Questionnaire Designing

and
Sampling
 Research Questions Questionnaire Questions

Keep in mind the required data for the analysis

Combined with measurements and scales of variables.

Required personal information and customer classifications

Prepare a dummy table.

 Questionnaire
An academic study need to be conducted across various Financial
companies to assess the employee satisfaction regarding salary, perks,
overtime payment, and similar benefits. Additional incentives are
intended according to the additional or more work they carryout for
the office. A questionnaire has to be prepared which will have many
questions. Write questions for the questionnaire giving options with
appropriate scales.
 Questionnaire Preparation
1.    Write a study protocol
2.    Draw a plan of analysis
3.    Draw a list of the information needed
4.    Design different parts of the questionnaire
5.    Write the questions
6.    Decide on the order of the questions asked
7.    Complete the questionnaire
8.    Verify the content and style of the questions
9.    Conduct a pilot study
10. Refine your questionnaire
 In the beginning of a Questionnaire clearly mention the purpose of the
questionnaire.

 The spelt-out research objective must be converted into clear questions.

 This will extract answers from the respondents.

 The questionnaire should be designed to engage the respondent and

encourage a meaningful response.

 Questionnaire should be self explanatory and not confusing as then the

answers one get might not be accurate or usable for analysis.
 SLIDE 8-1

The questionnaire method

DR This is the simplest and most often used method of
primary data collection
D R D E E PA K C H AW L A

There is a pre-determined set of questions in a

sequential format
NEENA SONDHI

Is designed to suit the respondent’s understanding

and language command

Can be conducted to collect useful data from a large

population in a short duration of time

RESEARCH CONCEPTS AND

 ess
Specify the information needed

oc
Pr
gn
es i
Specify the type of interviewing method
eD
air
n
ion

Determine the content of individual questions

est
Qu

Design the questions to overcome the respondent’s

inability and unwillingness to answer

Decide on the question structure

 Determine the question wording

Arrange the question in proper order

Identify the form and layout

Reproduce the questionnaire

Eliminate bugs by pretesting

Principles of constructing good questionnaire

• Be clear and precise.

• Response choices should not overlap.

• Use natural and familiar language.

• Do not use words or phrases that show bias.

• Avoid double-barreled questions.

• State explicit alternatives.

• Questions should be reliable and valid.

Type of questions:

Open-ended questions : completed in the respondents’ own words.

For example : How would you evaluate the work done by the present
government?

What is your age?

My career goal is _________

Closed -ended questions : Both the question and response format are
structured and defined.

There are three types of these questions.

1. Dichotomous questions

2. Multiple choice questions

3. Scaled response
Dichotomous questions: Respondents will have to choose only with
two answers.

Multiple choice questions: Respondents will have a number of

response alternatives.

Mostly respondents have to choose one of the alternatives.

They made make multiple responses.

Scaled questions:

Likert scale - Respondents shows a specific amount of agreement

/disagreement.

Semantic differential - Scale is inscribed between two bipolar words

and respondent selects the point that represents her/his opinion.

A semantic differential scale measures attitudes towards something

Top speakers - - - - - - - Poor speakers

Unnecessary - - - - - - - Necessary
Scaled questions:

Ranked scale: A scale that gives rank among the available choices.

Rating scale – A scale that rates some attributes from poor to excellent.

Constant sum scale – The respondents are asked to distribute a given

number of points across various items.

Intention-to-buy scale :A scale describes the respondent’s intention to buy.

1- definitely buy, 2. probably buy, 3 not sure,

4 probably not buy, 5 definitely not buy.

Constant Sum Scale
Preferences regarding various pre-defined criteria
Questionnaire Pretesting : Administer questionnaire to three respondents

Understanding, meaning

Arriving at the response

Opinion on alternatives

Continuity and flow.

Key Steps in a Structured Questionnaire

Step 1: Plan how to carry out the study.

Step 2: Define and identify the target group for the study.

Step 3: Develop the questionnaire(s) and guides.

Step 4: Select interviewers and other field team personnel.

Step 5: Refine question phrasing.

Step 6: Develop response format.

Step 7: Put the questions into an appropriate format.

Step 8: Pretest and revise.

Step 9: Train field personnel and pilot test the questionnaire.

Step 10: Prepare and conduct the field work.

