WAES data are, for the most part, binary.  This requires that arithmetic manipulation of scores be done with care, so as not to mislead users about what the resulting data mean.

Let's say you are analyzing one of the WAES areas that is marked by a 0 or 1, such as T1b, and you would like to average the scores to make the data more manageable. 

Mindful that the raw scores are binary numbers, you add all scores for a particular category, and divide by the number of agencies. This yields an "average" of 0.67. This means that 67% of the sites for that country have a 1 for their T1b score, while conversely 33% of the sites for that country have a 0 for their T1b score. In this case, that means 67% of the sites in that country have a different webmaster within the agency than the one listed on the main government page (remember, if you look above you can see that this is the characteristic which T1b measures). By inference, these sites then are more likely to have active organizational involvement in the construction and content of the site. 

Note again this type of analysis only works for the coding areas that are measured by a 0 or 1, not those attributes that are measured by increments of 0.1. For the latter, if there is an average score for I4e of, say, 1.4, this means that of the sites which allow the user to complete and submit forms online, the average number of different forms available to completed and submitted online is 14. By inference, the more forms available, the greater the commitment the agency has towards serving citizens effectively. 

The Transparency Totals and Interactivity Totals columns, which can be seen in both the database and the spreadsheet, are simply the sum of all the WAES areas for each particular area. For example, the Transparency Totals columns is the sum of the scores of T1a through T4d (T5a and T5b are non-numerical measures, thus cannot be included in the Totals column). Higher Transparency Totals mean the agency in question provides a good amount of data on their website.