Student Academic Performance and Employment Outcomes

Project Proposal: Data Transformation and Exploratory Analysis

Title: Student Academic Performance and Employment Outcomes: A Data Analysis Using R, SQL, and GitHub

1. Project Overview

This project examines the relationship between students’ academic performance and employment outcomes after graduation. It will investigate whether GPA, academic achievement, and college major are associated with graduates’ likelihood of securing employment.

The project will apply data cleaning, wide-to-long transformation, SQL querying, and exploratory statistical analysis using R. APIs and JSON may be incorporated to retrieve supplementary employment data. Quarto, GitHub, and RPubs will support reproducibility and publication.

2. Research Questions

Primary Question:

Is academic performance associated with graduates’ likelihood of securing employment?

Secondary Questions:

• Do students with higher GPAs have higher employment rates?

• Does employment status vary by college major?

• How does academic performance differ between employed and unemployed graduates?

3. Dataset and Variables

The project will use a publicly available student performance and employment dataset, preferably from Kaggle.

Response Variable: Employment status (employed/unemployed).

Explanatory Variables: GPA, academic assessment scores, college major, and internship experience (if available).

The dataset must contain academic performance indicators suitable for wide-to-long transformation.

4. Research Methodology

This observational, quantitative study will involve:

• Data Collection: Obtain a publicly available dataset and, if feasible, supplementary API data in JSON format.

• Data Cleaning: Prepare and validate the dataset using R and dplyr.

• Data Transformation: Convert wide-format academic performance data into long format using pivot_longer() from tidyr.

• SQL Integration: Use SQLite to store, query, and summarize the data.

• Exploratory Analysis: Calculate descriptive statistics and create visualizations using ggplot2.

• Statistical Analysis: Examine associations between academic performance and employment outcomes, potentially using logistic regression.

• Reproducibility: Document the analysis in Quarto and maintain project files on GitHub.

5. Expected Outcomes

The project will identify patterns between academic achievement, college major, and employment outcomes. It will also demonstrate how data transformation, SQL, and R support efficient data analysis.

Findings will describe statistical associations rather than establish causation.

6. Project Deliverables

• Datasets: Original, cleaned wide-format, and transformed long-format CSV files.

• R and SQL Code: Reproducible scripts for data cleaning, transformation, querying, and analysis.

• API/JSON Integration: Supplementary employment data retrieval and processing, if feasible.

• Statistical Results: Descriptive statistics, visualizations, and appropriate statistical analysis.

• Quarto Report: A reproducible .qmd document containing the methodology, code, results, and conclusions.

• GitHub Repository: Project files, documentation, permitted datasets, and reproducibility instructions.

• RPubs Publication: A publicly accessible HTML report generated using Quarto.

• Final Presentation: A summary of the methodology, findings, and conclusions.

7. Tools and Technologies

• R/RStudio: Data cleaning, transformation, and statistical analysis.

• SQL/SQLite: Database management and querying.

• APIs/JSON: Supplementary data collection and processing, if available.

• GitHub: Version control and project sharing.

• Quarto: Reproducible reporting.

• RPubs: Online publication.

8. Project Submission

The final submission will include two primary links:

• RPubs: Published Quarto HTML report presenting the analysis and findings.

• GitHub: Repository containing the source code, SQL queries, Quarto document, and supporting project files.

  9. Conclusion

This project will investigate associations between academic performance and employment outcomes while demonstrating a reproducible data science workflow integrating R, SQL, data transformation, GitHub, Quarto, and RPubs.