Predictive Analysis of Student’s Academic Performance at Virginia Run Elementary School

Using Supervised Learning

Chapter 1: Introduction

Sufiyan Ahmed

2106313

Bay Atlantic University

February 21, 2024

1
 Chapter 1: Introduction

1.0 Overview

The predictive analysis of student academic performance, particularly utilizing supervised

learning techniques, stands as a critical endeavor within contemporary education. With the

increasing emphasis on personalized learning and educational equity, understanding and

addressing the needs of individual students have become paramount (Dumont & Ready, 2023;

Zhang et al., 2020). The study at hand aims to delve into this realm by focusing on the academic

performance of students at Virginia Run Elementary School. This research is necessitated by the

gap in current educational practices, where proactive measures for identifying students who may

require additional support are often lacking (Syed, 2023). While educators strive to cater to the

diverse needs of their students, the ability to predict and preemptively address academic

challenges remains somewhat elusive. By leveraging the power of predictive analytics and

supervised learning, this study seeks to bridge this gap by developing a robust model that can

forecast students' academic trajectories based on past performance data, such as grades, test

scores, and attendance records.

The implications of this study are manifold. Firstly, it holds the potential to revolutionize the way

educational interventions are designed and implemented, shifting from reactive to proactive

strategies. By identifying students who are at risk of academic underperformance early on,

educators can tailor their support mechanisms accordingly, thereby fostering a more conducive

learning environment for all students (Yağcı, 2022). Moreover, the findings of this research could

contribute to the broader discourse on data-driven decision-making in education, highlighting the

efficacy of leveraging technology to optimize student outcomes.

2
1.1 Background of the Study

The background of the study provides a comprehensive overview of the existing research

literature concerning predictive analysis of student academic performance and supervised

learning techniques in education. It traces the evolution of this field, highlighting its significance

in addressing the ongoing challenges faced by educators in effectively supporting students'

academic growth.

1.1.1 Evolution of Predictive Analytics in Education

The evolution of predictive analytics in education represents a transformative journey marked by

significant advancements in data science, machine learning, and educational research. Initially,

educational research primarily focused on descriptive and inferential statistics to understand

patterns and trends in student performance (Picciano, 2012). However, the advent of predictive

analytics introduced a paradigm shift towards proactive decision-making and personalized

interventions (Sghir et al., 2022). Early efforts in predictive analytics within education can be

traced back to the late 20th century, where researchers began exploring the use of statistical

modeling techniques to predict student outcomes (Picciano, 2012). Studies by Baker et al. (2008)

and Romero et al. (2008) laid the groundwork for predictive modeling by identifying correlations

between various student attributes and academic success. These pioneering works demonstrated

the potential of predictive analytics in forecasting student performance based on demographic

factors, prior academic achievement, and socio-economic status (Baker et al., 2008; Romero et

al., 2008).

As computational power and data availability continued to expand, researchers turned towards

more sophisticated machine learning algorithms, particularly supervised learning, to enhance

3
predictive accuracy. Supervised learning algorithms, such as decision trees, logistic regression,

and neural networks, gained prominence due to their ability to learn from labeled training data

and make predictions on unseen instances (El Aouifi et al., 2021; Sarker, 2021). Kotsiantis et al.

(2007) and Sanchez-Pozo et al. (2021) are among the researchers who explored the efficacy of

supervised learning techniques in predicting student academic performance across diverse

educational contexts. The proliferation of learning management systems (LMS), student

information systems (SIS), and educational data warehouses further accelerated the adoption of

predictive analytics in education (Picciano, 2012; Sghir et al., 2022). These systems generate vast

amounts of data, including grades, attendance records, and assessment scores, providing rich

sources of information for predictive modeling (Maraza-Quispe et al., 2021). Educational

institutions began leveraging predictive analytics to identify students at risk of academic

underperformance, intervene proactively, and allocate resources more efficiently (Maraza-Quispe

et al., 2021).

Moreover, the emergence of open-source machine learning libraries, such as scikit-learn and

TensorFlow, democratized access to predictive modeling tools, enabling educators and

researchers to develop custom predictive models tailored to their specific contexts (Wang et al.,

2019). Collaborative initiatives, such as the Predictive Analytics Reporting (PAR) Framework

and the Society for Learning Analytics Research (SoLAR), facilitated knowledge sharing and

best practices in the application of predictive analytics in education (Wang et al., 2019).

