Re‐envisioning the purpose of early warning systems: Shifting the mindset from student identification to meaningful prediction and intervention

Abstract

Emphasis in school dropout literature has shifted from exploring wide‐ranging causes of dropping out to soliciting a smaller number of predictive indicators to identify students at increased risk for dropping out. However, much of the past decade's Early Warning research excludes indicators that do not add to the predictive nature of the model even if they might inform intervention. Considering this shift and the relative infancy of Early Warning research, this paper presents a critical systematic review of the literature surrounding dropout and Early Warning System(s) (EWS). Through a focus and evaluation of student‐level indicators, we present an argument of how an effective EWS can bridge the gap between the prediction of dropout events and underlying causes, while providing actionable information for schools to intervene. Finally, we present preliminary data that demonstrate the potential of elementary‐level student‐level indicators as an avenue for shifting the focus of EWS from student identification to meaningful prediction and intervention. The results show how the student‐level indicators explored in this paper can effectively identify students at these grade levels who are on‐ and off‐track to graduation. We discuss the implications of evidence warranting the position that dropping out of school can be predicted as early as the elementary grade levels and its potential to drive future research in this area.

Insights of Instructors and Advisors into an Early Prediction Model for Non-Thriving Students

“…The earliest and best predictor for non-thriving found by the model was missing two or more weekly preparation homework assignments. This model is in line with previous models in the sense that early grade points and early engagement cues were found to be predictive of the course final grade (Gray & Perkins, 2019;Williams et al, 2021), hence allowing for early prediction that is actionable and useful in assisting at-risk students (Balfanz & Byrnes, 2019;McMahon & Sembiante, 2020).…”

“…Recent literature reviews of early warning systems have argued that for an intervention to be effective, it should better be tailored to student needs. However, many schools lack the resources for such interventions (Balfanz & Byrnes, 2019;McMahon & Sembiante, 2020). Therefore, a short meeting may serve as the first low-cost step towards assisting the detected student.…”

“…Interestingly, our participants did not agree on whether those students who are predicted to be thriving should be notified. In practice, early warning systems are mostly used for detecting at-risk students (Balfanz & Byrnes, 2019;McMahon & Sembiante, 2020). The idea of notifying thriving students, however, is well established in many educational systems, particularly in higher education institutions where student honour systems have been implemented for decades and have proven success in maintaining students' quality of academic work (e.g., Seaver & Quarton, 1976).…”

Penanganan Peserta Didik Berkebutuhan Khusus dengan Hambatan Speech Delay melalui Model IPTERE di Tarbiyatul Athfal

“…Prediksi tentang perlakuan atau treatment yang tepat serta intervensi pada PDBK menjadi program awal pada sekolah inklusi (Sonnander, 2000). Identifikasi awal juga menjadi peringatan dini tentang hambatan PDBK yang harus dicarikan solusi agar mendapat terapi yang efektif (McMahon & Sembiante, 2020). Kurikulum modifikasi baru dapat dibuat setelah ada identifikasi awal yang terdata dengan jelas.…”

Early Prediction of a Team Performance in the Initial Assessment Phases of a Software Project for Sustainable Software Engineering Education

“…The performance prediction of students is one of the most challenging domains. Significant feedback is associated with the early prediction of student performance, such as modeling the learning behavior of students, adjusting and improving the academic environment, catering to students' issues, and engaging in decision-making practices based on data analysis [4]. The predictive analysis based on AI is a significant trait for the development of the sustainable conceptual framework for engineering projects.…”