Step 11: Analyze and interpret the data.

Unstructured questionnaire
- sequence is not precisely determined in advance.

- Open ended questions, free response /free answer questions.

Structured questionnaire
- Specify the set of response alternatives and response format.

- the questions asked are precisely decided in advance.

- the questions are asked exactly as they are written, in the same sequence.

- It could be multiple choice, dichotomous, or a scale.

Semi‑structured questionnaire
- is a mix of unstructured and structured questionnaires.
An unstructured questionnaire is an instrument or guide used by an
interviewer who asks questions about a particular topic or issue. The specific
questions and the sequence in which they are asked are not precisely determined
in advance.

A structured questionnaire is one in which the questions asked are precisely

decided in advance. When used as an interviewing method, the questions are
asked exactly as they are written, in the same sequence, using the same style, for
all interviews. However, sometimes it can be left a bit open for the interviewer
to amend to suit a specific context.

A semi‑structured questionnaire is a mix of unstructured and structured

questionnaires. Some of the questions and their sequence are determined in
advance, while others evolve as the interview proceeds.
Schedules : The method of data collection is very much like the collection
of data through questionnaire, with little difference which lies in the fact that
schedules (proforma containing a set of questions) are being filled in by the
enumerators who are specially appointed for the purpose.

These enumerators along with schedules, go to respondents, put to them the

questions from the proforma in the order the questions are listed and record
the replies in the space meant for the same in the proforma.
 Sampling
Population:

• Any group of objects or people that forms the subject of study in a

particular survey,

• this is the total of the items or units in any field of inquiry,

• refers to the total of items about which information is desired.

• The attributes that are the object of study are referred to as characteristics
and the units possessing them are called as elementary units. The
aggregate of such units is generally described as population.
• The population can be finite or infinite.

• The population is said to be finite if it consists of a fixed number of elements.

• For instance, the population of a city, the number of workers in a factory are
examples of finite populations.

• An infinite population is that population in which it is theoretically

impossible to observe all the elements.

• Occurrence of a species in a high-moisture soil environment is an example of

an infinite population.
Examples of finite population:

Employees of a company.

Customers who purchased cars of a company

All MBA students studying in Karnataka

Examples of infinite population

• All possible buyers (including in future) of a product.

• All possible tosses of coin that could be performed.

• Infinite population as an approximation of a very large finite population -

The number of germs in the body of a sick patient.
Sample: It is a subset of the population.

Sampling unit: A single member of the sample.

Sampling : A process of selecting an adequate number of elements

from the population.

Census: An examination of each and every element of the population.

It is complete enumeration of the elements of a population.
Element: An element comprises a single member of the population.

Sampling frame: Comprises all the elements of the population with

proper identification that are available to us for selection at any stage of
the sampling.

• It is a complete list of everyone or everything you want to study.

• A list containing all possible sampling units is known as sampling

frame.
Sampling design: A sample design is a definite plan for obtaining a
sample from the sampling frame. It refers to the technique or the
procedure the researcher would adopt in selecting some sampling units
from which inferences about the population is drawn. Sampling design
is determined before any data are collected. These are also called
sampling methods or sampling techniques.

Discuss Examples
 Sampling Design Process

Define the target population

Determine the sampling frame

Select the sampling techniques(s)

Determine the sample size

Execute the sampling Process

Errors: In order to find certain population characteristics we draw a
sample and estimate. These quantities differ from the actual. That is called
error.

There are two types of errors in sampling that is sampling error and non-
sampling error.
Sampling errors: The discrepancy between the sample data and true
population values that are attributed to random difference between the
sample and the population measures of interest.

Sampling errors are the random variations in the sample estimates around
its counterpart in the population.

Sampling error decreases with the increase in the size of the sample, and it
happens to be of a smaller magnitude in case of homogeneous population.

Sampling error depends on the sampling design as well.

Sampling error is the error caused by observing a sample instead of the
whole population. The sampling error is the difference between
a sample statistic used to estimate a population parameter and the actual
but unknown value of the parameter, provided no error occurred during
transferring and tabulating the data.
Non-sampling errors: This is the error arises due to the factors other than
that of sampling error while drawing conclusion based on the sample.

The error can arise while transferring the data from the questionnaire to the
spreadsheet on the computer.

There can be errors at the time of coding, tabulation and computation.

