Unit III Syllabus

Sampling Design: Questionnaire Design: Data Collection and

Fundamentals of Techniques Precautions Preparation
Sampling
P rimary and Secondary
P robability, Design, Non- Measurement and Scaling Sources of data; Data
probability and Sampling, Techniques: Types of Data; Tabulation, Editing and Coding.
Sample Size Determination, R ating Scale and R anking
R eliability and Validity. Scales .
 Questionnaire Design
A questionnaire is a research instrument that consists of a set
of questions or other types of prompts that aims to collect information
from a respondent.
A research questionnaire is typically a mix of close-ended
questions and open-ended questions.

Open-ended, long-form questions offer the respondent the

ability to elaborate on their thoughts. Research questionnaires were
developed in 1838 by the Statistical Society of London. The data
collected from a data collection questionnaire can be
both qualitative as well as quantitative in nature. A questionnaire may
or may not be delivered in the form of a survey, but a survey always
consists of a questionnaire.
Advantages of a Good Questionnaire
1 With a survey questionnaire, you can gather a lot of data in less time.

There is less chance of any bias(like selection bias) as the questionnaire will remain
2 standard for a group of respondents that fall in the same segment.

Quick and cost-effective. It offers you a rich set of features to design,

3
distribute, and analyze the response data.

The responses can be compared with the historical data and

4
understand the shift in respondents’ choices and experiences.

Respondents can answer the questionnaire without

5
revealing their identity.

6 It can be customized
 Characteristics of a Questionnaire
Your survey design depends on the type of information you need to collect from respondents. Qualitative questionnaires are used when
there is a need to collect exploratory information to help prove or disprove a hypothesis. Quantitative questionnaires are us ed to validate
or test a previously generated hypothesis. However, most questionnaires follow some essential characteristics:

UNIFORMITY EXPLORATORY QUESTION SEQUENCE

• A research form must have a uniform • It should be exploratory to collect • It typically follows a structured flow of
design and standardization. Every qualitative data. There is no questions to increase the number of
respondent sees the same questions. restriction on questions that can be in responses.
This helps in data collection your questionnaire. • This sequence of questions
and statistical analysis of this data. • For example, you use a data is screening questions, warm-up
• For example, the retail store collection questionnaire and send it to questions, transition questions, skip
evaluation questionnaire the female of the household to questions, challenging questions, and
template contains questions for understand her spending and saving
evaluating retail store experiences. classification questions.
habits relative to the household
Questions relate to purchase value, income. • For example, our motivation and
range of options for product buying experience questionnaire
• Open-ended questions give you more
selections, and quality of template covers initial demographic
merchandise. These questions are insight and allow the respondents to questions and then asks for time
uniform for all customers explain their practices. A very spent in sections of the store and the
structured question list could limit the rationale behind purchases.
data collection.
 Types of Questionnaires
Questionnaires can be either structured or free-flowing-

Structured Ques tionnaire Uns tructured Ques tionnaire

Structured questionnaires Unstructured questionnaires

collect quantitative data. The collect qualitative data. They use a
questionnaire is planned and basic structure and some
designed to gather precise branching questions but nothing
information. It also initiates a that limits the responses of a
formal inquiry, supplements data, respondent. The questions are
checks previously accumulated more open-ended to collect
data, and helps validate any prior specific data from participants.
hypothesis. A structured question For eg: Do you intend to buy a
may be multiple-choice, new car within the next six
dichotomous, or a scale. months?
 help increase responses to your research questionnaire as they tend to keep participants
You can use multiple question types in a questionnaire. Using various question types can
• help collect qualitative data in a questionnaire where the respondent
Open Ended Questions
Types of Questions in a Questionnaire
can answer in a free form with little to no restrictions.

• The dichotomous question is generally a “yes/no” close-ended question.

Dichotomous • This question is usually used in case of the need for necessary
validation. It is the most natural form of a questionnaire.

