L - Humor in Advertising - Fragen

Question 1

What is a meta-analysis? Name and explain two advantages of a meta analysis compared to individual primary studies.

Answer 1

• statistical analysis of large collection of analysis results from individual studies for the purpose of integrating the findings

Advantages:

• resolves conflicts in the literature -> inconsistent findings
• identifying gaps in literature
• generalization of findings -> guide the next wave of research
• integration of effects sizes

Question 2

Discuss the statement „Scholars are recommended to ask when humor in advertising is effective rather than if humor is effective.“ How did this statement motivate Eisend’s meta-analysis?

Answer 2

• „If humor is effective“ -> question for the search for generalizable results
• „when is humor effective“ -> question for moderators that contribute to variability of results

-> there are inconsistent findings -> to find out how the effect works, both tasks should be addressed

Question 3

According to previous reviews, which effects of humor in advertising are consistent across studies? Which effects are more ambiguous? What is the methodological difference between previews reviews and Eisend’s study?

Answer 3

consistent: 
Humor in advertising…
- creates attention and awareness
- enhances source liking (?), attitude towards ad, positive cognitions
- reduces negative cognitions

ambiguous:
Humor in advertising…
- enhancing vs. reducing comprehension?
- impact on recall and cognition?
- source credibility
brand attitude, purchase intention and behavior unclear

previous reviews:

• > literature reviews
• no generalization effect
• no integrated effect sizes

Eisend:

• > meta-analysis
• overall effect of humor in advertising
• generalization, integration of effect sizes

Question 4

From a theoretical point of view, different types of relationships between perceived humor and attitudes towards the brand are possible. Name and explain the possible relationships.

Answer 4

linear relationship:

• > positiv: affect transfer, affective mechanisms
• evoking a positive affect that is transferred to the brand (-> classical condition)
• > negativ: humor reduces credibility of a source, which reduces persuasion effects
• very strong humor might be seen as a weakness of the marketer to make a serious argument for the brand
• source might also be perceived as using humor because of a lack of ability to make her or his point seriously, impairs persuasion

curvilinear relationship:

• arousal theory: greater arousal results in greater pleasure up to a optimal point
• rising from beginning until extremum and after that, decreasing effect for too high levels of humor

no relationship:

• distracting consumers from benefits
• humor induces pleasant reactions and thus functions as an incentive to pay attention
• but because respondents are preoccupied with the humor they are less attentive to other parts of the message

Question 5

How did the author search for studies to be included in the meta-analysis? Did he use the systematic or cumulative approach? Did he include unpublished work? How do you evaluate the literature research procedure? Do you see any problems? Did the author run a test to solve the problem?

Answer 5

both approaches:

• systematic: keywords systematically used in different databases
• cumulative: google scholar, references from the paper

no real search for unpublished work
-> much work but worth it, otherwise publication bias

test:
fail safe N test for significant results
-> number says how many people you need to show results are not significant

Question 6

Which type of effect size (which coefficient) did Eisend use? Which dependent variable is most strongly affected by perceived humor? Is there a negative overall effect on any of the dependent variables? Does humor hamper comprehension?

Answer 6

• correlation coefficient
• attention
• negative effect on credibility
• no significant effect -> humor does not hamper comprehension

Question 7

Many primary studies of the effects of humor in advertisings have been run with student samples. Why did Eisend assume that they might be a bias? In terms that conclusion from these samples to the general population are not possible? Do the empirical reviews indicate a bias?

Answer 7

• students are highly educated, they can solve the incongruence easier and therefore have a better understanding of the humorous message
• Table 4: yes, there is a bias, because the findings show a significant effect for the student sample of humor and A(HD)

Question 8

Did Eisend expect some methodological factors that moderate the effects of humor in advertisings founding in academic studies? Describe these findings.

Answer 8

• characteristics of the audience
• humorous stimulus
• partially on product type and relatedness of humor and the product
  students
• laboratory experiments (usually not that humorous)
• print advertising (less humorous than real advertisements or ads in broadcast media)

Question 9

Name and explain the critics often raised with regard to meta-analysis.

Answer 9

• one number cannot summarize a research field
• filedrawer problem invalidates meta analysis (publication bias)
• mixing appeals and oranges -> depending on the question if it „fits“: totally different intentions in different papers in different countries
• garbage in, garbage out (quality of literature)
  (- language bias)

Question 10

Find 10 meta analysis terms.

Answer 10

Find 10 meta analysis terms.

sample size
moderator
garbage out
failsafe N
heterogeneity 
weight
meta analysis
random effects
publication bias
Q-test
effect size
attenuated 
coding