Upload a CSV, pick your outcome, groups, and covariate, and get the full analysis of covariance — raw vs adjusted means, the adjusted significance test, pairwise differences, and an assumption check. Free.
Free analyses run on up to 10,000 rows. Larger files are randomly sampled to that size — sign up to analyze your full dataset.
Adjusting your group comparison...
Sent to — raw vs adjusted means chart, covariate scatter, ANCOVA table, adjusted means with pairwise differences, the assumption check, R code, and AI insights.
Analyze another fileThe analysis fits a linear model of your outcome on the group and the covariate together (analysis of covariance). It reports each group's raw mean next to its adjusted (least-squares) mean — the model's prediction for that group at the covariate's grand mean — with standard errors from the model. The omnibus F-test asks whether the groups differ after adjustment; pairwise adjusted differences are tested with Holm-corrected t-tests; the covariate's slope is reported with a 95% confidence interval. A homogeneity-of-slopes diagnostic refits the model with a group-by-covariate interaction and warns prominently when the equal-slopes assumption underlying the adjustment looks violated.
Use it when you want to compare groups on a numeric outcome but the groups differ on a numeric variable that also drives the outcome — baseline scores, size, exposure, spend — so raw means would compare apples to oranges.
Not for categorical covariates (use stratified group comparison), for multiple covariates at once (use regression with the group as a predictor), or when the covariate is itself affected by the grouping — adjusting for a consequence of the treatment biases the answer.
Built for: Analysts, researchers, and operators comparing groups that start from unequal footing
Typical data source: Any spreadsheet or CSV with a numeric outcome, a group column, and a numeric covariate such as a baseline measurement
One numeric outcome, one group column, and one numeric covariate. For example, final exam scores by study program with each student's baseline score:
Minimum 10 rows · Best with 30-50,000 rows and 2-8 groups
Standard-library analysis: do your groups differ once you adjust for a covariate? Analysis of covariance fits outcome ~ group + covariate, reports raw versus adjusted (least-squares) group means side by side, the omnibus adjusted group test, pairwise adjusted differences with multiplicity correction, the covariate's slope with a confidence interval, and a homogeneity-of-slopes diagnostic that warns you when the ANCOVA assumption itself is questionable.
Each group's raw mean next to its covariate-adjusted mean — when the two disagree, the covariate imbalance was distorting the raw comparison.
The outcome against the covariate, colored by group — shows the imbalance being corrected and whether the group trends are parallel.
The model's F-tests: covariate effect, the adjusted group test (the headline), and the equal-slopes diagnostic.
Adjusted (least-squares) means with standard errors and confidence intervals — the fair comparison at a common covariate value.
The model, the adjustment, the Holm correction, the slopes diagnostic, and the assumptions the adjusted comparison rests on.
Plain-English interpretation — what the numbers mean, what's significant, and what to do next.
Did the program work, or did its group just start ahead?
Map the final score as the outcome, the program as the group, and the baseline score as the covariate. ANCOVA compares the programs as if every participant had started at the same baseline — raw and adjusted means are shown side by side so you can see exactly how much the starting-point imbalance was distorting the picture.
See our FAQ for details on pricing, data privacy, and how the analysis works. Every report includes a Methodology section showing the statistical test, assumptions checked, and diagnostics run.
Run any analysis on your own data — validated R analyses, interactive reports, AI insights, and PDF export.
Try Free — No Credit CardTell us what went wrong, in your own words. We capture the page you're on automatically, so no need to describe where you are.