Quantifying the User Experience

Chapter 4

Did We Meet or Exceed Our Goal?

Introduction

Confidence intervals are an excellent way for showing both the location and precision of an estimate. All other things being equal, estimates based on larger sample sizes will have a more precise estimate of the unknown population mean or proportion.

Once you get used to using confidence intervals with small sample sizes you may start to notice that the boundaries of the interval can be rather wide. For example, if eight out of nine users complete a task, we can be 95% confident that actual population completion rate is between 54.3% and 99.9%. While this interval is informative (e.g., there’s a very small chance the completion rate will be less than 50%) there is still a lot of uncertainty. ...