Exam:
Notes Survey Research Methods
Session 1: Introduction to Survey Research
WHAT IS SURVEY RESEARCH?
A survey is a systematic method for gathering information using
standardized questionnaires from a sample of entities to
construct quantitative descriptors of a larger population (Groves
et al., 2009).
Key Characteristics:
o Systematic & Standardized → Differs from qualitative
research.
o Sample of Entities → Often a selection of the population is
surveyed.
o Quantitative Descriptors → Requires statistical analysis.
Example: A survey of 1,000 customers to assess satisfaction with a new
airline service.
A SHORT HISTORY OF SURVEY RESEARCH
Ancient Egypt: Early census surveys.
1930-40s: Paul Lazarsfeld studied how radio influenced political
opinions.
Today: Used in business for customer insights and experimental
validation.
Example: Political polling before elections.
SURVEY VS. OTHER RESEARCH METHODS
Method Data Type Control Use Case
Level
Survey Self-reported Low Understanding behaviors,
Research opinions, or preferences.
Experimental Controlled High Testing causal
Research conditions relationships (A/B testing,
lab studies).
Modeling Existing data Moderate Identifying trends using
(Secondary large datasets.
Data)
Example: A company uses surveys to track brand perception over time,
while experiments test ad effectiveness.
TYPES OF SURVEY SAMPLING
Probability Sampling (random selection, results are generalizable)
o Simple Random Sampling: Every unit has an equal chance.
,Exam:
o Systematic Sampling: Every nth person is chosen.
o Stratified Sampling: Population divided into groups, then
sampled.
o Cluster Sampling: Entire groups (clusters) are randomly
selected.
Non-Probability Sampling (subjective selection, not generalizable)
o Convenience Sampling: Easily available participants.
o Quota Sampling: Targets specific groups (e.g., 50% women,
50% men).
o Snowball Sampling: Existing respondents recruit others.
o Purposive Sampling: Chosen based on specific traits.
Example: Using stratified sampling to ensure representation from
different age groups in a survey about social media use.
SURVEY TIME HORIZONS
Cross-Sectional Surveys: One-time snapshot of data.
o Example: A single survey measuring brand awareness in
2024.
Longitudinal Surveys: Track changes over time.
o Repeated cross-sectional: Different sample each time.
o Fixed-sample panel: Same sample surveyed multiple times.
o Cohort study: Follows a specific group over years (e.g.,
tracking 2010 marketing graduates).
SURVEY ADMINISTRATION METHODS
Personal Surveys (Face-to-face interviews)
Telephone Surveys (Often used for customer service feedback)
Online Surveys (Popular due to cost-effectiveness)
Written Surveys (Mail-in or paper-based)
Example: A university uses online surveys to evaluate student
satisfaction.
PROS & CHALLENGES OF SURVEYS
Pros
o Ease: Quick data collection from large groups.
o Simplicity: Easy to analyze.
o Reliability: Standardized format reduces variation.
Challenges
o Asking the right questions is difficult.
o Wording questions properly to avoid bias.
o Getting the right sample for representativeness.
o Motivating respondents to answer truthfully.
Example: A poorly worded survey question might lead to biased
responses:
❌ "Do you agree that electric cars are better for the environment?"
,Exam:
✅ "On a scale from 1-5, how environmentally friendly do you consider
electric cars?"
THE ROLE OF SURVEYS IN RESEARCH
Exploratory research: Identify key factors, generate hypotheses.
Descriptive research: Understand behaviours, measure variables.
Explanatory research: Test causal relationships.
Example: A company first explores why customers cancel subscriptions
(exploratory), then tracks dissatisfaction trends (descriptive), and finally
tests new service changes to reduce cancellations (explanatory).
WHAT IS SURVEY DATA GOOD FOR?
Correlational Insights: Analyzing relationships between variables.
o Example: Salary vs. job satisfaction.
Causal Insights: Requires additional manipulation (experiments).
o Example: Testing a new pricing model’s effect on sales.
Extending Data: Combining surveys with secondary data.
o Example: Merging survey responses with purchase history.
SURVEY RESEARCH PROCESS
Planning a Survey
o Define research objectives.
o Select research variables.
Designing a Survey
o Conceptualization (defining constructs).
o Operationalization (choosing how to measure variables).
o Construct and pretest the questionnaire.
