• No results found

Method for Obtaining Static Measures of Advertising Wearout .1 Overview

PRAC 1

CHAPTER 5: FEAR PATTERNS WITH REPEATED EXPOSURES (STUDY 2)

5.3 Method for Obtaining Static Measures of Advertising Wearout .1 Overview

An experimental design was the method used to address the research objectives in this study. Study 2 involved an advertising experiment that tested the effectiveness of two types of underlying "patterns" of fear arousal -fear-relief and fear-only - within anti-speeding advertisements. The four anti-anti-speeding advertisements that had been

previously fear patterned in Study 1 (refer to Chapter 4) were also used as the stimulus materials for Study 2.

The following sections detail two research investigations, both using experimental

designs, but consisting of separate groups of participants and separate measures (static and dynamic). The method and results for Experiment One of Study 2 are reported first, followed by the method and results for Experiment Two of Study 2. A comparison

between the results of each component of Study 2 is then provided.

5.3.2 Research Method - Experimental Design

Social marketers find it difficult to define advertisement effectiveness measures because of a large amount of uncontrollable variables, therefore an advertisement experiment is extremely useful for predicting the effectiveness of an advertisement. The major

advantage of using an experimental design to measure the effectiveness of various advertisements is the greater control of extraneous factors than if testing the

advertisements in their natural setting of in-home viewing. For example, Cacioppo and Gardner (1999, p. 89) stated that "laboratory studies can afford impressive control over relevant variables, an important feature when dissecting phenomena as complex and

multiply determined as the emotions". The effects of the advertisements are more likely to be isolated when other variables are controlled, such as the age of the viewing audience, number of advertisement exposures, similarity in time of measurement, and avoidance of competing advertisements. The major drawback of this experimental design is an artificial setting for viewing advertisements.

5.3.2.1 Data Collection Procedure

A post-only monadic experimental design was used for the static wearout study to avoid sensitising the audience to the purpose of the study. A total of 284 participants, from a first-year marketing class, divided into four experimental groups, took part in the

experiment. For three consecutive weeks each group viewed one advertisement (the same advertisement) per week and then completed a questionnaire pertaining to the advertisement after each viewing session. The advertisements were shown in tutorial

classes. The assignment of the four advertisements to various tutorial classes was based on convenience, given the relative randomness of how students were allocated to

tutorials in the first instance (that is, by a computer program that allocated tutorial places for students at the beginning of session). No quota sampling was undertaken, which resulted in an uneven gender balance in the experimental groups. There was also an attrition rate over the three weeks of the experiment, with many students not

attending all three sequential weeks of tutorials. Hence, while there was an initial group of 284 students in week 1, the number of usable responses dropped to 87 when

analysing only those participants who were present in all three weeks of the study. The drop-out participants from the study, who were students who missed one or two

tutorials within the three-week testing period, had no special characteristics or differences to those who remained in the study. Again, this study was given ethics approval by the University of Wollongong's Human Research Ethics Committee.

Participants were offered a $20 bookstore voucher if they completed the three week task. Every participant was also thanked with a $1 charity chocolate.

Social desirability bias was also considered for this study. Participants were asked to reflect their emotional reactions to the advertisement by using the dial, but they could not see the computer screen that showed the output of their movements of the dial,

hence without this visual feedback any social desirability bias w a s minimised. It is a

possibility that some participants may have tried to understate how reactive they felt.

However, any such reaction among individual participants was not considered likely to affect the general shape of the curves produced.

There was no survey instrument used for the patterning part of this study. However, the wearout study questionnaire is shown in Appendix 2. The participants used for the

studies comprising this thesis only took part in one of the studies, that is, there was no duplication of participants between studies.

The research process for the static measures component of Study 2 is shown in Figure 5.1.

Figure 5.1: Research Process for static measures of wearout

A d Group 1 -Pizza

Wkl-Ad&

survey

W k 2 -A d &

survey

W k 3 -Ad, survey

&VST

Ad Group 2 -Pram

A d Group 3 -Trike

Control group - V S T only

W k l -A d &

survey

W k 2 -A d &

survey

W k 3 -Ad, survey

&VST

W k l -A d &

survey

W k 2 -A d &

survey

W k 3 -Ad, survey

&VST

A d Group 4

-4WD

Wkl-Ad&

survey

W k 2 -A d &

survey

W k 3 -Ad, survey

&VST

5.3.2.2 VST Control Group

A separate group of 42 participants was used for the control group that undertook the VST. The participants were also University of Wollongong marketing students aged 18-25 years old and therefore were of a similar demographic composition to those participants in the experimental groups. Bloom and Novelli (1981) pointed to the need for a control group for studies. The control group was not shown an advertisement prior to undertaking the VST, as other advertisements may have included arousing elements that would influence speed choice.

5.4 Results: Static Measures of Fear Patterns with Repeated Exposures