1.1.2 Current Challenges and Limitations

One of the primary challenges in predictive analytics is ensuring the quality and availability of

data (Nunn et al., 2016). Educational data often exhibit variability, incompleteness, and

inconsistencies, stemming from diverse sources such as student information systems, learning

4
management systems, and standardized assessments (Nunn et al., 2016). Poor data quality can

compromise the accuracy and reliability of predictive models, leading to erroneous predictions

and misguided interventions (Nunn et al., 2016). Moreover, data silos and interoperability issues

across different systems pose obstacles to aggregating and integrating data for comprehensive

analysis. The use of student data for predictive analytics raises significant privacy and ethical

concerns (Al-Tameemi et al., 2020). Educational institutions must navigate complex legal and

regulatory frameworks, such as the Family Educational Rights and Privacy Act (FERPA) in the

United States, to ensure compliance with data protection laws (Al-Tameemi et al., 2020).

Moreover, there is a risk of unintended consequences, such as stigmatization and discrimination,

resulting from the use of predictive models to profile and label students based on their predicted

outcomes (Al-Tameemi et al., 2020; Nunn et al., 2016).

Another critical challenge in predictive analytics is the presence of algorithmic bias, where

machine learning models perpetuate or amplify existing societal biases present in the training

data (Akter et al., 2022). Biased predictions can disproportionately impact marginalized and

underrepresented student populations, exacerbating inequities in educational opportunities and

outcomes (Akter et al., 2022). Researchers and practitioners must employ fairness-aware

techniques, such as fairness-aware learning algorithms and bias mitigation strategies, to mitigate

algorithmic bias and promote equitable decision-making in predictive analytics (Yang et al.,

2023). The black-box nature of many machine learning algorithms poses challenges in

interpreting and explaining the underlying factors driving predictive outcomes (Yang et al.,

2023). Educators and stakeholders may be hesitant to trust and act upon predictions generated by

opaque models without understanding the rationale behind them.

5
The existing literature highlights the gap between identifying students at risk of academic

underperformance and implementing timely interventions to support their learning needs. While

predictive models can effectively identify high-risk students based on historical data, the

translation of these insights into actionable interventions remains a critical challenge. Educators

often lack the resources, expertise, and infrastructure required to implement targeted

interventions at scale. Consequently, there is a pressing need for research that not only identifies

students in need of support but also provides actionable recommendations for personalized

interventions tailored to individual students' needs (Reber et al., 2018).

In conclusion, the current state of research underscores the urgent need for proactive intervention

strategies informed by predictive analytics and supervised learning techniques. By addressing the

gap between prediction and intervention, this study aims to contribute to the ongoing efforts to

promote educational equity and student success. Through a nuanced understanding of the

challenges and opportunities associated with predictive analysis in education, this research

endeavors to empower educators with the tools and insights needed to foster a culture of data-

driven decision-making and personalized learning.

1.2 Problem Statement

Recent advancements in predictive analytics have promised to revolutionize education by

providing insights into student academic performance and enabling proactive interventions.

However, despite these technological advancements, a staggering statistic reveals the persistent

challenge facing educators: Data released by the Department of Education indicate that over 40%

of students in urban public schools are at risk of academic underperformance in Virgina,

highlighting the urgent need for targeted intervention strategies (Cline, 2023). Despite extensive

research in the last five years, a meaningful gap remains in understanding how to effectively

6
translate predictive insights into actionable interventions tailored to individual student needs. The

general problem is the disconnect between predictive analytics' potential to identify at-risk

students and the implementation of proactive intervention strategies to support their academic

success. Specifically, this study aims to address the gap in the current research literature by

investigating how supervised learning techniques can be leveraged to predict student academic

performance and inform personalized intervention strategies at Virginia Run Elementary School.

1.3 Statement of Purpose

This quantitative study aims to explore the predictive capabilities of supervised learning

techniques in forecasting student academic performance at Virginia Run Elementary School.

Specifically, the study intends to develop predictive models that describe the relationship

between independent variables such as grades, test scores, and attendance records, and the

dependent variable of student academic performance. Additionally, the study will explore the

potential moderating effect of socio-economic status as a covariate variable on the relationship

between predictor variables and academic performance. By employing a quantitative approach,

this research seeks to provide actionable insights for educators to identify at-risk students early

and implement targeted interventions to support their academic success.

1.4 Research Question

The primary research question guiding this quantitative study is:

RQ: What is the predictive capability of supervised learning techniques in forecasting student

academic performance at Virginia Run Elementary School, considering grades, test scores, and

attendance records as predictor variables?