Implementing a Survey
o Choose relevant population.
o Select sampling method.
o Conduct data collection.
Gaining Insights
o Data preparation & measurement evaluation.
o Data analysis (e.g., regression, correlations).
o Reporting results.
Example: A fitness brand surveys customers to find factors influencing
gym membership renewal.
SURVEY QUESTION TYPES
Open-ended: Respondents give detailed answers.
o Example: “What do you like most about our product?”
Closed-ended:
o Dichotomous (Yes/No).
o Multiple choice (Choose from options).
o Rating scales (Likert, Semantic Differential).
, Exam:
Example:
❌ "Do you use public transport?" (Too vague)
✅ "How often do you use public transport?"
a. Daily
b. Weekly
c. Monthly
d. Never
SAMPLING & DATA COLLECTION
Defining population (Who are we surveying?).
Sampling frame (List of potential respondents).
Choosing sampling method (Probability vs. Non-probability).
Determining sample size (Larger = more accurate results).
Executing data collection (Ensuring responses are valid).
STRUCTURAL EQUATION MODELING (SEM)
Analyzes relationships between multiple variables.
Used in marketing, psychology, business research.
Example: Examining how customer satisfaction influences loyalty.
INTERNATIONAL SURVEY RESEARCH
Consider cultural differences in interpretation.
Address response biases (e.g., tendency to agree with
statements).
Ensure measurement equivalence across countries.
Example: A survey on work-life balance might have different
interpretations in Japan vs. the Netherlands.
EXPERIMENTAL SURVEY RESEARCH
Controlled experiments where one variable is manipulated.
Goal: Establish causation, not just correlation.
Example:
Survey: “Do you prefer blue or red ads?”
Experiment: Expose one group to blue ads, another to red, then
measure engagement.
KEY TAKEAWAYS
Surveys provide structured, standardized data for analysis.
Choosing the right sampling method ensures valid results.
Good question design prevents bias and improves response quality.
Surveys can be combined with experiments for deeper insights.
EXAMPLES SURVEY QUESTION: WHY ARE THEY GOOD OR BAD?
Notes Survey Research Methods
Session 1: Introduction to Survey Research
WHAT IS SURVEY RESEARCH?
A survey is a systematic method for gathering information using
standardized questionnaires from a sample of entities to
construct quantitative descriptors of a larger population (Groves
et al., 2009).
Key Characteristics:
o Systematic & Standardized → Differs from qualitative
research.
o Sample of Entities → Often a selection of the population is
surveyed.
o Quantitative Descriptors → Requires statistical analysis.
Example: A survey of 1,000 customers to assess satisfaction with a new
airline service.
A SHORT HISTORY OF SURVEY RESEARCH
Ancient Egypt: Early census surveys.
1930-40s: Paul Lazarsfeld studied how radio influenced political
opinions.
Today: Used in business for customer insights and experimental
validation.
Example: Political polling before elections.
SURVEY VS. OTHER RESEARCH METHODS
Method Data Type Control Use Case
Level
Survey Self-reported Low Understanding behaviors,
Research opinions, or preferences.
Experimental Controlled High Testing causal
Research conditions relationships (A/B testing,
lab studies).
Modeling Existing data Moderate Identifying trends using
(Secondary large datasets.
Data)
Example: A company uses surveys to track brand perception over time,
while experiments test ad effectiveness.
TYPES OF SURVEY SAMPLING
Probability Sampling (random selection, results are generalizable)
o Simple Random Sampling: Every unit has an equal chance.
,Exam:
o Systematic Sampling: Every nth person is chosen.
o Stratified Sampling: Population divided into groups, then
sampled.
o Cluster Sampling: Entire groups (clusters) are randomly
selected.
Non-Probability Sampling (subjective selection, not generalizable)
o Convenience Sampling: Easily available participants.
o Quota Sampling: Targets specific groups (e.g., 50% women,
50% men).
o Snowball Sampling: Existing respondents recruit others.
o Purposive Sampling: Chosen based on specific traits.
Example: Using stratified sampling to ensure representation from
different age groups in a survey about social media use.
SURVEY TIME HORIZONS
Cross-Sectional Surveys: One-time snapshot of data.
o Example: A single survey measuring brand awareness in
2024.
