Purpose
The purpose of the pretest is twofold. We first seek to determine a suitable
dimensional range for our target, a hypothetical brand of apparel advertisement. As we will be testing for the effects of contextual stimuli both positive and negative of the target, simply one prime on each side of the target, we designed for the ad to be interpreted within neutral ratings, and the results of the pretest would verify if the target range would be suitable. The pretest also serves as a practice in defining a dimensional range. As the requirement of our survey is relatively new and differs from the conventional one-point rating questionnaires, results from the pretest will help us understand the level of clarity and practicality of the questions in our survey.
Participants
A total of forty people, both under- and postgraduate students from National Taiwan Normal University participated in the pretest. All of these participants were recruited
randomly from classes held in National Taiwan Normal University, and carried out during their respective class breaks.
Design
In our pretest, seven different dimensions for which university students would most likely be familiar with were selected as basis for testing the primes. Each dimension contains 2-3 primes, where each prime given within each dimension was ensured to be those highly familiar amongst our target participants, university students. These seven dimensions are namely:
1. Hardness of objects 2. Sweetness of food
3. Swiftness of animals’ movement 4. Prestige level of products
5. Durability of products 6. Fashion of products 7. Price of products
All participants answered the same pretest questionnaire, which consisted of two parts and included all seven dimensions of questions. In order to minimise interference of prime influence, dimensions chosen were meant to be as unrelated with one another; the first part of the questionnaire comprised of twelve questions on three dimensions, “Hardness of objects”,
“Sweetness of food” and “Swiftness of animals’ movement”. Primes within each of these dimensions were designed to cover both wide and narrow ranges in all low, high as well as moderately rated situations on a 0 – 10 rating scale. For example, “hardness of blanket” was given to participants as a range of 1.5-5 (negatively wide) while “hardness of cement” and
“sweetness of oranges” were given as a range of 9.5-10 (positively narrow) and 2.5-8.5 (moderately wide) respectively In addition, there were also questions that required
participants to think of items of hardness within 5-9.5 (positively wide) range, animals with movement within 0.5-1 (negatively narrow) range of swiftness, as well as to rate the
swiftness of giraffe’s movement, expected to be moderately narrow.
After participants have gone through the practice of giving representative value and range ratings, as well as providing items they can think of within a given range rating in the first part of the questionnaire, we then present our target, a hypothetical brand of apparel advertisement, for their review. Similarly, participants were asked to rate the target brand on dimensions as unrelated to one another as possible, namely, the target brand’s prestige level,
durability, fashion and price. The dimension of prestige level was presented first amongst the rest, in order to diminish possible influence of other dimensions.
Procedure
The written questionnaire of the pretest was completed by 40 random, undergraduates and post-graduates students of National Taiwan Normal University. We approached random classes and obtained permission from respective lecturing professors to distribute our paper questionnaires during class breaks. Participants who agreed to take part in the study were seated individually in the classrooms they were having lessons in, and asked to complete a ten-page booklet consisting of two sections.
In the first part of the questionnaire, participants were told that they were participating in a market research for a new apparel brand, and would first be tested for their level of observation and sensitivity towards objects they come across daily. They were then instructed to read six sample questions on 2 dimensions, “hardness of objects” and “sweetness of food”.
Three questions were asked on each dimension; the first question required participants to give a one-point score rating that best represents the item being evaluated, from a scale of 0 (e.g.
least sweet) to 10 (e.g. most sweet) in intervals of 0.5. Participants were next asked to give a range rating, a lowest and highest bound to indicate an interpretative range for items on the same 0 to 10 scale. In the final question, participants were assigned a range and asked to provide an item within the given range on a specific dimension.
Figure 4.1 An example of the rating scale on representative value most fitted for the dimension of “Hardness” in part 1 of the pretest.
Figure 4.2 An example of the rating scale on interpretative range for the dimension of
“Hardness” in part 1 of the pretest.
Figure 4.3 An example of an item required, given a specific dimensional range on the rating scale for the dimension of “Hardness” in part 1 of the pretest.
A spot-the-difference picture game designed to disrupt the participants’ concentration and eliminate any unexpected context effects was subsequently inserted between the two sections of the questionnaire booklet. Participants were asked to find seven differences from two similar pictures before proceeding to the second section.
The second part of the questionnaire was similar to the first, except only the one-point representative score and interpretative range rating on four dimensions of our target product brand advertisement were evaluated. Participants were told they were going to be evaluating a new apparel brand and before being presented an advertisement of our target apparel brand, Belìssimo. Participants were asked to take a look carefully at the advertisement and evaluate the representative value and interpretative range of Belìssimo on four dimensions, “prestige level”, “durability”, “fashion” and “price”, on a scale from 0 (e.g. least prestigious) to 10 (e.g.
most prestigious). After completing the questions, participants were debriefed, thanked and dismissed.
Results
The results of the pretest essentially gave us the mean and standard deviation values of our target advertisement’s best representative score, lower and higher bound values for the ranges of items tested on the seven dimensions. Essentially, the interpretative range of our target apparel brand advertisement was determined to be between 3.72 and 5.88 (Table 4.1), with both p-values showing that the range is not significantly different from our ideal range of 4 to 6 respectively, thereby allowing the target apparel brand to be used for the main experiment.
Based on the evaluations of our forty participants, it is shown that all participants were capable of defining interpretative ranges on the dimensions asked, indicating that the level of clarity and familiarity of the items asked on each dimension was sufficient for our main experiment. All items were subsequently included in the main experiment thereafter.
More importantly, results from the pretest served as an affirmation of the suitability of our target advertisement to be applied in our main experiment, falling at the range of lower bound, M = 3.72 (t(29) = -1.29, p = 0.21), in relation to 4, and upper bound M = 5.88 (t(29) = -0.42, p = 0.68), in relation to 6.