• are close-ended question type in which a respondent has to select one

more engaged

or many responses from a given list of options. The multiple-choice

Multiple Choice Questions question consists of an incomplete stem (question), right answer or
answers, incorrect answers, close alternatives, and distractors.

• These are based on the principles of the 4 measurement scales –

nominal, ordinal, interval, and ratio. Question types that utilize these
Scaling Questions are rank order questions, Likert scale questions, semantic differential
scale questions, and Stapel scale questions.

• This question type is easy to use and encourages respondents to answer.

• It works similarly to a multiple-choice question with answer choices are
Pictorial images. This helps respondents choose an answer quickly without over-
thinking their answers, giving you more accurate data.
 Types of Questions
Dichotomous Ques tions Multiple Choice Ques tions
– Only two response Do you intend to buy a new car within the
alternatives: yes or no, next six months?
agree or disagree, and so on.
____________Definitely will not buy
– Often, the two ____________Probably will not buy
alternatives of interest
are supplemented by a ____________Undecided
neutral alternative, such as ____________Probably will buy
“no opinion,” “don't know,”
“both,” or “none.” ____________Definitely will buy
____________Other (please specify)
– Do you intend to buy a new
car within the next six Scale
months? – Do you intend to buy a new car within
Yes the next six months?
No
Don't know Definitely Probably Undecided Probably Definitely will
Won’t buy Won’t buy will buy buy
1 2 3 4 5
 • Questionnaire is sent via email or • A phone call is made to a
Types of Questionnaire based on distribution other online mediums. Generally respondent to collect responses
cost-effective and time-efficient. directly. Responses are quick
Respondents can also answer at once you have a respondent on
leisure. Without the pressure to the phone. However, at times,
respond immediately, responses the respondents hesitate to give
may be more accurate. The out much information over the
disadvantage is that respondents Online Telephone phone. It is also an expensive
can easily ignore them. way of conducting research.

• Becoming obsolete but are still Mail In - House • Researcher visits the respondent’s
being used in some market home or workplace. The advantage
research. This method involves is that the respondent is in a
sending a physical data comfortable & natural environment,
collection questionnaire and in-depth data can be collected.
request to a respondent that The disadvantage, is that it is
can be filled in and sent back. expensive and slow .
 Technique of Questionnaire Designing
1. What is the information needed? Is the question necessary?
Planning what to Are several questions needed instead of one?
measure 2. Do respondents have the necessary information to answer?
Will respondents give you the necessary information?

Check for errors by pilot Formatting of a 1. Make a rough listing of questions

testing and correct the Pretest & Revise 2. Refine the question phrasing
problems Questionnaire 3. Develop Response Format

1. Neat and Clean ; Spacious but not

length; Proper numbering of questions Define the issue in terms of who, what,
must be done Sequencing & Questionnaire when, where, why, and way (the six Ws).
2. Divided into parts or sections for ease layout design Wording Who, what, when, and where are
of understanding and responding particularly important along with precautions
 Choosing Questionnaire Wording
Define the issue in terms of who, what, when, where, why, and way (the six Ws). Who, what, when, and where are
particularly important.
– Which brand of shampoo do you use? (Incorrect)
– Which brand or brands of shampoo have you personally used at home during the last month?
In case of more than one brand, please list all the brands that apply. (Correct)

Few Considerations
– Avoid using jargon, slang, or abbreviations
– Avoid ambiguity, confusion, and vagueness
– Avoid emotional language and prestige bias
– Avoid double-barreled questions
– Avoid leading questions
– Avoid asking difficult questions
– Avoid false premises
– Avoid asking about future intentions
– Avoid double negatives
– Avoid overlapping or unbalanced response categories
 Measurement
• It is the process of observing and recording the observations that are collected as part of research. The recording

may be in terms of numbers or other symbols to characteristics of objects according to certain prescribed rules.

The respondent’s, characteristics are feelings, attitudes, opinions etc.

