Juan Carlos Barahona and Andrey Elizondo
Components Information Attributes Latent and observed variables
Media quality
* Dichotomous: 1 / 0 ** Likert scale: 1‐7 throughout the site(**) 4.2.2 Title in the browser is meaningful(**) 4.3 Correct 4.3.1 Free of grammatical errors(**) 4.4 Current 4.4.1 The information is updated(*) 4.4.2 Degree of updating(**)
5. Efficiency 5.1 Navigability 5.1.1 Location of the site’ s pages are clearly shown(**) 5.1.2 The links are easily identified(**) 5.1.3 Interactive site map(*) 5.2 Flexibility 5.2.1 The files for downloading are provided in different formats(*) 5.3 Verifiability 5.3.1 The site’ s pages show author and source(**) 5.4 Usability 5.4.1 The domain works without using www in the address(*) 5.4.2 Type size can be adjusted by users(*) 5.4.3 There are no pages under construction or lacking content(*) 5.4.4 The content is displayed on the same page(*) 5.4.5 The site provides a 404 error page to guide the user(*)
6. Media reliability 6.1 Accessibility 6.1.1 Content in other languages(*) 6.1.2 No broken links exist(*) 6.1.3 Option to use hotkeys rather than mouse(*) 6.1.4 The site facilitates access from a mobile device(*) 6.1.5 Images and objects with Alt Text(*) 6.1.6 The site is compatible with the most recent versions of the three main browsers on the market(*) 6.2 Security 6.2.1 Digital certificate(*) 6.2.2 Data validation(*) 6.3 Visibility in search engines 6.3.1 The site has a“ sitemap” file(. txt or. xml)(*) 6.3.2 The site has description and keyword metadata(*) 6.3.3 The site has a " robots” file(. txt)(*) 6.4 Speed 6.4.1 Average load time(**) 6.4.2 Page weight(**)
Currently, there are 55 observed categorical variables( 16 are 7‐item Likert scales and 39 are dichotomous). The 24 factors are actually a mix of observed( 11) and latent variables( 13). A weighted average of groups of factors is used to build a score for each informational attribute. Equation( 1) describes this framework.
4. Evaluation process
Quality =
24
∑ i = 1
i
∑
⎡⎛
1 ⎢⎜ ⎢⎣ ⎝ M
*
M i
∑
= 1 i j = 1
⎞ ⎤ x ⎟ j
* Weighti
⎥; ⎠ ⎥⎦
Where i is the number of factors; M is the number of observed variables; X j is the observed variable; j is the number of observed variables for each factor variable; and W i is the factor weight.
Two independent evaluators performed separate appraisals on each institutional website. Their evaluation was compared and if both agreed on or assigned a similar value, the average of both independent evaluations
( 1)
58