Abstract
Introduction
Experiment
Results
Discussion
Conclusions

Acknowledgements
References
Appendices
Credits
Feedback

Back To Main

Logins & Passwords - What Users Want

Results

Since there are two sets of comparisons taking place, we will examine them one at a time. The first issue is the comparison of the time it took users to fill out the short registration form as opposed to the long registration form. This is illustrated in the chart below.

Column one represents the average time it took users to fill out the short form. Column two represents the average time it took users to fill out the long form. The vertical bars represent one positive and negative standard deviation from the average for each form. Although there is an apparent difference between the two times, it is not enough to be statistically significant, as a T-test yields a probability of only 0.612, below the range of significance. The rest of the calculations for this comparison can be seen here.

The other comparison examines the time differentials between the complex password and the simple password. The chart can be seen below.

Password type one is the simple password, and password type two is the complex password. Again, there is a noticeable difference between the two average times, but not enough to be statistically significant. A T-test yields 0.580, the rest of the calculations can be viewed here.

Strangely enough, there were more errors when subjects were given the simple passwords. One possible reason for this, which also lends itself to an explanation of the time variance between simple and complex passwords is the probability that users given a complex password actually stopped to commit their password to memory whereas users with a simple password assumed that they would remember it properly and went to the registration page too quickly. Analysis of this theory would require several questions on the subjective evaluation questionnaire that we had not thought of prior to the experiment.

The number of errors for the long form and simple form were almost identical.

Department of Computer Science: Direct questions and comments to the student editorial team