University, the response rate from students was lower than that in paper-based

surveys administered at the university. Despite strategies put in place to make the system

effective, overall response rates for each semester were 30% in comparison with 65%

achieved for paper forms.

Many factors contribute to low response rates. For example, research suggests that people

do not like to participate in surveys unless they feel it is beneficial to them in some way

[Kitchenham & Pfleeger, 2001]. Incentives are usually introduced to increase response rates.

Cummings and Ballantyne used a "cash prize draw as an incentive for students who

completed surveys for all their units online." This improved the response rate from 30% to

54%.

Due to the "nature of online survey process, response rates need particular attention"

[Cummings & Ballantyne, 2000]. Cummings and Ballantyne demonstrate that there are a

"number of useful strategies to improve response rates." These include but are not limited to

encouraging staff to promote the survey, discussing feedback and any consequential course

changes with students, and offering incentives, particularly cash. These methods have been

shown to contribute to improved response rates.

The following steps were highlighted by Kitchenham and Pfleeger [2002b] to improve

response rates:

1. Ensure that respondents are able to answer the survey questions (questions should

be unambiguous and answerable)

 17

2. Ensure that questions cannot be perceived by respondents as intrusive or

impertinent

3. Ensure that respondents are motivated to answer the questions (show clearly that

there is some benefit to them in answering the survey questions

Kitchenham and Pfleeger [2001] also noted that people are more motivated to complete

surveys if they can see that the results are going to be useful to them within the education

environment, students need to be motivated that it is worthwhile to complete the evaluation

for a class they have taken. This can be done by informing them of changes or things that

have been put in place as a result of previous survey responses.

2.5 Overview of related work / Comparable Survey Systems in Current Use

There are few systems already in existence implementing the same basic goals as the system

that will be produced as a result of this thesis. The first is the Web-Online Feedback system

(WOLF). This was implemented as part of research at Queensland LIniversity of Technology

[Nulty, 2000]. The research aimed to "overcome the common problem with student

evaluation of teaching" [Nulty, 2000]. The research found that "qualitative components of

student evaluations (in particular student comments) were not being made available to

teaching staff concerned until sometime after the teaching has occurred." The system

developed offered a user-friendly, web-based interface that allowed users to generate web-

based questionnaires. Access to the system was 24 hours a day. The main goals for the

system were:

1. Reduce/remove delays in receiving feedback from paper-based surveys.

 18

2. Improve on the range of different ways that people have to obtain feedback on

conceivably any aspect of their work or service.

3. Support the gradually changing ethos surrounding evaluation of teaching and units

from one-off sporadic evaluations to that which evaluation is seen as an integral part

of the daily pracdce.

4. Enable people to obtain context-specific feedback as and when it is needed.

5. Ability to review pre-existing questions available in item banks (for paper and pencil

questions) to ensure that issues explored quesdons posed are adequately represented

in those item banks.

It was found that academic staff opposed questionnaires which made use of a "fixed bank of

optional items or consisted of a fixed set of compulsory items limiting them from

conducting meaningful evaluations in relation to their context" [Nulty, 2000]. Hence, as part

of the fourth goal of the system, WOLF allows people to specify their own questions.

The system was not used as expected, so the success rate was low. However, there was

success with the response rate of questionnaires created on the system "by lecturers in the

faculty of Information Technology and Brisbane Graduate School of Business" [Nulty,

2000].

The second system was a pilot program for an evaluation system implemented by the

College of Computer Science, Mathematics and Physical Science at the University of

Maryland. Instead of building a system from the ground up, the researchers made use of the

features available in WebCT [Denman, Robinson &White, 2004]. The feasibility of moving
 19

from classroom paper-based evaluation to web-based evaluation forms was investigated by

the Physics department. The paper-based forms were mimicked within a web interface to a

database. The success of the pilot was measured by the amount saved on paper and scanning

cost (scan of paper forms). The pilot also eliminated sources of potential error (damaged or

misplaced scan sheets). This also reduced the amount of time required to produce the final

reports.

The University made use of the survey tool available on WebCT. The tool offered a

confidential electronic means of collecting evaluation data from students on courses at the

university. This information could then be transferred to the statistics lab for processing. It

was the university's desire to have a system that would send introductory emails and

reminder emails to students. The frequency of the reminders would be controlled by each

college. Staff desired a mechanism where a list was provided at the end of the evaluation

period of students who had completed evaluations; this would then be used to allocate extra-

credit to those students.

The system was set up in a way that one WebCT "course" was created per department, this

was known as the evaluation space, and one 'survey' (a tool provided by WebCT) was

created for each university course. Department Representatives (DRs) were set up, assigned

and given the responsibility for very basic troubleshooting [Denman, Robinson &White,

2004]. DRs also had a tool available to them which allowed them to receive student response

rates upon request for each evaluation and overall for the department. Students had access

to surveys for the courses they were enrolled in for the semester. Access to surveys was

given to students based on criteria of the course number and section of the course(s)
 20

enrolled in, in a particular department. Dates where also set for the release of the survey

ending the last day of the term [Denman, Robinson &White, 2004]. Icons that linked to

surveys were provided on the WebCT homepage. Students were notified of surveys via

emails which were sent from a web site set up for DRs. Reminder messages were placed in

queues with information regarding the beginning and end dates and frequency set by the

DRs. Reminders were processed once a day and any messages which matched the criteria for

the specific day were sent to students who had not yet responded to all their surveys in the

different departments.

Denman, Robinson & White [2004] focused on student response rate as the dominant

measure the college used to determine the success in the transfer from paper to online

course evaluation. The response rate for the student population for the summer, fall, and

spring semesters were 38%, 44%, and 31% respectively, the research did not provide

response rates for previous paperdoased surveys administered. The disparity of the response

rates in the spring as well as the lower percentages may have been related to the fact that

reminder emails sent during the fall semester were limited and dependency was placed on

introductory emails and a mid-way email. Certain departments had higher response rates due

to incentives such as extra credit and $50 gift certificate being offered.

The last system reviewed was developed as "part of a joint venture between Hong Kong

University of Science and Technology, University of Hong Kong and Hong Kong

Polytechnic University" [Ha & Mars, 1998]. The project saw the development and

implementation of two web-based systems, COSSET and OSTEI. These aimed to support

"student evaluation of teaching in local higher education institutes" [Ha & Mars, 1998].
 21

COSSET is a centrally controlled system designed for collecting and processing data for

summative evaluation of teaching. It is a system for formative evaluation of teaching,

allowing instructors to construct their own questionnaires, gather student responses and

view the evaluation results online. During the project, the team also focused attention on

"evaluating the viability of the web as a valid and reliable medium for student evaluation of

teaching" [Ha & Mars, 1998].

OSTEI allows instructors to access its web site and set the necessary configurations to

conduct a questionnaire survey on the web. Once an instructor creates a questionnaire, with

the help of a question bank, students are able to access the OSTEI student site and complete

questionnaires.

The OSTEI system uses a "registration system to control access by instructors" [Ha & Mars,

1998J. When an instructor builds a questionnaire, he or she is assigned a questionnaire ID.

The ID along with the instructor's username uniquely identifies the questionnaire. For the

system to retrieve the correct questionnaire from the database, a student must supply the

questionnaire ID and instructor's username. No personal information from a student is

required to complete a questionnaire. This is limiting to the system as it allows students to

complete a given questionnaire more than once. "It is also possible for anyone aware of a

questionnaire ID and instructor username to complete a questionnaire" [Ha & Mars, 1998],

even if it was not intended for them to complete it. For this reason, OSTEI is not

recommended for formal teaching evaluations [Ha & Mars, 1998].

3. Requirements Analysis

3.1 Introduction

This chapter describes the system in terms of its functional and non-functional

requirements. It also presents the functional model of the system. The funcdonal model

represents the use cases, which elaborate on the requirements of the system by describing

the funcdonality of the system from an actor's point of view. An actor is an external endty

that needs to exchange information with the system. An actor can represent either a user role

or another system. The analysis object model is presented in this chapter and is represented

by class diagrams for the proposed system. The diagrams describe the entities manipulated

by the system.

This chapter begins by introducing the scope of the system and goes on to discuss the

objectives and success criteria for the proposed system. The proposed system is then

detailed along with the funcdonal and non-functional requirements. The functional model

and an abstraction of the object model are also provided.

3.1.1 Scope of the system

The Online Survey System is an application that is aimed at improving the convenience of

the student evaluation process, both for students and for faculty. It aims to alleviate the

time and cost incurred when collecting data from course evaluations and to provide staff

with the opportunity of constructing evaluations themselves. The system will be a web-based

application which will be hosted in the CSU Studentwebs server. The system will allow

faculty, survey administrators, to create evaluation surveys by specifying the questions they
 23

want on the survey and the answer formats for those questions. These criterions will be

stored on a database along with the intended audience as selected by the survey

administrator. The survey administrator is able to create as many surveys as he or she

requires and the selected target audiences for these surveys to be sent to. Once the surveys

have been created, the survey administrator will need to indicate to the system that the

surveys should be sent to the intended audience. The system will provide a list of all surveys

that the survey administrator has created along with the title they provide for the survey, the

survey creation date, the survey expiration date, which will also be provided by the survey

administrator, and the date the survey was sent to the intended audience. On selecting a

created survey, the survey administrator will be able to choose to send the survey to the

intended audience, send a reminder email or update questions or answers on the survey

provided that a notification email has not been sent out. On receipt of a survey notification,

a survey taker can click the web link contained in the email. This will bring the survey taker

to the system login and on authentication the survey taker will be provided with a list of

surveys he or she needs to complete or have already completed. When a survey is selected,

the survey will be dynamically created and displayed on the survey taker's screen. Survey

takers will be able to view responses of surveys they have previously completed. A survey

administrator will not be able to access individual student responses but can access statistical

information on the overall responses for each survey question on selection of a survey.

3.1.2 Objectives and success criteria of the project

Online surveys provide a faster and more cost effective way of obtaining feedback from

audiences [Madge, 2006]. The objective of the application is to create a system that will allow
 24

a certain group of users (survey administrators) to create surveys that other users can

complete online. Web-based surveys will be generated dynamically from the retrieval of

questions stored on a database. Once a survey taker selects a survey to complete, the

questions for the survey will be retrieved from the database and the survey will be

dynamically generated. A survey taker needs to be authenticated by the system before he or

she can complete a survey.

The system will therefore allow for the following:

— Create user login functionality

— Create a survey display page that displays "user specific" survey questions based

on the group the users has been allocated to

— Create a mechanism for certain users (survey administrators) to create surveys, by

providing questions and the user groups to whom the survey should be assigned

— Create a set of user groups, allowing users to fall into more than one group

3.2 Proposed System

3.2.1 Overview

The system will be a web-based application that will allow students and staff members of

CSU to take surveys online. The proposed system has features from each of the systems

described in section 2.5. This system will be similar to the OSTEI, but will implement a few

additional features that will eliminate the limitations identified with the OSTEI system.

These additional features are user authentication, which will restrict survey takers from

completing a survey more than once and make surveys available to only those that they are
 25

intended for, notification and reminder email, which will give the survey takers the

notification and reminders they need of the availability of the surveys, and provision of

stadsdcal results of the responses of survey.

3.2.2 Functional Requirements

The following section provides an overview of the funcdonal requirements of the system.

Functional requirements deal with what the system should do or provide for the users. The

funcdonal requirements will detail what facilities are required and what acdvides the system

should carry out. In other words, functional requirements define the required funcdonal

support or funcdonality that the system to be created should provide.

3.2.2.1 User Authentication

1. Authendcadon of all users should be provided by the system.

3.2.2.2 Querying Database for previous survey questions

1 . Allow survey administrators to retrieve quesdons used in previous surveys, which are

stored in a question pool on the database. The quesdon pool should contain all

previous survey questions from all surveys stored on the database.

3.2.2.3 Dynamic Survey Creation

1. The system should build surveys dynamically at run-time. That is, the survey is

derived from questions stored in a database and the question response format and

question numbers as specified by the survey administrator.

 27

3.2.2.6 Generation of statistical survey result report

1. Survey administrators should be able to view statistical results of the responses to the

survey.

2. Qualitative results should be made available for surveys before and after the cut-off

date, depending on the responses at the time of query.

3.2.2.7 Allocation of Survey Administrators and user groups

1. A system administrator should be able to provide access to survey administrators.

For example, if the department secretary wanted to create a survey, the system

should have a mechanism for allowing him or her to do that.

2. System administrator should be able to create user groups as requested by survey

administrators.

3.2.2.8 Viewing completed surveys

1 . Survey takers should be able to view surveys that they have previously completed.

3.2.2.9 Other aspects of the system

1. Provide Internet client/server application that will allow users to connect via a login

system in order to complete surveys and create surveys.

2. Inform users of any errors detected while using the system or writing to the

database, loss of connection to the server, database, etc.

3. Survey Administrators should be able to modify survey questions and answers before

it has been sent to users.

 28

4. Survey Administrators should have the ability to delete created surveys.

5. System should protect user anonymity.

3.2.3 Justification of functional requirements

3.2.3.1 User Authentication

Users will be authenticated to avoid multiple submissions from the same user. The user

authentication will be the same as their Novell authentication. This will enable survey takers

to use the same authentication they use for logging onto the school network. Survey takers'

email addresses will also be retrieved from the information held on the Novell database.

3.2.3.2 Querying Database for previous survey questions

Questions used in previous surveys should be made available for all survey administrators to

save time in typing and avoid duplication of data in the database.

3.2.3.3 Dynamic Survey Creation

The system should be able to generate surveys dynamically so that survey administrators are

not restricted in the number of questions they can have on a survey.

3.2.3.4 Email Notification

Survey introduction and reminder notifications should be sent to inform users that the

survey is available and remind those who have yet to take the survey to do so.
 29

3.2.3.5 Generation of statistical survey result report

The system will aggregate survey responses and present that aggregate information in the

form of charts.

3.2.3.6 Protection of anonymity

To protect anonymity of the student taking the survey, the system must encapsulate all

information regarding the students and surveys that they have completed. Survey takers'

email addresses will be encapsulated in group names so when reminders are sent, the survey

administrator is only able to view the group name of the survey takers who are being sent a

reminder. Survey takers' completed surveys are only available to the survey takers themselves

upon logging in the system.

3.2.4 Non-functional requirements

The Online Survey System must fulfill the following non- functional requirements. Non-

functional requirements describe the user-visible aspects of the system that are not direcdy

related with the functional behavior of the system [Bruegge & Dutoit, 2000]

3.2.4.1 User interface and human factors

The user interface should be similar to that of a standard online survey application and

satisfy the ten usability heuristics (see Appendix A).

