MSFT204-Unit 2 - Sampling and Data Collection
MSFT204-Unit 2 - Sampling and Data Collection
T-1
UNIT II
Sampling: Rationale, characteristics- meaning, concept of population and sample, and utility.
Types of sampling Data Collection: Tools for Data Collection, Primary and secondary
methods of data collection Planning for data analysis-coding of responses. Construction of
tools for data collection.
INTRODUCTION
Sampling is a process used in statistical analysis in which a predetermined number of
observations are taken from a larger population. The methodology used to sample from a larger
population depends on the type of analysis being performed but may include simple random
sampling or systematic sampling.
Objectives
After reading this unit the learner will be able to:
Understand the meaning of sampling
Understand sampling Design
Understand characteristics of Sampling Design
Understand aims in selection a sample
Understand the various types of sample design.
Understand role of data in research
Understand Types of data
Understand various methods of collecting primary data-observation, interview and
questionnaire
Understand Sources of secondary data.
MEANING OF SAMPLING
As per Merriam Webster Dictionary it is the act, process, or technique of selecting a
representative part of a population for the purpose of determining parameters or characteristics
of the whole population. Sampling is a process used in statistical analysis in which a
predetermined number of observations are taken from a larger population. The methodology
used to sample from a larger population depends on the type of analysis being performed but
may include simple random sampling or systematic sampling.
1 Sampling Design
1
This is identified as a fixed plan or system to enable a researcher to obtain data in smaller sizes
from parts of a larger population known as samples. It also includes the modules, techniques
or procedures to be used in identifying the items for the sample. Sample design also defines
the number of items to be included in the sample i.e., the size of the sample. Sample design is
determined before data is collected. The choice of determining sampling design depends on the
individual and the type of research undertaken. Researcher must select/prepare a sample design
which should be reliable and appropriate for the research study.
2 Characteristics of a Good Sample Design
1. Proportional: Sample design must result in a truly representative sample. This means
that the sample selected should be exactly or almost similar to the population it
represents I terms of data and characteristics.
2. Error Free: Sample design should reduce the probability of errors. The minimum
numbers of errors in any sample ensure correct data obtained and analyzed.
3. Budgeted: Sample design must be practical and be within the limits of funds available
for the research study.
4. No Bias: Sample design should be able to control systematic bias.
5. Generalization of Results: Sample should be such that the results of the sample study
can be applied, in general, for the universe with a reasonable level of confidence.
2
An individual.
A geographical one such as state, district, village, etc.,
iii. Source list: It is also known as ‗sampling frame‘ from which sample is to be drawn. It
contains the names of all items of a finite universe. If source list is not available, researcher has
to prepare it. Such a list should be comprehensive, correct, reliable and appropriate. It is
extremely important for the source list to be as representative of the population as possible.
iv. Size: The sample size should be justified, not be excessively large nor it should be too small.
Preferably the sample size should be optimal which fulfills the requirements of efficiency,
representativeness, reliability and flexibility and representative of the population to obtain
dependable outcomes. Population variance, population size, parameters of interest, and
budgetary constraints are some of the factors that impact the sample size.
This refers to the number of items to be selected from the universe to constitute a sample which
at times may be an important issue for a researcher. While deciding the size of sample,
researcher must determine the desired precision as also an acceptable confidence level for the
estimate. A larger variance usually needs a bigger sample size to assure correct results. The
parameters of interest in a research study must be kept in view, while deciding the size of the
sample.
v. Consideration of interest: In determining the sample design, one must consider the question
of the specific population stricture which is of interest. E.g. we may calculate the number of
walk in guest‘s from total arrivals at a hotel on daily basis to understand the proportion and
then to leave an optimum number of unreserved rooms everyday for such guest.
vi. Budgetary limitation: Funds available guide us to decide the size, variation and quantum
of samples. This fact can even lead to the use of a non-probability sample.
vii. Sampling procedure: Finally, the researcher must decide the type of sample he will use
i.e., he must decide about the technique to be used in selecting the items for the sample. In fact,
this technique or procedure stands for the sample design itself. An ideal design is the one that
for a given sample size and for a given cost, has a smaller sampling error.
TYPES OF SAMPLE DESIGNS
1. Non-probability sampling: Non-probability sampling is a sampling technique where the
samples are gathered in a process that does not give all the individuals in the population
equal chances of being selected. A core characteristic of non-probability sampling
techniques is that samples are selected based on the subjective judgment of the researcher.
It is also known as deliberate sampling, purposive sampling and judgment sampling.
3
With non-probability sampling methods, we do not know the probability that each
population element will be chosen, and/or we cannot be sure that each population
element has a non-zero chance of being chosen.
The choice of researcher is considered to be the topmost in priority. The sample construction
depends upon purposive selection i.e. deliberate and is such that it represents the entire
population in that smallest unit.
This offers the advantages of convenience and cost but the disadvantage is that non-probability
sampling methods do not allow the estimation of the extent to which sample statistics are
possibly varying from population parameters.
E.g. To study the average spending or average number of days stayed by tourists visiting
religious destinations the researcher has the freedom to choose destinations and state them to
be representative of all other religious destinations.
Non-Probability Sampling Methods: Voluntary and convenience samples are two major
types discussed here.
Voluntary Sampling: This constitutes of people who have keen interest in the topic of
survey being conducted and are themselves getting involved to contribute as
respondents. E.g. for a survey or online poll being conducted on a social site like
Facebook attracts volunteers with common interests and they participate in it.
Convenience Sampling: A convenience sample consists of people who are easily
approachable and can be reached out to in shorter time.
E.g. To study the popularity of handmade products or traditional goods the researcher may
choose to visit a local fair where
2. Probability sampling: This Sampling technique uses randomization to make sure that
every element of the population gets an equal chance to be part of the selected sample. It‘s
alternatively known as ‗random sampling‘ or ‗chance sampling‘. Selection of winner of a
lottery selected through mechanical process gives all ticket holders an equal chance of
winning.
