Quantifying the User Experience: Practical Statistics for User ResearchElsevier, 2012 M03 16 - 312 pages Quantifying the User Experience: Practical Statistics for User Research offers a practical guide for using statistics to solve quantitative problems in user research. Many designers and researchers view usability and design as qualitative activities, which do not require attention to formulas and numbers. However, usability practitioners and user researchers are increasingly expected to quantify the benefits of their efforts. The impact of good and bad designs can be quantified in terms of conversions, completion rates, completion times, perceived satisfaction, recommendations, and sales. The book discusses ways to quantify user research; summarize data and compute margins of error; determine appropriate samples sizes; standardize usability questionnaires; and settle controversies in measurement and statistics. Each chapter concludes with a list of key points and references. Most chapters also include a set of problems and answers that enable readers to test their understanding of the material. This book is a valuable resource for those engaged in measuring the behavior and attitudes of people during their interaction with interfaces.
|
Contents
1 Introduction and How to Use This Book | 1 |
2 Quantifying User Research | 9 |
3 How Precise Are Our Estimates? Confidence Intervals | 19 |
4 Did We Meet or Exceed Our Goal? | 41 |
5 Is There a Statistical Difference between Designs? | 63 |
Part 1 Summative Studies | 105 |
Part 2 Formative Studies | 143 |
Other editions - View all
Common terms and phrases
adjustment alpha analysis available for discovery average benchmark binary binomial confidence interval cell Chapter chi-square test comparing comparison complete the task completion rate compute confidence level correlation critical difference critical value degrees of freedom Design Enterprise.com example Excel function Figure formula groups heuristic evaluations Interaction interface iteration Jeff Sauro large sample level of confidence levels of measurement margin of error McNemar exact test measure median methods Nadj Net Promoter Score normal distribution null hypothesis number of problems observed overall p-value paired t-test participants population probability problem discovery proportion PSSUQ psychometric range recommend reliability responses sample mean sample size estimation satisfaction scores small sample SMEQ standard deviation standard error standardized usability questionnaires subscales SUMI SUPR-Q t-distribution Tullis two-proportion test two-sided two-tailed Type I error typically usability metrics usability problems usability testing User Experience user research variability variance Wald WAMMI within-subjects z-score