7
Null Hypothesis (H0): There is no significant relationship between grades, test scores, attendance

records, and student academic performance at Virginia Run Elementary School.

Alternative Hypothesis (H1): There is a significant relationship between grades, test scores,

attendance records, and student academic performance at Virginia Run Elementary School.

1.4.1 Variables

Independent Variables: Grades, test scores, and attendance records.

Dependent Variable: Student academic performance.

Covariate Variable: Socio-economic status.

Grades will be measured on a numerical scale (e.g., GPA), test scores will be measured as

numerical scores (e.g., standardized test scores), attendance records will be measured as the

percentage of days attended, and student academic performance will be measured using

standardized assessments or GPA.

1.4.2 Theoretical Foundation

The theoretical foundation of this study is grounded in Social Cognitive Theory (SCT), proposed

by Albert Bandura (1986). SCT posits that individuals learn through observation, imitation, and

modeling, emphasizing the reciprocal interaction between personal factors, environmental

influences, and behavior (Bandura, 1986). In the context of academic performance, SCT suggests

that students' beliefs in their capabilities (self-efficacy), environmental factors such as family

support and school resources, and behaviors such as study habits and attendance contribute to

their academic success (Bembenutty et al., 2016).

8
This study aligns with SCT as it investigates how external factors (grades, test scores,

attendance) influence students' academic performance, reflecting the interplay between personal,

environmental, and behavioral determinants of learning outcomes. By employing supervised

learning techniques to analyze historical academic data, the study seeks to elucidate the

predictive relationships between these variables, thereby contributing to our understanding of

how environmental factors shape academic achievement. Furthermore, by considering socio-

economic status as a covariate variable, the study acknowledges the socio-cultural context in

which learning occurs, echoing SCT's emphasis on the influence of environmental factors on

behavior and cognition.

1.5 Significance

By applying supervised learning techniques to predict student academic performance, this study

contributes to the advancement of educational theory, specifically in the areas of predictive

analytics and personalized learning. The findings of this research have the potential to enrich our

understanding of the complex interactions between student characteristics, environmental

factors, and academic outcomes, thus enhancing existing theoretical frameworks such as Social

Cognitive Theory and the Student Success Framework. From a practical standpoint, the insights

gained from this study can inform evidence-based decision-making and intervention strategies in

educational settings. Educators and administrators can utilize predictive models to identify

students at risk of academic underperformance early on and implement targeted interventions to

support their learning needs. Additionally, the development of a predictive analytics framework

tailored to the context of Virginia Run Elementary School can serve as a model for other

educational institutions seeking to leverage data-driven approaches to enhance student success.

9
This study addresses a significant gap in the current research literature by focusing on the

practical application of predictive analytics in predicting student academic performance at the

elementary school level. While existing studies have explored predictive modeling in higher

education contexts, there is limited research examining its effectiveness in elementary schools.

By filling this gap, the study provides valuable insights into the feasibility and efficacy of

employing supervised learning techniques to support early intervention efforts and promote

academic achievement among elementary school students.

1.5.1 Assumptions

Several assumptions underlie this study, which are critical to its meaningfulness and validity.

Firstly, it is assumed that the historical academic data collected from Virginia Run Elementary

School is accurate, reliable, and representative of students' academic performance. While efforts

will be made to ensure data quality and integrity through rigorous data cleaning and validation

processes, the accuracy of the data cannot be guaranteed. Secondly, the study assumes that the

predictive models developed using supervised learning techniques will generalize well to new

data and maintain their predictive accuracy over time. Although cross-validation techniques and

model evaluation metrics will be employed to assess the robustness of the models, there is

inherent uncertainty in their performance on unseen data. Lastly, it is assumed that the

implementation of proactive intervention strategies based on predictive insights will lead to

improvements in student academic performance. While theoretical frameworks and empirical

evidence support the efficacy of targeted interventions, the actual effectiveness of these

interventions may vary depending on various contextual factors and implementation fidelity.

1.5.2 Limitations

10
Despite the rigorous methodology employed in this study, several limitations warrant

consideration. Internal validity refers to the degree to which the observed effects can be

attributed to the independent variables rather than confounding factors or measurement errors

(Andrade, 2018). One potential limitation is the inability to control for all possible confounding

variables that may influence student academic performance, such as teacher quality, family

dynamics, and individual student characteristics. To mitigate this limitation, statistical techniques

such as multivariate analysis and propensity score matching will be employed to account for

known confounders and minimize their impact on the study outcomes (Andrade, 2018). External

validity pertains to the generalizability of study findings to other populations, settings, and

contexts (Andrade, 2018). The generalizability of the study findings may be limited by the

specific characteristics of Virginia Run Elementary School and its student population. To

enhance external validity, efforts will be made to ensure the representativeness of the sample and

the transparency of the research methodology, allowing other researchers to replicate the study in

different educational settings (Andrade, 2018).