Longitudinal Surveys: Track changes over time.
o Repeated cross-sectional: Different sample each time.
o Fixed-sample panel: Same sample surveyed multiple times.
o Cohort study: Follows a specific group over years (e.g.,
tracking 2010 marketing graduates).
SURVEY ADMINISTRATION METHODS
Personal Surveys (Face-to-face interviews)
Telephone Surveys (Often used for customer service feedback)
Online Surveys (Popular due to cost-effectiveness)
Written Surveys (Mail-in or paper-based)
Example: A university uses online surveys to evaluate student
satisfaction.
PROS & CHALLENGES OF SURVEYS
Pros
o Ease: Quick data collection from large groups.
o Simplicity: Easy to analyze.
o Reliability: Standardized format reduces variation.
Challenges
o Asking the right questions is difficult.
o Wording questions properly to avoid bias.
o Getting the right sample for representativeness.
o Motivating respondents to answer truthfully.
Example: A poorly worded survey question might lead to biased
responses:
❌ "Do you agree that electric cars are better for the environment?"
,Exam:
✅ "On a scale from 1-5, how environmentally friendly do you consider
electric cars?"
THE ROLE OF SURVEYS IN RESEARCH
Exploratory research: Identify key factors, generate hypotheses.
Descriptive research: Understand behaviours, measure variables.
Explanatory research: Test causal relationships.
Example: A company first explores why customers cancel subscriptions
(exploratory), then tracks dissatisfaction trends (descriptive), and finally
tests new service changes to reduce cancellations (explanatory).
WHAT IS SURVEY DATA GOOD FOR?
Correlational Insights: Analyzing relationships between variables.
o Example: Salary vs. job satisfaction.
Causal Insights: Requires additional manipulation (experiments).
o Example: Testing a new pricing model’s effect on sales.
Extending Data: Combining surveys with secondary data.
o Example: Merging survey responses with purchase history.
SURVEY RESEARCH PROCESS
Planning a Survey
o Define research objectives.
o Select research variables.
Designing a Survey
o Conceptualization (defining constructs).
o Operationalization (choosing how to measure variables).
o Construct and pretest the questionnaire.
Implementing a Survey
o Choose relevant population.
o Select sampling method.
o Conduct data collection.
Gaining Insights
o Data preparation & measurement evaluation.
o Data analysis (e.g., regression, correlations).
o Reporting results.
Example: A fitness brand surveys customers to find factors influencing
gym membership renewal.
SURVEY QUESTION TYPES
Open-ended: Respondents give detailed answers.
o Example: “What do you like most about our product?”
Closed-ended:
o Dichotomous (Yes/No).
o Multiple choice (Choose from options).
o Rating scales (Likert, Semantic Differential).
, Exam:
Example:
❌ "Do you use public transport?" (Too vague)
✅ "How often do you use public transport?"
a. Daily
b. Weekly
c. Monthly
d. Never
SAMPLING & DATA COLLECTION
Defining population (Who are we surveying?).
Sampling frame (List of potential respondents).
Choosing sampling method (Probability vs. Non-probability).
Determining sample size (Larger = more accurate results).
Executing data collection (Ensuring responses are valid).
STRUCTURAL EQUATION MODELING (SEM)
Analyzes relationships between multiple variables.
Used in marketing, psychology, business research.
Example: Examining how customer satisfaction influences loyalty.
INTERNATIONAL SURVEY RESEARCH
Consider cultural differences in interpretation.
Address response biases (e.g., tendency to agree with
statements).
Ensure measurement equivalence across countries.
Example: A survey on work-life balance might have different
interpretations in Japan vs. the Netherlands.
EXPERIMENTAL SURVEY RESEARCH
Controlled experiments where one variable is manipulated.
Goal: Establish causation, not just correlation.
Example:
Survey: “Do you prefer blue or red ads?”
Experiment: Expose one group to blue ads, another to red, then
measure engagement.
KEY TAKEAWAYS
Surveys provide structured, standardized data for analysis.
Choosing the right sampling method ensures valid results.
Good question design prevents bias and improves response quality.
Surveys can be combined with experiments for deeper insights.
EXAMPLES SURVEY QUESTION: WHY ARE THEY GOOD OR BAD?
