Known-group Validity

In scale research, it is a common practice to regard the theoretical identification of groups that are hypothesized to score low/high on the focal construct and subsequently compare the scores of these groups on the measure as additional evidence for construct validity (e.g., Heimberg & Holaway, 2007; Netemeyer et al., 1995; Tian et al., 2001). In this research, it is expected that two convenient populations, namely university students and faculty members, are likely to differ on their tendency to trust consumer reviews. More precisely, it is assumed that faculty members are more likely to show lower levels of eWOM trust due to several reasons which correspond to the proposed nomological network. For instance, faculty members are older, are more skeptical, have higher self-esteem and do have more marketplace experiences. Hence, members of this group are more likely to be aware of potential manipulations by online reviews and perceive heightened skepticism for marketing communication in general. Also they have been subject to a different socialization process, providing them with cultural folk wisdom. Further, faculty typically exhibits different Internet and computer usage patterns as, for instance, its members less regularly participate in social networks, like Facebook, where they could socialize with others. Accordingly, the following research question was proposed:

RQ 5: Do students and faculty members differ significantly in their trust in customer reviews?

Significant difference between these two groups would provide evidence for known-group validity. However, also differences within these two groups were checked. Therefore faculty members were categorized by disciplines (natural sciences, business sciences, humanities, law, social sciences, others) and academic qualifications (PhD, professorship). On the other hand, the (business) students were grouped into undergraduates and graduates.

Data for the student sample were collected from two Austrian universities during the reliability stage (samples 3a, 3b). In total, 850 usable questionnaires were obtained (574 undergraduates; 276 graduates). For the faculty sample, 1,182 invitations were sent to qualified respondents in academic institutions in Austria, Germany, and Switzerland. The emails sent in several waves included a link to an online questionnaire which was accessible for three weeks in August 2013. The final convenience sample comprised 205 academics, resulting in a response rate of 17.34%. Of these, 112 persons (54.63%) held at least a PhD title and 93 a professorship (45.37%). The sample included 47 persons from natural sciences, 62 from business sciences, 29 from humanities, 20 from law, 32 from social sciences, and 15 from other disciplines.

As always, all respondents were requested to indicate their personal level of generalized eWOM trust with the 22 items on a 7-Point Likert scale ranging from 0 to 6. Here, higher numbers indicated heightened trust. Overall, the mean trust in online reviews for students was 3.82 (SD = 1.84), with undergraduate students averaging at 3.82 (SD = 1.86) and graduates at 3.84 (SD = 1.78). The mean eWOM trust for faculty members was 3.12 (SD = 1.11), with a break-down of 2.85 (SD = 1.18) for natural sciences, 3.31 (SD = 1.04) for business sciences, 2.87 (SD = 1.24) for humanities, 3.26 (SD = .96) for law, 3.03 (SD = 1.15) for social sciences, and 3.66 (SD = .80) for faculty belonging other disciplines. The difference between students and faculty was statistically significant, t(1,053) = 5.24, p < .001, with students having more trust in online reviews than members of the faculty, as expected. Thus, known-group validity is supported.

The comparison eWOM trust for undergraduate and graduate students revealed no significant differences between these groups (t(848) = .14, n. s.) and there was also no significant difference in means between the students from the University of Vienna (M = 3.88, SD = .11) and the Vienna University of Economics and Business (M = 3.76, SD = .04); t(554) = 1.03, p > .05. However, there was a significant difference in the trust level for professors (M = 2.84, SD = 1.17) and for academics, who possessed only a PhD title (M = 3.35, SD = 1.01); t(203) = 3.37, p < .05, implying that professors are more skeptical towards online customer reviews. The difference across the academic qualifications was not predicted, but it is consistent with the assumption that eWOM trust decreases with a number of personal characteristics, such as advanced age, widespread market place knowledge and awareness of manipulation strategies that the consumer becomes aware of over time. By applying one-way ANOVA, the mean scores on the eWOM trust scale of the respondents belonging to different academic disciplines were also compared. This revealed no significant effect on the level of trust caused by the membership of one of the six disciplines (F(5,199) = 2.05, p > .05). Similarly, there was no significant difference in the means of the faculty members belonging to universities (M = 3.05, SD = 1.13, n = 173), universities of applied sciences (M = 3.49, SD = 1.04, n = 23), or other academic institutions (M = 3.59, SD = .75, n = 9); (F(2,202) = 2.51, p > .05).

< Prev   CONTENTS   Source   Next >