Construct validity refers to the degree to which the operationalizations of the variables accurately

measure the theoretical constructs they are intended to represent (Westen & Rosenthal, 2003).

While measures of academic performance, such as grades and test scores, are commonly used

indicators, they may not fully capture the multidimensional nature of student achievement. To

address this limitation, multiple measures of academic performance will be utilized, including

standardized assessments, teacher evaluations, and longitudinal academic records, to provide a

more comprehensive assessment of student achievement. Biases inherent in the data collection

and analysis process may influence study outcomes. For instance, selection bias may arise if

certain groups of students are disproportionately included or excluded from the analysis

11
(Pannucci & Wilkins, 2011). To mitigate selection bias, random sampling techniques will be

employed to ensure the representativeness of the study sample. Additionally, researcher bias may

occur if subjective judgments or preconceived notions influence data interpretation (Baldwin et

al., 2022). To minimize researcher bias, inter-rater reliability checks and independent validation

of data coding will be conducted to enhance the reliability and objectivity of the study findings

(Baldwin et al., 2022).

To address these limitations, transparency and rigor will be prioritized throughout the research

process. Detailed documentation of the study methodology, data collection procedures, and

analytical techniques will be provided to enhance the transparency and reproducibility of the

study findings. Additionally, sensitivity analyses and robustness checks will be conducted to

assess the robustness of the results to different methodological assumptions and data

specifications. By acknowledging and addressing these limitations, this study aims to enhance

the validity and reliability of its findings, thereby contributing to the advancement of knowledge

in the field of educational research.

1.5.3 Definition of Terms

Confounding Variables: Confounding variables are extraneous variables that may influence the

relationship between the independent and dependent variables, leading to spurious or misleading

associations. In quantitative research, confounding variables are sources of bias that can distort

the true effect of the independent variable on the dependent variable. Controlling for

confounding variables is essential to establish causal relationships and ensure the internal

validity of the study (Pearl, 2010).

12
Construct Validity: Construct validity refers to the extent to which a measurement accurately

assesses the theoretical construct or concept it intends to measure. It involves demonstrating that

the operationalization of the construct aligns with its theoretical definition and that the

measurement instrument effectively captures the intended construct. Establishing construct

validity is essential for ensuring the meaningfulness and relevance of research findings. (Westen

& Rosenthal, 2003).

Predictive Analytics: Predictive analytics refers to the process of using historical data, statistical

algorithms, and machine learning techniques to forecast future events or outcomes. In the context

of this study, predictive analytics involves analyzing past academic performance data, such as

grades, test scores, and attendance records, to predict students' future academic performance.

(Al-Tameemi et al., 2020).

Supervised Learning: Supervised learning is a machine learning paradigm where the algorithm

learns from labeled training data, with each example being a pair consisting of an input object

(typically a vector) and a desired output value (also called the supervisory signal). The algorithm

then generalizes from the training data to make predictions or decisions about unseen data. In

this study, supervised learning techniques, such as decision trees, support vector machines, and

neural networks, are utilized to develop predictive models based on past academic performance

data (Kotsiantis et al., 2007).

Validity: Validity refers to the extent to which a research study measures what it claims to

measure. In quantitative research, validity encompasses various aspects, including internal

validity, external validity, and construct validity. Internal validity relates to the accuracy of the

causal inferences drawn within the study, external validity pertains to the generalizability of the

study findings to other populations or contexts, and construct validity concerns the extent to

13
which the study accurately measures the theoretical constructs under investigation. Ensuring

validity is crucial for establishing the credibility and trustworthiness of research findings (Patino

& Ferreira, 2018).