With probability sampling methods, each population element has a known (non-zero)
chance of being chosen for the sample.
A facility to measure the errors of estimation or the significance of results obtained from a
random sample by providing same chance to all options proves that random sampling design
is better than deliberate sampling design. The law of Statistical Regularity which states that if
on an average the sample chosen is a random one, the sample will have the same composition
4
and characteristics as the universe is truly applied here. This is the reason why random sampling
is considered as the best technique of selecting a representative sample.
a) It gives each element in the population an equal probability of getting into the sample;
and all choices are independent of one another.
b) It gives each possible sample combination an equal probability of being chosen.
it is easy to reach out to buyers of such goods and services.
Probability Sampling Methods: The main types of probability sampling methods are simple
random sampling, stratified sampling, cluster sampling, multistage sampling, and systematic
random sampling. The basic advantage of probability sampling methods is that they assure that
the sample chosen is representative of the population thereby ensuring that the statistical
conclusions stand valid.
i. Simple random sampling. Simple random sampling refers to any sampling method that
has the following properties.
The population consists of N objects.
The sample consists of n objects.
If all possible samples of n objects are equally likely to occur, the sampling method
is called simple random sampling.
There are many ways to obtain a simple random sample. One way would be the lottery
method. Each of the N population members is assigned a unique number. The numbers
are placed in a bowl and thoroughly mixed. Then, a blind-folded researcher selects n
numbers. Population members having the selected numbers are included in the sample.
ii. Stratified sampling. With stratified sampling, the population is divided into groups,
based on some characteristic. Then, within each group, a probability sample (often a
simple random sample) is selected. In stratified sampling, the groups are called strata.
E.g. For a survey carried our across a state the population may be divided age wise into
groups or strata, like infants, children, minors, adolescents, teenagers, adults, etc.
Within each stratum, we might randomly select survey respondents.
iii. Cluster sampling. With cluster sampling, every member of the population is assigned
to one, and only one, group. Each group is called a cluster. A sample of clusters is
chosen, using a probability method (often simple random sampling). Only individuals
within sampled clusters are surveyed.
With stratified sampling, the sample includes elements from each stratum. With cluster
sampling, in contrast, the sample includes elements only from sampled clusters.
5
iv. Multistage Sampling: In this method of sampling, we select a sample by using
combinations of more than one sampling method.
For example, in Stage 1, we might use cluster sampling to choose clusters from a
population. Then, in Stage 2, we might use simple random sampling to select a subset
of elements from each chosen cluster for the final sample.
v. Systematic Random Sampling: This begins with creation of a list of each member of
the population. From the list, we randomly select the first sample element from the first
k elements on the population list. Thereafter, we select every kth element on the list.
This method is different from simple random sampling since every possible sample of
n elements is not equally likely.
The fact that each guest in the sample was randomly sampled is a necessary condition for a
simple random sample, but it is not sufficient. Similarly, the fact that each guest in the sample
had an equal chance of being selected is characteristic of a simple random sample, but it is not
sufficient. The sampling method in this problem used random sampling and gave each guest
an equal chance of being selected; but the sampling method was actually stratified random
sampling.
The fact that hotel guests of each of the brands were equally represented in the sample is
irrelevant to whether the sampling method was simple random sampling. Similarly, the fact
that population consisted of guests of different hotel brands is irrelevant.
DATA IN RESEARCH
Data collection is the process of gathering and measuring information on targeted variables in
an established systematic fashion, which then enables one to answer relevant questions and
evaluate outcomes. Data collection is a component of research in all fields of study including
physical and social sciences, humanities, and business. It is a component of research in all
fields of study including physical and social sciences, humanities, and business.
While methods differ by discipline, the importance of collecting accurate and honest data
remains in place. The goal for all data collection is to capture quality evidence that allows
analysis to lead to the formulation of convincing and credible answers to the questions that
have been posed.
Importance of accuracy in Data Collection
Despite the variation of the field of study or preference for defining data
(quantitative or qualitative), accurate data collection is essential to maintaining the
integrity of research.
6
Both the selection of appropriate data collection instruments (existing, modified, or
newly developed) and clearly delineated instructions for their correct use reduce the
likelihood of errors occurring.
A formal data collection process is necessary as it ensures that the data gathered are
both defined and accurate and that subsequent decisions based on arguments
embodied in the findings are valid.
The process provides both a baseline from which to measure and in certain cases an
indication of what to improve.
Consequences from improperly collected data include:
Inability to answer research questions accurately;
Inability to repeat and validate the study.
distorted findings resulting in wasted resources
misleading other researchers to pursue fruitless avenues of investigation
compromising decisions for public policy
causing harm to human participants and animal subjects
Types of Data
a) Primary Data: Primary data means original data that has been collected specially for the
purpose in mind. It means someone collected the data from the original source first hand. Data
collected this way is called primary data. The people who gather primary data may be an
authorized organization, investigator, enumerator or they may be just someone with a
clipboard. Those who gather primary data may have knowledge of the study and may be
motivated to make the study a success. These people are acting as a witness so primary data is
only considered as reliable as the people who gathered it.
b) Secondary Data: Refers to data which is collected by someone who is someone other than
the user. Common sources of secondary data for social science include censuses, information
collected by government departments, organizational records and data that was originally
collected for other research purposes. Secondary data analysis can save time that would
otherwise be spent collecting data and, particularly in the case of quantitative data, can provide
larger and higher-quality databases that would be unfeasible for any individual researcher to
collect on their own. In addition, analysts of social and economic change consider secondary
data essential, since it is impossible to conduct a new survey that can adequately capture past
change and/or developments. However, secondary data analysis can be less useful in marketing
research, as data may be outdated or inaccurate.
7
DATA COLLECTION METHODS / TOOLS
8
population) fill to give the researcher information needed for the study. The data given from a
questionnaire cannot be achieved from the secondary resources (Pandey & Pandey, 2015).