Taken from http://www.useit.com/papers/heuristic/heuiistic_list.html

 30

3.2.4.2 Hardware and software considerations

The required hardware for the system is as follows:

The system should operate on the studentwebs server in the Department of Computer

Science, CSU.

The time taken for the system to load and retrieve database data will depend on the

network connection over the Internet and the performance of the server.

The user interfaces with the system via a web-based interface in the user's browser.

3.2.4.3 Performance characteristics

The system should be able to retrieve data from, and write data to, the database, and

generate and display surveys for completion with minimum delay. The fact that a database

connection is made on creation of surveys or retrieval of surveys should be transparent to all

users. The response time between a request being made by the user and the response

displayed on the user's screen/browser should be a factor in this transparency.

3.2.4.4 Error Handling and Extreme Conditions

Two possible errors could occur on the system.

1. Error in writing to the database. In the case of this error, the user will be directed to an

error page and instructed on steps they need to take.

2. The client may not successfully connect to the server. The client will be informed that

connection to the server was refused and advised to try again.

 31

3.2.4.5 System Modifications

The bear minimum of the system is to allow survey administrators to create surveys and

administer them through email notification to the intended survey takers. The system could

be modified in the future by adding a 'find' functionality and also making all the surveys

created by other survey administrators available to all survey administrators.

3.2.4.6 Operating Environment

The system will be implemented using ASP.NET C# and an SQL Server 2000 database. The

system will perform best using an Internet Explorer browser. Mozilla Firefox browser can

also be used.

3.2.4.7 Security Requirements

The use of the system will be user authenticated. Pages within the system will also be

authenticated causing restrictions to certain pages for certain users. The Survey Taker will

only be able to access the page where they can complete a survey and survey administrators

will be able to access pages for survey creation purposes. The system administrator will have

access to all pages.

3.3 System models

3.3.1 Scenarios

This secdon highlights the scenarios identified for the proposed system. A scenario is an

informal, concrete, focused description of a single feature of the system from the viewpoint

of a single actor [Bruegge & Dutoit, 2000].

 32

Scenario name: Creation of Survey

Participating actors: Survey Administrator: User

Flow of Events

1. The user accesses the main page of the System.

2. The user is redirected to the system login page, if they are not authenticated.

Once the user logs in successfully, they are redirected to the main page, where

they are presented with surveys they have previously created, with the options of

editing, deleting or creating a new survey.

3. The user selects creation of a new survey. The user is forwarded to a page where

he/she can create a new survey.

4. The user goes through the steps of defining the survey criteria. A textbox is

available to input new questions, a button to add the new question to the survey

being created is next to the textbox.

5. The user inputs a question, selects a question category (if none selected, question

is given the default 'General' category will be used) and presses the add button,

this displays the question in a table on the page labeled 'Survey Questions.'

6. The user selects a question category from the table which is used to display

existing database questions. This lists questions from the question pool in the

database from the selected category in the table labeled 'Existing Questions.'

From this the user checks all questions required from the chosen category. The

user then clicks the 'Add' button, which adds it to the list of questions for the

survey being created. These questions are displayed in the table labeled 'Survey

Questions.'
 33

7. The user clicks the 'Continue' button and is brought to a page where he/she can

configure question answers. Questions already in the database are listed with

their answers displayed below them in a table (future system, user can modify

answers). New questions (questions not already in the database) do not already

have a selection of answers attached to them. A user selects the question from

the table of questions to provide an answer for and selects the format of the

answer(s), then enters the relevant answers for that particular question (user can

go back and add or delete questions from the survey).

8. User then clicks the 'Create Survey' button and is directed to a page that displays

the created survey. If the user is happy with the survey, he/she can click the

finish button to create the survey permanently on the database. The user must

provide a start and expiration date for the survey to be taken by and a survey title

in the textbox provided. Once the 'Submit' button is clicked, the user is brought

back to their 'Main' page where they can select a survey to send to the specified

users group (future system, can modify user group and expiration date, select

multiple user groups for survey, this should send duplicated notifications to

users) by clicking the 'Distribute' for a selected.

9. User logs out of the system.

Scenario name: Completing a Survey

Participating actors: Survey Taker: User

Flow of Events

1. User receives an email requesting he or she complete a survey.

 34

2. The email details the survey to be taken and provides a link to the page where

the survey can be accessed.

3. User accesses the main page of the system via the link in the email notification

and is redirected to the login page for authentication.

4. Once authenticated, the user is sent to the main page of the system for the user

with a survey taker role. Here the user is presented with a list of surveys they

have previously taken and those they are required to take, with dates taken and

expiration dates displayed. Expired surveys and surveys already taken are

inactive.

5. The user selects a survey to complete via the survey title and is taken to a page

that displays the survey. Once completed, the user clicks the 'Submit' button; and

is redirected to the main page.

6. User logs out of the system.

7.

3.3.2 Use Case Model

This section establishes the use cases for the system and goes on to describe each use case.

"A use case represents a complete flow of events through the system in the sense that it

describes a series of related interactions that result from the initiation of the use case"

[Bruegge & Dutoit, 2000].

The use case diagram for the Online Survey System is shown below in Figure 1. There are

three actors for the system, Table 1 ; an actor is an external entity, which interacts with the

system. In this case the actor is the user.

 35

System Administrator User is responsible for setting-up and

configuring (allocating survey
administrators) and for the maintenance of
the system
Survey Administrator One of the main users of the system. These
users are able to logon to the system, create
surveys and administer them to survey
takers, create survey taker user groups and
allocate users to those groups (specify what
type of users will take the survey).

Survey Taker These are one of the main users of the

system. These users use the system to
complete surveys. These users should be
able to log into the system and take a given
survey within the time allocated.

Table 1: Actors for the system and their roles

 36

Online Survey System

Assign Survey Administrator

Complete Survey

Survey Taker

-
System Administrator

Survey Administrator

Figure 1: The use case modelfor the propose system

3.3.3 Use Case model Descriptions

In the sections following are detailed descriptions of the use cases displayed in Figure 1,

These use cases were derived from the functional requirements listed in section 3.2.2.

3.3.3.1 The 'Log On' Use Case

Name: Log on
 37

Description

A User (a System Administrator, Survey Administrator or Survey Taker) logs onto the

Survey System by entering the username and password. The system identifies the user as a

system administrator, survey administrator or a survey taker from login credentials.

Entry Condition:

The user accesses the main web page, which checks user authentication before entering the

system.

Normal Flow

1. User directed to login page.

2. Enters user id and password.

3. Successful login.

4. The system displays the main page for the appropriate user.

Alternative Flow

The 'Unsuccessful Login' Alternative

1. At step 2, if the user is not authenticated, the logon screen says displayed.

The 'Connection Refused' Alternative

1. At stepl, if the system fails to connect to the database to pull the user information

for any reason, e.g. the server is not running, the system informs the user that his or

her connection to the server was refused.

3.3.3.2 The 'Create Survey' Use Case

Name: Create Survey

 38

Description

A survey administrator creates Survey to be stored in system database and notifies target user

group of the availability of the survey.

Entry Condition:

Authentication of Survey Administrator and selection of create survey option from main

page.

Normal Flow

1. Survey administrator clicks the create button on main page.

2. A form is displayed where the survey administrator can create or choose survey

questions from the database.

3. User types a question(s) into the provided text box, selects question category and

clicks the add button.

4. Question(s) is displayed in the table of questions for survey being created.

5. User clicks the 'continue' button and is taken to a page where he/she can assign

answers to the given questions.

6. User selects each question at a time and enters the answer(s) for that question

selecting the format of the answer and clicks the 'add' (answers) button.

7. Questions are displayed with their possible answers.

8. User clicks the continue button and the survey is displayed.

9. User survey title and description and provides survey expiration date.

10. User clicks the create survey button. It is compulsory for the user to provide a survey

title for the survey to be created.

Alternative Flow

The 'Use existing questions' Alternative

 39

1. At step 3, user can click the display button after selecting a question category.

2. Questions for the selected category are displayed.

3. User selects questions he/she wants for the survey.

4. Clicks continue.

5. Questions are displayed with their answers, nothing to modify.

The 'Remove/Add question' Alternative

1 . At step 6 or 9, if the user wants to modify, change, or remove the selected questions,

they click the back button and make the necessary changes.

3.3.3.3 The 'View Survey' Use Case

Name: View Survey

Description

User accesses the system to delete or edit a previously created survey.

Entry Condition:

Surveys created by user are on display.

Normal Flow

1 User selects a survey from the list of surveys they have previously created, using the

select link.

2. The survey questions for the chosen survey are displayed in a table, with the answers

to those questions

Alternative Flow

The 'Delete survey' Alternative

1. At step 1, user clicks the 'Delete' link to delete the survey.

The 'Modify Survey' Alternative

 40

1. At step 2, user selects the question he/she wants to change, the survey selected has

not yet been sent to any recipients

2. User make changes to question

The 'View Results' Alternative

1. At step 1, user clicks the "View Statistics' link to view the response rate of the survey.

2. A chart is displayed indicating the amount of responses for the survey, and the

amount of responses for each question on the survey.

3.3.3.4 The 'Notify Survey Takers' Use Case

Name: Notify User Group

Description

Survey Administrator sends email to targeted survey takers for a particular survey to inform

them that survey is ready to be taken.

Entry Condition:

Survey Administrator authenticated and is on the main page where surveys he or she has

created are displayed.

Normal Flow

1. User clicks the select link for the desired survey in the table of created surveys.

2. User is directed to a page where the selected survey is displayed.

3. User clicks the distribute button for a survey.

4. A textbox is displayed where user can selected the targeted user group for the survey.

5. A textbox with a prewritten message is displayed; the message contains a link to the

survey takers main page. The user modifies the message as they wish and presses the

send button.
 41

6. User receives a confirmation that the emails have been sent to all users in the user

group selected for the survey.

Alternative Flow

The 'Reminder Alternative

1 At step 3, user clicks reminder button

2. A textbox displaying a prewritten reminder message is displayed and the user group

to send the reminder to is entered in the 'To' field.

3. The user presses the send button.

4. Confirmation emails are sent to those who have not yet taken the survey from the

selected user group.

3.3.3.5 The 'Complete Survey' Use Case

Name: Complete Survey

Description

Survey taker logs on to the system having received an email(s) saying he/she has a survey to

complete.

Entry Condition:

User authenticated and is on the main page where surveys to be taken are displayed along

with previous surveys already taken and expired surveys.

Normal Flow

1. User clicks the 'Select' link for the survey to be completed.

2. System displays survey for user to take.

3. User completes survey.

4. Clicks submit button.

 42

Alternative Flow

The 'Expired survey' Alternative

1. At Step 1, if the survey expiration date has passed, the survey displayed is not

editable.

The 'Before Start Date of Survey' Alternative

1. At Step 1, if the survey start date is after the current date, an error message is

displayed.

3.3.3.6 The 'View Completed Survey' Use Case

Name: View Completed Survey

Description

Survey taker logs on to the system to view a survey he/she has completed previously or take

a new survey.

Entry Condition:

User authenticated and is on the main page where surveys to be taken are displayed along

with previously completed surveys with the date of completion indicated.

Normal Flow

1. User clicks the select link for the survey he/she wants to view.

2. System displays survey, the survey is disabled so that the user cannot modify previously

selected answers
 43

3.3.4 Object Models

The class diagram for the Online Survey System shown in Figure 2 shows the initial classes

that will implement the system". The diagram is an overview of the classes (objects)

discovered and includes inidal attributes and methods. The diagram also shows the

relationships (associadon) between the object classes. Table 2 describes each object in terms

of its responsibilities.

Class Description
Quesdon This class represents the question object that is pulled
from the database or written to he database
Answer This class represents the answer object that is allocated
to a question when a question is created. A question
possesses an answer arraylist to hold multiple answers
for a particular question.
SurveyUser The class represents the user who is logged onto the
system. It holds the user's credentials.
QuesdonCategory This is a class representing the category a question is

assigned on creation.
UserAuth This class is used for authenticating the user that is

trying to log onto the system. It connects to the

database to check that the credentials the user supplied
are valid.
SurveyObject This is a base class that holds error managing
exceptions that may occur on the system. These
exceptions are written to the system log.
UserGroup This class represents the group that a user can be
assigned to. A user can be assigned to one or more
groups and this is held in the SurveyUser object
Survey This class represents a survey object that is created
when a Survey Administrator creates a survey. This
survey is written to the database on the request of the
creator.
SurveyPageBaseClass This class is a base class for most of the above classes
that need to access the database and the interface (web
pages) classes that need database access. It holds a
database connection object and methods for writing
cookies for the application.
SurveyControlBaseClass This is a base class for the web controls used in the

New classes will be discovered during the design stage.

 44

web pages for this application. It also contains a

database connection and methods to write cookies for
the application

Table 2: Initial objects Descriptions for the Online Survey System.

3.3.4.1 Class Diagram

Figure 2 shows the classes (objects) identified for the system. In the design section, this

diagram will be refined showing class dependencies and any other fields and methods which

are yet to be identified for the system.

Question

Survey Category
has assigned to

1 1 1

has

Answer
assigned to
takes

UserGroup

Survey User

assigned to

Figure 2: Abstract object modelfor proposed system.

 45

User Interface

The graphical user interface for the system is solely for the user. Figure 3 shows the Login

interface the actors of the system will interact with in order to use the system. Figure 4 show

the survey creation interface that the survey administration will interact with in order to

create a survey.

EEOHB Favorites
JifTIhHTOni
Tods Hefc
tan
Q | Bari
Bad- • m
S
j!"
\
Search Favorites -f>
, s- i %
£J http://sfoxJeftfwetK.colstate.edu/ifowuJatm^ - a<*
Go jlc - v' |G Search • < ^i 10 blacked "*' Check - s AutoUnr - jrj Options

Columbus State
1 1
\ I V I: II S I 1

Survey System Login

User ID

Password
Login

*lDoi» • Internet

Figure 3: Login Interface of Proposed system

 46

F4e Edt view Favorites Tools Help

MEM
Q** .' ;' ; *,,* f,v„„ ( •
. [71 -
, .'»