1.6 Organization of the Remaining Chapters

In Chapter 2, the literature review will delve into the existing research and scholarly works

relevant to predictive analytics in education, supervised learning techniques, and their

application in forecasting student academic performance. The chapter will synthesize key

findings from recent studies, theories, and frameworks related to the predictive modeling of

student outcomes. It will explore theoretical foundations, methodological approaches, and

practical implications of predictive analytics in educational settings. Additionally, the literature

review will identify gaps, inconsistencies, and areas for further investigation, setting the stage for

the current study's research questions and objectives. Chapter 3 will provide a detailed overview

of the methodology employed in conducting the study. It will outline the research design,

sampling procedures, data collection methods, and operationalization of variables. The chapter

will describe the specific machine learning algorithms and techniques utilized for predictive

modeling, including data preprocessing, model selection, and evaluation metrics. Additionally, it

will discuss ethical considerations, potential biases, and limitations inherent in the methodology,

along with strategies to mitigate these challenges. The methodology chapter aims to provide a

transparent and systematic framework for replicating the study and ensuring the validity and

reliability of the findings.

In Chapter 4, the focus will be on the data analysis process, where the predictive models

developed in Chapter 3 will be applied to the collected data from Virginia Run Elementary

School. The chapter will present descriptive statistics, model performance metrics, and

14
interpretation of results, highlighting significant findings and patterns identified through the

analysis. It will discuss the implications of the findings for theory, practice, and future research,

addressing the research questions posed in Chapter 1. Additionally, the chapter may include

visualizations, tables, and graphs to enhance the presentation of results and facilitate

comprehension. Chapter 5 will serve as the culmination of the dissertation, providing a

comprehensive summary of the study's objectives, methodology, findings, and implications. The

chapter will revisit the research questions and hypotheses, summarizing the key findings and

contributions of the study to the field of education and predictive analytics. It will discuss the

theoretical and practical implications of the findings, as well as their limitations and areas for

future research. Finally, the chapter will conclude with reflections on the significance of the

study and recommendations for educators, policymakers, and researchers interested in leveraging

predictive analytics to support student success.

15
 References

Akter, S., Dwivedi, Y. K., Sajib, S., Biswas, K., Bandara, R. J., & Michael, K. (2022).

Algorithmic bias in machine learning-based marketing models. Journal of Business

Research, 144(144), 201–216. https://doi.org/10.1016/j.jbusres.2022.01.083

Al-Tameemi, G., Xue, J., Ajit, S., Kanakis, T., & Hadi, I. (2020, June 1). Predictive Learning

Analytics in Higher Education: Factors, Methods and Challenges. IEEE Xplore.

https://doi.org/10.1109/ICACCE49060.2020.9154946

Andrade, C. (2018). Internal, external, and ecological validity in research design, conduct, and

evaluation. Indian Journal of Psychological Medicine, 40(5), 498–499.

https://doi.org/10.4103/IJPSYM.IJPSYM_334_18

Baker, R. S. J. d., Corbett, A. T., Roll, I., & Koedinger, K. R. (2008). Developing a generalizable

detector of when students game the system. User Modeling and User-Adapted

Interaction, 18(3), 287–314. https://doi.org/10.1007/s11257-007-9045-6

Baldwin, J. R., Pingault, J.-B., Schoeler, T., Sallis, H. M., & Munafò, M. R. (2022). Protecting

against researcher bias in secondary data analysis: Challenges and potential solutions.

European Journal of Epidemiology, 37(1), 1–10. https://doi.org/10.1007/s10654-021-

00839-0

Bandura, A. (1986). Social foundations of thought and action: A social cognitive theory.

Prentice-Hall.

Bembenutty, H., White, M. C., & DiBenedetto, M. K. (2016). Applying Social Cognitive Theory

in the Development of Self-Regulated Competencies Throughout K-12 Grades. The

16
 Springer Series on Human Exceptionality, 215–239. https://doi.org/10.1007/978-3-319-

28606-8_9

Cline, N. (2023, September 8). Despite post-pandemic efforts, Virginia student test scores remain

mostly flat. Virginia Mercury. https://virginiamercury.com/2023/09/07/despite-post-

pandemic-efforts-virginia-student-test-scores-remain-mostly-flat/

Dumont, H., & Ready, D. D. (2023). On the promise of personalized learning for educational

equity. Npj Science of Learning, 8(1). https://doi.org/10.1038/s41539-023-00174-x

El Aouifi, H., El Hajji, M., Es-Saady, Y., & Douzi, H. (2021). Predicting learner’s performance

through video sequences viewing behavior analysis using educational data-mining.

Education and Information Technologies, 26(5), 5799–5814.

https://doi.org/10.1007/s10639-021-10512-4

Kotsiantis, S. B., Zaharakis, I. D., & Pintelas, P. E. (2007). Machine learning: a review of

classification and combining techniques. Artificial Intelligence Review, 26(3), 159–190.

https://doi.org/10.1007/s10462-007-9052-3

Maraza-Quispe, B., Valderrama-Chauca, E. D., Cari-Mogrovejo, L. H., & Apaza-Huanca, J. M.

