Description
Breaks metrics down by participant groups in a table to see where experiences differ.
Cross-tabulation lays one variable against another, for example task success by device, or NPS category by customer tenure, to show how outcomes vary across segments. It is the first step in "why" analysis for benchmarks. Sauro advises collapsing sparse categories (such as merging 7-point responses into top-box and the rest) so cells have enough data, then testing differences with chi-square or N-1 tests. Done carelessly, slicing by many variables generates false positives, so pre-planned comparisons and multiple-comparison corrections matter.
Attributes
- Discipline
- Statistics & Modelling
- Loss type
- Active
- Goal
- Segment users
- Stage
- Evaluation, Post-launch
- Data
- Quantitative
- Effort
- Hours
Steps
- Choose the outcome and the grouping variables you planned to examine.
- Collapse sparse categories so each cell has adequate counts.
- Build tables with row or column percentages as appropriate.
- Test differences with chi-square, N-1 or Fisher tests, correcting for multiple comparisons.
- Present only meaningful splits, with counts shown.
Sources
- Benchmarking the User Experience (Sauro)
- Practical Statistics for Data Scientists (Bruce & Bruce)
- Myatt, G., Making Sense of Data: A Practical Guide to Exploratory Data Analysis and Data Mining (2007)
- Gratton, C. & Jones, I., Research Methods for Sports Studies (2010)
- Veal, A. J. & Darcy, S., Research Methods in Sport Studies and Sport Management (2014)
Also known as
- Cross tabs
- Contingency table analysis
Categories
- Analytics & Behaviour
- Statistics & Modelling