Pre-test

Before the main experiment, a pre-test was conducted to appraise the validity and reliability of the experiment concerning the manipulation levels (Berry, 2008). The pre-test will be conducted to ensure that the different advertisement claims are perceived as environmental and financial focused, and the sources of the advertisements (public and private) are perceived as different by the participants.

In total forty participants took the pre-test, and they were randomly assigned to one of the four treatment groups; each experimental group would be disclosed to one of the four

advertisements only. The questionnaire started with two questions to check whether the participants had been exposed to the concept of car sharing and the frequency of usage of the service. Afterward, participants were randomly shown one of the advertisements and asked to take their time to look at the advertisement thoroughly. At the beginning of the survey, the participants were notified that they would not be able to see the advertisement later on again.

Subsequently, the participants were asked to indicate how much they agreed/disagreed with the following statements: “This advertisement is focused on the environment,” “This advertisement is focused on cost savings,” “I think this advertisement was communicated by a governmental instition,” “I think this advertisement was communicated by a corporation,” and “I am familiar with the source of the advertisement.” Lastly, gender and age were incorporated into the

questionnaire.

3.4.1 Manipulation Check Pre-test

A one-way ANOVA test showed the focus on the environment and the cost savings in the advertisement claims were significantly different (F(1, 38) = 29.45, p < .05). In total 19

participants were exposed to an advertisement focused on the environment, and 21 participants were exposed to an advertisement focused on cost savings. Looking at the results of the question

““This advertisement is focused on the environment,” the advertisements focused on the environment were perceived by the participants as environmentally focused (M = 4.37, SE= 0.83) and the advertisements focused on financial were, on the other hand, correctly perceived as focused on financial (M = 2.57, SE = 1.21). Even in more depth, condition group 1 had a higher mean score than group 2 (M = 4.55 and M = 4.13 respectively), meaning the group by which the advertisement was communicated by a public source scored higher on mean. Another one-way ANOVA test showed that the source of the advertisement was appropriate to use for the main test, which means that the sources were significantly different as well (F(1, 38) = 63.12, p < .05).

In total 22 participants were exposed to an advertisement communicated by a public source, and 18 participants were exposed to an advertisement communicated by a private source. The results of the question “I think this advertisement was communicated by a governmental authority”

revealed that the advertisements communicated by the public source were perceived as coming from a governmental institution (M = 4.77, SE = .43). On the other hand, the advertisements communicated by the private source were perceived as coming from a corporation (M = 2.44, SE

= 1.29). Considering this, it shows that the manipulation was successful, and the advertisements could be used in the main study.

The use of two statements for each manipulation was considered unnecessary. If the participants were only provided with one statement about the focus of the advertisement claim,

participants would perhaps not agree on both options. This is what a few respondents did. The same goes for the source of the advertisement. So, for the main study, some minor alterations in the questioning of the manipulation check are made. In addition, some feedback provided by participants included the presence of the color green in the advertisement. People can

(unconsciously) associate the color green with the environment. Despite that the text of the advertisement claim focused on the financial benefits is clear to participants, they can question if the environment does not also play a role since half of the claim is in green. Therefore, to prevent deception, the whole text will be in black instead of partly green in the main experiment.

However, the green area at the bottom will be kept since green is one of the main colors of the visual brand of Greenwheels.

3.4.2 Familiarity

In general, the results of the pre-test showed that the manipulation was successful. However, striking was that a quarter of the participants were not familiar with the term car sharing. Hence, in the main test, a description of the term car sharing will be provided to everyone after the introduction. Moreover, participants were asked to indicate their familiarity with the Dutch source which was provided to them on a 5-point Likert scale (1 = completely unfamiliar, 5 = completely familiar). The pre-test revealed that participants were familiar with the sources. There was significant difference in familiarity between groups (F(1, 38) = 4.12, p < .05). The public source had a high mean score for familiarity (M = 4.09, SE = 1.15), whereas the private source scored lower (M = 3.22, SE = 1.56). The results indicated that Greenwheels is not so well known as Rijksoverheid. There is a chance that people not living in The Netherlands took part in the pre-test since the pre-test was conducted in English. This can cause a lower familiarity score than if just Dutch citizens would take part in the test. The main study is focused on the Dutch

Millennial consumers. Therefore, the main test will be conducted in Dutch. Due to the notable fact that a significant number of people are not familiar with car sharing and two-thirds never used the service, familiarity will be considered as a control variable in the main study.

Familiarity is related to the knowledge of something. For marketers, it all starts with expanding the knowledge about the product or service. Knowledge is essential to cause effective changes in the behavior of consumers. People need to know and be aware of it to form an attitude, increase intentional behavior, and in the end, change behavior (Ishak & M. Zabil, 2012). This is

summarized in the knowledge-attitude-behavior (KAB) model. The variable familiarity aims to measure the extent to which participants are familiar with car sharing. This is measured based on familiarity scales used by Dursun et al. (2011). The reported Cronbach’s Alpha scores was .78.

The questionnaire regarding familiarity in the main study would consist of three items. The participants were asked to rate how familiar, experienced and knowledgeable they see themselves concerning car sharing.