(2021). Predictive Model of Student Academic Performance from LMS data based on

Learning Analytics. ICETC ’21: Proceedings of the 13th International Conference on

Education Technology and Computers, 13–19. https://doi.org/10.1145/3498765.3498768

Nunn, S., Avella, J. T., Kanai, T., & Kebritchi, M. (2016). Learning Analytics Methods, Benefits,

and Challenges in Higher Education: A Systematic Literature Review. Online Learning,

20(2). https://doi.org/10.24059/olj.v20i2.790

17
Pannucci, C. J., & Wilkins, E. G. (2011). Identifying and Avoiding Bias in Research. Plastic and

Reconstructive Surgery, 126(2), 619–625.

https://www.ncbi.nlm.nih.gov/pmc/articles/PMC2917255/

Patino, C. M., & Ferreira, J. C. (2018). Internal and external validity: Can you apply research

study results to your patients? Jornal Brasileiro de Pneumologia, 44(3), 183–183.

https://doi.org/10.1590/S1806-37562018000000164

Pearl, J. (2010). Confounding, and Collapsibility In Causality: Models, Reasoning and Inference.

Cambridge University Press.

Picciano, A. G. (2012). The Evolution of Big Data and Learning Analytics in American Higher

Education. Online Learning, 16(3). https://doi.org/10.24059/olj.v16i3.267

Reber, R., Canning, E. A., & Harackiewicz, J. M. (2018). Personalized Education to Increase

Interest. Current Directions in Psychological Science, 27(6), 449–454.

https://doi.org/10.1177/0963721418793140

Romero, C., Ventura, S., & García, E. (2008). Data mining in course management systems:

Moodle case study and tutorial. Computers & Education, 51(1), 368–384.

https://doi.org/10.1016/j.compedu.2007.05.016

Sanchez-Pozo, N. N., Mejia-Ordonez, J. S., Chamorro, D. C., Mayorca-Torres, D., & Peluffo-

Ordonez, D. H. (2021). Predicting High School Students’ Academic Performance: A

Comparative Study of Supervised Machine Learning Techniques. 2021 Machine

Learning-Driven Digital Technologies for Educational Innovation Workshop.

https://doi.org/10.1109/ieeeconf53024.2021.9733756

18
Sarker, I. H. (2021). Machine Learning: Algorithms, Real-World Applications and Research

Directions. SN Computer Science, 2(3), 1–21. Springer. https://doi.org/10.1007/s42979-

021-00592-x

Sghir, N., Adadi, A., & Lahmer, M. (2022). Recent advances in Predictive Learning Analytics: A

decade systematic review (2012–2022). Education and Information Technologies, 28.

https://doi.org/10.1007/s10639-022-11536-0

Syed, D. (2023). Predictive Analysis of Students’ Learning Performance Using Data Mining

Techniques: A Comparative Study of Feature Selection Methods. Applied System

Innovation, 6(5), 86–86. https://doi.org/10.3390/asi6050086

Wang, Z., Liu, K., Li, J., Zhu, Y., & Zhang, Y. (2019). Various Frameworks and Libraries of

Machine Learning and Deep Learning: A Survey. Archives of Computational Methods in

Engineering, 31(9). https://doi.org/10.1007/s11831-018-09312-w

Westen, D., & Rosenthal, R. (2003). Quantifying construct validity: Two simple measures.

Journal of Personality and Social Psychology, 84(3), 608–618.

https://doi.org/10.1037/0022-3514.84.3.608

Yağcı, M. (2022). Educational data mining: prediction of students’ academic performance using

machine learning algorithms. Smart Learning Environments, 9(1).

https://doi.org/10.1186/s40561-022-00192-z

Yang, J., Soltan, A. A. S., Eyre, D. W., & Clifton, D. A. (2023). Algorithmic fairness and bias

mitigation for clinical machine learning with deep reinforcement learning. Nature

Machine Intelligence, 5(8), 884–894. https://doi.org/10.1038/s42256-023-00697-3

19
Zhang, L., Basham, J. D., & Yang, S. (2020). Understanding the implementation of personalized

learning: A research synthesis. Educational Research Review, 31, 100339.

https://doi.org/10.1016/j.edurev.2020.100339

20