£ 'WIpil/sfAjde^webs. coWato.edu/riowuJatrna/AppiicatioT/5jveYApplication/create. asp/ - Q^o

Co Jlk - v i., Search • < 'C,' 10 blocked ^ Check - v AutoUnk -

LOIUMBUS MATH
II \ I V I: II S I I 1

Survey Syste

Select Question from Database

Add New Question

Enter Survey Question I Select Question Category

^
Please enter a question to acid to survey General
I
Add

Selecl Category

Building
Previous Next

New Survey Questions

How big are your class sizes'? General

Previous Next

Remove Selected

Figure 4: Create Survey Interface oj Proposed system

3.5. Functional Requirements Cross-referenced and prioritized

Table 3 details all the system functional requirements prioritized and cross-referenced with

the use cases. The abbreviations are as follows:

Must Have - MH
Should Have - SH

Could Have - CH
 47

functional Requirements Priority Cross Reference

Allow survey administrators MH Use case:
to retrieve questions used in Create Survey
previous surveys.
The system should build MH Use case:
surveys dynamically at run- Create Survey / Complete
time. Survey

The system should MH Use case:

authenticate all users. LogOn

Survey administrators should MH Use case:

be able to create surveys, Create Survey
specify the questions, their
types, their possible
responses, and their groups.
Survey Administrators should SH Use case:
be able to modify survey View Survey
questions and answers.

Survey Administrators should SH Use case:

have the ability to delete View Survey
created surveys.

Surveys should be MH Use case:

inaccessible for completion Complete Survey
after expiration date.

Survey takers should be MH Use case:

notified via email when they Notify Survey Takers
have surveys to complete.
Reminder emails should be SH Use case:
sent to those who have not Notify Survey Takers
yet completed the survey as
the survey end date
approaches.
Survey Administrator should SH Use case:
be able to view statistical View Survey Results
results of the responses to the
survey.
Qualitative and quantitative SH Use case:
results should be made View Survey Results
available for surveys after the
cut off date.
A system administrator MH Use case:
should be able to give access Assign Survey Administrator
 48

to survey administrators.
Survey takers should be able SH Use case:
to view surveys that they have View Survey
previously completed.

Table 3: Functional requirements cross-rejere need with use cases.

Summary

This chapter has dealt with requirement elicitation and requirements analysis. The functional

and non-functional requirements of the system were captured. The functional requirements

concerned with the functionality of the system were used to develop the functional model;

this is represented in UML by use case diagrams. The use cases were then described in detail
using natural language; this is so that anyone reading the report without UML knowledge will
understand the purpose of the use case model. The non-functional requirements focused on

the user visible aspects of the system that are not directly related with the functionality of the

system, i.e. the platform in which the system should operate. After identifying scenarios and

use cases, the initial classes required for the system functionality were identified. This chapter

established the user interface for the system and prioritized and cross-referenced the

functional requirements with the use cases.

4. Design and Implementation

4.1 Introduction

This chapter is concerned with the design and implementation of the system. It discusses the

objects (classes) used to implement the system and the various techniques used to build the

system. This chapter focuses on defining the subsystem interface, also referred to as the

application programmer interface (API).

The web-based application being developed will have a 3-tier architecture. The three layers

of the architecture will be highlighted and their functionality will be discussed. In particular,

the user interface layer is presented and the business logic layer will be defined by the names

of classes, operations, parameters, types, and return values.

4.2 Design decisions

This section lists all decisions made for the system to provide the services it proposes

effectively.

Providing the title field and description field for created surveys

As with any survey that one takes, a title is necessary to provide some indication of the

purpose of the survey. The provision of a survey title and description by survey

administrators on creation of a survey will serve as a way of informing the students of the

purpose of the survey they are required to take. The survey title will be required, but the

survey description will be optional. These will be used in the email notification to students.
 50

Allowing survey administrators to view all answers that have previously been used for

a selected question

Survey administrators will have access to answers that have been used for a selected question

by any other survey administrators. The database will record all answers that have ever been

used for a question. This will save time on survey creation and decrease the chances of data

duplication.

All answers should not be attached to all questions

The application will only have answers that have already been selected for a question

previously to be displayed when a user is formatting answers for a question. Having all

answers displayed will cause a problem once the answer pool gets large as users will have to

search through hundreds of answers to find one that they wish to use. Instead, allow the

users to create new answers if they cannot find the one that they wish to use and when

writing the new answer to the database ensure that it does not already exist.

New questions and answers written to the database during survey creation

New questions will be written to the database when the survey administrator adds them to

the survey being created. New answers however will not be written until the survey

administrator has actually attached the answer to the question for the survey. This will help

the user to retrieve new questions and answers if the connection is lost, and limit data

duplication, as the survey administrator will only be able to view answers which have been

attached to a question previously.

 51

Three step process to configure questions and answers

The application will provide a three step process for selecting answers for questions. For

step 1 the user will select a question number. For step two, the user will select an answer

format. For step 3 the user will select responses for the selected question. The three-step

process should be visible only when 'Set Answer' link button for a question is selected. The

'add new answer' input control should remain invisible until the user clicks the 'add new

answer' button.

Predefined answers that do not need configuring

The application will provide answer formats that do not need to give any answer. These are

Open Ended (One Line), Open Ended (Multi Lines), Dichotomous (Yes/No), Likert

(Agree/Disagree). If a survey administrator chooses one of these formats when configuring

an answer for a question, he or she will not need to choose answers for the question, these

will be already defined.

Store Surveys and questions in a session object

To avoid null pointer exceptions and limit the amount of connections to the database, store

relevant survey objects and question objects in the session once data for them is retrieved

from the database.

Use Dot Net Chartings to display results

.netCHARTING is a .NET control that will enable the application to display dynamically

generated data quickly and easily through a visual interface. This control is written in C# and
 52

will be integrated with the application to enable statistical view of survey results. The

control was obtained from wwww.dotnetcharting.com .

4.3 Proposed software architecture

Overview

The proposed architecture for the system is the three tier architecture usually used in web

applicadons. The tiers are:

1 The presentation layer or user interface layer

2. The business rules layer, and

3. The data access layer

The presentation layer consists of HTML and ASP.NET pages; these create the look and
feel of the user interface. The business layer uses the code-behind classes to control the flow

of the application; these classes are written in the C# programming language. These code-

behind classes call other C# classes to store and retrieve data from the database and at times

forward the results to the ASP.NET pages or other code-behind objects. The data layer

works with data that is stored in the database.

These three layers are relatively independent and should be kept as separate and independent

as possible.
 53

Presentation Layer
This is user interface of the application. This lavcr is

responsible for translating tasks and results into

something visible and understandable by the user

R,,o;«» M ^<v^

#
T T ^, TO , jM.

This layer handles the application process commands,

makes logical decisions
presentation layer and the data access
and transfers data between the
layer.
#

Query Data

Data Access Layer

This layer stores and retrieves data from a database. Data is

received from the logic layer is stored m the database. Upon

recruest data is passed to the lojjic layer tor processing and
eventually displayed on the user interface
Database

Figure 5: Three Tier architecture for web applications

4.4 Subsystem decomposition

The system can be divided into three subsystems that correspond to the 3-tier architecture.

The model of the 3-tier architecture can be seen in Figure 5.

Figure 6 is a diagram of the subsystem decomposition for the online survey system.

The diagram is an overview of the subsystems and their containing classes.

 54

i 1
Survey AppObject Survey Application

+UserGroup +survey
+SurveyUser +answers2
+SurveyPageBaseClass +SurveyError
+Survey +Main
+UserAuth +create
+SurveyControlBaseClass < +AdminMain
+SurveyObject +page
+QuestionCategory +SendMail
+Pager +results
+Answer +messagesent
+Question +paging
+SurveyConnection
Security

Survey Application This subsystem initializes interaction with the user by

subsystem providing the system interface. It contains the .aspx files that
are responsible for the user interface and the code-behind files
that handle the interaction. This subsystem also contains the
security subsystem, which is used in authenticating all users of
the system. Within the 3-tier architecture, this subsystem is the
Presentation Layer.
This subsystem is responsible for interfacing between the
Survey Object Survey Application subsystem and the database. It holds all

subsystem objects used when the user interacts with the system. This is

the Business Logic Layer.

Database Subsystem This subsystem is responsible for storing and retrieving data
used by the application. This is the Data Access Layer.

Figure 6: The subsystems for the Online Survey System.

 55

4.5 Subsystem services

As mentioned in the previous section (subsystem decomposition), the system is divided into

three subsystems: the survey application subsystem, survey object subsystem, and the

database subsystem. This section describes the services these subsystems provide for other

subsystems. 'A service is a set of related operations that share a common purpose' [Bruegge

& Dutoit, 2000].

Survey Application Subsystem

This subsystem is concerned with the initialization of the system and is responsible for

interfacing with the user. The subsystem contains all the user interface files for all users of

the system. It contains the interface for the survey takers to complete surveys and view

completed surveys, for survey administrators to create and send out email notifications for

created surveys and also view survey results.

Survey Object Subsystem

This subsystem interfaces the Survey Application Subsystem with the Database Subsystem.

It contains twelve classes which work together to provide database connection for the survey

application subsystem. This subsystem is responsible for establishing database connections

when needed by the application subsystem and holding data retrieved from the database in

objects for use by the survey application subsystem.

Database Subsystem

This subsystem is responsible for holding data and querying the data store for use by the

Survey Object Subsystem. The entity relationship diagram can be found in Figure 24.
 56

4.6 Survey Application Subsystem

This section provides the interface screen shots and class diagram for the Survey Application

subsystem. The class descriptions for each class with this subsystem can be found in

Appendix B. The diagram contains the attributes and operations of each class and the

association, which relate the objects. The class descriptions describe in detail each object in

terms of attributes and operations and their visibility. The following are definitions for the

visibility assigned to methods or field in the following classes:

Public- makes the element visible to code outside the class

Private- private methods and fields are visible only inside the class they are defined,

Static- static methods and fields belong to the class as a whole, rather than to any individual

instance.
 57

»Oack System WBb Ul WebConiiois Button S urveyf^goBaeCliM

•Questioned System Web Ul WebContiols OeHOnd Survwy/'stw da wT la u
•olnGue shon*dd System Web Ul WebC ontrois Button SwvuyApph jHon-AdmniMari
•AnswerGrid System Web Ul WebContiols OalaC.rld •newSuryey System W bUtWebC onbols Butlon
•eont System Web
WebConiiois Button
Ul fSurvayt mii System.W
•Label! SyslemWeoUl WebContiols Label

•LaoeU SyslemWedUIWBtiC ontrois Label

•answeifoimalSiik"' v\ei. jiwob' uiilrois OiopDownUsI n ewS u rvev_ C II c k.yo id
•QNoddl System web n web'! ontn.K OiopPownusI
'

Wbone System web UlWeCC ontrois Label

•SunreyOesopbon System Web •one System Wab Ul WebContiols TertBoi
iuryeyOuestionsList System Wi
Kwo System web Ul WebConlrois TertBoi
on! System Web Ul WebC ontrc
ilartdale System Web Ul WebC •tniee System web ui webConiiois TertSo
•tbfour System Web Ul WebC onbols Label Seiuirty
•comparedale System JlWebControls Compar •(our Syslem Web Ul WebContiols TertSoi Component »oi0
fupaaieO System wen •IoiKb System Web Ul WebContiols Label
(UDdaleA: System Web

•updates System Web •Sit Syslem We ibC ontrois TertBoi

..'jr^HvP*^ (.I*-..* i.yi
•Laben System web <_ •IbsewnSystei
•ReauiiedFieidVaiKiati •seien System, WebConbolsTertBo.
StMveyAn^ul urLSendMai
•lueigbi System web Ul WebC ontrois Labe •;,U'I I idle iyslcm rtPO '
H Hlfi-.h .onliols HlmlTable •MToAddress System Web Ul rVebContiois TertBoi

Hid -.-..Men, -Vet •eijni SvSiefii weo ui vyebConuols TertBo Web Ul WebContiols Button
•blnSriowAttOFoim System

•Ibmne System Web Ul WebContiols Label •DSOuestionGnd Syslem wed Ul WebConiiois OalaOMc •trtSudiectSystem Weo Ul We
•nine Syslem Web Ul WebC onbols TertBoi MbCluestAddSTSlem Web Ul WebC (introls.euHon •baa Syslem.Web Ul WebCor jolsfiutlon
•idten System wed Ul WebConlrois Label •lemovebtn Syslem web Ul WebC onliols Button •curl Syslem Web Ul WebC on IOIS Button
-inrnaieeC omponentvoia
weln oiiliuls TertBoi •10 System web UlwebConrro
•FcrmalType Syslem Web Ul WebC onliols I

cnangeSysli rfcju b stio nC atAdd System Wei Jl WebContiols DropDownL flfom Syslem Web Ul WebC on
treateSurvey^Clickyold
febCoi •etnopdovm System Wed Ul ebConiiois OiopOownust »sub|ecl Syslem Weo Ul WebC
•geWnsyveis AirayUsi #*-..!. .

JlWebControls Butlon •btnSend Syslem Web Ul wed

•gelAnsweisFiomDB Aria.USI
rMbOueslionsArrayLiSl •message string
•suryeyOuesliu'uLM ItcnT- il.itin.ii
»',urv9yijueslions Wiaynsl
•categories. AriayLlsl •goodbye sizing

•t alegory string •audiencelbi Syslem Web Ul WebContiols Label

*qiOij|i- System web Ul WebC onliels LiStBot
•useAnsyyerBuL System Web Ul WebC onbois Button •setecl System Web Ul WebC c nliolS Butlon
*oo. system web UiweBControis TertBo.
•MBodyFieeTertBoiConiiois reeTertBo>

•cutienlQuestron Question
SirutyApp u l m ta l p age --.-l„IMd.l

•WPageSlieSyster -Page Loadyold

•ibiPages SyslemW •selAnswerFoimal System Web Ul WebConiiois Button
•ibiRecords System Jl WebCoMrols Label •bolder Syslem Web Ul WebC onliols Table initialize' iirii|iiiiie"l «ito

•o»l System weo ui i ontrois HtmlTableCeli •lesporiseCeii ".i.i-''iWi,h n™i ir,,K Tshiu. eli •sBlRecipienlsstring
fiOPac.es System Wi -<lr[,ii-|.nlr.,l, ntpi.IT-ir.le' f •Orid Pagemoert angea rold -btoSend Clitk.void
•surveyTille System' •LoadDalavoid
•dbQueslAdd_Ciic yoid xsr
•btnBack Ciickroi
PageSize ml •btnContinue Cllr.k-.oid

-Panes long •eOcpaown Seie

-WhoiePages long •ijut"-liori"*(iiJ_tleniComm •DBOueslionGrid llemOalaeoundw
•gel' dietaries yol
•getOuesttonyoid
-SlanOtPaoe Inl •btnShowAddForm .Click*
-EnflutPage tot •LoadOalavoid
b UI Pstfe
StaveyAppHcallon
•blrijut- _tji>nAdO_L lick.von
•administralorSyslem AHl.^HWel

sets ess lonCwesut

•conllgBul_Ciickioi Sui wy App b t jT kM i.p aguig
•Onlnrlyoid
useAnswe'6u1_CH(
role_Commandyold
'ibiPages Svslem .
A,.- t .. .„

It Systen
•tiiF jijie::.;e vvilem Weii'ii ^el>' ...r.ii...l ,
>.1Bl.
•hdnOyerallPeicent Syslem Web Ul HlmlConljcls H
-dsDalaSel
• o.l System Web Ul HtmIC onliols HlmlTableCell
riveiaiiPiogiess ml

BuiiaTables .1

Figure 7: Object Model (class diagram) of the Survey Application subsystem

 58

4.6.1 User Interface

The user interface is the presentation layer of the system. This section highlights the web

pages of the application and details how the pages are used in order to interact with the

system.