These forms are suitable to gather both qualitative and quantitative data. Although they are not
the most common methods used in qualitative research, they are useful in case of facing a large
sample in a study.
Sir Francis Galton designed a questionnaire for the first time. A questionnaire is utilized for
different purposes, although it is commonly used to gather statistical data. It can be designed
for measuring separate variables such as behaviors, preferences, and facts (Kabir, 2016).
Although the preparation and administration of a questionnaire are not hard, specific points in
these processes should be required. This form is used, normally, when it is not possible to
discuss each participant personally (Pandey & Pandey, 2015). Thus, it helps to gather data from
different individuals, groups, and companies easily. Questionnaires can be categorized based
on different aspects such as types of questions and administration modes.
A) Types of Questions
First, questions can be designed to measure variables for example in a survey. On the other
hand, questions can be based on aggregating into indexes or scales, for instance in tests.
Second, question types can be categorized into closed-ended and open-ended questions. In
close-ended questions, the respondents face a specific range of answers to choose from, but the
respondent is asked to provide formulated answers using open-ended questions. Qualitative
questions are open-ended (Taherdoost, 2019). In this type, then, the answers should be coded
into a response scale. Therefore, in comparison to the open-ended questions, close-ended ones
are pre-coded to make the work quickly be implemented (Olsen, 2012).
For close-ended questions, there are four types of options to respond the questions:
You can have a two-option as the responses possibilities which are known as
dichotomous scales.
If you add more than two options for the respondents, the scale is known as nominal-
polychromous.
In ordinal-polytomous scales, you prepare more than two options which are also
ordinal.
Finally, you can use continuous or bounded types which use a continuous scale as a
possible response case.
B) The Mode of Administration
9
Questionnaires can be implemented in different ways. A face-to-face questionnaire mode can
be used which provides the chance of presenting the questions orally, paper-and-pencil types
can be utilized with the items presented in the paper or computerized questionnaires for data
collection (Kabir, 2016). Questionnaires can be also utilized through telephone, online, or even
posting. An online questionnaire is a cost-efficient option; however, you should consider the
possibility of missing samples due to problems with internet access. In these types, different
online survey services can be used which provide questionnaires for the purpose of study, and
then the collected data can be easily added to the analyzing software (Taherdoost, 2021). In all
these choices, it is important to secure ethical concerns such as the confidentiality of
participants. On the other hand, participants should try to answer the questions politely and
clearly (Kabir, 2016).
C) General Rules for Constructing a Questionnaire
Use simple and short questions as much as possible;
Navigate respondents clearly to avoid any difficulty and motivate participants through
answering questions (Olsen, 2012);
Use understandable, simple, and clear statements for all respondents with different
educational levels;
Utilize positive sentences;
Do not use more than one question (double-barreled) in one item;
Add an open-answer possibility after providing the listed answers and where possible;
Avoid making assumptions for the respondents (Kabir, 2016);
Try to increase reliability by appropriate word selection;
Avoid directing the respondent to any answer using objective questions including clues,
suggestions, and hints) (Taherdoost, 2021);
Explain the importance of the questionnaire in its content as well as its cover letter
(Pandey & Pandey, 2015);
Use sensitive questions after some innocuous or concrete ones. The order of questions
is an important point. For example, those questions which are about sexual activities,
drugs, incomes, etc. should be added to the final parts of questions to help the researcher
gain participants’ trust and avoid their embarrassment (Kabir, 2016; Taherdoost,
2016a).
There are also specific challenges and concerns that may be faced through designing an
appropriate questionnaire. First, the maximum respondents’ rates should be guaranteed
10
together with securing maximum reliability and validity as much as possible. The respondent
rates can be maximized when you:
● Can convince them you secure their information and keep their side;
● Can reward their cooperation.
On the other hand, you can gain an accurate data set considering two points:
● Prepare a suitable set of questions;
● Select appropriate sample size and type which can avoid biases and non-responded
questions (Taherdoost, 2021).
D) Advantages of Questionnaires
Questionnaires provide several merits in comparison to other survey methods as listed in the
following:
● Collecting a large amount of data from a large sample size;
● Time saver;
● Cost-effective options;
● Highly structured;
● The possibility of gaining high accurate data;
● The possibility of being carried out by other people instead of the researcher regardless of
affecting the reliability and validity term, and the possibility of group administrations;
● Analyzing the results easily by entering the achieved data to the software quickly in the
majority of cases;
● The opportunity of more objective and scientific analysis;
● The achieved quantitative data can be used to compare and contrast the results of the study
with others to measure the changes;
● The possibility of achieving comprehensive design and tests, and administrating the
research with required details;
● Creating novel theories or/ and testing an existing hypothesis using the achieved
quantitative data (Frechtling, 2002; Kabir, 2016; Taherdoost, 2021);
11
● Suitable in a wide range of study fields;
● Suitable and reliable in special cases (Pandey & Pandey, 2015).
E) Disadvantages of Questionnaires
However, there are also several demerits that are not negligible. There are several difficulties
researchers may face using questionnaires as the following:
● Hard or inadequate to perceive gathered data in some cases such as emotional, feelings,
and behavioral changes;
● Human errors for example if the respondent is forgetful and cannot consider the whole
concept truly;
● Determining the reliability of answers is not possible;
● The possibility of misunderstanding the questions which can overshadow the answers;
● The effects of differences in human beliefs on their answers in some cases since even a
standard subject can be considered good for one group and bad for others (Kabir, 2016);
● Facing difficulties when participants need clarifications for particular questions in
impersonal administrations and the possibility of failing to answer those questions
(Taherdoost, 2021);
● Low response rates if respondents’ low interests cannot be addressed to answer questions
(Frechtling, 2002);
● The possibility of illegible answers;
● Useless and wrong answers are prevalent (Pandey & Pandey, 2015).