• For example, you may assign ‘1’ for Male and ‘2’ for Female respondents. In response to a question on whether

he/she is using the ATM provided by a particular bank branch, the respondent may say ‘yes’ or ‘no’. You may wish

to assign the number ‘1’ for the response yes and ‘2’ for the response no.

• We assign numbers to these characteristics for two reasons. First, the numbers facilitate further statistical analysis

of data obtained. Second, numbers facilitate the communication of measurement rules and results.

• The most important aspect of measurement is the specification of rules for assigning numbers to characteristics.

The rules for assigning numbers should be standardized and applied uniformly. This must not change over time or

objects
 Scaling
• It is the assignment of objects to numbers or semantics according to a rule. In scaling, the objects are text

statements, usually statements of attitude, opinion, or feeling.

• For example, consider a scale locating customers of a bank according to the characteristic “agreement to the

satisfactory quality of service provided by the branch”. Each customer interviewed may respond with a semantic like

‘strongly agree’, or ‘somewhat agree’, or ‘somewhat disagree’ or ‘strongly disagree’.

• We may even assign each of the responses a number. For example, we may assign strongly agree as ‘1’, agree as

‘2’ disagree as ‘3’, and strongly disagree as ‘4’. Therefore, each of the respondents may assign 1, 2, 3 or 4.
Measurement Scales
The level of measurement refers to the
relationship among the values that are assigned to
the attributes, feelings or opinions for a variable.
For example, the variable ‘whether the taste of fast
food is good’ has a number of attributes, namely,
very good, good, neither good nor bad, bad and
very bad. For the purpose of analyzing the results
of this variable, we may assign the values 1, 2, 3, 4
and 5 to the five attributes respectively. The level
of measurement describes the relationship among
these five values. H
Measurement S cales
Data Type
 Characteristics or Goodness of Instruments /
Measurement Scales
A measurement scale has to have certain
desirable characteristic or criteria to judge its Accuracy & Precision Reliability

“goodness” so that one could have faith or

trust in the scale that it will measure what it is • The characteristics of accuracy in • Reliability indicates the confidence one
measurement scale means it should be could have in the measurement obtained
intended to measure. The following are the
a true representative of the observation with a scale. It tests how consistently a
main characteristics of measurement scales. of underlying characteristic. The measuring instrument measures a given
precision, however, means the power to characteristic/attitude is measured again
They are discriminate/distinguish and indicate the and again, leading to about the same
extent of accuracy that can be achieved conclusion. However, it may be
•Accuracy and Precision with the measurement scale. emphasized that reliability does not
•Reliability • Eg: The examination is conducted to necessarily imply that the measuring
measure the knowledge and instrument is also accurate. All it means
•Validity understanding of the student. The marks is consistency in drawing conclusion.
scored out of say 100, would provide
•Practicality better accuracy and precision than
simply grading the students A+, A, B+, B
and C
 Validity Practicability
• The extent to which an instrument scale tests or measures what • From theoretical viewpoint, a measure is ought to be reliable and
it is intended to measure. For example, if we intent to valid. However, from practical viewpoint, the measure should be
measure intelligence, the instrument, say question paper, ought
• Economical
to be such that it results in measuring true intelligence; if the
paper tests only general knowledge, the instrument is not valid. • Convenient
• Interpretable
• Types of validity
• Content validity- extent to which a measuring instrument
• The economic consideration leads to a comparison between
provides adequate coverage of the issues that are under study.
ideal research project and availability of budget for a study. Thus,
• Criterion validity - These are two types. One indicates the the measuring instrument has to take cognizance this aspect and
success of the measuring instrument used for predicting. The designed accordingly.
other, also called concurrent validity, is used to estimate the
present status. • The convenience implies the ease with which an instrument like
• Construct validity- It links psychometric notions and practices to questionnaire could be easily administered to the
theoretical notions. It attempts to explain the variation observed subjects/participants/respondents. In this one should give due
on several individuals. attention to the proper layout of the measuring instrument. This
• Eg: If a test of intelligence is conducted on individuals, and if poses more challenge in the situation whereas the concepts and
the test scores obtained by a measurement scale vary from constructs are rather difficult to understand.
individual to individual, one would like to know the factors to
construct behind this phenomenon. • The interpretability of an instrument, like questionnaire, is the
ease with which the researcher is able to interpret the responses
from the subjects/respondents/participants.
 Scaling Technique

