Logo
ScienceStack
Table of Contents
Fetching ...
Sign In

Analysis of ELSA COVID-19 Substudy response rate using machine learning algorithms

Marjan Qazvini

TL;DR

RF outperforms other models in terms of balanced accuracy, KNN in terms of precision and test accuracy, and logistics regressions in terms of the area under the receiver operating characteristic curve (ROC), i.e. AUC.

Abstract

National Statistical Organisations every year spend time and money to collect information through surveys. Some of these surveys include follow-up studies, and usually, some participants due to factors such as death, immigration, change of employment, health, etc, do not participate in future surveys. In this study, we focus on the English Longitudinal Study of Ageing (ELSA) COVID-19 Substudy, which was carried out during the COVID-19 pandemic in two waves. In this substudy, some participants from wave 1 did not participate in wave 2. Our purpose is to predict non-responses using Machine Learning (ML) algorithms such as K-nearest neighbours (KNN), random forest (RF), AdaBoost, logistic regression, neural networks (NN), and support vector classifier (SVC). We find that RF outperforms other models in terms of balanced accuracy, KNN in terms of precision and test accuracy, and logistics regressions in terms of the area under the receiver operating characteristic curve (ROC), i.e. AUC.

Analysis of ELSA COVID-19 Substudy response rate using machine learning algorithms

TL;DR

RF outperforms other models in terms of balanced accuracy, KNN in terms of precision and test accuracy, and logistics regressions in terms of the area under the receiver operating characteristic curve (ROC), i.e. AUC.

Abstract

National Statistical Organisations every year spend time and money to collect information through surveys. Some of these surveys include follow-up studies, and usually, some participants due to factors such as death, immigration, change of employment, health, etc, do not participate in future surveys. In this study, we focus on the English Longitudinal Study of Ageing (ELSA) COVID-19 Substudy, which was carried out during the COVID-19 pandemic in two waves. In this substudy, some participants from wave 1 did not participate in wave 2. Our purpose is to predict non-responses using Machine Learning (ML) algorithms such as K-nearest neighbours (KNN), random forest (RF), AdaBoost, logistic regression, neural networks (NN), and support vector classifier (SVC). We find that RF outperforms other models in terms of balanced accuracy, KNN in terms of precision and test accuracy, and logistics regressions in terms of the area under the receiver operating characteristic curve (ROC), i.e. AUC.

Paper Structure

This paper contains 19 sections, 40 equations, 17 figures, 4 tables.

Table of Contents

  1. Introduction
  2. Data
  3. Models and algorithms
  4. K-nearest neighbours (KNNs) algorithm
  5. Decision trees
  6. Ensemble methods
  7. Logistic regression
  8. Support Vector Classifier
  9. Neural Networks (NN)
  10. Classification metrics
  11. Results and discussion
  12. Model evaluation and interpretation
  13. Discussion
  14. Conclusion
  15. Appendix
  16. ...and 4 more sections

Figures (17)

  • Figure 1: The number of participants and proportion of non-participants per cohort
  • Figure 2: The number of participants and proportion of non-participants by interview mode
  • Figure 3: Employment status before and during the pandemic among participants in wave 1 and non-participants in wave 2
  • Figure 4: Activity levels during the pandemic among participants in wave 1 and non-participants in wave 2
  • Figure 5: The distribution of age among participants and proportion of non-participants at different ages
  • ...and 12 more figures