2. Interviews
In interviews, as a fundamental way of social interaction, questions are asked and data is
collected using provided answers and it is in contrast to the questionnaire with indirectly
collected data methodology. Thus, the chance of getting confidential data from interviewees is
also possible; however, it requires special skills which are not necessary for questionnaires.
Researchers can employ different methods to conduct an interview (Pandey & Pandey, 2015;
Taherdoost, 2021) and perform them in individual, or group face-to-face interviews, as well as
not personally for example using telephone, computer, etc. (Kabir, 2016). Both forms bring
along merits and demerits which can be listed in Table.1.
There are different ways to record interviews including paper-based, self-reports, and tape-
record interviews, etc. It can be also noted that they are useful to gain both qualitative and
quantitative data using open-ended and close-ended questions. Researchers rely on interviews,
generally, when they face complex or sensitive concepts and need detailed and high-status
information (Frechtling, 2002).
12
A) Types of Interviews
In addition to the administration mode of interviews which was discussed before, interview
types can be structured, semi-structured, and unstructured.
▪ Structured interviews:
In these kinds of interviews, interviewees face the same set of standardized questions which
are pre-prepared before the interview session. The possible responses are limited, and
participants may just face a few open-ended questions. In case of gaining a good
understanding of the research topic, it is possible to provide a questionnaire, or an interview
guide based on the structural interview characteristics. A suitable place to use these kinds is
when there is a comprehensive literature review about the topic of study or to utilize it
following a less structural method.
▪ Semi-structured Interviews
Semi-structured interviews are formal and are conducted based on a guide. The interviewers
ask questions considering the guidance; however, when researchers or interviewers need
extra information, they can continue the conversation based on the questions provided ahead
of time. For example, it is an appropriate choice when you do not have the chance of
interviewing with individuals more than once and you should gather clear, comparable, and
reliable qualitative data in the first meeting. Thus, training of the interviewers and the design
of open-end questions are important points. It can also be used when you aim to develop a
subject such as a culture, setting, or experience which is not completely understood, and as
an initial step for developing another more-structured method like a survey or an interview.
13
▪ Unstructured interviews
Unstructured interviews are informal methods of interviewing without using a specific
structure. There is no guide in this type, and they just conduct casual conversations.
Interviewers collect data using brief notes and try to memorize the responses. This process
is a part of field observation and can be a good choice when there is limited literature about
the field of study.
In these forms of interviews as they are "on the fly", the chance of tape-recording or writing
detailed notes is slim, thus, the researcher should try to participate in the setting directly as
soon as possible and should approve the notes and achieve the final data quickly after the
observation to avoid losing information. Like semi-structured methods, informal ones can
be also a fundamental step for preparing structured interviews. These methods can enhance
the trust of the interviewees for the next levels as well, as both sides speak openly and freely
to each other. Achieving qualitative data using informal interview methods can help the
researcher to understand other individuals' experiences. These types also allow asking
questions to the interviewees as a follow-up response to the interview questions (Kabir,
2016).
B) Important Notes in Conducting Interviews
As running interviews can be both rewarding and challenging in a study, different points should
be considered to conduct the process as far as possible. Some important points are as the
following:
● The interviewer plays an important role as they are responsible to motivate participants to
answer questions and judge whether the answers cover the expected quality or not. They
conduct and guide interviewees when needed. Thus, training the interviewers is essential in
these kinds of data collection types.
● The setting should be completely ready for the interviewers. They should address all their
needs to the materials (known as interviewer's kit) including identifications, phone numbers
for later contacts, maps, copies of guides, cover letters for example provided by the sponsors
and main researcher, notebooks, etc.
● Another important point is that interviewers’ biases can affect the findings of the study.
Thus, this bias and judgmental thoughts should be minimized to gain the best result. This
factor can be also addressed by interviewers’ training. These problems can mainly happen
through job interviews.
The process of executing an interview is shown in Figure 3.
14
C) Merits and Demerits
The main advantages and disadvantages of conducting interviews as a method of collecting
data are provided in the following. The merits can be listed as:
● Gathering rich, in-depth, and detailed data directly
● The opportunity of obtaining knowledge about past and future for particular events and
features
● The flexibility of administration of interviews
● The possibility of an explanation of the questions to the interviewees to clarify the
questions
On the other hand, there are some difficulties that may be faced when conducting an interview
method including:
● Hiring and training interviewers
● Complex process
● Scheduling where and when to meet people and the possibility of changing plans at the
last minutes
● The possibility of missing information
● The coding process can be difficult and long
● Being expensive (Frechtling, 2002; Kabir, 2016; Pandey & Pandey, 2015)
3. Focus Groups Discussion (FGD)
This method, simply, is a mixture of interviewing and observation. This method is used to
discover human behavior, attitudes, and respondents facing a particular concept. This in-depth
field method gathers a group of individuals, normally between 6-12 people in each group,
commonly with a shared characteristic such as sex, age, and educational status to discuss a
specific study field. The mediators aim to stimulate and discover the behavior of the
15
participants and the reasons of each behavior using social dynamic of the group (Frechtling,
2002) (Kabir, 2016).
For this purpose, data regarding a particular subject is collected using a semi-structured
interview and utilizing a specific guide in a relaxed and friendly atmosphere which helps the
interviewees to easily trust the facilitators of the program and answer them honestly. As in
interviews which interviewers should be trained to be supportive and good listeners, the
training of the mediators is essential also here to gain especial skills such as analyzing the
participants’ behavior, facilitating, controlling, and managing the meeting sessions, listening
and observing the events. Furthermore, to record data and provide detailed notes an observer
often helps the mediator (Kabir, 2016).
This method is used in times of aiming to pretest a concept or an idea, generate an idea, discover
a topic further, discover new areas, clarify the findings of other methods, and achieve a full
understanding of a commonly social concept using small groups by stimulating discussions. It
can also assist researchers to develop the interviews and surveys. When a concept is hard to
observe, it can also be a useful method to obtain a concentrated observation set in a short period
of time. When the results of a concept cannot be discussed statistically, this method helps to
gain meaning for the findings of a survey (Frechtling, 2002) (Kabir, 2016).