Population of the study is not properly defined.

Respondent may refuse to be part of the study.

There may be a sampling frame error (nonsampling error)

Plain lying by the respondent (nonsampling error)

Incomplete coverage of the population or sample (nonsampling error)

Defective methods of data collection (nonsampling error)

 Total Survey Error

Sampling Error Nonsampling Error

Nonresponse Error Measurement Error

Systematic Random
CHARACTERISTICS OF A GOOD SAMPLE DESIGN

(a) Sample design must result in a truly representative sample.

(b) Sample design must result in a small sampling error.

(c) Sample design must be viable in the context of funds available for the
research study.

(d) Sample design must be such so that systematic bias can be controlled in a
better way.

(e) Sample should be such that the results of the sample study can be applied,
in general, for the population with a reasonable level of confidence.
Probability sampling: Probability sampling is also known as ‘random
sampling’ or ‘chance sampling’.

Under this sampling design, every item of the population has an equal
chance of inclusion in the sample.

So to say, a lottery method in which individual units are picked up from

the whole group not deliberately but by some mechanical process.
 Random sampling is also called as probability sampling.

 Observations are obtained based on random numbers or by lottery method.

 Non-random sampling units need not satisfy the equal probability structure of inclusion in the
sample.

 In probability sampling or random sampling there will be a list of to be sampled units whereas
in non random sampling there is no such list is prepared and they can be selected according to
certain rules of investigator.

 The random samples will have less sampling error, samples are unbiased representatives of the
population.

 Statistical analysis is meaningful when the sampling is random. Many statistical analyses can
be applied. Satisfies many required statistical properties.
Refer a Random Number Tables
 SLIDE 9-4
Sampling Design
Probability Sampling Design - Probability sampling designs are used in
DR

conclusive research. In a probability sampling design, each and every element

of the population has a known chance of being selected in the sample.
D R D E E PA K C H AW L A

Types of Probability Sampling Design

Simple random sampling with replacement
NEENA SONDHI

Simple random sampling without replacement

Systematic sampling
Stratified random sampling
Cluster sampling

RESEARCH CONCEPTS AND

 A simple random sample is a sample selected in such a way that every
possible sample of the same size is equally likely to be chosen.

Simple Random Sampling With Replacement (SRSWR)

Simple Random Sampling Without Replacement (SRSWOR)

• The example in which the names of 25 employees out of 250 are chosen out
of a hat is an example of the lottery method at work. Each of the 250
employees would be assigned a number between 1 and 250, after which 25 of
those numbers would be chosen at random.
• In SRSWR, a unit once selected will have a chance that it is included
in the sample again

• In SRSWOR, a unit once selected will not be included in the sample

again
SRS: Steps

1. A list of all the members of the population is prepared initially and then
each member is marked with a specific number (Sampling frame).

2. Figure out what your sample size is going to be.

3. From this population, random samples are chosen using one of the two
ways: random number tables or random number generator software.

4. Select only corresponding members selected in step 3 as your sample and

collect information or measurements from them.
Exercise for you

List advantage and disadvantage of simple random sampling

 Stratified random sampling
Stratified random sampling is a method of sampling that involves the division of a
population into smaller groups known as strata. In stratified random sampling 
stratification or the strata are formed based on members' shared attributes or
characteristics. Samples from each stratum is selected randomly.
Let’s consider a situation where a research team is seeking opinions about purchase
of a mobile phones in Karnataka, amongst various age groups. Instead of collecting
feedback from about six crore citizens, random samples of around 1000 can be
selected for research, from a list of citizens of Karnataka aged 10 or more. These six
crore citizens can be divided into strata according to age, i.e, groups of 10-16, 17-
24, 25-34, 35-49, 50-65 and 66 and above. Each stratum will have distinct members
and number of members. (Question: Why such a stratification?)
Continued…

Another example:

An agency wants to find out and analyse the mistakes in filing income tax
returns and they are likely to get notices from IT department. The sampling
frame can be obtained from the list of all tax payers available at IT
department. These members can be divided according to the type of IT
form submitted. A sample can be obtained from each stratum.
Simple Random Versus Stratified Random Sample

• A simple random sample is drawn uniformly from the entire data

population. A stratified random sample divides the population into
smaller groups, or strata, based on shared characteristics.

