Survey Research Methods
Lecture notes
Lecture 1 | Survey Research at DVJ Insights
Campaigns are essential to support brand growth but only a few campaigns can generate brand
impact
Only 20% is able to significantly increase advertising recall
o which can be considered the first step in realizing the effects on the brand
Campaigns can drive both short- and long-term growth for a brand across domains, markets,
and brand sizes. Where a combination of sales activation and brand-building campaigns has
shown to be the most successful campaign strategy to increase sales.
The DVJ pre-test methodology is based on a combination of breakthrough, brand building,
and sales activation to determine the potential of an ad.
Never use forced exposure
Traditionally the testing of communication (both ads and packaging) is conducted using a 'forced
exposure' methodology. In doing so, several fundamental laws of advertising are disregarded:
80% of the TV commercials in a TV reel are zapped away by 80% of the consumers.
Print ads are scanned in 2 to 3 seconds.
Online ads only get noticed within the 3rd and 5th second.
Social ads are viewed for less than 2 seconds.
Outdoor ads need to be able to attract attention within 1 second.
This implies that using a forced-exposure methodology in a copy test leads to incomplete or even
invalid conclusions and recommendations.
Channel-specific unforced exposure
To determine breakthrough potential and ad impact we use a natural environment. For each media,
respondents are shown the ad in the correct media environment along with other ads and move from
ad to ad with the ability to behave naturally. Eg zaps away with a TV commercial or browse a
website.
How do you measure memory effects?
• Natural exposure
• Dummy questions
Brand cues: the way of trigger associations with the brand
• Music
• Characters
• Personalities
• Voice
• Memory questions
• Consistent story approach
• Tagline
1
,Lecture 2 | Introduction to Survey Research
Survey research
A survey is a systematic method for gathering information with standardized questionnaires from (a
sample of) entities for the purpose of constructing quantitative descriptors of the attributes of the
larger population of which the entities are members.
Sampling procedures of survey research
Probability sampling:
Selection of sampling techniques in which the chance of
each case (e.g., individual) being selected from the population is known and not zero.
→ Randomization is used instead of deliberate choice
Nonprobability sampling:
Selection of sampling techniques in which the chance of each case (e.g., individual) being selected is
NOT known. → The researcher deliberately picks the sample
Cross-sectional design: opinion polls/thesis survey
Pros survey:
Ease: Questionnaires are relatively easy to administer to large number of respondents
Simplicity: coding, analysis, and interpretation of data is relatively simple
Reliability: Standardization reduces variability in the answers
Challenges:
Asking the right questions is not easy
Properly wording questions is not easy
Getting the right sample (sampling) is not easy
Motivating respondents to respond (candidly) is not easy
Measurement and representation
Measurement: what is the survey about, or what does the survey measure?
Representation: whom does the survey describe, or who provides the desired data?
Analysis and reporting: what can we learn from the survey?
The four phases of the survey research process
1. Planning a survey
a. Define research objectives
b. Selection of research variables
2. Designing a survey
a. Conceptualization and operationalization
b. Construct and pretest questionnaire
3. Implementing a survey
a. Choose relevant population
b. Sampling
c. Data collection
4. Gaining insights from a survey
a. Data preparation and measurement evaluation
b. Data analysis
c. Reporting of results
2
, Lecture 3 | Planning a survey
Why conduct a survey instead of experiment?
We are interested in experience not their behavior. Survey research questions are too complicated to
do an experiment for.
Typical goals of academic survey research
- Scale development
- Segmentation and typologies
- Relationships between variables
Scientific research
Experimental research -> RIGHT SIDE
Survey research -> can be both
4 criteria for scientific research
- Precision: theoretical concepts, which are often hard to measure, must be defined with such
precision that others can use those definitions to measure those concepts and test that theory
- Falsifiability: a theory must be stated in a way that it can be disproven
o Theories that cannot be tested are not scientific theories and any such knowledge is
not scientific knowledge
o A theory with imprecise terms or whose concepts are not accurately measurable
cannot be tested
- Replicability: others should be able to independently replicate or repeat scientific study
3
Lecture notes
Lecture 1 | Survey Research at DVJ Insights
Campaigns are essential to support brand growth but only a few campaigns can generate brand
impact
Only 20% is able to significantly increase advertising recall
o which can be considered the first step in realizing the effects on the brand
Campaigns can drive both short- and long-term growth for a brand across domains, markets,
and brand sizes. Where a combination of sales activation and brand-building campaigns has
shown to be the most successful campaign strategy to increase sales.
