The validity of Engagement and Feedback Assessments (EFAs): identifying students at risk of failing

BACKGROUND

Imperial College School of Medicine, London UK, introduced a new curriculum in 2019, with a focus on the GMC outcomes for graduates, and pedagogy best practice. The new curriculum included formative assessments, named engagement and feedback assessments (EFAs), to support learning, and attainment in the summative examinations. The aims of this study were to assess the validity of EFAs and to determine whether they have utility as a modified form of programmatic assessment to inform decision-making regarding possible interventions by measuring and analysing attendance at and performance in these formative events.

METHODS

Seven hundred and sixty-one students were included in the study and assessment results were included for academic years 2019/20 to 2020/21. Forty-one data points per student, (27 in Year 1 and 14 in Year 2) were used, to compare EFA scores with the summative performance. Attendance was monitored through engagement with the EFAs.

RESULTS

Cohort 1 (enrolled 2019): In year 1, EFAs were associated with summative exam scores (overall r = 0.63, p < 0.001). Year 2, EFA scores were also associated with summative scores (overall r = 0.57, p < 0.001), including the clinical practical assessment (r = 0.45, p < 0.001). Missing two or more EFAs was associated with a significant increase in the likelihood of failing one or more summative examinations in the first year (OR: 7.97, 95% CI 2.65-34.39) and second year (OR: 3.20, 95% CI 1.74-5.95). Missing more than two EFAs in their first year was also associated with a higher risk of failing a summative examination in the second year (OR: 2.47, 95% CI 1.33-4.71). Students who increased their attendance between year 1 and 2 fared better in summative assessment than those who maintained poor attendance, whereas those that reduced their attendance fared worse than those that maintained high attendance. Cohort 2 (enrolled 2020): Analysis of cohort 2 supported these findings and in this cohort missing two or more EFAs was again associated with an increased likelihood of failing a summative examination (OR = 4.00, 95% CI = 2.02-7.90).

CONCLUSION

Our EFA model has validity in predicting performance in summative assessments and can inform prospective interventions to support students' learning. Enhancing attendance and engagement can improve outcomes.

Predicting failure before it happens: A 5-year, 1042 participant prospective study

PURPOSE OF THE ARTICLE

Students who fail assessments are at risk of negative consequences, including emotional distress and cessation of studies. Identifying students at risk of failure before they experience difficulties may considerably improve their outcomes.

METHODS

Using a prospective design, we collected simple measures of engagement (formative assessment scores, compliance with routine administrative tasks, and attendance) over the first 6 weeks of Year 1. These measures were combined to form an engagement score which was used to predict a summative examination sat 14 weeks after the start of medical school. The project was repeated for five cohorts, giving a total sample size of 1042.

RESULTS

Simple linear regression showed engagement predicted performance (R²_adj = 0.03, F(1,1040) = 90.09, p < 0.001) with a small effect size. More than half of failing students had an engagement score in the lowest two deciles.

CONCLUSIONS

At-risk medical students can be identified with some accuracy immediately after starting medical school using routinely collected, easily analysed data, allowing for tailored interventions to support students. The toolkit provided here can reproduce the predictive model in any equivalent educational context. Medical educationalists must evaluate how the advantages of early detection are balanced against the potential invasiveness of using student data.

Student Performance Prediction with Short-Term Sequential Campus Behaviors

As students’ behaviors are important factors that can reflect their learning styles and living habits on campus, extracting useful features of them plays a helpful role in understanding the students’ learning process, which is an important step towards personalized education. Recently, the task of predicting students’ performance from their campus behaviors has aroused the researchers’ attention. However, existing studies mainly focus on extracting statistical features manually from the pre-stored data, resulting in hysteresis in predicting students’ achievement and finding out their problems. Furthermore, due to the limited representation capability of these manually extracted features, they can only understand the students’ behaviors shallowly. To make the prediction process timely and automatically, we treat the performance prediction task as a short-term sequence prediction problem, and propose a two-stage classification framework, i.e., Sequence-based Performance Classifier (SPC), which consists of a sequence encoder and a classic data mining classifier. More specifically, to deeply discover the sequential features from students’ campus behaviors, we first introduce an attention-based Hybrid Recurrent Neural Network (HRNN) to encode their recent behaviors by giving a higher weight to the ones that are related to the students’ last action. Then, to conduct student performance prediction, we further involve these learned features to the classic Support Vector Machine (SVM) algorithm and finally achieve our SPC model. We conduct extensive experiments in the real-world student card dataset. The experimental results demonstrate the superiority of our proposed method in terms of Accuracy and Recall.