A) General Rules
After identifying the suitable target group considering shared required characteristics, a short
explanation about the concept should be provided to help the participants get familiar with the
background and the meaning of the concept. Then, the standardized questions should be
discussed based on the protocol and the responses of the group to issues should be written.
Discussions about the concept should be also facilitated to gain more than just normal question-
answer interactions. To obtain a comprehensive investigation, more than one focus group
(minimum three) should be managed. After summarizing the results of the discussions, the
meeting can be finalized, although sometimes more than one session is needed. General steps
of conducting an FGD are shown in figure 4.
B) Interview or Focus Group
There are general circumstances to decide which method can be effective. First, when the
interaction between members is not productive or may negatively affect the findings, an
interview is a good choice. On the other hand, if the group pressure can positively overshadow
the challenges of the participants' thoughts, a focus group can be conducted. For covering a
large volume of issues, interviews are more appropriate choices. In focus groups, a single
behavior is commonly examined regardless of considering connections between behaviors and
16
attitudes. Additionally, the participants do not have a depth understanding of the concept and
it is assumed that individuals can discuss all their known relevant knowledge in less than 10
minutes (Frechtling, 2002).
C) Advantages and Disadvantages
The merits and demerits of the focus group method are focused on in this section as these types
also possess both. They have several merits as:
● They can help to discover social, health, and cultural concepts for example by considering
humans' behavior facing different situations;
● Literacy of individuals is not an issue;
● They are suitable to explore complex subjects;
● They are useful for the development of hypotheses.
On the contrary, they are also:
● Can be expensive and time-consuming;
● Face privacy lacks;
● Need trained facilitators;
● Can face the issues due to the domination of limited individuals in the focus groups
(Frechtling, 2002; Kabir, 2016).
4. Observational Methods
In these techniques, first-hand data is gathered through the observation of events, behaviors,
interactions, processes, etc. directly to obtain an understanding of the concepts. For example,
observation is an appropriate technique to evaluate teaching methods in the classes. It can be
used when focus groups and interviews cannot help to gather data due to the different reasons
including times that participants:
17
● Are not aware of the concept;
● Are not able to talk about the concept;
● Do not prefer to discuss the concept.
It can be also utilized to explore whether a study is progressing as planned, or whether the study
has been successful or not. In the evaluation of studies, these two phases are known as
formative and summative, respectively (Frechtling, 2002). It also can be helpful when the
concept is unexplored or not well-known. If it is required to explore a subject in the natural
setting and the reported information can be different from the findings of the real setting, an
observational technique should be used.
This method can collect both qualitative and quantitative data. The qualitative data is gathered
as a description of events in the setting. The quantitative data can be obtained by using the
duration or frequency of the particular subjects. During this kind of systematic observation,
formal and structured instruments and protocols nominal, ordinal, ratio, and interval scales are
utilized. Thus, it can be used to record the findings template coding sheets with specific guides
if the observer is not the main researcher. On the other hand, data achieved through this method
can be used in conjunction with the quantitative findings of other methods.
Generally, observation helps the researcher to find out what is going on (Kabir, 2016) in the
surrounding environment; however, as a data collection method, it is further than just listening
and looking. This method includes an engagement with the setting, a clear expression of the
events, technical improvisations, high attention, and good recording.
A) Advantages and Disadvantages
The observational method also possesses several pros and cons. In this section, the most
important ones are listed. The advantages are as the following:
● Gathering direct information;
● The participation of evaluators in the natural setting;
● Flexible and natural atmosphere;
● Free from biases;
● Can be generalized as large samples can be covered in the studies;
● High reliable and precise data can be achieved.
These techniques also provide some difficulties as:
● They can be time-consuming and not economical
● The training of observers is effective
● Observers can be selective and distort data
18
● It can be sometimes unreliable due to the misrepresenting of the qualitative data
measurement
● It does not consider processes and the changes during them and cannot be appropriate
for fresh concepts (Frechtling, 2002; Kabir, 2016).
B) Special Notes
After conducting an observation, the researcher should first analyze achieved data. For this
purpose, data is summarized in a process known as data reduction and it is coded based on
particular criteria to specific categories. The reliability of data according to the agreements of
independent observers should be also considered to show how the behaviors are measured
accurately. It should be noted that the participants can act differently when they are in the
research setting. These acts should be controlled using techniques of controlling reactivity such
as indirect observation, the adaptation of participants, and unobtrusive measurements.
Biases of the observers are another important point to consider what happens when an
observer's bias can affect what behaviors to choose and record; however, it can be minimized
by keeping them unaware about the aims (Kabir, 2016).
5. Survey Methods
A survey simply is an appropriate method to determine feelings, opinions, and thoughts. The
aim of the survey can be both globally and specifically. They can provide a large volume of
data using telephone calls, emails, or face-to-face interviews.
On the other hand, data can be collected in self-completion surveys or by the interviewer. A
survey can be used to explore social behaviors such as measuring the behavior of political
candidates and professional people in educational institutions. However, it is not useful when
evaluating people for government programs since in these programs, all members of the
population should be studied (Kabir, 2016). Overall, in both formative and summative phases
of a study, surveys are useful when it is required to collect information from a large target
population, and detailed and in-depth data are not necessary in the project (Frechtling, 2002).
In a survey, a set of questions are provided to give a sample that is chosen from a specific target
population. This sample presents the characteristics and behaviors of the population. Surveys
are conducted to explore the populations' attitudes, the differences between different
populations' behavior and discover the possible changes over time by repeating surveys in
regular time intervals. Thus, the sample selection is an important stage in this process which
can highly affect the findings. Sample sizes should be chosen based on the possibility of
selecting every participant with a non-zero chance. Therefore, samples need to be chosen using
a non-volunteer and non-haphazard selection technique (Kabir, 2016).