• Unlike simple random samples, stratified random samples are used with
populations that can be easily broken into different subgroups or subsets.
These groups are based on certain criteria. Randomly choose elements
from each in certain proportion. For instance, the sample size versus the
population.
 Elements are selected from each stratum by a random procedure, usually
SRSWOR. That is probability sampling within each stratum.

 Stratified sampling increases precision without increasing cost.

 The elements within a stratum should be as homogeneous as possible, but

the elements in different strata should be as heterogeneous as possible.

 The stratification variable is closely related to the characteristic of

interest.
• Variables commonly used for stratification include

- demographic characteristics, type of customer, size of firm, or type

of industry.

• We can use more than one variable for stratification.

• The number of strata is matter of judgement.

An example for stratified sampling: A company is considering an on-site
kindergarten facility. But before taking further steps, it wants to get the reactions of
four groups to the idea: (1) Employees who are parents of kindergarten-age children,
and where both are working outside of the home, (2) employees who are parents of
kindergarten-age children, but one of them is not working outside of the home, (3)
single parents of kindergarten-age children, and (4) all those without children of
kindergarten-age. If the four groups are expected to represent 60%, 7%, 23% and
10% respectively, in the population of 480 employees in the company, then a
stratified sampling with proportional allocation will be appropriate to represent the
four groups with total sample size of 60.
Systematic Sampling

To draw a sample of size n ,

- select a random number between 1 and k, where k is approximately N/n.

- Suppose it is i .

- Select the first unit whose serial number is i .

- Select every kth unit after ith unit.

- Sample will contain i, i +k, i+2k, i+(n-1) k , serial number units.

A cluster sample is a simple random sample of groups or clusters of
elements.

The population first divided into mutually exclusive sets and

collectively exhaustive subpopulations, or clusters. Then a random
sample of clusters is selected.

In one-stage cluster sampling all the elements in each of the selected cluster
are included in the sample.

In two-stage cluster sampling a sample from each of the selected cluster

are drawn probabilistically.
A common form of cluster sampling is area sampling, in which the
clusters consist of geographic area, such as countries, housing tracts, or
blocks.
Difference between cluster sampling and stratified sampling

1. Clusters are formed based on proximity of the observations within

clusters, whereas strata are formed based on homogeneity of the
observations.

2. In stratified sampling observations from all the strata are taken but in
cluster sampling observations from only a few (selected) clusters are
taken.

3. In stratified sampling only a sample from a strata is taken not all units are
observed but in cluster sampling all the observations from a cluster are
taken.
 Non-probability Sampling

Convenience Sampling: The only criteria for selecting sampling units in this
scheme is the convenience of the researcher.

For example: People interviewed in shopping mall,

Interviews conducted by TV channel

 Easy to collect
 Self selected
 Ease of availability
 Not a representative sample
 Prone to biased sample
 Non-probability Sampling

Snowball Sampling: This is generally used when it is difficult to

identify the members of the desired population. Under this design, each
respondent, after being interviewed, is asked to identify one or more in
the field. This will result in a sample.

The first element is selected preferably at random. The subsequent

elements are selected based on references. This group possess similar
characteristics and who in turn, identify others. The sample will have
demographic and psychographic characteristics.
 Non-probability Sampling

Snowball Sampling is used in industrial buyer-seller research to

identify buyer-seller pairs.

These methods are used to study drug cultures, teenage gang activities,
community relations, insider trading, and other applications where
respondents are difficult to identify and contact.
 Non-probability Sampling

Purposive Sampling is a non-representative subset of some population,

and is constructed to serve a very specific need or purpose. Two major
types – judgement sampling and quota sampling.

Judgement Sampling: Under this sampling experts in this filed choose

what they believe to be the best sample for the study of the question.

The researcher exercising judgement, or expertise, choose the elements

to be included in the sample.
 Non-probability Sampling

Examples: Test markets selected to determine the potential of a new product.

Purchase engineers selected in industrial marketing research because they

are considered to be representative of the company.

Expert witnesses used in court, and

Department stores selected to test a new merchandising display system.

Business to business marketing.

 Non-probability Sampling

Quota Sampling: The sample includes a minimum number from each

specified subgroup in the population. The sample is selected on the basis of
certain demographic characteristics such as age, gender, occupation, education,
income, etc.