4.6.1.1 Login

This is the main login page of the system. From this page users are authenticated and

redirected to the page they initially requested. If the user came directly to the system login

page, the user will be directed to the default. aspx page, Figure 8.

4.6.1.2 Default

The user is directed to this page from the login. aspx page, Figure 3. Here the user is

presented with three links, one for System Administrators, one for Survey Administrator,

and one for Survey Takers. The user selects the appropriate link depending on which of the

three user roles the user falls into. If the user selects a role to which he/she is not assigned,

the user is redirected to the surveyerror.aspx page, Figure 22. When the System

Administrator link is clicked, the user is redirected to the sysadminmain page. When the

Survey Administrator link is clicked, the user is redirected to the adminmain.aspx page,

Figure 9. When the Survey Taker link is clicked, the user is redirected to the main. aspx page,

Figure 19.
 59

3 default Microsoft Internet Explorer LIICI®

File Edit View Favorites Tools Help

^J Back • »] S" ,
:

Search Favorites 4p ii *
4»] http://studentwebs.colstate.edu/idowuJatima/Application/SurveyApplication/default.aspx v flGo
Go gle » v ( , Search $$S §1 10 blocked ""/ Check - N AutoLink - [rj Options

Survey System

L Mi^ijljt

User Pages Description

System These are pages accessible to the system administrator for the purposes of maintainence
Administrator and creating user account
These are pages accessible to the survey administrator, those responsible for creating
Survey surveys. From these pages you can create surveys, administer surveys to students, send
Administrator reminders, view survey response rates and survey statistics for surveys the survey
administrator has administered.
These are pages accessible to survey takers From these pages you can complete surveys,
'Survey Takt
view previously completed surveys and view survey statistics for surveys you have completed

4-j Done Internet

Figure 8: System default page

4.6.1.3 AdminMain

The user is directed to this screen either from the login. aspx page, Figure 3, or from the

default.aspx page, Figure 8. This page displays all surveys that have been created by the user

in a table. The information displayed about each survey is the survey title, the start date of

the survey, the expiration date of the survey, and the date the survey was emailed to
 60

students. For each survey in the table, the user can choose to view the survey by clicking the

'Select' link; in this case the user is redirected to the survey. aspx page, Figure 13. The user

can choose to delete the survey by clicking the 'Delete' link, this will delete the survey from

the database and from the user's view. The user can also choose to view the response rate of

the survey by clicking the "View Statistics '

link, in this case the user is redirected to the

results. aspx page, Figure 17 . Finally, on this page the user can use the 'Create Survey' button

to start the survey creation process. Once the 'Create Survey button is clicked the user is

redirected to the create. aspx page, Figure 4.

 61

3 Survey Administrator Main Page Microsoft Internet Explorer Qdi®

File Edit View Favorites Tools Help

Q Back - u) Z^ \ Search Favorites & -

g|| * i I
ss |£j http://studentwebs.colstate.edu/idowuJatima/Application/5urveyApplication/adminmain,aspx
Co '^ - v Q search *
$ §3 10 blocked *% Check » N AutoLink Q Options

LoqOut

SiJ r ey
Select Delete l Start Date Expiration Date Sent

April 16 4/18/2006 4/29/2006 4/23/2006

Select Delete Statistics
2006 12 00 00 AM 1 2.00 00 AM 12 00:00 AM

Create Survey

m # Internet

Figure 9: Survey .Administrator Main page

4.6.1.4 Create

Figure 10 shows the first page in the survey creation process. The user has the option of

selecting questions from the database to add to the current survey. To do so the user must

select a question category from the dropdown menu in the table labeled 'Select Question

from Database.' Once a question category is chosen, all questions in that category on the
 62

database are displayed in the grid. The user can select the questions he or she wants to use

by checking the check boxes or the user can select all the questions by clicking the 'select all'

link. The user then needs to click the 'add selected' button to add the selected questions to

the current survey. These questions are then added to the grid labeled 'New Survey

Questions.' The user can add a new question to the survey by clicking the 'add new question'

button. A panel is displayed, Figure 10, containing a textbox where the user can input the

question they wish to add to the survey. By default the 'General category is selected. The

user can choose another category and press the add button. This writes the new question the

database and adds it to the grid displaying questions for the current survey. In the 'New

Survey Question' grid, the user can remove questions by selecting the 'remove all' link or

checking the questions they want to remove and clicking the 'Remove selected' button. The

continue button takes the user to answer2.aspx page, Figure 11, which is the second stage of

the survey creation process. The back button takes the user to the adminmain.aspx page.
 63

File Edit
!«»»
View Favorites tools
mm Help
Hi sr

Q Back * *"
m , 5e*ch FavorJes & - ~~\ -
f %
• 1
-
http://studentwet>s.colstaCe.edu/idowu_l aCrna/ApplitdtKxVS'Jti'eYAppkceSion/create.aspi «CJc
Co jt. - v t search • *' ^ 10 blocked '"- Check -
s AutoUnk -
'J Options

f
....„...,....,...
iaai

Select Question from Database

Ado New Question

Enter Survey Question I Select Question Category

Uhat Year do you gra

What i
s youi maioi :
'
P^JP^TMsI^Pb^bJb^BbI ^Tl Selecl C ateqor*

Student
MT avnmmi
D
What is you student status' Student 1
n
Previous Next

New Survey Questions

Question Category Select] )

What is your major? TaaaVal

What is you student status-' Student a
Previous Next

Remove Selected

£_ Done

Figure 10: Add New Question panel on the create page

4.6.1.5 Answer2

This answer2.aspx page, Figure 12, is used by a survey administrator. Here the user selects a

question from the grid by clicking the 'Set Answer" link and is presented with three steps.

Step one is the selection of the survey question number, step two is the selection of the

question response format, and step three is the selection of the response. Once the user

selects an answer format, the 'use selected' button is displayed. The user can select answers

from the answer grid or add new responses by clicking the 'add new answer' button. The
 64

'add new answer' button displays a panel where the user can select the amount of answers he

or she wants to add, Figure 12. Once this is selected and the 'configure' button is clicked, a

textbox appears where the user can input the desired answers. The user can then press the

add button, these responses are now added to the answer grid, where the user can select the

answers they want for the question and click the 'use selected' button. This information is

updated in the question grid, which now displays the question number, the question and the

answer format. To change the question number, the user can select the question then select

the empty choice from the question number dropdown list and click the 'use selected'

button. This will make the previous question number available and give the user an

opportunity to choice a new question number. Once all questions are configured, the user

clicks the 'create survey' button and is redirected to the survey page.
 :

65

E
He-

Q
GO
p— tdrt

Back

:*'* -
*
Vie* Favorites

*
Tods

£" |
Hefc

Sea/ch
^M
Favor»«

#Thttp://studcntwetjs.cofstate.edu/riowu_t'dtm s/ApplicatK>a/SijtvevAppkation/df

v t. search
i

-
f £b
^
10 blocked ^f Check tj Options
vflc

LiiJJJI

|Number| Question |Category| Answer | Answer Type

Multiple Choice (Mult
1 What 15 your major? Student Change
selection)

Multiple Choice (Single

2 What is you student status? Student Change
selection)

' ||What Year do you graduate fstudent | Chance ||

Open Ended (One Line)

Multiple Choice (Single

4 What was your SAT Score? Classes Change
selection)

How muc h would you say you have learnt from thi s class? Classes Not Set
Answer
Set
How often do you use the library? Facilities Not Set
Answer
Do you find the books in the book store more expensive than outside of ej
Facilities Not Set
campus bookstores? Answer
|Previous Next

Stepl:

Question Number: 5 »

Step 2

Answer Type : Multiple Choice (Single selecti

Step 3:

have learnt a lot from this class

have leanrt a little from this class

have learnt nothing from this class

had to do extra reading for this class

C'Done # Internet

Figure 11: Configuration of survey question responses

 66

HlifrmfllFM'MillilHli'llflM
File Edt View Favorites Tools Help

** • ~~\
(^ Bad. * , Search Farartes 4P -
f ',

• 1 Mtp://shxlertfwebs.ccdstate.echj/dowuJatima/^
»a<*
Co git- " ( Search - f Z,:. 10 Nocked '-* Check - i, Autolink * [rjoptions

Number Question Category Answer Answer Type

|What was your SAT Score? _||Classes Set Answer ||
Not Set
|How much would you say you have learnt from this class? ||Classes Set Answer |[
Not Set
Previous Next 1

Creat New Response

Number
Responses
of 3 1 N/A
Si
2 Vol 10 lake SAT
3.1
i.nt.yiiip

Step 1:

Question Number: SbIbci »

Step 2:

Answer Type: Select

Step 3:

HBHMml" «
*500
:

HHSSl
D
500- 1000 II

—T
tl'Done

I , rMBf llr'i

Figure 12: Creating a new answerfor a survey question

4.6.1.6 Survey

The survey. aspx page, Figure 13, displays the survey to the user. It displays the questions and

the responses with the web controls that correspond to the answer format chosen for the

question. From this page the user can choose to submit the survey, in which case the survey

is written to the database and the user is redirected to the adminmain.aspx page. To submit a

survey the user needs to provide a survey tide and start/end dates for the survey. The user

can also choose to update questions or answers by clicking one of the update buttons or
 67

administer the survey to students by clicking the 'distribute' button. Finally, the user can

choose to send out a reminder for the survey by clicking the 'reminder' button. If the user

clicks an update button, he or she is redirected to the create. aspx or the answer2.aspx page.

If the user chooses the 'distribute' of 'reminder' button, the user is redirected to the

sendmail.aspx page.

EB
FJe Edt View Favorites rods He£

Q ' ["] [ft ,

Seaich Favortes «p Eh ft *
[g http://shjoentwebs.cobtate.edU/dcwu_tatrYia/Appkation/SijveyAppii:ation/sijivev asp>>Paoe-u
Add.es - a*
Google - v C Search - $i §5 10 blocked "5 Check - ~\ Autolink ___ Options

1. What is your major?

r '
Phsycology
CiSocioloy

D Computer Science
r Mathematics
Engineering

C Biology

2 What is you student status?

Freshman
Junior

Senior

3. What Year do you graduate

"
I
Next

Start Date

ffHB MKWRJfTfSBI
bun Hon rue wed r„u m bat Sun Men lue wed Ihu in SJt

1 2 3 i 5 1 2 3 4 5

2 a 2 la 11 12 13 7 S 2 IB 11 12 13

14 IS Ui 12 IB IS 2Q 14 15 142 1Z IB 12 2E

21 22 23 24 25 26 2Z 21 22 23 24 25 26 22

2fi 22 3D 31 28 22 IB 31

Figure 13: Survey created by survey administrator on survey page

 68

4.6.1.7 SendMail

The sendmail.aspx page, Figure 14, is used to send email notifications to students. Here the

user can select user groups to send a survey notification. This is done by pressing the 'To'

link, which displays a list of survey taker groups. Once the groups are selected, the user

replaces the text in the text area with a message and presses the send button. This distributes

the survey notification. An email is received by a survey taker, Figure 1 8, providing a link to

log on to the system and complete the survey. For anonymity purposes, the survey

administrator cannot see the email addresses of recipients. If a reminder is being sent, Figure

15, the user only needs to type a message and click the send button. The user is then

redirected to the messagesent.aspx page, Figure 16.

 69

3 Administer Survey Microsoft Internet Explorer iTlfni®

File Edit View Favorites Tools Help

Favorites #^ S- * ^
PI http://>tudentwebs.colstate.edu/idowu_fatima/Application/5urveyApplication/sendmail.aspx v Qgo
Go glc ( , Search • AutDLink » Options

Columbus State
i \ i \ i: li s i i v

,urvcy System

Select Survey
Li Audiencelctil
to select
From idowu_tatima@colstate edu
multiple)
Subject Survey Notification [Survey title will be inserted by system]
Staff

Please complete survey EH-SSMBHI

Tenured Facul
Non-Tenured F
Graduate Assi j
Select
|

Computer Sck
Nursing Major
Engineering

[Send j

Figure 14: Email notification on sendmail page

 70

nEmamammmsmmmmmmmmmmmm
File Edt View Favorites Tods Heb

Qr3aoV - n £ \
Search FavoHes $* '* 3 "
J
'"•

#1 http://stuctenr.vwbs. colstate.eckj/idowuJatima/Ap^ -Qgc

CooglC'i v (/, Search - ,f V 10 blocked '7 Check - N AutoUnk -
£_ Optic*

Columbus Statk
U N I V i: Jt S 1 T 1

From [email protected]

Subject Remindet to complete survey (Survey title will be inserted by system|

this 13 to remind you that the above survey has yet to be completed.
Please complete it at your earliest convenience You can access the .

survey by clicking the link belou.

Sincerely [Survey Creator name will be inserted by system]

[
Send |

4jDone Internet

Figure 15: Reminders notification on sendma.ilpage

4.6.1.8 Message Sent

The messagesent.aspx page, Figure 16, displays a confirmation of the email sent and

provides a link back to the AdminMain page.

 71

HimffllMiMHIfllflffl
File EcK View Favorites Too* Heb

Q Back - U
X
£ \
Search Favorites & - ~~\ •
i
"»

• 1 http://srudei^websxotet3teechj/diwuJatiTia/Apphca^cWSufveyAppkation/rriessagesent.asp>

CoOglt- v ( search -
f Q 10 blocked
f Check • i, AutoUnk »
|
Options

COLUMBUS STATi:
l \ I \ I II S I I ^

Survey Administered.
your Survey Message has been sent to:

To start to thQ main pagg, go here .

CDone # Internet

Figure 16: Email confirmation

4.6.1.9 Results

The results. aspx page, Figure 17, displays a graph of the response rate of the selected survey.

