Public Service Announcements: Does Deception in Advertising Affect the Public's Intent to Donate?

Public Service Announcements: Does Deception in Advertising Affect the Public's Intent to Donate?

Technical Business Research

GENB 5321

Fall 2005

December 12, 2005

Table of Contents

I. Executive Summary 3

II. Introduction 6

III. Research Methods and Procedures 8

IV. Managerial Implications 9

V. Data Analysis and Findings 10

A. Scale Reliability 10

B. Factor Analysis 13

C. Analysis of Variance (ANOVA) and Results 14

D. Multiple Regression Analysis and Results 16

VI. Conclusions and Recommendations 17

VII. Limitations 18

VIII. Appendix 20

A. References 21

B. Model Results 22

C. Responses to Surveys 23

D. Reliabilities 24

E. Assessment of Scale Dimensions 31

F. Demographic Results 35

G. Explorative Interviews 39

H. Surveys 54

Executive Summary

The objective of the research is to ascertain whether or not one's intent to donate to a non-profit organization (NPO) is affected by deception in an anti-smoking advertisement. More specifically, the objective is to determine if intent to donate is dependent

upon the affective response to the ad, the level of trust and social responsibility one has towards the NPO, and the perceived ethicality of the NPO. The results of the research will help management to make an educated decision regarding the use of actors in anti-smoking campaigns.

Three separate hypotheses were formulated to explain the relationship between the dependent variable, "Intent to Donate", and the independent variables, "Trust and Social Responsibility", "Affective Response to the Ad", and "Ethicality". The first hypothesis (H1: "Trust and Social Responsibility") states that a person's intent to donate is impacted by the level of trust the potential donor has in the non-profit organization and the person's perceived degree of social responsibility attributed to that organization. The second hypothesis (H2: "Affective Response to the Advertisement") states that a person's intent to donate is dependent

upon the affective responses solicited from the advertisement. The greater the emotional attachment the respondent felt toward the "victim" in the ad, the great the intent to donate. The third hypothesis (H3: "Ethicality") states that the perceived ethicality or "moral compass" of a non-profit organization significantly influences intent to donate.

A quantitative method was selected for the research project. A survey was composed and distributed to a sample population that was based on the perceived local demographics. The responses of initial probing interviews were used as the framework for the questions in the survey. Secondary data was also used in composing the survey. Previous research, and the scales that resulted from that research, were used to help define constraints for the project and establish the questions and scales for the survey. The scales utilized in the survey for collecting primary data were taken from several different sources and revised for the purpose of the research at hand. All coefficient alphas from the scales used in the survey for this research reported acceptable values for reliability. Primary data was collected through non-probability (convenience) sampling. Respondents completed the survey both in writing and on-line.

The survey was divided among four treatments. The first treatment to the advertisement was the control treatment; there was no factual information provided and no actors were used. The second treatment was the addition of factual information (date of death) and no actor was used. The third treatment was the use of deception; an actor was used in the advertisement and not revealed initially. The final treatment to the advertisement was the revealed use of an actor ("Actor Portrayal" disclaimer).

An analysis of variance (ANOVA) was used to determine the statistical differences between the means of our dependent variable (Intent to Donate) and the three independent variables (Trust and Social Responsibility, Ethicality, and Affective Response to the Ad). The ANOVA results showed that the null hypothesis, there is no statistical difference in the intent to donate when the independent variables are considered together, should be rejected.

Multiple regression techniques were used to analyze the relationship between the independent and dependent variables. If the regression coefficients are not significant, this means that the independent variable does not have a relationship with the dependent variable. In our analysis, "Affective Response to the Ad" and "Trust and Social Responsibility" were shown to be significant predictors of "Intent to Donate". "Ethicality" was not proven to be a significant and should not be used as a predictor for "Intent to Donate".

The research findings indicate that the perceived "Trust and Social Responsibility" and the "Affective Response to the Ad" contribute to 94.5% of the variation in "Intent to Donate." The most effective way in which to positively affect a person's "Intent to Donate" in an anti-smoking advertisement is through the use of actual "victims". The use of actors, even if that fact is divulged, decreases the level of trust a person feels towards a non-profit organization and has a negative effect on their "Intent to Donate". The recommendation for future marketing endeavors is to eliminate the use of actors portraying victims in anti-smoking campaigns. The use of actual "victims" is