Comparative Scale Non Comparative

(Ranking Scale) Scale (Rating Scale)

Graphic Rating Scale

Paired Comparison Constant Sum Scale (Continuous Rating Itemized Rating Scale
Scale)

Rank Order Q-Sort and others Likert

Semantic Differential

Staple
 Comparative Scale
Rank Order Scale Constant Sum Scale
Please rank the following detergents Divide 100 points among following
Tide Wheel Tide Wheel
Ariel Surf
Ariel _ Surf
Paired Comparison Scale Q-Sort
Which of the two brands you would
prefer? Cheap Good Easily Conve- Good
Tide | Wheel Wheel | Surf Qlty Available nient Smell
Tide | Surf Wheel | Ariel 1 2 3 4 5
Tide| Ariel Surf | Ariel
 Non – Comparative Scale

Continuous Rating Scale

Graphic Rating Scale How would you rate Tide?
Cheap II Costly
Scale Slider Cheap Costly
25 50 60
0 100

Pictorial Scale
Likert Scale
Semantic Differential
 Staple Scale
Surf

+5 +5 +5
+4 +4 +4
+3 +3 +3
+2 +2 +2
+1 +1 +1
Quality Price Availability
-1 -1 -1
-2 -2 -2
-3 -3 -3
-4 -4 -4
-5 -5 -5
 Data Collection & Preparation
Data collection is the process of collecting and evaluating information or data from multiple sources to find answers to research
problems, answer questions, evaluate outcomes, and forecast trends and probabilities. It is an essential phase in all types of research,
analysis, and decision-making, including that done in the social sciences, business, and healthcare.
Accurate data collection is necessary to make informed business decisions, ensure quality assurance, and keep research integrity.
During data collection, the researchers must identify the data types, the sources of data, and what methods are being used.

Before an analyst begins collecting data, they must answer three questions first:
•What’s the goal or purpose of this research?
•What kinds of data are they planning on gathering?
•What methods and procedures will be used to collect, store, and process the information?

Data preparation is the process of gathering, combining, structuring and organizing data for use in business intelligence,
analytics and data science applications. It's done in stages that include data preprocessing, profiling, cleansing,
transformation and validation. Data preparation often also involves pulling together data from both an organization's internal
systems and external sources.
IT, BI and data management teams do data preparation work as they integrate data sets to load into data warehouses, data
lakes or other repositories. They then refine the prepared data sets as needed when new analytics applications are
developed. In addition, data scientists, data engineers, data analysts and business users increasingly use self-service data
preparation tools to collect and prepare data themselves.
 Data Preparation Process
Preparing Statistically
Preliminary plan Coding adjusting the
of Data Analysis data

