Univariate Tests
Dependent variable: PostTest
| Source | Sum of Squares | df | Mean Square | F | Sig. | Partial Eta Squared |
|---|---|---|---|---|---|---|
| Contrast (Program) | 145.640 | 2 | 72.820 | 40.057 | < .001 | .589 |
| Error | 101.802 | 56 | 1.818 | - | - | - |
Lesson 06.07 | Printed pages 644-658
Build the full ANCOVA conclusion
The omnibus test tells you whether adjusted group differences exist. The rest of the output explains why the adjustment matters, what the fair comparison looks like, and where specific differences occur.
Novice rule
Do not read ANCOVA as a pile of separate tables. Follow one reasoning chain from raw patterns to a careful final conclusion.
Output 6
Read the adjusted omnibus test
The Univariate Tests table asks one overall question: after controlling for Pretest, do the three Program groups still differ on PostTest?
p < .001. The adjusted Program effect is statistically significant.
In this output, Contrast represents the overall Program test.
Partial η2 = .589, described in the workbook as a large effect.
What this supports
Program differences remain after controlling for Pretest. The table does not identify which program pairs differ. Pairwise comparisons answer that later question.
Decision rule
Know when the analysis must stop
The adjusted omnibus result controls whether follow-up comparisons are meaningful.
Open the contrasting non-significant example
A researcher compares three wellness programs on PostStress while controlling for BaselineStress.
| Source | df | F | Sig. | Partial Eta Squared |
|---|---|---|---|---|
| Program | 2 | 2.41 | .098 | .079 |
| Error | 57 | - | - | - |
Interpretation: After controlling for baseline stress, Program differences are not statistically significant. The workbook treats the effect as small, and pairwise follow-up should stop.
Integration
Give every output one job
The academic enrichment study uses one factor, Program; one outcome, PostTest; and one covariate, Pretest. Read the results in this order.
Descriptive Statistics suggest how the groups differ before adjustment. They are clues, not the final verdict.
Pretest strongly predicts PostTest, confirming that starting scores matter and should be controlled.
Program remains significant after controlling for Pretest, so adjusted group differences exist overall.
At the same Pretest level, Summer Bridge is highest, Saturday Academy is next, and After School is lowest.
Bonferroni comparisons determine which adjusted program pairs differ. They are interpreted only because the omnibus test is significant.
1Adjusted omnibus test
2Covariate effect
3Adjusted means
4Pairwise details
5Raw means
Avoid overreach
Statistical control does not make the groups equal, and ANCOVA alone does not prove causation. It makes the comparison fairer by accounting for the covariate.
Final synthesis
Turn the chain into one conclusion
A strong conclusion follows the analysis logic instead of listing tables one at a time.
Model conclusion
The academic enrichment story
Pretest scores strongly predicted PostTest performance, showing that baseline differences mattered. After controlling for Pretest, the overall Program effect remained statistically significant and large. Adjusted means placed Summer Bridge highest, followed by Saturday Academy and After School, while pairwise comparisons identified the specific program differences that remained after adjustment.
End-of-lesson checkpoint
The adjusted omnibus test is the anchor. It tells you whether to continue to adjusted means and pairwise comparisons or stop with a cautious conclusion.