71  Data Collection and Questionnaire Design

71.1 Concept

Data Collection = the process of gathering and measuring information on variables of interest in a systematic, established manner — enables answering of research questions and testing hypotheses. Donald Cooper & Pamela Schindler (Business Research Methods), Naresh Malhotra, Earl Babbie, C.R. Kothari are the leading textbook authors.

71.2 Types of Data

TipPrimary vs Secondary

Primary data — collected first-hand by researcher for specific purpose. - Methods: Survey · Observation · Experiment · Interview · Focus group.

Secondary data — already-collected data used for new research. - Sources: govt publications, industry reports, databases, internet.

71.3 Data-Collection Methods

TipMajor data-collection methods

Method	Description
Survey / Questionnaire	Structured questions
Personal Interview	One-to-one
Telephone Interview / CATI	Computer-Aided Telephone Interview
Mail / Postal Survey	Self-administered
Online Survey	Web-based — Google Forms, SurveyMonkey, Qualtrics
Mobile Survey	SMS, WhatsApp, app
Observation	Direct or participant
Experiment	Lab or field
Focus Group	6-10 participants discussion
Depth Interview	Long, qualitative
Projective Techniques	Indirect (word association, TAT, Rorschach)
Ethnography	Immersive observation
Content Analysis	Documents, social media
Big Data / Web Scraping	Digital footprints

71.4 Survey Errors

TipCooper-Schindler — survey error types

• Sampling errors — random.
• Non-sampling errors:
  • Response bias — social desirability, acquiescence.
  • Non-response bias.
  • Interviewer bias.
  • Measurement / instrument bias.
  • Processing errors.

71.5 Questionnaire Design

A questionnaire is a formalised set of questions to obtain information from respondents. Naresh Malhotra’s 10-step questionnaire design process is widely taught.

71.5.1 Malhotra’s 10-step Questionnaire Design

TipQuestionnaire design — 10 steps (Malhotra)

1. Specify information needed.
2. Type of interview method.
3. Determine content of individual questions.
4. Design questions to overcome respondent’s inability and unwillingness to answer.
5. Decide question structure.
6. Determine question wording.
7. Arrange questions in proper order.
8. Identify form and layout.
9. Reproduce the questionnaire.
10. Pretest, revise and prepare final.

71.6 Types of Questions

TipQuestion types

Type	Description
Open-ended	Free response
Close-ended	Predefined options
Dichotomous	Two options (Yes/No)
Multiple choice	Several options
Rating scale	Likert, Semantic Differential
Ranking	Order preferences
Filter / Screener	Routing
Contingency	Conditional

71.7 Scaling Techniques

TipMajor scaling techniques

Scale	Description	Inventor
Likert	5/7-point agreement (Strongly Disagree → Strongly Agree)	Rensis Likert (1932)
Semantic Differential	Bipolar adjectives (Good-Bad)	Charles Osgood (1957)
Thurstone	Equal-appearing intervals	L.L. Thurstone (1928)
Guttman	Cumulative scale	Louis Guttman (1944)
Stapel Scale	+5 to −5 single adjective	Jan Stapel
Bogardus Social Distance	Acceptance closeness	Emory Bogardus (1925)
Q-Sort	Forced-choice ranking	William Stephenson (1953)
Constant Sum	Distribute fixed points
Paired Comparison	Pairs ranked

71.8 Validity and Reliability

TipValidity

• Content validity — coverage.
• Construct validity — Convergent + Discriminant.
• Criterion validity — Concurrent + Predictive.
• Face validity — appears valid.

TipReliability

• Test-retest — consistency over time.
• Internal consistency — Cronbach’s Alpha (1951) — > 0.7 acceptable.
• Parallel forms.
• Inter-rater reliability — Cohen’s Kappa.
• Split-half.

71.9 Pretest and Pilot Study

Run questionnaire with small sample (15-30) before main study. Identifies — confusing questions, time required, response patterns.

71.10 Errors to Avoid in Questions

TipQuestion wording errors

• Leading questions — bias response.
• Loaded questions — emotional charge.
• Double-barrelled questions — two ideas in one.
• Ambiguous questions — vague terms.
• Negative wording.
• Jargon and complex vocabulary.
• Assumptive questions.
• Generalisation questions.

71.11 Indian Research Agencies

TipIndian market research agencies

• IMRB / Kantar India.
• Nielsen India.
• AC Nielsen ORG-MARG (now Nielsen).
• GfK India.
• Hansa Research.
• Ipsos India.
• TNS India.
• C-Voter (political).
• CMIE (economic data).
• Crisil Research.

71.12 Modern Trends

TipModern data-collection trends

• Online and mobile surveys dominant.
• Big data and web scraping.
• Social media listening.
• Real-time / Always-on panels.
• Eye-tracking and biometrics.
• Neuro-marketing tools.
• AI-generated personas and synthetic data.
• Conversational surveys (chatbots).
• Behavioural analytics.
• Geo-location surveys.
• Privacy-first / consent management.
• Mobile ethnography.

71.13 Practice Questions

Q 01LikertEasy

The Likert scale (1932) was developed by:

• ARensis Likert
• BCharles Osgood
• CThurstone
• DGuttman

View solution

Correct Option: A

Rensis Likert (1932).

Q 02SemanticMedium

Semantic Differential scale uses:

• ABipolar adjective pairs
• B5-point agreement
• CForced ranking
• DYes/No

View solution

Correct Option: A

Osgood (1957) — Good-Bad, Hot-Cold etc.