19
Sampling process steps can be simply listed as the following:
● Defining target population such as the number of individuals that are living in the
country;
● Selecting a frame for sampling as the actual cases that we select a sample from them;
● Choosing the method of sampling which can be either a random or non-random
technique;
● Measuring the appropriate sample size to avoid biases and sampling errors using the
related formula (Taherdoost, 2016c).
Biases of participants can commonly happen, especially when they need to answer about
sensitive subjects, or when the individuals need to trust the team before giving the right answer.
Questions can be designed in different ways as the following:
● Open-ended questions in which participants answer questions in their own ways;
● Close-ended questions mostly based on yes/no or true /false answers;
● Multiple choices that provide the opportunity to choose a favorite topic by
participants (Kabir, 2016).
Here, as discussed in the questionnaires, the questions should be written considering several
aspects ranging from their language to their length and their presenting order. For example,
sensitive questions should be added among final questions as well. Cover letter and
introduction should be also provided as discussed in other types.
A) Survey Process Steps
Figure 5 summarizes main process steps in surveys; however, it should be noted that although
some steps such as pilot testing are essential, some steps are not necessary in conducting all
surveys, and as the surveys are complex, the process does not need to follow the order and
these sequential steps are provided as a common example. First, the design and plan of the
study should be provided. It includes defining the purpose of study, sample selection method,
its size, and designing and providing questionnaires.
At the next level, you go through the testing and modifying processes which include examining
pilot tests and pre-tests if possible, analyzing the tests, and modifying them if needed, and
repeating the whole process from the first step after all modifications, then you should conduct
the survey, finalize the documents and questionnaire forms, select your sample, train the
interviewers and conduct the survey. Finally, the gathered data should be analyzed by entering
data, coding it, preparing the result reports and lastly, finalizing them as technical reports
(Kabir, 2016).
20
B) Common Issues and Errors
Different issues and errors may happen during the survey process due to the several possible
resources. These errors can stem from the following origins:
● Inappropriate planning due to the lack in well-understanding of the study concept;
● Selecting unappropriated survey method including emails, face-to-face, etc.;
● Inappropriate questionnaire design;
● Interviewers issues such as misunderstanding the answers, biases, etc.;
● Sample selection issues such as biased selection, small size;
● Problems of the respondents, for example, memorial mistakes;
Errors can also happen during the processing of the gathered data. For example, in
coding processes and also when researchers estimate the results of the survey. To
prevent the errors or minimize them as much as possible, researchers should consider
different points considering error types, they need to:
● Gain a comprehensive definition about all concepts;
● Use sample selection methods;
● Test the surveys using suitable methods;
● Consider the points of designing a questionnaire;
● Train the interviewers;
● Use introductions and guides for the interviewees;
● Use trained staff for data analysis;
● Use professional statisticians for estimation processes.
C) Advantages and Disadvantage
21
As discussed in other proposed methods, this type also possesses several advantages and
disadvantages. The important merits and demerits of the survey method can be summarized in
Table 2.
6. Case-Studies
Case studies give you the opportunity to investigate issues deeply and descriptively. It covers
different concepts ranging from individual(s) to organizations and sectors when the number of
sites is small. As they develop a real-world context by observing what happens there and also
interviewing the participants, they can help the researcher to study the site precisely and obtain
deep explorations. It is basically an empirical investigation that can study different phenomena
in a real-life context when the boundaries between them are not clear and also needs a
comprehensive investigation. Thus, brief and occasional visits cannot be helpful to achieve this
complex task. For example, detailed data gathered through these types are mainly used by a
psychologist who needs to gather information about their patients' lives to gain the best
treatments for their diseases (Frechtling, 2002; Taherdoost, 2021). Although case studies are
not considered as research methods by themselves, researchers conduct different data
collection and analysis methods to gain the material that they need for their study. For example,
to gather data in qualitative methods, interviews, and diaries can be used. Furthermore,
different official and personally provided notes can be utilized to collect information. Methods
such as grounded theory can be used to analyze the achieved data (Kabir, 2016). The common
process steps in a case-study method are shown in Figure 6.
22
A) Types of Case Studies
Case studies generally can be categorized into single and multiple ones. Although, these two
types can be either narrow or broad. Single case studies are single experiments that can be
utilized to analyze a concept deeply, gain an initial result for the next extended multiple groups
of cases, and study long-standing theories. Multiple case studies cover different units, and the
number of units can positively affect the results of the study. The results of the considered units
can be identical or opposite. Holistic cases analyze one unit and embedded analyze two or more
sub-units and, in these methods, the number of units is an important factor. Furthermore,
considering these four discussed types, case studies can be broad or narrow-based on Table 3
(Taherdoost, 2021).
In addition to the above categories which are based on the number of cases and units of the
study, case studies can be categorized as the following:
✓ Explanatory: they study the surface and deep levels of data to gain a comprehensive
explanation of the phenomenon in data. They are used to match a pattern for a
phenomenon in complex casual cases.
✓ Exploratory: this type of case study is used to obtain an introduction for further
complete studies. It does not consider in-depth factors and helps to collect data to
achieve a protocol before developing the research question and hypothesis of the study.
Pilot studies are considered as an example of this case study type.
✓ Descriptive: the researcher in this case study type aims to achieve a description of
data as occurring or in other words the occurred phenomena within the data. Thus, the
researcher needs a description basis theory to gain the description of the phenomenon
as a result. This can make challenges as the theory can fail from the first step and it can
bring different problems during the next steps of the study. Journalistic descriptions are
considered descriptive case studies.
23
✓ Prospective: the outcome of the process is the aim of this type of case study. For
example, if researchers observe a group of patients to discover what happens to them
after a particular period of time, it is a prospective case study type.
