|
2007 Annual International CHRIE Conference & Exposition
468
websites promoting visitor attractions in Wales. Each of these content analysis studies identified a different set of
features against which the websites were evaluated. Benckendorff and Black (2000) evaluated site planning, design,
content and management characteristics while Blum and Fallon (2002) reviewed product, price, promotion, place,
customer relations and technical aspects. Among content analysis studies, several specialized evaluations have
focused on specific features of websites. For example: Law and Leung (2000) evaluated the online reservation
features of 30 airline websites, Jeong and Lambert (2001) evaluated information quality of eight hypothetical
websites, and Holubek (2003) evaluated the content of images that featured on websites.
Other evaluation approaches have been proposed to more comprehensively measure the effectiveness of
websites; the Balanced Scorecard (BSC) is one such approach. Sterne (2002) in his book 'Proven methods for
measuring website success' acknowledges the growing popularity of the BSC tool for evaluating websites.
Developed by Kaplan and Norton in 1992, the BSC is a tool that "… translates an organization's mission and
strategy into a comprehensive set of performance measures that provide the framework for a strategic measurement
and management system …” (Balanced Scorecard Collaborative, 2002). The BSC measures performance from the
four perspectives: customer, financial, internal, and learning and growth. Morrison, Taylor, Morrison and Morrison
(1999) modified the BSC tool in order to evaluate the effectiveness of websites of small hotels in Scotland. Adapting
the four perspectives they used: customer, technical, marketing and internal. The authors decomposed these four
major perspectives into Critical Success Factors (CSF) that a website should meet to be effective. CSF is a term for
an element which is necessary for an organization or project to achieve its mission (Critical success factor, 2006).
Morrison et al. (1999) based their BSC tool on 25 CSFs for effective websites. Subsequently, other researchers
adopted a BSC approach for website evaluations (for a more comprehensive summary of website studies, see
Morrison, Taylor and Douglas, 2004).
Content analysis and BSC may be classified as “… heuristic usability methods …" that rely on the
subjective judgment of the website evaluators (Corigliano & Baggio, 2006; p. 320). According to Corigliano and
Baggio (2006), heuristic methods have been used in approximately 95% of website studies as compared to
“automated methods” that use readily available software tools for a more objective evaluation. Automated methods
include measures such as download time, browser compatibly and number of bad links. These are common
measures that facilitate comparisons across websites. In this study, a combined approach was used. For a detailed
and comprehensive evaluation the BSC tool as modified by Morrison et al. (1999) was selected. In addition, the
readily available automated methods were used to assess technical features of the websites. The next section
discusses the methods used for the competitive benchmarking.
METHOD
The modified BSC approach has been acknowledged to have the potential to provide a standardized
approach for the evaluation of tourism and hospitality websites (Morrison et al. 2004). A website evaluation
questionnaire based on the modified BSC approach and including perspectives (legal compliance, trip planner
assistance and link popularity) was used as the primary data gathering tool. Questionnaires with closed questions,
such as those used in this study, make it easier to interpret and compare answers since the data are structured
(Cooper & Schindler, 2003). Closed questions are also appropriate for collecting information that can be expressed
in statistical formats. The questionnaire had a numerical scale for measuring evaluators' opinion of specific features
in the assessed categories. The numerical scale has been defined by Zikmund (2000) as an attitude rating scale that
uses numbers as response options to identify response positions rather than verbal descriptions. Numerical scales
with numbers to indicate scale positions are often selected when the researcher wishes to assume interval-scale data
(Cooper & Schindler, 2003). Unlike an ordinal scale, adopting an interval scale does not limit the statistical
techniques that may be used for data analysis. For this reason the numerical scale was preferred to a Likert scale,
also an attitude rating scale, that indicates how strongly respondents agree or disagree with carefully constructed
statements relating to an attitudinal object. This avoided the main disadvantage of a Likert scale, namely that it is
hard to judge what a single summated score means in terms of attitudes of the respondents (Zikmund, 2000).
The seven main categories used in the questionnaire included: technical aspects, user friendliness, site
attractiveness, marketing effectiveness, link popularity, trip planner assistance and legal compliance. Technical
aspect scores such as link checks, html checks, spell checks and load-time were obtained directly from Net
Mechanic (accessible at www.netmechanic.com). These technical aspects were scored on a scale of 1 to 5. All the
other categories, excluding link popularity were rated on a scale of 0 to 4, with 0 indicating the absence of that
particular feature on the website and 4 representing the highest rating possible. Link popularity was measured by
Do'stlaringiz bilan baham: |