It is viewed as two stage sampling. The first stage consists of developing

control categories, or quotas, of population elements. The control
characteristics such as age, gender, occupation, education, income, etc.

Idea is that the quotas ensure the composition of the sample is the same as the
composition of the population with respect to the characterises of the interest.
 Non-probability Sampling

Suppose the student body in a campus of a big educational institute

consists of 55 percent female and 45 percent male wants to survey to find
out commuters in bus. The sampling quota will call for sampling students
at a 55:45 percent ratio. This would eliminate distortions due to
nonrepresentative gender ratio.
 Classification of Sampling Techniques
Sampling Techniques

Probability Nonprobability
Sampling Techniques Sampling Techniques

Simple random Systematic Stratified Cluster Other sampling

Sampling Sampling Sampling Sampling Techniques

Convenience Snowball Judgmental Quota

Sampling Samplin Sampling Sampling
g
 Multi-stage sampling can be used

That is, first sampling is applied to subpopulations or

subgroups and a sampling design is used to each of these
selected sub-population.
Examples:

1. In order to carryout quantitative research to measure the relationship between

job characteristics and job satisfaction in all Nationalised banks in Karnataka.

2. Consumption of packed sugarcane juice in a city.

3. A study has to be carried out to risk behaviour of drug users infected with HIV.

4. A Microsoft-owned Web site gives vehicle information to consumers. It

conduced a poll to find out if people currently use or would consider using cell
phone hands-free devices. A sample from the visitors of the Web site. Which
sampling procedure could be used?
Probability sampling is also known as random sampling and

non-probability sampling is also known as non-random sampling

Advantage of probability sampling over non-probability sampling

• Under probability sampling the subjects of the population get an equal

opportunity to be selected as a representative sample. In non-
probability sampling it is not known which individual from the
population will be selected as a sample observation.

• This process saves cost and time to a great extent. We can take any
required number of samples from this process.

• Probability sampling is more effective and more accurate.

• Almost all statistical analysis assumes independence which is satisfied by
probability sampling observations, whereas non-probability sampling
observations are not independent.

• Probability sampling does not involve any complex and long process. Thus
this is an easier way for sampling.

• Probability sampling uses random numbers which ensures that the samples
vary as much as the population itself.
Source of Data:Primary data and Secondary data.
 Qualitative Methods of Data Collection
Observation Method

• Observation means the careful and systematic watching of facts as they

occur in course of nature. In the strict sense, observation implies more use of
the eyes and the ears than the mouth.
• Observation involves three processes, i.e. (i) sensation (ii) attention
(iii) perception
• Observational research typically provides qualitative data as the researcher
is observing the subject in their natural setting. The output of Observational
research, is sometimes followed with a quantitative survey to support certain
behaviors/observations, correlate and derive more meaningful insights.
Content Analysis
Content analysis is a research technique used to make replicable and valid
inferences by interpreting and coding textual material. By systematically
evaluating texts (e.g., documents, oral communication, and graphics),
qualitative data can be converted into quantitative data.
Content analysis is a method for summarizing any form of content by counting
various aspects of the content. This enables a more objective evaluation than
comparing content based on the impressions of a listener. For example, an
impressionistic summary of a TV program, is not content analysis.
 Qualitative Methods of Data Collection

• Personal Interviews.

• Telephone Method
Computer assisted, and Mechanical Observation

• Mail interviews

• Electronic Methods

E-mails, Internet interviews

 Further on Data Collection Methods

• Questionnaires and Surveys.

• Oral History and Case Studies.

• Documents and Records.

 Focus Groups
• Focus group is a panel of people led by a trained moderator. They will meet for
about two hours (may be for more duration). They facilitate dynamics principles
to focus or guide the group in an exchange of ideas, feelings, and experience on
specific topic.
• It is constituted by different experts who have knowledge in the area. Different
persons are chosen with different expertise, with a maximum number of 12
members.
• Their objective may be new product or product concept, a new employee
motivation program, or improve production-line organisation
• They obtain general background about the topic or issue
• They generate research questions that are to be explored

• They interpret previously obtained quantitative results

• They stimulate new ideas for products, sales and programs

• They highlight area of opportunities for specific managers to pursue

• They diagnose problems that managers need to address

• They generate impressions and perceptions of brands and product ideas.