The back button takes the user back to the AdminMain page.
 )

72

3 Complete Survey
File Edit View Favorites
Microsoft Internet Explorer

Tools Help
Q
Qfiack -
|*J Z\ '. Search ': Favorites 4p i. ^
ss ^jhtcp://scudentwebs.colstate.edu/idowu_fatima/Applicacion/SurveyApplic3tion/re5ults.aspx?=48 a
Co gle v G Search » $ §1 10 blocked "J Check • v AutoLink » » -| Optic

Survey System

Li.".l'JUt

Responses Rate

www.dotnetcharting.com
Development Version: Not for production use.
:
<

1 6 -

14 .

1 2 -

0.8

Survey

-^ ngt Charting For more information visit http Jfwww dotnetcharting

, . com

.netCHARTING Mentor ( 3 Suggestion!. j taie

St\ Done £ Internet

Figure 17: Survey response rate graph

4.6.1.10Main

This is the main page for survey takers. This page displays all surveys that have been

assigned to the user for completion and those the user has previously completed in a table.

The information displayed is the survey tide, the start date of the survey and the expiration

date of the survey. The user can select a survey by clicking the 'Select' link; in this case the
 73

user is redirected to the page.aspx page, Figure 20, where the user can complete the survey.

The system will warn the user if he or she tries to select a survey before the start date.

3 Yahoo! Mail [email protected] Microsoft Internet Explorer frim®

File Edit View Favorites Tools Help *
t J Back ' W j
,

jjj | Search Favorites #^ s- li a

«j http://us.f605. mail. yahoo. com/ym/ShowLetl:er?MsgId=3498_39950428_1057981_1949_946_0_10146_-l_08Jdx=08<YY=4493'l&Jnc=25&order=down8is v'Qgo
Google » v ( ; search • %-' [_? 10 blocked *y; Check » \ AutoLink |fj Options

Yahoo! My Yahoo! Mail Make Yahoo! your home page

Search

"XXHoof mail ^^ . r^p,:; Mail Home - Hail Tutorial; - Heli

NHUIX >
Free Shipping No Late r ees 60,000- Titles
MHIHiiiTlH
New
Adiliesses Calendai Notepad What's - Moil Foi Mobile Upcnades
"
Options

Check Mail . Compo Search t+ie Web

Previous I
Ne>t I Back to Me-raae;
f^Vonage 1 FREE Month
V>"Ofer expires 5/15/00 Delete |
Reply Forward [..
Spa

Check Other Mail [Edit] This message is not flagged ( Flag Message - Mark as Unread |

" p0p3stU.C0lstat. Date: Sat, 06 May 2006 22:49:0-? -0400

From: [email protected] review Contact Details 9 add Mobile Alert

Folders [Add - Edit]
Subject: Survey Notification: Field Survey
©Inbox (18)
To: rf.arieidowulaiyarioo.com
GjDraft

C_Sent Dear Mane,

Please complete survey
©Bulk (9) [Empty]

L_-Trash [Empty] Please Click lieie and login to the system to complete the survey

My Folders [Hide]
Sincerely
Jobs |17| Fatima Idowu

Thesis (3)
Delete [ Reply Forward Spam Mov

^" See your

Previous I Next I Back to Ne-s ave Messaoe Text I Full Heade

i # Internet

Figure 18: Email notification received by survey taker

 74

3 Survey Taker Main Page Microsoft Internet Explorer (TlfnlrSl

File Edit View Favorites Tools Help :r

$Back • ,«] j", ',

Search Favorites «£ ' ,. [~3 '
|J. ?l
4f 1 http://studentwebs.colstate.edu/idowuJatima/Application/SurveyApplication/mairi.aspx v Qgo
Gogle* v c, Search - j! £i 10 blocked *J
:
Check * N AutoLink » ^2 Options

Si Columbus State

HSZSL^LMHHHl
LoaOut

Select Survey Title Start Date Expiration Date Date Taken

Select April 18 2006 4/18/2006 12 00 00 AM 4/29/2006 12 00 00 AM Not Taken
Select Field Survey 5/6/2006 12 00:00 AM 5/25/2006 12:00.00 AM Not Taken
< >

Q Done <C Internet

Figure 19: Survey Taker Main page

4.6.1.11 Page

The page.aspx page, Figure 20, displays a survey for completion by a survey taker. Five

survey questions are displayed per page. The user can click the previous and next links to

navigate through survey pages. Once the user has completed the survey, the user clicks the

'Submit' button which writes the users response to the database. This page informs the user
 75

of the number of questions on the survey and the page number they are currently viewing.

Users can also view surveys that they have previously taken on this page, Figure 21.

However, the controls will be disabled, preventing users from modifying their responses.

Fde Edt View Favorites Took Met .V

Q Back • m" £ ',

Search Favartw «P "43* I "»

• lhttp;//s^oen( webs. cobi ate. edu/«lcwuJatfTid/Ap[)(Kdtion/SL»^eyAppk-ition/pdge.a5pi '-48 ;

Go gle- * ,, Search - f C 10 blocked *3 Check - s AutoLint -

April 18 2006
1 Pages Fust Prey Mext LdSI 5 Questions
Page 1

How big are your class sizes?

Hour many hours a week do you study for each

class?

How many classes do you take a semester?

What Is your major?

What is you student status?

Figure 20: Survey completion page

 76

3 Complete Survey Page Microsoft Internet Explorer

File Edit View Favorites Tools Help

Back " "1 m •

Search sorites f< -
_
"^ » |J ;^

Addiess jfc] http://studentwebs.colstate.edu/idowuJatirna/Application/SurveyApplication/page.aspx ;'=50 - g^Go

Co 'J|If » v ( , Search r^" 10 blocked J"
Check Optic

La y [Ut

Field Survey
1Pages First F'rey NM Last 2 Questions
Page 1

Whats is your intended career field?

1

What is your major?

2

First Prey !
i t

Figure 21: Previously completed survey

4.6.1.12Survey Error

The surveyerror.aspx page, Figure 22, is used to display an error message when users try to

access a page that does not correspond to their user role.

 77

3 System Login Microsoft Internet Explorer (7©®

File Edit View Favorites Tools Help

QBack - ,«) ]£| i

Search Favorites £ *
@§ -
jgj \
flj http;//studentwebs. colstate.edu/idowu_fatima/Application/5urveyApplication/surveyerror.aspx flGo
Google- v I Search * $ §] 10 blocked *^ Check - v AutoLink » Q Options

Columbus Staff

LoqQut

An error has occured.

Error message: You do not have access to the page requested

To try again, use your back button. To start over, go here,

ftj Done

Figure 22: Page access error page

4.7 The Survey Application Object Subsystem

This subsystem is concerned with managing data from the database, placing them in objects

that can be manipulated by the Survey Application subsystem. Figure 23 illustrates the class

diagram for this subsystem. The class descriptions for each class within this subsystem can

be found in Appendix C.
 78

c5 SurveyApp. Objects. Answer SurveyApp.Objecls.UsefAuth

SurveyApp.Object s. SurveyObjed
answerlD int

-description string
formal string -LogErrorvoid UserAuth
-idint wnteToLog void IsUserVaiidbool
GelErrorsFormatted string
ToString: string
Answer LJ System Web Ul UserConl/ol
•Answer .Web.SurveyControlBaseClass
LoadFromReader void
GelAnswerBvQuestionlD AnayList SurveyApp.Objects. Quest lonC at egory
C^D ...Database.SurveyConnection #m_conn SurveyConnectjon
GetAnsweiS ArrayLiSt
categorytD ml
• GetAnswerBylO «Vnswei SurveyControlBaseClass
-description siring
_conn SqlConnectlon
GetFormatlDByDescription ml #WnleEventLogEntryvoid
GetlDByDescnption ml SurveyConnection #WnteCookie void
ToString string
GelAnswerFormats ArrayLisl
Open bool iFWrileCookievoid
QuestionCategory
AddAnswer void
Isupen bool *ReadCookie string
QuestionCategory
Close bool #OpenDBConnection bool
QuestionCategory
Format string #CloseDBConnection void
AnswerlD int
LoadFromReader void
Connection SqIConnection -PageJJnioad void
•GetAHC ategones ArrayLisl
Description siring #ShowErrorPagevoid
'.'etc alegoryByQuestionlD Questions ategory
+ GetC alegoryBylD QuestionCategory
DBConnection SurveyConnection
SurveyApp. Objects. Question
C ategorylD in!

•questionID int Description string

SurveyApp.Objects.Pager
surveyQuestionID int

-(ategorylD QuestionCategory
-pg PagedDataSource
-questionNumber string
-description string S urveyApp.Objecls.Survey
Pager
- answers ArrayLiSt
C^ System Web Ul Page Get_NextLink string -surveylD int
-answerSet AnayList
.Web.SurveyPageBaseClass Get_PreviousLink string -creation DateTime
•caldes string
GetNewData Source PagedDataSource -expire DateTime
-answerForm at siring
#m_conn Survey: unntHttion -laken string
-id int
DataSource iCollectlon -title string
-result obiect
SurveyPageBaseClass PagedSource PagedDataSource -description string
-selecledResponse ArrayLiSt
#WnteEventLogEntn/ void -creatorlD string
responseStnng string
JWnteCookie void -userGroups ArrayLisl
#WnteCookie void -questions ArrayLiSt
Question
#ReadCookie string -sent string
geIC ategoryOesctiptlon void
#OpenDBConnection bool
ToStnng string
#CloseDBConnectionvoid Survey
Question
Page_Unloadvoid du ... Object s.SurveyUser Survey
LoadFromReader void
#SnowErrorPage void ToString string
GetQuestionsBylatecjory AnayList
-firstName string Survey
OetAIIQuestions ArrayLiSt
DBConnection SurveyConnection -lastName siring LoadFromReader void
GetQuestionBylD Queslion
-emailAddress string
getSurveyBylD Survey
AddQuestion void
-userlD stung getSurveyByC r eatorlD ArrayLiSt
AttachAnswervoid
SurveyApp Obiects.User Group -password string -' ^):'^
wnteSurveyQuestionResponse void L^J *a.t, I
.
l

'C'
, l

l
l

^.'JP_^I ^y_kL
,
: ,

-rolelD int AddSurveyvoid

-userGroupID mt tfgnjuplDs ArrayLiSt AddResultvoid
Description string
-descnplion string updaleDateSentvoid
CatDesc string
SurveyUser deleleSurveyvoid
AnswersToUse AnayList
UserGroup SurveyUser AddSurveyOuestion void
AnswerFormat string
LoadFromReader void LoadFromReader void AddSun/eyQuestionAnswer void
Result obiect
*^el'rouplDs Qrrayt is!
getUser Sun/e/L'ser GetSurveyQuestions ArrayLiSt
Answers ArrayUst
QuestionID ml
getGrouplDs Array! is! GetAnswerlDsBySQlD ArrayLisl
•' -vV • MjupfciyJD '
'-t-r 1
m.hijj UserlDslring GetUserSurveyQuestionResponsevoid
ResponseStnng string
*GetAll'_.ioups AndyLi^l Group ArrayLiSt
SelectedResponse ArrayLisl
y.-<tV -lO upEmails AnayLiil FirstName string Description string
SurveyQuestionID int
AddUserOioupToSurvey void Password string Creation DateTime
QuestionNumber string
LastName string CreatorlD String
Category QuestionCategory
UserGroupID ml Role int
Title string

Descnplion string Email siring Questions ArrayLisl

Taken string
Expiration DateTime
SurveylD int

Sent string

Figure 23: Object Model (class diagram) of the Survey Application Object subsystem
 79

4.8 The Database Subsystem

The database subsystem is the data access layer of the application. The tables held in the

database and the relationships can be seen in Figure 24, which shows the associations

between each table where data from the system will be stored.
 *

-5

er>

5
.1°

L. pan
0)

* tu
H
c
< 1) Q T>
r c
a g Q
, S
V
1 n J > > o
oj 5 d _
3
3 >• > in
1

<a £
la

> 1
w
s
c
3 o
0) C 3 pr -<B C0H ID
> < i/i
u
J
CO
Op > s
> 5
3 X 1; 1)
> in U h
3 O
p oj to
Summary

This chapter has discussed the design and implementation process for the online survey

system. It talked about the design decisions made during the implementation of the system.

The 3-tier architecture of the system was discussed. The subsystems of the system were also

detailed, in particular the responsibility of each subsystem was outlined and the object

models for each subsystem were presented. The entity relationship diagram for data access

layer of the system was also presented.

 82

5. System Evaluation

After implementation, an evaluation of a system is carried out. This chapter focuses on the

evaluation carried out on the system developed as part of this thesis. The evaluation

considers the functional analysis of the system, in which the specified system functionalities

are tested one by one. The results of this evaluation can be found on section 5.2.

5.1. Functionality Testing

Functionality testing ensures that the complete system complies with the functional and non-

functional requirements of the system. The testing carried out in this section is comprised of

functional testing. This uses black box techniques, and the test cases were derived from the

use case model. The functional tests were identified by inspecting the use case model in

Chapter 3 and identifying use case instances that are likely to cause failure.

5.1.1. Choosing Test Cases

The test cases were chosen by going through the use case description of each use case in the

use case model in Figure 1 and finding the features of the system which are likely to fail and

should be tested. What follows is a list of features from each use case which are likely to fail.

5.1.1.1. The 'Log On' Test Case

1. The user have been authenticated and is directed to the default page of the

application, where they are provided three links and should choose the link

appropriate for their user role. The System Administrator has access to all pages.

A User tries to access a page without the appropriate user role.

 ..

83

2. The user may try to access a page on the system without logging onto the

system.

5.1.1.2. The 'Create Survey' Test Case

1 The user tries to create a new question which is already in the database.

2. When configuring answers for a question, the user selects an answer format that

needs answers to be supplied for the question, but the user does not provide any

answers. For example, if the user selects a multiple choice answer format for a

question, the user needs to indicate what responses are to be used for the

multiple choice options

3. The user selects the create survey button without providing a start or end date

for the survey. Alternatively, the user selects a start date which is greater than the

end date.

4. The user tries to submit a survey for creation without supplying a survey title.

5.1.1.3. The *View Survey' Test Case

1 The user logs onto the system with the intention of modifying a survey that has

already been sent to recipients.

2. User selects a survey that has not been sent to recipients and tries to send a

reminder.

3. User selects the 'View Statistics' link to view statistical information on a survey

that has not been sent to recipients.

4. User selects the 'View Statistics' link to view statistical information on a survey

that has been sent to recipients, but has no responses.

 ..

84

5.1.1.4. The 'Complete Survey' Test Case

1. User selects a survey that they have previously submitted in order to change their

responses and resubmit.

