aceModerate evidence
More
AI models predicted school dropout with 91% accuracy in a review of 15 studies
Study title: Effectiveness of Artificial Intelligence Models for Predicting School Dropout: A Meta-Analysis
- Evidence type
- Systematic review or meta-analysis
- Sample
- n = 199,015
- Sources
- 1 sourcePeer-reviewed
- Published
- 2 min read
- Truvace record
- TRV-2026-2854cite or check it
Study title: Effectiveness of Artificial Intelligence Models for Predicting School Dropout: A Meta-Analysis
A meta-analysis of 15 studies with 199,015 participants found that AI models predicted school dropout with a pooled accuracy of 91% (95% CI 89-93%). The studies were published up to May 2022 and drawn from seven databases.
Read the study abstract
School dropout is a major concern in the educational systems of all countries. In recent years, artificial intelligence is playing an important role in predicting school dropout in the different educational stages of formal education. In this context, it is crucial to know that these predictions are accurate and understandable. This meta-analytic study aims to investigate the effectiveness of dropout prediction models conducted until May 2022. The databases used are Web of Science, Scopus, PubMed, ERIC, PsyInfo,…
The gain
A meta-analysis of 15 studies covering 199,015 students found AI models predicted school dropout with a pooled 91% accuracy (95% CI 89-93%), with decision trees reaching 95.3%.
In brief
Decision tree models did best, at 95.3% (95% CI 93-98%), ahead of random forests, artificial neural networks, support vector machines, logistic regression and stacking ensembles. The authors say dropout is a major concern in education systems in every country, which is why the accuracy of such predictions matters.
The authors say more models should be tested on larger groups of students to confirm the findings.
Main points
- The review pooled 15 studies with 199,015 participants on AI dropout prediction.
- Searches covered Web of Science, Scopus, PubMed, ERIC, PsyInfo, Dialnet and Scielo, with studies included up to May 2022.
- Pooled accuracy across all AI models was 91% (95% CI 89-93%).
- Decision trees scored 95.3% (95% CI 93-98%), better than the other model types compared.
- The comparison models were random forest, artificial neural network, support vector machine, logistic regression and stacking ensemble.
- The analysis used a random-effects proportions model with 95% confidence intervals.
What the review found when it combined the studies
These are the review’s own pooled results, copied from its abstract. Truvace did not recalculate them.
Accuracy of artificial neural network models in predicting school dropout
Proportion 91.8% (95% CI 90% to 93%)
3 studies · search to May 2022
The review’s own words
The ANN model has a mean proportion effect size of 91.8% (95% CI = 90-93%) with significant variability
Accuracy of decision tree models in predicting school dropout
Proportion 95.3% (95% CI 93% to 98%)
5 studies · search to May 2022
The review’s own words
The DT model has a mean proportion effect size of 95.3% (95% CI = 93-98%) with significant variability
Accuracy of artificial intelligence models in predicting school dropout
Proportion 91% (95% CI 89% to 93%)
15 studies · 199,015 people · search to May 2022
The review’s own words
The accuracy shows a mean proportion effect size of 91% (95% CI = 89-93%).
Accuracy of random forest models in predicting school dropout
Proportion 90.6% (95% CI 88% to 93%)
3 studies · search to May 2022
The review’s own words
The RF model has a mean proportion effect size of 90.6% (95% CI = 88-93%) with significant variability
Code checks every number above against the review’s text before it is shown. The short labels are an AI model’s reading of that text, so the review’s own sentence sits under each one. A pooled figure is only as strong as the studies behind it. How Truvace uses pooled results
The full story
The researchers searched seven academic databases for studies of dropout prediction models published up to May 2022. They analysed 15 studies with a combined 199,015 participants, using a random-effects proportions model with 95% confidence intervals.
Across all models the pooled accuracy was 91% (95% CI 89-93%). Decision trees scored 95.3% (95% CI 93-98%) and significantly outperformed random forests, artificial neural networks, support vector machines, logistic regression and stacking ensembles.
The authors conclude that more models should be applied in dropout research with larger numbers of participants to confirm these findings and improve the quality of education.
Sources
1 on file- Multidisciplinary Journal of Educational Researchdoi:10.17583/remie.13342
Talk it over
Who gets this benefit, and who was left out of the study or story?
AI models predicted school dropout with 91% accuracy in a review of 15 studieshttps://truvace.com/item/effectiveness-of-artificial-intelligence-models-for-predicting-school-dropout-a-
Who gets this benefit, and who was left out of the study or story?
The gain A meta-analysis of 15 studies covering 199,015 students found AI models predicted school dropout with a pooled 91% accuracy (95% CI 89-93%), with decision trees reaching 95.3%.
Answer in the threadHow do we know? What can this kind of evidence show, and what can’t it?
The evidence This rests on a systematic review or meta-analysis with a sample of 199,015, from Multidisciplinary Journal of Educational Research.
Answer in the threadWhat does it leave open? What else would you want to know before acting on it?
The record’s own caveat The authors say the findings need confirmation, calling for more models to be applied to dropout prediction with larger numbers of participants.
Answer in the threadHave you seen AI used this way where you study, work or live? Did it go the same way?
Answer in the thread
The questions are the same on every record. The details under them are taken from this record as written.
What readers say