Time Series Forecasting of Students’ Mathematics Performance in Certificate of Secondary Education Examinations in Tanzania: Evidence from Mbeya Region Using ARIMA Models
Journal of Policy and Development Studies · 30.03.2026 · 20 (2) · 192-207
DOI: 10.4314/jpds.v20i2.14 ↗Kurz gefasst
Fachartikel in der Zeitschrift Journal of Policy and Development Studies, online erschienen am 30.03.2026. Verfasst von Boyd Anyigulile. Die Arbeit wurde bisher 1-mal zitiert (Crossref).
Abstract (Originalsprache)
This study examines and forecasts students’ performance in mathematics in the Certificate of Secondary Education Examination (CSEE) in Mbeya Region, Tanzania, using a time series modeling approach. The study utilized secondary data obtained from the National Examinations Council of Tanzania (NECTA), covering a 22-year period from 2000 to 2022 across 31 public secondary schools. A longitudinal time series research design was adopted, and the Autoregressive Integrated Moving Average (ARIMA) model was applied to analyze trends and generate forecasts of pass and fail percentages. Preliminary analysis revealed that the data series were non-stationary; however, first-order differencing achieved stationarity as confirmed by the Augmented Dickey-Fuller (ADF) test, along with ACF and PACF plots. Model selection based on the Bayesian Information Criterion (BIC) identified ARIMA (0,1,1) as the best-fitting model. Diagnostic tests confirmed the adequacy of the model, with residuals exhibiting no significant autocorrelation and satisfying normality assumptions. Forecast results indicate a declining trend in students’ mathematics performance, with pass rates projected to decrease from approximately 19.06% in 2023 to 14.53% in 2027, while failure rates are expected to increase over the same period. The findings highlight persistent challenges in mathematics education and underscore the need for timely, data-driven interventions. The study provides valuable insights for policymakers, educators, and stakeholders to improve teaching strategies, allocate resources effectively, and implement targeted support programs aimed at enhancing students’ academic outcomes in mathematics
Zitieren
APA
Anyigulile, B. (2026). Time Series Forecasting of Students’ Mathematics Performance in Certificate of Secondary Education Examinations in Tanzania: Evidence from Mbeya Region Using ARIMA Models. Journal of Policy and Development Studies, 20(2), 192-207. https://doi.org/10.4314/jpds.v20i2.14
Vancouver
Anyigulile B. Time Series Forecasting of Students’ Mathematics Performance in Certificate of Secondary Education Examinations in Tanzania: Evidence from Mbeya Region Using ARIMA Models. Journal of Policy and Development Studies. 2026;20(2):192-207. doi:10.4314/jpds.v20i2.14
BibTeX
@article{Anyigulile2026,
author = {Anyigulile, Boyd},
title = {Time Series Forecasting of Students’ Mathematics Performance in Certificate of Secondary Education Examinations in Tanzania: Evidence from Mbeya Region Using ARIMA Models},
journal = {Journal of Policy and Development Studies},
year = {2026},
volume = {20},
number = {2},
pages = {192--207},
doi = {10.4314/jpds.v20i2.14}
}
EUVIMED