5.1.2. Test cases and Results of Testing

Test case Name Log_On_Test (1)

Entry Condition User has been authenticated by the system

Flow of test events 1 Survey laker attempts to access System Administrator or

Survey Administrator main page via the link on the default
page.
2. Survey Taker attempts to access System Administrator main
page via the link on the default page.

Expected system User should be directed to the system error screen and informed
response that he or she does not have access to the requested page. The
system should deny access to the Survey Administrator if he/she
tries to access the main page System Administrator. The Survey
Taker will be denied access to the System and Survey Administrator
main page.
Observed system System responds as expected.
response

Test case Name LogOn_ Test (2)

Entry Condition User types in the URL to access a user main page

Flow of test events 1 The user is redirected to the system login page
2. The user logs onto the system
Expected system The system redirects the user to the page initially requested.
response
Observed system System responds as expected.
response
 .

85

Test case Name Create_Survey_ Test (1)

Entry Condition User is on the create. aspx page selecting quesdons to create a survey

Flow of test events 1 . User add new quesdon button

clicks the
2. A textbox appears where user can input the quesdon.
3. User Clicks the add button
Expected system 1. The system informs the user that the question or one similar
response already exists in the database via an error message and
should identify what category the question exist in order for
the user to retrieve the question from the database.
Observed system System adds question to new question grid and also writes it to
response the database.

Test case Name Create_Survey_ Test (2)

Entry Condition User has selected survey questions and is on the answer2.aspx page
configuring question responses.
Flow of test events 1. User selects a question from the question grid.
2. The selected question ishighlighted on the grid and a
response configuration panel is displayed.
3. The user selects a question number to assign the question on
the survey.
4. The user selects the multiple choice option from the answer
format dropdown list.

5. The user clicks the use selected answers button

Expected system The system should not display the chosen answer format in the grid

response until the user selects responses for the format.

Observed system The system makes the configuration panel invisible and sets the

response chosen answer format for the selected question

Test case Name Create_Survey_ Test (3)

Entry Condition User is on the survey page viewing the survey.

Flow of test events 1 User enters the survey title in the textbox provided
2. User enters a description of the survey.
3. User clicks the create survey button
4. User selects a start date but does not select an end date for
the survey-
Expected system The system should inform the user of the error and should not write
response the survey to the database until the date issue has been rectified by
the user.
Error 1 : start and end date must be provided for the survey
 86

Error 2: start date cannot be later than end date

Observed system System responds as expected.

response

Test case Name Create_Survey_ Test (4)

Entry Condition User has configured answers for the survey questions and is looking
at the survey on the survey screen.
Flow of test events The user clicks the submit button to write the survey to the database

Expected system The system should inform the user that a survey title is required,
response before the survey can be submitted to the system for creation.

Observed system System responds as expected.

response

Test case Name Vie\v_Survey_ Test (1)

Entry Condidon User in on AdminMain.aspx page, a list of surveys which the

the
user has previously created is on display.
Flow of test events The user selects a survey that has already been sent to students with
the intention of modifying the survey.
Expected system The system should direct the user to the survey page where the
response survey is displayed and disable all update buttons.

Observed system System responds as expected.

response

Test case Name View_Survey_ Test (2)

Entry Condidon User in on AdminMain.aspx page, a list of surveys which

the the
user has previously created is on display.
Flow of test events User selects a survey that has not been sent to students with the
intendon to send out a reminder
Expected system The system should direct the user to the survey page where the
response survey is displayed and disable the reminder button but enable the
distribute button.

Observed system System responds as expected.

response
 .

87

Test case Name Vie\v_Sutvey_ Test (3)

Entry Condition User is on the AdminMain.aspx page, a list of surveys which the
user has previously created is on display.

Flow of test events 1. User clicks the 'View Statistics' link for a survey that has not
been sent out to students.
2. User is directed to the results page

Expected system The system should display that no data could be retrieved for the
response survey.
Observed system System responds as expected.
response

Test case Name Vie\v_Survey_ Test (4)

Entry Condition User is on the AdminMain.aspx page, a list of surveys which the
user has previously created is on display.

Flow of test events 1 User clicks the View Statistics' link for a survey that has no
responses from students.
2. User is directed to the results page

Expected system The system should display that no data could be retrieved for the
response survey.
Observed system System responds as expected.
response

Test case Name Complete_Survey_ Test

Entry Condition User log on to the system and is authenticated

Flow of test events User selects a survey that has already been completed with the
intention of resubmitting with different responses.
Expected system The system should display the survey with the responses chosen by
response the user and not allow any modification of these responses.

Observed system System responds as expected.

response
 88

Summary

This chapter has described the testing carried out on the developed system. Functional

testing was carried out on the whole system. This involved finding the differences between

the functional requirements and functionality of the system. Test cases were derived from

the use case model developed in Chapter 3 and the tests were recorded. Two of the test

cases failed, details of these can be viewed in the Create_Survey_Test (1) and

Creat_Survey_Test (2) test table.

 89

6. Conclusion

6.1. Future Enhancements

The development of the Online Survey system was successful and provided the

functionalities initially proposed. However, when considering the further development and

improvements for the system, a few new features were realized. These features are outlined

below with an explanation of their purpose.

Updating of questions and answers to the database

Currently the option to update questions and responses are available, but when these

changes are written to the database a new survey is created. It is desirable that updates to

survey questions and answers be written to the database as updates rather than new records.

Ability to resend surveys to different groups of students than those which they were

originally sent

Once a survey has been distributed to students, a survey administrator can only send

reminders to the group of students the survey was inidally distributed to. The system does

not allow a survey administrator to include another group of students to receive the survey.

Additional answer formats

Currently the system supports six answer formats, these are Likert (Agree /Disagree),

Dichotomous (Yes/No), Open Ended (multi lines), Open Ended (one line), Multiple Choice

(multi selection), and Multiple Choice (single selection). An extension to the system would
 90

be to support Table (multiple choice), Table (open ended), and Multiple Choice (dropdown

list).

Easier process for changing question numbers

The process for changing a survey question number is a little tedious. An improvement to

the system would be to make the survey question number selection easier by allowing users

to be able to drag and drop survey questions to the position in the grid that they want the

question to be numbered on the survey.

Utilize survey descriptions

Currently survey descriptions are not being used by the system. On improvement, the

descriptions can be sent out in the notification emails.

Usability & Effectiveness Testing

Usability testing tests the users' understanding of the use case model. Effectiveness testing

will test the user satisfaction that the ease of use of the system for survey administrators

when creating and administering surveys and the ease of use and convenience of use for

survey takers. All use cases identified during the development of the analysis model will be

cross-referenced by the system requirements; therefore, the usability testing will aim to

derive from the users whether or not the requirements of the system were met.

System administrator interface

The interface would provide the system administrator with the ability to create users

accounts for both survey administrators and survey takers.

 .

91

Confirmation survey delete request

This will avoid the case of a user mistakenly deleting a survey. With a delete confirmation the

user will be asked if the delete operadon is what was intended.

Rearrange the survey creation process

The survey process should be changed to ask for survey dtle at the first step rather than at

the last step.

6.2. Project Summary

Although it would be ideal for every survey sent or distributed to students for the purpose of

educational evaluadon to be completed and returned in a timely manner, this is usually not

the case. In academia, and the same is true for the non-academia world, there are many

factors that contribute to recipients not completing surveys; these were discussed in Chapter

2. Due to these factors, it is difficult to get students to complete surveys in their entirety

without some sort of incentive being offered.

Despite the fact that there are several influential factors that tend to prevent students from

completing surveys, the key factors are:

1 Convenience to take the surveys

2. Anonymity of responses
 92

As mentioned in Chapter 2, the success of student evaluations is dependent upon the

faculty's support of issuing student surveys; whether online or paper-based. Because faculty

members are responsible for distributing the surveys, the following are some factors that

hinder faculty from promoting paper-based surveys:-

1. Time taken to distribute surveys

2. Resources needed to analyze survey data.

As previously mentioned, the objective of this project was to develop a web-based

application that would allow faculty members to adopt the role of survey administrators and

administer surveys to students that they teach. As part of this, the intent of this project was

to produce a system aimed at providing CSU Computer Science faculty members with a tool

to create surveys and administer these surveys online. The system developed took into

consideration the factors mentioned above and implemented functionality that would

protect anonymity, provide the convenience of allowing survey administrators to create and

administer surveys and the convenience for survey takers to complete surveys in their own

time. The system also provides an analytical tool to interpret survey responses. The attributes

of the system focused on were ease of use, ease of distribution and provision of data analysis

tools. The system developed utilizes user authentication in order to restrict users from

completing a survey more that once and makes surveys available only to those students

specified by faculty The system does not enforce that students should answer all questions

on a survey; this eliminates factors that could discourage student from completing the

surveys administered to them.

 93

On evaluation of the finished product of this project, it was evident that the objectives were

met successfully. All the functional requirements listed for the proposed system in Chapter

Three with a MH (Must Have) priority were implemented.

References

[Brueggc & Dutoit, 2000] Bruegge Bernd & Dutoit Allen H, Object-Oriented
Software 1 Engineering. Prentice Hall, 2000.

[Cummings & Ballantyne, Gumming Rick & Ballantyne Christina., Online student
2000] feedback surveys: Encouraging staff and student use,
Refereed Proceedings of Teaching Evaluation Forum,
p29-37, August 2000

[Denman, Robinson & Robinson Paulette, White Jason, & Denman Daniel W.,
White, 2004] Course Evaluation Online: Putting a Structure into Place,
Proceedings of the 32nd annual ACM SIGUCCS
conference on User services, p 52-55, October 2004

[Gaide, 2005] Gaide Susan, Evaluating Distance Education Programs

with Online Surveys, Distance Education Report, p4-5,
October 2005

[Gunn] Gunn H., Web-Based Surveys: Changing the Survey Process,

First Monday Peer-Reviewed journal on the internet,
h ttp:/ /www, firs tmondav.org/issucs/issuc7 12/gunn/#n
ote4

[Ha & Mars, 1998] Ha Tak S. &

Mars fonathon, Using the Web for Student
Evaluation of Teaching (COSSET & OSTEI), December
1998

[Kelly & Marsh, 1999] Kelly M & Marsh J,

1999, Going online with student evaluation
of teaching. Hong Kong: I Evaluation of Student Experience
Project, City University Hong Kong

[Kitchenham & Pfleeger, Pfleeger Shari Lawrence & Kitchenham Barbara,

2001] Principle of Survey Research Part 1: Turning Lemons into
Lemonade, ACM SIGSOFT Software Engineering Notes
vol. 26 no. 6, pi 6- 18, November 2001
 95

[Kitchenham & Pfleeger, Kitchenham Barbara & Pfleeger Shari Lawrence,

2002a] Principles of Survey Research Part 4: Questionnaire
Evaluation, ACM SIGSOFT Software Engineering Notes
vol. 27 no. 3, p20-23, May 2002

[Kitchenham & Pfleeger, Kitchenham Barbara & Pfleeger Shari Lawrence,

2002b] Principles of Survey Research Part 2: Designing a Survey,
ACM SIGSOFT Software Engineering Notes vol. 27 no.
1, pi 8-20, January 2002

| Kitchenham & Pfleeger, Kitchenham Barbara & Pfleeger Shari Lawrence,

2002c] Principles of Survey Research Part 3: Constructing a
Survey Instrument, ACM SIGSOFT Software
Engineering Notes vol. 27 no. 2, p20-24, May 2002

[Madge, 2006] Madge C. 2006, Online questionnaires, Exploring online

research method in a virtual training environment,
University of Leicester,

^
http://www.geog.le.ac.uk/orm/questionnaires/quesprint
^
3.pdf

[Nulty, 2000] Duncan Nulty, Web On-Line Feedback (WOLF):

Intentions and evaluation, Refereed Proceedings of
Teaching Evaluation Forum, p38-41, August 2000

[Wright, 2005] Wright K. B. (2005), Researching Internet-based

populations: Advantages and disadvantages of online
survey research, online questionnaire authoring software
packages, and web survey services. Journal of Computer-
Mediated Communication, 10(3), article 11.

http://jcmc.indiana.edu/voll0/issue3/wright.html
Appendix A
Ten Usability Heuristics

Visibility of system status

The system should always keep users informed about what is going on, through

appropriate feedback within reasonable time.

Match between system and the real world

The system should speak the users' language, with words, phrases and concepts

familiar to the user, rather than system-oriented terms. Follow real-world

conventions, making information appear in a natural and logical order.

User control and freedom

Users often choose system functions by mistake and will need a clearly marked

"emergency exit" to leave the unwanted state without having to go through an

extended dialogue. Support undo and redo.

Consistency and standards

Users should not have to wonder whether different words, situations, or actions

mean the same thing. Follow platform conventions.

Error prevention

Even better than good error messages is a careful design, which prevents a

problem from occurring in the first place. Either eliminate error-prone

conditions or check for them and present users with a confirmation option

before they commit to the action.

Recognition rather than recall

Minimize the user's memory load by making objects, actions, and options visible.

The user should not have to remember information from one part of the
 97

dialogue to another. Instructions for use of the system should be visible or easily

retrievable whenever appropriate.

Flexibility and efficiency of use

Accelerators — unseen by the novice user — may often speed up the interaction

for the expert user such that the system can cater to both inexperienced and

experienced users. Allow users to tailor frequent actions.

Aesthetic and minimalist design

Dialogues should not contain information, which is irrelevant or rarely needed.

Every extra unit of information in a dialogue competes with the relevant units of

information and diminishes their relative visibility.

Help users recognize, diagnose, and recover from errors

Error messages should be expressed in plain language (no codes), precisely

indicate the problem, and constructively suggest a solution.

Help and documentation

Even though it is better if the system can be used without documentation, it may be

necessary to provide help and documentation. Any such information should be easy to

search, focused on the user's task, list concrete steps to be carried out, and not be too large.
Appendix B

Survey Application Class Descriptions

The following sections describe the classes and attributes of the Survey Application

subsystem. Figure 7 depicts the object model of the survey application subsystem. Any

classes that are not included in these descriptions can be found in the class diagram, Figure

7.

Class _default

public class _default implements System. Web. UI. Page

This object handles the functionality of the default page of the application

Field Summary
protected administrator
utton
Link to the system administrator main page

Protected surveyadmin
LinkButton
Link to the Survey Administrator main page

protected taker
LinkButton
Link to the survey taker main page

Method Summary
private void InitializeComponent Q
Used to initialize components on the webpage and load event
handlers.

private void Page Load fobject sender, System.EventArgs e)

Loads the web page

private void role Command fobject sender, System.Web.UI.WebControls.CommandEventArgs e)

Handles the 'Administrator', 'SurveyAdministrator' and 'SurveyTaker'

commands of the links buttons, redirecting the user to the appropriate main
page that matches the user's role.
 99

Class AdminMain

public class AdminMain implements SurvcyPagcBascClass

This object handles the functionality of the main page for the Survey Administrator. It

populates a DataGrid with all surveys created by the Survey Administrator and allows for the

creadon of new surveys.

