Svein Ølnes
The central guidelines for the development of quality indicators for public web sites in Norway have been:
• the Governmental ICT policy as expressed by the Ministry of Government Administration, Reforms, and Church Affairs, especially the Government’ s recent new digitalisation programme( Ministry of Government Adm. 2012)
• Relevant laws, regulations and principles for public administration
• Publicly accepted standards and guidelines on the Net, formulated by the W3C, especially their recommendations for web accessibility, expressed through the recent Web Content Accessibility 2.0 Guidelines( WCAG)( W3C 2008)
Sørum et al.( 2009) analyzed 10 Scandinavian frameworks for website quality evaluation, among them the framework discussed in this paper, and mapped them to Delone and Mclean’ s Information Systems Success Model. The IS Success Model consists of the following quality dimensions: systems quality, information quality, and service quality( Delone and McLean 2002). The main quality dimensions influence the use and user satisfaction of the system and the result will be the net benefits of the system. The proposed metrics are therefore divided in these six categories.
Sørum et al. found that most of the frameworks covered the systems quality dimension best, with emphasis on accessibility issues, and that the coverage of the two other quality dimensions was weaker.
In the information systems literature SERVQUAL is a well known model for measuring service quality. The original SERVQUAL model had 10 dimensions of service quality, but this was later reduced to five: Tangibles, Reliability, Responsiveness, Assurance, and Empathy( Parasuraman et al. 1988). With more than 20 years of use the model has withstood a number of“ attacks” and citics( Ladhari 2009). In order to better capture the specific characteristics of the web the model WebQual has been developed on basis of the SERVQUAL model( Barnes and Vidgen 2000).
Although the SERVQUAL and WebQual models were developed for businesses and the private sector, they have also been used for public sector services, despite the obvious differences between the sectors( Boulter and Bendell 2010). Both support for the SERVQUAL model and caution against using it for public sector has been put forward( ibid.).
2.1 Heuristic models for measuring quality
Measuring quality of websites often rely on using heuristic methods and models. It has become the most used method for expert‐oriented evaluations( Donker‐Kuijer et al. 2010). Kahneman( Kahneman 2011) gives this( technical) definition of heuristics:“ A simple procedure that helps find adequate, though often imperfect, answers to difficult questions. The word heuristic has the same root as eureka“ I found it!””. A more practical definition concerning the use in website evaluations is“ all the sets of process guides, principles, criteria, tips and tricks, and guidelines that are available to support web designer”( De Jong and Van der Geest 2000). It should be added that in the ongoing process of developing and broadening the scope of public website evaluation the focus has shifted from almost solely assessing usability issues to add more and more governmental policy based issues such as the provision of digital services to the citizens.
The heuristic method for evaluating website quality was developed by Nielsen and Molich in the early 1990s( Jakob Nielsen 1994). In a heuristic evaluation, one or more experts check a given website using a predefined set of evaluation criteria, the heuristics( Donker‐Kuijer et al. 2010). The heuristics developed by Nielsen and Molich were primarily aimed at evaluation of user interfaces and consist of ten basic principles derived from studies of problems found in dealing with user interfaces. This basic set has since been applied to a wide variety of ICT applications including websites( ibid).
However, although the heuristic model involving an expert evaluation is much used, we do not know very much about how heuristics function( Donker‐Kuijer et al. 2010). Donker‐Kuijer et. al analysed five e‐ Government heuristics with respect( to a) context of use,( b) the information they cover,( c) their validity, and( d) their presentation format( ibid). Their conclusions were that the government heuristics are very complex documents difficult for( end) users to read and comprehend. Also information about the foundations of the heuristics is often missing making it difficult to judge the quality of the heuristics. Compliance with the heuris‐
391