✓ Retrospective: in this type, researchers consider the historical information for their
study case. For example, they gather historical data regarding the patients to collect the
risk factors when studying the outcomes of a specific illness.
✓ Intrinsic: In this form of case study, the subject of the study is the researcher's interest.
✓ Instrumental: the individuals provide observers with more information than just what
happens in the setting.
B) Information Sources
Several methods are used to collect data in a case study. Researchers can use a direct
observation method in which they observe participants in the natural setting and gather
information regarding the reactions and interactions of the individuals. In this method, a single
or a group of observers can participate. Another method is participant observation in which the
researcher participates in the setting like other under-study people and observes the happenings
from a closer perspective.
Interviews are also among the most common methods that can be conducted based on survey-
type questions which are structured or based on more open-ended question sets. Finally,
researchers can use different census and survey records, newspapers, letters, instruments, tools,
etc. (Kabir, 2016).
C) Advantages and Disadvantages
Several merits can be noted in case-study methods as they:
● Assist specific human sources cases by gathering in-depth information to gain a
hypothesis;
● Can utilize different research methods based on researchers’ preferences;
● Consider past, present and future aspects together (Kabir, 2016);
● Provide an explanation about the changes and impacting factors that are not always
driven from the data. This can bring positive consequences in social units;
● Consider both individuals and their interactions (Taherdoost, 2021).
However, they also provide some challenges as:
● They are complex processes, time consuming and expensive;
● There is no limit for collecting data and it is a challenge for the researcher to realize
when it is time to stop the data collecting process;
24
● The taken basic assumptions are not always realistic and eventually, the gathered data
should be tested in these situations;
● The personal documents and letters which are used by the researcher can face
distortion problems (Kabir, 2016);
● Case studies need to use expert and trained conducting teams;
● Over-interpreting and over-generalizing issues can also happen in these types
(Taherdoost, 2021).
7. Experimental Methods
Experiments are also one of the main methods and even the prime method of data collection.
In this method, the researcher can create a research situation, choose the participants and
manipulate the independent variables (often limited and small numbers) which can affect a
dependent one and explore how they overshadow the dependent variable (Hox & Boeije, 2005).
Briefly, it discovers a causal relationship between them known as effect and causes
relationships (Taherdoost, 2021). These methods are objective, and the thoughts of the
researchers do not affect the results. Therefore, their validity is high and bias effects are
minimized. However, the possible biases can still stem from different aspects including:
● Selection, when you can recognize the differences of the groups at the beginning of
the research;
● The placebo effect, when a belief about a study subject can affect the findings, for
example, you think the understudy medicines can help you to be better;
● Experimenter, when you as the researcher have a specific belief about the subject
(Kabir, 2016).
A) Types of Experiments with Their Merits and Demerits
Experiments can be conducted in three ways:
✓ Laboratory/controlled settings: This provides you with high control over study design
and processes. You can gain precise and accurate data although it may cause some
problems. For example, in the term of ecological validity (how much the results of our
study can be generalized to real-life?) in laboratory settings, according to the artificial
degree of the studies. Thus, these methods are better choices when a phenomenon is
manageable in the laboratory, not about the variables which are seen in everyday
activities in humans’ real lives (Hox & Boeije, 2005).
25
✓ Field Experiments: you can conduct it where your participants are in their real-life
situations. Although you can still manipulate the variables, the ecological validity of
your research is higher as your control is less than the first category.
✓ Natural Experiments: These types are done in the participants' natural setting;
however, in contrast to two other types, here you have no control over the setting. They
can assure a very high ecological validity as it is less likely that different parameters
can affect the results. However, on the other hand, the effects of biases are possible due
to the control of less extraneous variables which are not independent but can
overshadow the result). In addition, other researchers cannot conduct the same study
with the exact parameters as the study as well (Kabir, 2016).
8. Diaries
these self-administered questionnaires ask participants to fill out the forms in two different
situations. First, event-based ones ask to fill in when something happens. The second type is
known as time-based diaries which should be filled at a particular time frame or interval. On
the other hand, diaries can also be elicitation and feedback studies. Elicitation types let the
participants utilize media to help their memory. Feedback studies use predesigned questions to
get immediate responses.
This method can be used in the following situations:
• Tracking people over time;
• Making a cross-sectional comparison of individuals;
• Investigating processes among families and people.
9. Activity Sampling Technique
This method utilizes consecutive observations on individuals, groups, and even machines in a
specific time frame to record the happening events and achieve a rating if needed. Here, the
main aim is to record delays and percentages of the works that are done during a specific period.
It employs sampling techniques to define the activity of the understudy subjects and then
analyze those using statistical methods. The number of observations is a key factor since it can
affect the accuracy of the study. However, as there are limits due to the time and costs, an
optimum number should be utilized based on the approximate number of the occurrence of the
factor and acceptable accuracy of the study.
10. Memo Motion Study
26
This method is also known as spaced-shot photography, and uses a camera to analyze a long
process. As a tool of time, it uses film material for study purposes. Nowadays, this technique
is used commonly, as many of the current settings can address the required conditions.
11. Process Analysis
This analysis is used to gain higher efficiency and decrease wastes by improving understanding
of processes. In this method, the process is broken down into different steps including inputs,
operational, and output steps. In this method, first, you need to define the process, design a
flowchart, define steps' capacity, explore the lowest capacity as the bottleneck of the process,
and quantify its impact on the process. Then, you can utilize the gathered data and findings
from the sections to enhance the process.
12. Link Analysis
This is a data analysis method that can be used to explore the links, connections, and
relationships between different objects. This analysis can be based on relational, physical, and
digital aspects. Thus, data collection in this method can be also based on digital data for
example from websites. This method is also utilized in networking. For this, the data that pass
the links are analyzed to explore the integrity of the links between each network node.
Generally, this method uses known violated patterns to discover anomalies, considers linked
objects to match the known patterns of interest, and finally discovers new patterns of interests.
