GEE (Generalized Estimating Equations)
GEE (Generalized Estimating Equations) is one of the statistical analyses applied automatically in MerQur. This page illustrates — on a Education Sciences sample dataset — how the analysis is run, what the MerQur output looks like, and how the result is reported in APA 7 format.
🎯 What is it for?
GEE (Generalized Estimating Equations) automatically performs, in the background, all the assumption checks required for the relevant data type (normality, homogeneity of variance, etc.) and presents the results with a clear table and chart. Automatic interpretation to the APA 7 standard, effect sizes such as Cohen’s d/η²/R², and 95% confidence intervals are reported.
📌 When is it used?
- Statistical analysis of measurements in the Education Sciences domain
- To produce APA 7-compatible result tables for academic publications
- Hypothesis testing and decision-making processes
- Undergraduate / master’s / PhD theses after the appropriate method has been selected
📐 Assumptions
- Appropriate scale — Variables must be at the measurement level required by the analysis (nominal/ordinal/interval/ratio)
- Independent observations — Observations must come from individuals independent of one another
- Sufficient sample size — The minimum n requirement for the analysis must be met
- Outlier check — Outliers must be detected and evaluated
If assumptions are violated, MerQur automatically suggests a non-parametric or robust alternative.
🛠 How to do it in MerQur
Load the data. Select the sample file from File → Open. MerQur auto-detects column types.
Select the analysis. From the left side select GEE (Generalized Estimating Equations).
Panel assignments (form fields in the program):
- error:
cats() got an unexpected keyword argument 'max_unique'
Optional settings. Effect size ✓ · 95% confidence interval ✓ · Assumption checks (automatic).
▶ Run — click the button. Results are produced automatically as a table + chart.
📄 Export to Word. APA 7-formatted report with italic statistical symbols.
📊 Sample Dataset — Education Sciences
ℹ Note: The scenario, MerQur output and interpretation below were produced by actually running the real example dataset in MerQur. Numeric results on your own data will differ; the goal is to show how the analysis is set up and interpreted end-to-end.
🎬 Example File
This analysis is demonstrated on the following example dataset for Education Sciences:
Egitim_Bilimleri/55_gee_motivation_group.xlsx
🎬 Scenario
Here we ran a longitudinal intervention study where 300 student records track motivation across repeated visits, with each student assigned to either a control or an intervention condition. Because the same students are measured at several visits, their motivation scores are correlated over time. Generalized estimating equations let us model the population-averaged effect of group and visit on motivation while accounting for this within- subject correlation through a working correlation structure, giving us robust estimates of how the intervention shifts motivation on average.
⚙️ Variable Selection
- Dependent variable: motivation
- Subject (cluster id): student_id
- Within-subject (time): visit
- Predictor (group): group (control/intervention)
Data Preview (First 5 Rows)
| student_id | visit | group | motivation |
|---|---|---|---|
| 1 | 1 | control | 3.24 |
| 1 | 2 | control | 3.46 |
| 1 | 3 | control | 3.03 |
| 1 | 4 | control | 3.66 |
| 1 | 5 | control | 2.69 |
n = 300 · Columns: student_id, visit, group, motivation
📈 MerQur Output
💬 Interpretation
We took repeated (multi-visit) motivation measurements from the same students — one student’s measurements are dependent. GEE estimates the relationship at the “population average” level in such clustered data, accounting for the correlation structure. The visit effect is non-significant (p = 0.24): motivation did not change systematically over time. A non-significant result is information too — the intervention/time may not have changed motivation. GEE is an alternative to LMM: rather than modelling random effects, it corrects the correlation among observations and gives the marginal (average) effect.
⚠ Common Mistakes
- Misidentifying the data type (e.g., loading a categorical variable as numeric)
- Skipping assumption checks and going straight to the p-value
- Failing to report effect size — APA 7 requires both p and effect size
- Failing to apply a Type I error correction (Bonferroni/Tukey) in multiple comparisons
- Not switching to a non-parametric alternative when n is insufficient
📹 Video Walkthrough
Watch the video below for an end-to-end walkthrough of this analysis on a Education Sciences file.
▶ GEE (Generalized Estimating Equations) — video walkthrough
This analysis is demonstrated on a sample dataset from Anaesthesiology (the steps are identical across disciplines). The link jumps straight to 17:27. Narration is in Turkish.
🧪 Natural Sciences & Mathematics · 🏛 Architecture, Planning & Design · ⚙ Engineering · 🏥 Health Sciences · 📊 Social, Humanities & Admin Sciences · 🏃 Sport Sciences · 🌾 Agriculture, Forestry & Aquatic
📚 If You Used This Analysis, Cite MerQur
If you performed this analysis using MerQur in a scientific study, please use the citation below as part of your academic citation obligations (APA 7):
Örücü, Ö. K. (2026). MerQur: Integrated Academic Data Analysis & Reporting Platform [Computer software] (Version 1.0.0). https://doi.org/10.53463/merqur.2026001
For BibTeX, RIS, EndNote and the English citation form: all citation formats →
- American Psychological Association. (2020). Publication manual of the American Psychological Association (7th ed.).
- Field, A. (2018). Discovering statistics using IBM SPSS Statistics (5th ed.). Sage.
- Cohen, J. (1988). Statistical power analysis for the behavioral sciences (2nd ed.). Lawrence Erlbaum.