Field Summary
protected nevvSurvey
button
Button used for creadon of new surveys.

Protected SurveyGrid
DataGrid which holds survey object items retrieved from the database
for the Survey Administrator. These are all surveys that the survey
administrator has created. The DataGrid displays the survey title, creadon
date, expiradon date and the date sent for each item.

Method Summary
private void newSlirvey Click (object sender, System.EventArgs e)

This method handles the create survey button click. Once the button is

clicked the user is redirected to the create page.

private void Page Load (object sender, System.EventArgs e)

Makes a request to the SurveyApplication subsystem to obtain surveys
created by the Survey Administrator the first time the page is displayed in the
user's browser.

private void SurveyGrid ItemCommand fobject source,

System. Web. UI.WebControls.DataGridCommandEventArgs e)

Handles the 'Select' and 'Delete' and 'Stadsdcs' commands for the
DataGrid. When an Item is selected using the 'Select' link on the DataGrid,
this method sets a session object indicating whether the selected survey had
been sent to recipients and then calls the survey page passing the survey id as
a query string.

When an item is selected using the 'Delete' link, this item is deleted from the
grid and the database.

Class answers2

public class answers2 implements SurveyPagcBascClass

 100

Page used to configure survey question response

Field Summary
protected addAnswerbtn
utton
Button used to add a new answer to the Answergrid control
protected Table addTable
Table that holds controls for creatine new answers

protected AjisNo
DropDownList
Dropdown list that holds numbers used to create textboxes for
answers

protected answerFormat
DropDownList
DropDownList that holds the different answer formats that can be
assigned to a question.

protected AnswerGrid
nd
DataGrid that displays all answers attached to the selected question

protected mt AnswerlD
Used to hold an Answer object answered

protected back
Button
Back button
protected Table border
Table that holds answer configure controls

protected btnQuestionAdd
Button
DropDownList that holds question numbers
protected configBut
Button
Configure new answer(s) button

protected cont
Button
Continue button

protected currentQuestion
Question Holds the currently selected question
protected Label FormatType
Label that holds the response format of a question

protected QNoddl
DropDownList Question number dropdown list, used to select a question number for
the selected question

protected QuestionGrid
ta
DataGrid that holds and displays the question objects for the survey
being created

protected Label questionNumber

Label that displays question number
 101

protected responseCell
TableCell
TableCell that holds all controls used to create a new answer
protected uscAnswctBut
Button
jj se answer button

Method Summary
protected void addAnswerbtn Click fobject sender, System.EventArgs e)

Sets the addTable control to visible

protected void answerFormat SelectedIndexChanged (object sender, EventArgs e)

Makes the useAnswerBut visible

protected void back Click fobject sender, System.EventArgs e)

Redirects user to create page

private void bindQuestionNos O

Puts question numbers in the QNoddl DropDownList
protected void btnQliestionAdd Click (object sender, System.EventArgs e)

Adds new answers to the AnswerGrid and attached the answers to the
question in the database

protected void configBut Click fobject sender, System.EventArgs e)

Used to dynamically generate textboxes for new answer input, the

number of textboxes to create is retrieved from the AnsNo dropdown list
protected void cont_Click(object sender, System.EventArgs e)

Redirects user to the survey page

protected void CreateChildControls O

Maintains view state of dynamically created textboxes used to get new
answers from user

private void CreateTextBoxes Q

Dynamically creates textboxes for new answers and adds the to the
responseCell control

stnngD formatS (ArrayList arr)

Puts the "Select" string at the first index to the answerFormat control

protected void getQuestion Ont id)

Sets the 'CurrentQuestion' session object to the selected question

private void InitializeComponent f)

Used to initialise components on the webpage and load event handlers.

protected void LoadData Q

Binds the 'NewSurveyQuestion' session object to the QuestionGrid

stnngQ numbers Q
Returns a string array containing the numbers 1 to 1 and the string
"Select" in the first index. Returned array is bound to the AnsNo control.
 102

private void Page Load fobject sender, System.EventArgs e)

Loads the page and binds the data sources for the QuestionGrid,
answerFormat DropDownList and the AnsNo DropDownList
protected void QuestionGrid ItemCommand fobject source, DataGridCommandEventArgs e)
Handles the '
Select' command for the QuestionGrid which displays
the border control and binds any answers that are attached to the selected
question to the AnswerGrid

private void QuestionGrid ItcmDataBound fobject sender, DataGndltemEventArgs e)

Sets the text for the Format'lype of each question in the QuestionGrid
if the question has a format type the text of the edcon link button is changed
to 'Change' is not it remains as 'Set Answer'

protected void QuestionGrid PagelndexChanged fobject source,

DataGndPageChangedEventArgs e)

Changes the page number of the QuestionGrid

bod setSessionQuestions Q
Updates the 'NewSurveyQuestions' session object with the currently
configured question.

private void useAnswerBut Click fobject sender, System.EventArgs e)

Sets the configured answers to be used by the selected question. Sets

the answer format type of the question and the responses for the question
object.

Class create

public class createimplements SurveyPageBascClass

Page where questions for a survey being created are selected.

Field Summary
protected addTable
HtmlTable
Table that holds the control used to create a new question

protected back
Button
Back button
protected btnQuestionAdd
Button
Buttons used to add new questions to the SurveyGrid and database

protected btnShowAddForm
Button
Button used to show the addTable control
protected categories
ArrayList
ArrayList that question categories from the database

protected string category

 103

String that holds the description of a category

protected COIll
Button
Continue button
protected dbOuestAdd
Button
Button used to add database questions the SurveyGnd

protected DBQuestionGrid
DataGnd DataGrids that hold questions from the question pool in the database

protected dbOuestions
ArrayList
ArrayList that holds questions retrieved from the database

protected e Dropdown
DropDownList DropDownList that holds question categories

protected Label exists

Label which indicates whether a new question already exists in the

database

protected question
Question Question object

protected questionCatAdd
DropDownList DropDownList that holds question categories

protected removebtn
Button
Remove question button, use to remove questions from the
surveyGrid and the survey being created

protected SurvevGrid
DataGrid
Datagrid that holds the questions for the survey being created.

protected survevOuestions
ArrayList
ArrayList that holds survey questions

protected txtOuestionAdd
TextBox Textbox input for new survey questions, questions which do not exist
already

Method Summary
protected void btnBack Click fobject sender, EventArgs e)

Redirects the user to the AdminMain page

protected void btnContinue Click fobject sender, EventArgs e)

Puts the SurveyGrid in a the 'QuestionGrid' session object and

redirects the user to the answer2 page
protected void btnQuestionAdd Click fobject source, EventArgs e)
Adds the question that the user has entered to the database and
SurveyGrid
 104

protected void btnShowAddForm Click fobject sender, EventArgs e)

Makes the addTable control visible

protected void dbQllCStAdd Click (object sender, System.EventArgs e)

Handles the click event of the dbQuestionAdd button, adds the

selected questions to the SurveyGrid

protected void DBQuestionGfid ItemDataBound (object sender, DataGridltemEventArgs e)

Binds category data to the eDropdown control on the

DBQuesdonGrid control
protected void eDropdown SelectedlndexChanged fobject sender, EventArgs e)
Handles the index changed event of the eDropdown control.
protected void getCategories Q
Sets the first index of the 'Category 1' session object to "Select
Category"

private void Grid ItemCommand (object source, DataGndCommandEventArgs e)

Handles the 'SelectAlT and 'RemoveAlT commands of the two
DataGrids. Checks all the questions in the datagrid to be added to the survey
or removed from the survey

protected void Grid PagelndexChanged fobject source, DataGndPageChangedEventArgs e)

Changes the page index of the calling grid

protected void LoadData fDataGrid source)

Binds question data to the DBQuesdonGrid and the surveygrid

private void Page Load fobject sender, System.EventArgs e)

Instandates the 'NewSurveyQuesyions', 'Categories' session objects.

Binds data to the DBQuesdonGrid, SurveyGrid and quesdonCatAdd
controls. Makes the continue button visible if the SurveyGrid has questions
in it otherwise sets the visibility to false

protected void removcbtn Click fobject sender, System.EventArgs e)

Checks the SurveyGrid for selected questions and removes them from
the grid

Class Main

public class Main implements Sun eyPageBaseClass

This object handles the functionality of the main page for the Survey Takers. It populates a

DataGrid with all surveys that the Survey Taker has to complete and those already

completed.
 105

Field Summary
protected Label beforeStartDatc
Error message label. Use to display error when a user selects a survey

to complete before the survey start date.

protected SurveyGrid
DataGrid which holds survey object items retrieved from the database
for the Survey Taker. These are all surveys that have been assigned to the
Survey Taker, whether they have been completed or not. The DataGrid
displays the survey title, expiration date and the date that the survey was
completed.

Method Summary
private void Page Load fobject sender, System.EventArgs e)
Makes a request to the SurveyApplication subsystem to obtain surveys
assigned to the Survey Taker the first time the page is displayed in the users
browser.

private void SurveyGrid ItcmCommand fobject source,

System. Web. U I. WebControls.DataGndCommandEventArgs e)

Handles the 'Select' command for the DataGrid. When an Item is

selected using the 'Select' linkon the DataGrid this method checks the start
date of the selected survey and if the current date is before the state date sets
the error message of the 'beforeStartDate' label and sets the table to visible. It
also sets a session object indicating whether the selected survey had been has
expired or has already been completed by the user and then call the 'page'
page passing the survey id as a query string.

Class messagesent

public class messagesent implements System.Web.UI.Page

Displays a email sent confirmation message, listing the user groups that the survey email has

been sent to.

Field Summary
protected recipients
t st
Datalist of recipients that the survey email has been sent to
 106

Method Summary
private void Page Load fobject sender, System.EventArgs e)

Loads the messagesent page, binding the 'SendTo' session object to the
recipients dataiist.

Class page

public class page implements SurveyPageBascClass

This object handles the functionality of the 'page page for the Survey Taker. Its dynamically

builds the selected survey to be completed by the survey taker.

Field Summary
protected back
Button
Back button
private int EndOfPage
The number of the last question on the current survey page

protected Label lblPages

Label for the number of pages fro the survey

protected Label lblRecords

Label for the number of questions on the survey

private int Leftover

The amount leftover upon division of the number of questions on the
survey and the page size

private int Numltems

Number of questions on the selected survey

private long Pages

Number of pages for the survey

private int PageSize

Number of survey questions to be displayed per survey page

private int StartOfPage

Number of the first question on a survey page

protected submit
Button
Submit survey button
protected Label surveyTitle
Survey title label

protected tdPages
HtmlTableCell
HTMLCell that holds survey questions

private long WholePages

 107

Number of pages that have the amount of questions displayed equal to ..

the pagesize

Method Summary
protected void back Click fobject sender, System.EventArgs e)

Returns to the survey taker main page

private void BuildTables O

Dynamically generates tables to hold survey questions. Determines the
number of pages that a survey will have.

private void createAnswers ( Questlon question, HtmlTableCell cell)

Dynamically creates the answers controls for questions on the survey.

Disables the controls if the session object 'Taken' or 'Expired' at true.

protected void CreateChildControls Q

Maintains the viewstate of dynamically generated controls for the page

P nvate FillPages f Questlon Record, int tableNumber, int pageNumber, int numOfRecords)
Html table
Places the questions in the dynamically generated tables and puts these
tables in the tdPages cell.

protected getAnswers (ArrayList answends, ArrayList answers)

Array
Returns an ArrayList Answer objects for a Question object

protected getAllSwersFromDB (ArrayList answends)

ArrayList
Queries the database for answers. Returns an ArrayList of Answer
objects.

pnvate void InitializeComponent Q

Used to initialise components on the webpage and load event handlers.

pnvate void Page Load fobject sender, System.EventArgs e)

Loads the survey page. Also sets the Survey tide for the survey. Checks
whether the survey has expired or has been taken if so disables the submit
button.

protected void RenderScript fint Pages, int Items)

Generates the JavaScript to handle paging for the table holding the
survey questions

protected void setResponse fWebControl control, Question q)

Used to set the response of survey quesdons for surveys which have
already been taken.

protected void submit Click fobject sender, System.EventArgs e)

Collects the user responses for each question in the current survey and
creates a response entry in the database for the user.

protected void writeRcsults f)

Writes the responses of the survey for the user to the database .
 108

Class SendMail

public class SendMail implements SurveyPagcBascClass

This object handles the functionality of theSendMail page for the Survey Administrator.

Field Summary
protected Label audiencelbl
Labels used to identify the ListBox of groups

protected back
Button
Back button
protected btnScnd
Button
g encj button, used to send email

protected emails
Array t
ArrayList that holds email recipient addresses

protected Label from

Label used to identify the from inputbox

protected string goodbye

String that hold the closing message of the email with the survey
administrator's name.

protected groups
ListBox
ListBox that holds all the users groups in the system.

protected string message

Message to be sent to Survey Taker

protected string salutation

Greeting that will be used in the email.

protected select
button
Button used to select survey recipients from the groups listbox.

protected Label subject

Subject of the message

protected to
L Button
LinkButton used to display the groups listbox of survey recipient
groups

protected txtBody
ox
Textbox used to hold the body (message to be sent) of the email
protected txtFromAddress
lextx Textbox which holds the survey administrators email address
 109

protected txtSllbject
TextBox
Textbox which holds the subject of the email message
protected txtToAddress
ox
Textbox which holds the group names of the survey recipients

protected usergfOlips
HtmlTable
HTMLTable that holds the groups listbox

Constructor Summary
SendMail O
Sets the SmtpServer to localhost.

Method Summary
private void back Click fobject sender, System.EventArgs e)

Redirects the user to the survey page, where the current survey is being
displayed.

private void btnSend Click fobject sender, System. EventArgs e)

Handles the event of the send button. Creates a mail message, gathers
the 'to', 'from', subject and body of the message from the appropriate
controls and send an email message.

private void Page Load fobject sender, System.EventArgs e)

Sets the email message body and from fields the first time the page is

loaded.

private void select Click (object sender, System.EventArgs e)

Handles the event of the select button. Takes all the selected user
groups from the groups listbox and sets the text for the txtToAddress
control. Also stores the selected groups in a 'SendTo' session object

protected stnng setRecipients Q

Traverse through the 'SendTo' session object and requests the email of
the groups contained from the database via Survey Application Subsystem.
Writes the email addresses to a 'GroupLTnails' session object. Returns a
semicolon delimited string of usergroup names.

private void to Click (object sender, System.EventArgs e)

Handles the event of the 'to' linkbutton. Sets the usergroups

HTMLTable to visible and populates the groups kstboxwith group names
from the database.
 110

Class survey

public class survey implements SurvcyPagcBascClass

Displays the selected survey for the survey administrator.