Questionnaire Selecting a data

Transcribing
Checking analysis strategy

Editing Data Cleaning

 Data Preparation Process
1. Preparing Preliminary Plan of Data Analysis:
Objective: Define the goals of the analysis and the hypotheses to be tested.
Steps:
Identify the key variables and data needed.
Choose the statistical techniques to be used (e.g., regression analysis, ANOVA).
Determine how results will be presented (tables, graphs, etc.).
Create a timeline for the analysis.
2. Questionnaire Checking:
Objective: Ensure that the questionnaire is clear, complete, and capable of collecting the necessary data.
Steps:
Review questions for clarity and relevance.
Check for logical flow and avoid leading or biased questions.
Pre-test the questionnaire with a small sample to identify issues.
Revise the questionnaire based on feedback.
3. Editing:
Objective: Correct errors and omissions in the collected data.
Steps:
Check for completeness (no missing answers).
Correct inconsistent responses.
Verify that all instructions were followed by respondents.
 Data Preparation Process
4. Data Cleaning:
Objective: Prepare the data set for analysis by handling errors and inconsistencies.
Steps:
Identify and address missing data (e.g., imputation or deletion).
Detect and correct errors (e.g., outliers, impossible values).
Ensure uniformity in data entry formats (e.g., date formats).
5. Transcribing:
Objective: Convert collected data into a digital format for analysis.
Steps:
Manually enter data into a software program or database if it was collected on paper.
Verify the accuracy of the transcribed data.
For qualitative data, convert audio or written responses into text.
6. Coding:
Objective: Assign numerical or categorical codes to qualitative data for analysis.
Steps:
Develop a coding scheme (codebook) that defines categories and their codes.
Apply codes to responses consistently.
Double-check coding for accuracy and consistency.
For Example : Education – Post Graduation : 1; Graduation: 2; High School:3
 Data Preparation Process
7. Statistically Adjusting the Data:
Objective: Modify the data to account for biases, outliers, and other anomalies.
Steps:
Use statistical techniques to handle outliers (e.g., winsorization).
Apply weighting to adjust for survey sampling biases.
Normalize or transform variables if necessary for analysis.
8. Selecting a Data Analysis Strategy:
Objective: Choose the most appropriate methods to analyze the data.
Steps:
Review the preliminary plan of analysis.
Select appropriate statistical tests based on data types and research questions.
Determine software and tools to be used (e.g., SPSS, R, Excel).
Plan for validation and verification of analysis results.
By following these steps, researchers can ensure their data is accurate, consistent, and ready for thorough analysis
 Data Collection Methods
Secondary Data Primary Data

• Secondary Data • Surveys and Experiments

• Newspapers • Interviews and Observation
• Television and Radio Studies

• Internet • Other Methods- Warranty

Cards, Distributor
• Libraries
audits, pantry audits,
• Word-of-Mouth
consumer panels, using
• Published Databases –
mechanical devices, through
World Bank Sites
projective techniques, depth
interviews, content analysis
Factors : Nature, Scope and Object of Enquiry ; Availability of Funds; Time Factor ; Precision Required
 Secondary Data
• Data that has already been collected for purposes other than the problem at hand
• Advantages

• Identify the problem

• Better define the problem

• Develop an approach to the problem

• Formulate an appropriate design

• Answer certain research question and test some hypothesis

• Interpret primary data more significantly

• Disadvantages
• Usefulness to current problem may be limited in several
• important ways, including relevance and accuracy
• The objectives, nature, and methods used to collected secondary
• data may not be appropriate to the present situation
 Primary Data
• It is originated by the researcher for the specific purpose of addressing the
problem at hand

• Can be qualitative or quantitative in nature

• Advantages

– It is pertinent to the problem at hand

– It is useful to get to a conclusive decision

• Disadvantages

– Costly to obtain

– Time required to collect primary data is very large

– Difficult to get true responses from the respondents

– Largely based on the perception of the researcher

 Editing
A process of examining the collected raw data to detect errors and omissions and to correct these
when possible
1. Which records are problematic?
2. Field Editing – At the time of recording the respondent’s responses
3. Central Editing - Correction of obvious errors in the office

Important Considerations for an editor-

1. Familiar with the instructions given to the interviewers and coders
2. Corrections made must be legible
3. Make entries in some distinct color and in a standardized form
4. Initial all answers which they change or supply
5. Initial the form with date of editing
• You may go through this link for more insights on data collection -
https://www.simplilearn.com/what-is-data-collection-article

• Few PDFs are attached in the folder for unit 3 study material.

Thank You !!