Log in / Register
Home arrow Engineering arrow Measuring Electronic Word-of-Mouth Effectiveness: Developing and Applying the eWOM Trust Scale

Application Stage

The purpose of the sixth study was to cross-validate an English version of the eWOM trust scale (RQ12) and to introduce a new typology of online consumers considering their trust in eWOM/Online advertisements (RQ13). It was concluded that the aforementioned objectives will be best achieved by surveying a representative, adequately sized sample of US online consumers. Therefore, an online survey among members from the Research Now panel was conducted. This study shared some reasonable similarities with this research’s main study concerning its execution and structure. For instance, when accessing the survey, respondents first faced an introductory page which informed them about the rough content of the following questionnaire and the privacy policy. On the same page, a similar description of online reviews and some examples were also provided. Before being able to answer the actual questionnaire, respondents were prompted with a question concerning their review/online ads habits within the last 6 months. On the following page, subjects were asked to answer a series of sociodemographic questions. The following pages were only accessible to persons who had recent contact with eWOM/Online advertising and who had the desired socio-demographic characteristics. Similarly to the main study, the US study also enforced quotas on the respondents’ age and gender which correspond to the demography of the US Internet population (US Census Bureau, 2011): age group 18-34 yrs (male: 17%, female: 18%), 35-44 yrs (male: 9%, female: 10%), 45-64 yrs (male: 17%, female: 18%), 65 yrs and older (male: 5%, female 6% of total population).

The questionnaire was structured as follows: (i) Internet usage behaviour with items taken from Corbitt et al. (2003), Barnes et al. (2007), and Lee et al. (2011), (ii) perceptions and attitudes towards online customer reviews in general. This section included measurement items for quantifying the constructs of attitude toward reviews (with 3 items taken from Park and Kim (2008)), review usefulness (4 items adapted from Park and Lee (2009)), review avoidance (3 items mainly adapted from Park et al. (2007)), and perceived review usage risk (5 items adapted from Jarvenpaa and Tractinsky (1999)).

Another part was dedicated to the new eWOM trust scale. As the items were originally developed in German, they were subject to a rigorous forward-backward translation process. Four persons who had American English as their mother tongue and were proficient in German (all academics from two Austrian universities) were asked to translate the items in the target language. For every item, they were supposed to provide at least one word or phrase that captures the nature of the item, but they were also able to provide two optional translations. After collecting the answers, two researchers identified recurring word and phrases. They agreed on 100% of the items. Finally, two other persons being fluent in English and German translated the English version again back to German. Again, there was a high agreement on the meaning of the terms. eWOM trust was measured, as the items mentioned before, on a 7-Point Likert scale ranging from 0 (“I strongly disagree”) to 6 (“I strongly agree”). Finally, the questionnaire’s second section also included items measuring respondents’ earlier eWOM experiences and habits. These items were adapted from the research of Park and Lee (2009), Obermiller and Spangenberg (1998), and Chu and Kim (2011). Here, a 7-Point rating scale was used (0 = “never”; 6 = “always”).

The next section (iii) addressed consumers’ attitudes towards the Internet. More specifically, respondents were asked to answer 4 items taken from Morse et al. (2011) measuring Internet attitude and 3 items adopted from Barnes et al. (2007) measuring online shopping attitude. All were measured on a 7-Point Likert scale from 0 (“I strongly disagree”) to 6 (“I strongly agree”). In addition, measures of active and passive usage behaviours of Web 2.0 applications were included. Section (iv) targeted perceptions and attitudes towards online advertising in general. Here, online advertising trust was measured with 9 items (taken from Obermiller and Spangenberg (1998)) and online advertising attitude with 7 items (taken from Bauer and Greyser (1968)). (v) Perceptions of business ethics and market offerings in general. The 10- item scale for consumer alienation was adopted from Burns (2010), consumer attitude towards product quality was measured with 5 items introduced by Ferdous and Towfique (2008), 2 items taken from Lyonski et al. (1996) were used to measure product price attitude. The final section (vi) focused on selected consumer psychographics including disposition to trust (5 items taken from Gefen (2000)), self esteem (4 items taken from Rosenberg (1965)), self-confidence (12 items taken from Bearden et al. (2001)), and CSII (8 items taken from Bearden et al. (1989).

All constructs were measured on a 7-Point Likert scale with 0 (“I strongly disagree”) and 6 (“I strongly agree”). As in the main study, the order of the items was randomized in each section.

In total, 1220 panellists visited the online survey. Of these, 635 persons were excluded from the study due to unfamiliarity with the trust objects (eWOM and online advertising) or full quotas. During the time the questionnaire was available (July 30, 2013 - August 2, 2013), 547 participants completed the survey. Subsequent review of data led to the exclusion of several respondents and a total number of 517 usable responses suitable for data analysis. The sociodemographic profile of the total sample was as follows. 53.2 % of the sample were females. The respondents’ age ranged from 18 to 79 yrs (M = 43.9, SD = 15.0). The majority were married or lived together with another person (62 %). Singles compromised 24.8 % of the sample. Almost three-fourths of the sample (72.9 %) had an annual income below USD 75,000. The majority of the sample were office workers or public servants (36.6 %), followed by retirees (16.9 %), and blue-collar workers (13.8 %). European Americans represented 81.5 % of the total sample (African American: 8.2 %; Asians: 5.1 %; Hispanics: 3.9 %; Native Americans: 1.4 %). Almost a third had a 4-year college degree (31.6 %) and 16.4 % a master’s degree. 48.7 % of the sample lived in the suburban area, while the remainder reside in urban (27.6 %) or rural (23.7 %) areas. Further sample characteristics are provided in Chapter 5.

The next chapter discusses the findings of the surveys outlined above.

Found a mistake? Please highlight the word and press Shift + Enter  
< Prev   CONTENTS   Next >
Business & Finance
Computer Science
Language & Literature
Political science