Chi-Square Test Calculator

It tests whether there is a statistically significant relationship between two categorical variables, such as which landing page variant a visitor saw and whether they converted. It works on counts organized into a table, one row per group and one column per outcome, and tells you whether the pattern of counts looks different from what pure chance would produce.

A p-value below 0.05 conventionally means the relationship between your groups and outcomes is statistically significant, meaning it is unlikely to have happened purely by chance. A p-value at or above 0.05 means the data does not give you enough evidence to rule out chance as the explanation.

It is a small adjustment applied specifically to 2x2 tables that makes the chi-square test slightly more conservative, since it compensates for count data being treated with a continuous statistical distribution. It is standard practice for 2x2 tables, which is why this calculator applies it by default whenever your table has exactly two rows and two columns, and you can turn it off if you want the uncorrected result.

Up to 5 rows and 5 columns, which covers the large majority of real comparisons, from a simple 2x2 test comparing two groups on two outcomes up to a larger table comparing several groups across several outcome categories.

There is no single number, but a common guideline is that every expected cell count in your table should be at least 5. If your groups are small or one outcome is rare, some cells can fall below that threshold, which makes the p-value less trustworthy even if the calculation itself still runs.

No, on its own it only tells you that a relationship exists somewhere in the table. To find out which specific groups or outcomes are driving that relationship, you typically need to look at the individual cell contributions or run pairwise comparisons between specific groups.