Field Summary
protected answerCell
Cell which holds question answer controls

protected Table answerTablc

Table which holds answer controls

protected back
Burton
Back button
protected chooseDates
HtmlTable
Table thaf holds ca l en d ars

protected cont
Button Create survey button

protected createdSurvey
Purvey Survey object that has just being created

protected Label dateerror

Tabel used to display error message for survey creation and expiration
dates

protected Label Descriptionln

Survey description label

protected distribute
Button
Button used to go to the SendMail page

protected enddate
Calendar
Calendar used to choose an expiration date for the survey

protected hlNext
HyperLink
Tink button used forgoing to the next page of the survey

protected hlPrevious
HyperLink
Link button used forgoing to the previous page of the survey

protected pager
'
a K cr Object used for separating the survey questions into multiple pages

protected reminder
Button
email reminder button

protected
RequiredFieldValidatorl
RequiredFieldVa
 Ill

lidator Checks whether a survey title has been provided. If a title has not been
provided and error message is displayed

protected startdate
Calendar
Calendar used to choose an start date for the survey

protected SurvcyDescription
TextBox
Textbox for survey description

protected surveyQuestionsList
DataList
DataList which holds the survey question objects

protected surveyTable
HtmlTable
Parent control of the surveyQuestionList

protected SurveyTitlc
TextBox
Textbox for survey dtle

protected Label SurveyTitlelb

Survey title label

protected updateA
Button
Update answers button
protected updateQ
Button
Update question button
protected updates
HtmlTable
Table used to hold the update and email buttons

Method Summary
protected void back Click fobject sender, System.EventArgs e)

Redirects the user to the page that called it.

private void createAnswers ( Question question, TableCell cell)

Dynamically creates the controls for each question in the survey being
created

protected void cfCateSurvcy Click (object sender, System.EventArgs e)

Handles the create survey button. Check that the start and end dates
chosen for the survey are valid and writes the survey to the database
private void distribute Click (object sender, System.EventArgs e)

Redirects user to the sendmail page

protected getAnswers(ArrayList answerids, ArrayList answers)

Array
Returns an ArrayList of Answer objects that have the AnswerlDs in
the answerids arrayList

protected getAnswersFromDB (ArrayLrst answerids)

ArrayList
Returns an arrayList of Answer objects which have the answerids in
the answerids arraylist
 112

private void Page Load fobject sender, System.EventArgs e)

Loads the survey page, setting the 'Referrer',

'SurveylDVCurrentSurvey', 'NewSurveyQuestions' session objects. Checks
whether the current survey has been sent, if so disables the update buttons, if
the user was redirected from the answer2 page the chooseDates table is set to
visible so start and end dates can be chosen for the survey being created.

protected void silfveyQuestionsList ItemDataBound fobiect sender,

System.Web.UI.WebControls.DataListltemEventArgs e)

private void updateA Click fobject sender, System.EventArgs e)

Sets the 'UpateAnswer' session variable to true and redirects the user
to the answer2 page

private void updateQ Click fobject sender, System.EventArgs e)

Sets the 'UpateQuestion' session variable to true and redirects the user
to the create page

Class SurveyError

public class SurveyError implements System. Web.UI.Page

Handles the functionality of the Error page for the Online Survey Application. Any errors

with the system is redirected to this page

Field Summary
protected Label lblEi rror
Label that holds the error message

Method Summary
private void InitializeComponent Q
Used to initialize components on the webpage and load event
handlers.
Appendix C

Class Descriptions

The following are all the relevant classes that make up the survey application object

subsystem. Any classes not included can be found in the class diagram in Figure 23.

Class Answer

public class Answer extends SurvcyObjcct

Represents question responses retrieved from the database.

Field Summary
private int answerlD
ID of an answer object

private string description

Response string

private string format

Response format

Constructor Summary
Answer (stnng desc)
Default constructor

Answer fSqlDataReader reader)

This constructor takes a SqlDataReader and attempts to load the object from it

Method Summary
void AddAnswer SurvcyConncction
(
conn)
This method is used to add an Answer the DB.

static Answer GetAnswerBylD f SurveyConncction conn, int answerlD)

Call a stored procedure to get the answer with the given ID, returns an
Answer object

static ArrayList GetAnswerByQuestionlD Sun eyConnection

f conn, int questionID)
Call a stored procedure to get the answer for the given quesdon,
 114

returns an ArrayList of Answers

static ArrayList GctAnswer Formats Sun (
T
ev( Connection conn)
Call a stored procedure to get all answer formats, returns an ArrayList
of answer format descriptions
static ArrayList GctAnswcrs Survcy(
(
C< >nncction conn)
Call a stored procedure to get all answers in the database, returns an
ArrayList of Answers

static int GetFormatIDByDescfiption Sun eyConnection (

conn, string format,
Sql Transaction trans)

Call a stored procedure to get format ID by the given format

description, returns the format id

static int GetlDByDescription Sun f ey( Connection conn, string answer)

Call a stored procedure to get answer id for a given description,

returns the answer id

vol d LoadFromReader fSqlDataReader reader)

Extract the values from the reader and sets the Answer properties

string ToString Q
Returns the Answer description

Class Pager

public class Pager implements SurveyPagcBascClass

Object used to set pages for the survey object on the user interface.

Field Summary
private
pg
PagedDataSource
p ages datasource

Constructor Summary
Pager (int pagesize)

Constructor, sets the page size of the pager object

Method Summary
string Get NextLink fPage P)

Returns a string identifying the next page of the pager object

 15

stnng Get PreviousLink fPage P)

Returns a string identifying the previous page of the pager object

PagedDataSource GetNewDataSource fPage P)

Returns the PagedDataSource object for the Pager

Class Question

public class Question implements SurvevPagcBascClass

Represents Question objects retrieved from the database.

Field Summary
private string answerFormat
String representing the answer format

private ArrayList answers

Holds all possible answers available for this question

pnvate ArrayList answerSet

Holds answer ids that will be used for this quesdon on a particular
survey

pnvate stnng catdes

Descripdon of the category a quesdon belongs to

pnvate categorylD
QucstionCatcgory QuesdonCategory object that the quesdon belongs to

private string description

Descripdon of the quesdon
private int questionID
Quesdon id

pnvate stnng questionNumber

The number that the quesdon will be on a particular survey

private stnng responseString

String representing a user response for the given quesdon on a
survey

pnvate ArrayList selectedRcsponse

Holds the responses selected for a given quesdon on a survey

pnvate int surveyQuestionID

Idendfies the survey in which the quesdon is used
 116

Constructor Summary
Qliestion fstnng description)

Constructor, sets the question description

Question (SqlDataReader reader)

This constructor takes a SqlDataReader and attempts to load the object from it

Method Summary
void AddQuestion SurvevConncction
f conn)
This method is used to add a Question to the Database.

void AttachAnswer SurvcyConncction conn, Answer a)

f

This method is used to add an answer to a Question in the Database

void getCategoryPescription Q
Sets the category description for the Quesdon object

static ArrayList GetQuestionByDescription SurveyConnection conn, string description)

f

Call a stored procedure to get the quesdon by descripdon given,

returns an ArrayList containing the retrieved quesdon

static GetQuestionByID SurvcvConnection(

conn, int questionID)
Question ca ]} a stored procedure to get the quesdon from the database with the
given id, returns the Quesdon object

static ArrayList GetQuestionsByCategory SurvcyConnection conn, int intCategorylD)

(

Call a stored procedure to get the quesdon by category ID, returns all

an ArrayList of all Questions in the given category

void LoadFromReader (SqlDataReader reader)

This method is used to populate the members of this object from a

SqlDataReader

string ToString O
Returns a description of the question

void writcSurveyQiiestionResponse Sun-evConnection conn, int

(

userSurveyResultID, string response)

This method writes the answer selected for the question in a survey to
the database

Class QuestionCategory

public class QuestionCategory extends SurvcyObjcct

Represents a Question Category object

 17

Field Summary
private int categorylD
Category id

private string description

Category description

Constructor Summary
QiiestionCategory fstnng des, int id)

Constructor instantiates the category description and QuestionCategory id

QuestionCategory (SqlDataReader reader)

Constructor takes in a SqlDataReader and instantiates the properties of this object

Method Summary
static ArrayList GetAllCategories f SurveyConnection conn)
Call a stored procedure to get all of the question categories,
returns an ArrayList of QuestionCategories

static GetCatCgoryBylD SurveyConnection

f conn, int categorylD)
QuestionCategory- Call a stored procedure to get the category with the given id,
returns a QuestionCategory

sta tic GetCategoryByQuestionlD Sun eyConnection f conn, int questionID)

QuestionCategory ca ll a stored procedure to get the question categories for the given
question id

void LoadFromReader fSqlDataReader reader)

This method is used to populate the members of this object from
a SqlDataReader
string ToString Q
Returns the question category description

Class Survey

public class Survey implements SurveyPagcBascClass

Represents a survey object.

Field Summary
private creation
DateTime
Survey creation date
 18

private string creatOfID

ID of the survey creator

private string description

Description of the survey

private expire
DateTime
Expiration date of the survey

private questions
Array t
ArrayList of Questions that make up the survey

private string sent

String indicating whether the survey has been sent to recipients

private int surveylD

Survey id

private string taken

String indicating whether the survey has been taken for a particular

survey taker

private string title

Survey dde

private userGroups
ArrayList
ArrayList of user groups to whom the survey has been assigned

Constructor Summary
Survey (ArrayList questions, string title, string description, DateTime creationdate, DateTime expirationdate,
string creatorlD)

Constructor, sets the survey questions, dtle, description, creationdate, expirationdate

and creator id

Survey (int id)

Constructor, set the survey ID

Survey (SqlDataReader reader)

This constructor takes a SqlDataReader and attempts to load the object from it

Method Summary
void AddResult SurvcyConncction conn)
(

This method is used to add a result entry to the database to indicate

the user has taken the survey.

void AddSurvey SurvcyConncction conn)

(

This method is used to add a new survey to the database.

void AddSurveyQuestion SurvevConncction conn, Question q)

(

This method is used to add a new survey question to the survey.

 119

void AddSurveyQuestionAnswer (
Survcy( Connection conn, int answerlD, int

surveyQuestionID)
This method is used to add an answer to a survey question.

static void deleteSurvcy SurveyConnection conn, int surveylD)

(

Deletes survey from the database

ArrayList GetAnswerlDsBySQID fSqlConnection conn, int sqlD)

Get Answer by surveyQuestionid, returns an arraylist of answers
static ArrayList getSurveyByCreatorID Suivcy( Connection conn, string creatorlD)
(

Returns an arraylist of surveys for the given creator id

static Survey getSurveyByID ( Survcy( Connection conn, int surveylD)

Returns a Survey object with the given survey ID
static ArrayList getSlirveyByUserlD f SurveyConncction conn, string userlD)
Returns an arraylist of surveys for the given user id

ArrayList GetSurveyQuestions SurveyConnection

( conn, int surveylD)
Gets the question survey questions for the given survey id, returns an
arraylist of questions
vol d GetUserSurveyQuestionResponse SurvcvConncction f conn. Question
question)

Gets the given question response for the user for a survey

vold LoadFromReader fSqlDataReader reader)

This method is used to populate the members of this object from a
SqlDataReader

string ToStringO
Returns the survey description

v °id updateDateSent SurvevConnection conn, int surveylD)

(

This method is used to add the sent date of a survey to the database.

Class SurveyConnection

public class SurveyConnection implements SurveyObject

Represents a Survey Connection object.

Field Summary
private m conn
SqlConnection
Sql connection object
 120

Constructor Summary
SurveyConnection (stnng strConnectionStnng)
The constructor takes a connection string and automatically opens the connection

Method Summary
bool Close O
Closes the database connection

bool IsQpen Q
Returns a boolean indicating whether the connection is open
bool Qpen fstring strConnectionStnng)
Opens a database connection

Class SurveyUser

public class SurveyUser implements SurveyPagelSaseClass

Represents system all system users.

Field Summary
private string emailAddress
User's email address

private string firstName

User's first name
protected static groupIDs
ArrayList
Arraylist of groups user is assigned to

private string lastName

User's last name
private string password
User's password

private int rolelD

User's role id

private string userlD

User's login id
 121

Constructor Summary
SurveyUser fSqlDataReader reader)

constructor loads the user object

Method Summary
static getUser(stnng strUid, SurveyConncction conn)
Survey User Call a stored procedure to get user, returns a user object

void LoadFromReader fSqlDataReader reader)

Retrieved values for attribute of the user object from an
SqlDataReader

Class UserAuth

public class UserAuth extends Survey-Object

Object used to authenticate a user.

Field Summary
private string err
Error string

Method Summary
stnng GetErrorsFormatted Q
Returns the error string

bool IsUserValid fstnng strUid, string strPwd, bool IsValid, Survey-Connection conn)
Returns a boolean indicating whether the user is a valid user

Class UserGroup

public class UserGroup

Represents a user group object

Field Summary
private stnng description
 122

User group description

private int userGfOUpID

User group id

Constructor Summary
UserGroilp fSqlDataReader reader)

This constructor takes a SqlDataReader and attempts to load the object from it

Method Summary
static void AddUserGroupToSurvey SurvcvConnection (
conn, int surveylD, int

userGroupID)
This method is used to add a usergroup to a given survey.

static ArrayList GetAllGroups SurvcyConncction

(
conn)
Returns an arraylist of all user groups

static GetGroupByID SurveyConnection

(
conn, int groupID)
UscrGroup Returns a user group object for the given usergroup id

static ArrayList GetGroiipEmails SurveyO>nncction conn, int groupID)

(

Get the email address and name of recipients format returned in

ArrayList 'emaikname'

static ArrayList getGroupIDs (stnng strUid, SurvevConnection conn)

Returns an array List of Usergroups for the given user id

static ArrayList getGroupIDs fint surveylD, SurveyConnection conn)

Returns an ArrayList of Usergroups for the given survey id

volci LoadFromReader fSqlDataReader reader)

This method is used to populate the members of this object from a

SqlDataReader