Alexopoulos Charalampos et al.
Figure 1: Evaluation dimensions and criteria of the PSI e‐Infrastructures evaluation model( continues)
4.2 Evaluation procedure
In order to finalise the construction of the evaluation framework, a comprehensive evaluation procedure was developed for the use of the evaluation model that was presented in section 4.1, which includes both quantitative and qualitative evaluation methods to get deeper insights.
The evaluation approach is shown in Figure 3. It consists of internal evaluation( by individuals within the e‐ infrastructure development consortium) and external evaluation( by individuals not belonging to the development consortium). The internal evaluation includes: i) a semi‐structured discussion in a group of experts, ii) Web Analytics and iii) a SWOT( Strengths‐Weaknesses‐Opportunities‐Threats) analysis. The external evaluation includes: i) a similar semi‐structured discussion in a group of social, political and management sciences researchers( potential users of the platform) not belonging to the project’ s scientific committee, ii) a structured usability test measuring user’ s performance on specific tasks( using a 5‐Likert scale from very difficult to very easy), iii) a quantitative and structured questionnaire and iv) a Qualitative Discussion with users.
The structured questionnaire follows the complete structure of the evaluation model and could be used for both traditional and advanced PSI e‐infrastructures with the appropriate changes. The users who fill the questionnaire are asked to enter the extent of their agreement or disagreement with the statements it includes, answering the question:“ To which extend do you agree with the following statements?”. A five point Likert scale is used to measure agreement or disagreement with( i. e. positive or negative response to) such a statement( 1 = Strongly Disagree, 2 = Disagree, 3 = Neutral, 4 = Agree, 5 = Strongly Agree).
The semi‐structured discussions in groups of experts( e. g. researchers from the social, political and management sciences who would be interested in using public sector data), on the topics / questions shown in Figure 4.
106