The DVJ pre-test methodology is based on a combination of breakthrough, brand building,
and sales activation to determine the potential of an ad.
Never use forced exposure
Traditionally the testing of communication (both ads and packaging) is conducted using a 'forced
exposure' methodology. In doing so, several fundamental laws of advertising are disregarded:
80% of the TV commercials in a TV reel are zapped away by 80% of the consumers.
Print ads are scanned in 2 to 3 seconds.
Online ads only get noticed within the 3rd and 5th second.
Social ads are viewed for less than 2 seconds.
Outdoor ads need to be able to attract attention within 1 second.
This implies that using a forced-exposure methodology in a copy test leads to incomplete or even
invalid conclusions and recommendations.
Channel-specific unforced exposure
To determine breakthrough potential and ad impact we use a natural environment. For each media,
respondents are shown the ad in the correct media environment along with other ads and move from
ad to ad with the ability to behave naturally. Eg zaps away with a TV commercial or browse a
website.
How do you measure memory effects?
• Natural exposure
• Dummy questions
Brand cues: the way of trigger associations with the brand
• Music
• Characters
• Personalities
• Voice
• Memory questions
• Consistent story approach
• Tagline
1
,Lecture 2 | Introduction to Survey Research
Survey research
A survey is a systematic method for gathering information with standardized questionnaires from (a
sample of) entities for the purpose of constructing quantitative descriptors of the attributes of the
larger population of which the entities are members.
Sampling procedures of survey research
Probability sampling:
Selection of sampling techniques in which the chance of
each case (e.g., individual) being selected from the population is known and not zero.
→ Randomization is used instead of deliberate choice
Nonprobability sampling:
Selection of sampling techniques in which the chance of each case (e.g., individual) being selected is
NOT known. → The researcher deliberately picks the sample
Cross-sectional design: opinion polls/thesis survey
Pros survey:
Ease: Questionnaires are relatively easy to administer to large number of respondents
Simplicity: coding, analysis, and interpretation of data is relatively simple
Reliability: Standardization reduces variability in the answers
Challenges:
Asking the right questions is not easy
Properly wording questions is not easy
Getting the right sample (sampling) is not easy
Motivating respondents to respond (candidly) is not easy
Measurement and representation
Measurement: what is the survey about, or what does the survey measure?
Representation: whom does the survey describe, or who provides the desired data?
Analysis and reporting: what can we learn from the survey?
The four phases of the survey research process
1. Planning a survey
a. Define research objectives
b. Selection of research variables
2. Designing a survey
a. Conceptualization and operationalization
b. Construct and pretest questionnaire
3. Implementing a survey
a. Choose relevant population
b. Sampling
c. Data collection
4. Gaining insights from a survey
a. Data preparation and measurement evaluation
b. Data analysis
c. Reporting of results
2
, Lecture 3 | Planning a survey
Why conduct a survey instead of experiment?
We are interested in experience not their behavior. Survey research questions are too complicated to
do an experiment for.
Typical goals of academic survey research
- Scale development
- Segmentation and typologies
- Relationships between variables
Scientific research
Experimental research -> RIGHT SIDE
Survey research -> can be both
4 criteria for scientific research
- Precision: theoretical concepts, which are often hard to measure, must be defined with such
precision that others can use those definitions to measure those concepts and test that theory
- Falsifiability: a theory must be stated in a way that it can be disproven
o Theories that cannot be tested are not scientific theories and any such knowledge is
not scientific knowledge
o A theory with imprecise terms or whose concepts are not accurately measurable
cannot be tested
- Replicability: others should be able to independently replicate or repeat scientific study
3