Q 03CronbachMedium

Cronbach's Alpha (1951) measures:

• AInternal-consistency reliability
• BValidity
• CSkewness
• DStatistical significance

View solution

Correct Option: A

Acceptable α ≥ 0.7.

Q 04BogardusHard

Social Distance scale (1925) is by:

• AEmory Bogardus
• BLikert
• CThurstone
• DGuttman

View solution

Correct Option: A

Bogardus 1925.

Q 05PretestEasy

Pretest is conducted to:

• AIdentify problems in questionnaire
• BInflate sample size
• CAvoid sampling
• DForecast results

View solution

Correct Option: A

Catch confusing questions, timing issues.

Q 06Double-barrelMedium

"Do you find the website fast and easy to use?" is an example of:

• ALeading question
• BDouble-barrelled question
• CLoaded question
• DAmbiguous question

View solution

Correct Option: B

Two attributes (fast + easy) in one question.

Q 07Q-SortHard

Q-Sort technique is by:

• AWilliam Stephenson
• BLikert
• COsgood
• DGuttman

View solution

Correct Option: A

William Stephenson (1953).

Q 08Open-endedEasy

"What did you like about our product?" is:

• AOpen-ended
• BClose-ended
• CDichotomous
• DMultiple choice

View solution

Correct Option: A

Free response.

Q 09Focus groupMedium

Optimal size of a focus group is:

• A2-3
• B6-10
• C20-30
• D50+

View solution

Correct Option: B

6-10 participants is the norm.

Q 10Secondary dataEasy

Census of India data is:

• APrimary data
• BSecondary data
• CSample data
• DExperimental data

View solution

Correct Option: B

Pre-collected; used by other researchers → secondary.

Q 11ValidityMedium

"Does the scale measure what it claims to measure?" refers to:

• AReliability
• BValidity
• CSensitivity
• DStability

View solution

Correct Option: B

Validity = accuracy of measurement.

Q 12ProjectiveHard

Word association and TAT are examples of:

• AProjective techniques
• BLikert scaling
• CStratified sampling
• DQuantitative methods

View solution

Correct Option: A

Indirect / qualitative.

Q 13ThurstoneHard

Equal-appearing intervals scale is by:

• AL.L. Thurstone
• BLikert
• CStapel
• DBogardus

View solution

Correct Option: A

L.L. Thurstone (1928).

Q 14CATIHard

CATI stands for:

• AComputer-Assisted Telephone Interview
• BCumulative Attitude Test Index
• CComputer Algorithm for Test Interpretation
• DCentralised Audit Tool

View solution

Correct Option: A

Phone survey software.

Q 15Match scalesHard

Match:

(i)	Likert	(a)	Bipolar
(ii)	Semantic Differential	(b)	Cumulative
(iii)	Guttman	(c)	Agreement
(iv)	Bogardus	(d)	Social distance

• A(i)-(c), (ii)-(a), (iii)-(b), (iv)-(d)
• B(i)-(a), (ii)-(b), (iii)-(c), (iv)-(d)
• C(i)-(b), (ii)-(c), (iii)-(d), (iv)-(a)
• D(i)-(d), (ii)-(c), (iii)-(a), (iv)-(b)

View solution

Correct Option: A

Likert — Agreement; Semantic — Bipolar; Guttman — Cumulative; Bogardus — Social Distance.

71.13.1 Advanced Format Questions

AR 1Assertion-ReasonHard

A: Likert is summated rating.
R: It uses bipolar adjective pairs.

• ABoth true; R explains A
• BBoth true; R does not explain A
• CA true, R false
• DA false, R true

View solution

Correct Option: C

Bipolar pairs are used in Semantic Differential, not Likert.

S 1Statement-basedMedium

Reliability measures: (i) Test-retest. (ii) Internal consistency. (iii) Parallel forms. (iv) Inter-rater.

• AAll four
• B(i) and (ii) only
• C(iii) and (iv) only
• D(ii) only

View solution

Correct Option: A

S 2Statement-basedHard

Questionnaire pitfalls: (i) Leading. (ii) Double-barrelled. (iii) Ambiguous. (iv) Loaded.

• AAll four
• B(i) and (ii) only
• C(iii) and (iv) only
• D(i), (ii), (iii) only

View solution

Correct Option: A

71.14 Quick Recall

ImportantQuick recall

• Data: Primary vs Secondary.
• Methods: Survey · Interview · Observation · Experiment · Focus Group · Projective · Ethnography · Web scraping.
• Errors: Sampling vs Non-sampling (response, non-response, interviewer, measurement, processing).
• Malhotra’s 10-step questionnaire process.
• Question types: Open · Close · Dichotomous · MCQ · Rating · Ranking · Filter · Contingency.
• Scales: Likert (1932) · Semantic Differential (Osgood 1957) · Thurstone (1928) · Guttman (1944) · Stapel · Bogardus Social Distance (1925) · Q-Sort (Stephenson 1953) · Constant Sum · Paired Comparison.
• Validity: Content · Construct (Convergent/Discriminant) · Criterion (Concurrent/Predictive) · Face.
• Reliability: Test-retest · Cronbach’s α (1951) ≥ 0.7 · Parallel · Inter-rater (Cohen’s κ) · Split-half.
• Question errors: Leading · Loaded · Double-barrelled · Ambiguous · Negative · Jargon · Assumptive.
• Indian agencies: IMRB/Kantar · Nielsen · Ipsos · CMIE · Crisil · C-Voter · Hansa · GfK.
• Modern trends: online/mobile · big data · social listening · biometrics · neuro · AI/synthetic data · conversational · geo-location · privacy-first.