WAMMI Numeric Summary


Web Site: Demo Evaluation of a Web Site
Date of Analysis: 17:41 on 06 Apr, 2005

Return to Summary Page




Scale Median Mean Standard Dev.
Attractiveness 44 43.72 27.41
Controllability 53 ½ 52.06 28.03
Efficiency 63 63.12 25.94
Helpfulness 70 61.53 25.36
Learnability 51 ½ 52.66 23.44
Global Usability 56 ½ 54.25 21.34

How to read the table

WAMMI scores are expressed as percentiles which means that a score of 50, for instance, represents the average score for the scale: 50% of the websites will get a score of less than 50, and 50% of websites will get a score of 50 or more. A score of 70 represents the 70th percentile: 70% of the websites will get a score of less than 70, and 30% will get a score of 70 or higher. Thus the higher your scores, the more usable your web site is.


Notes


WAMMI Scales:
Please see the Graph of WAMMI Results page for a description of what the WAMMI scales mean.

Median:

This is the center-point of the array of the individual scores for each scale. You can see a listing of the actual scores the users gave on the individual user profiles page. Don't forget, a score of 50 is an average score, below 50 is below average, and above 50 is above average.

Mean:

This is the numerical average of the individual scores your users have rated you at.

Standard Deviation:

The standard deviation expresses the amount of variability in your data. For this kind of data, a reasonable value for the standard deviation is 20.00 If all your users are agreed on their evaluations of your web site, the standard deviations will be smaller. If your users have divergent opinions, the standard deviations will be much greater. Standard deviations over 30 suggest you have have two or more groups of users with very different opinions about the usability of your web site. It is not uncommon to find that the standard deviations are larger for some scales than for others. This indicates that there are differences in the amount of agreement between users about these scales.

Discrepancy between Median and Mean:

It may happen that there is a discrepancy between the median and the mean for one or two scales. What this indicates is that you have a small number of users who gave you some artificially high or low scores. It is wisest not to include these, and the median is the appropriate way to make the adjustment.

Return to Summary Page