13. Time and Motion Study
This method uses different techniques and standards to simplify the industrial processes,
improve employees' motivation, and gain better efficiency. For this purpose, you need to look
at your activities well and find the areas which can improve to gain higher efficiency. Then,
you need to make changes and see what the achieved results are. You can rinse and repeat the
former steps to gain better results. Therefore, the next step is to make small changes that can
bring important merits and consider what and how you conduct the process. Then, you should
think about broad items and spot the points that can be achieved in enhancements. Finally,
make the change and find out the results.
14. Statistical Method
These methods are a combination of processes including data collection, summarizing the
information, analyzing them, and reporting the findings as quantitative data. The data collection
step includes deciding what should be observed and making observations to address the
questions using samples that should represent the whole population and be adequate in number.
The summarization process includes measuring suitable statistics and displaying them in terms
of tables, graphs, etc. Then, using statistical analysis methods such as regression the gathered
27
statistical data can be analyzed and related. Different models can also determine to what extent
the gathered information can be utilized in similar projects (Kabir, 2016).
28
Location is a critical part of data collection considerations. If you want to conduct a reliable
process, you need to use a neural location. Because the participants need to feel free to provide
their responses. Otherwise, there is a high possibility of getting biased answers. For example,
when you interview the students in the school atmosphere, they may feel they can face
difficulties and troubles if they discuss the issue with their school staff.
2. Literacy of the Participants and the Language of the Questions
You need to ensure that the design of the questions is appropriate considering the literacy level
of the participants, or you need to select a target sample with a suitable level to respond to your
questions. Pilot tests must be conducted to understand whether these cases are considered
successful or not. Also, audio assistance methods can be used to address these kinds of issues.
3. Timing
As discussed in other sections, different criteria can impact the required time for the data
collection process. Here, it can be also added that a suitable schedule is necessary for research
projects as the following:
✓ First, an adequate time can help participants to be engaged with the project easier
and better without stress and other issues, and in contrast, their discomfort can bring
different challenges.
✓ On the other hand, long study periods can be also problematic as there are
possibilities in changing the circumstances for some participants if the length of study
is so long.
✓ To provide an approximate required time, pilot tests can be also useful.
4. Exhaustion of the Research Team
This factor is important as it can negatively affect the results. It can affect the ability of the
researchers to:
✓ Conduct the data collection processes efficient;
✓ Manage the interviews;
29
It is a vital point that participants should feel comfortable during interviews. They should not
be distracted or feel embarrassed responding to some sensitive topics for example about their
sexual experiences. In these situations, they can also refuse or hesitate to provide their personal
information such as their addresses. To address these challenges, you can start interviews with
icebreaking conversations, give them the rights of using alias names, and provide an
approximate address instead of their exact locations (Rimando et al., 2015).
ETHICAL COMMITMENTS
Ethical considerations should be noted during the data collection processes. It means, for
example, the confidentiality of data must be considered. Thus, where it is needed, the data
should be presented by an unknown respondent. In other words, there are different legal
problems that can stem from research studies. Thus, you need to ensure you consider the
necessary points to avoid them. The legal issues can be health risks happening for the volunteer
and participants, risks against the reputation of the organizations, and injuring the team. These
issues are affected by the laws of the country in which the study is conducted.
Ethical clearance processes are used to assure a project is designed based on the required ethical
rules. In this process, it must be examined what kinds of considerations are applied based on
the provided guidelines. It can be a simple or complex process for example:
• A process can be simple if there is no participant, and the project is not experimental;
• A process can be complex if you need to pay the participants or provide other kinds
of awards (Olsen, 2012).
There are different points that should be considered to conduct the data collection ethically.
First, if you need to get personal identifications of the participants like their telephone number,
postal address, and email address you should provide the reasons clearly and make them
consent about this. Secondly, you need to make it clear whether there is any possibility that
respondents can face a specific difficulty if they participate in the research or not. It means they
must be aware of it; otherwise, it can put them in jeopardy in some ways. Furthermore, an
identification method should be provided for the participants to be sure about who the
researchers are. In addition, it must be noted that, if participating in that specific research
project is voluntary, participants should feel free about answering the questions as it should not
be compulsory.
For the under 14 groups of participants, the consent of their parents should be achieved before
their participation (Kabir, 2016). Considering the above points, your provided statement for
the committees should cover the subjects such as the following:
30
● A summary of the processes; helps the committee to discover the positive
achievements of the project;
● If the data is provided for specific organizations for example universities, you need
to provide the commitment of giving them full access, and you will lodge the data in
them by considering the legal aspects for protecting data as well;
● An assessment document listing the risks which participants can face;
● A list of the considered activities can minimize the risks;
● The insurance benefits’ possibilities;
● A list providing the participants' activities during the research;
● A list of activities can ensure the confidentiality of the participants for example for
sensitive information;
● Finally, informed-consent forms ensure the participants are not forced to participate
until the end of the data collection process and they can withdraw when they want.
After providing the documents for the committee, other considerations should be
approved. An interview must be required to discuss the processes, methods, and
convince them that your research is useful and can provide positive results. The other
important point you need to convince them about is why putting participants in trouble
or danger is rational and to what extent it can be beneficial.
Furthermore, there are some objects that can cause data-protection issues, and to avoid them
some general points can be considered. For example, you should not hold information about
the participants in the appointment stage before approving that you have this right. You also
should not type the actual names on the questionnaires without their agreements. You also can
not hold their MP3 files in your archive without getting their consent. These processes are time-
consuming and can take months as discussed; however, there are specific circumstances that
this period can be even more. For example, if you conduct a data collection for an organization
and your co-researcher is from outside of this organization, the process can be longer, as:
● Differences in regulations can impact the risk factors.
● The other researcher is not conformed to the organization's rules and procedures, and
it takes time to address this situation.
To sum, ethics are very complex, and you need to investigate not only the processes but also a
variety of different factors which can overshadow them (Olsen, 2012).
31