Skip to content

Latest commit

 

History

History
80 lines (55 loc) · 5.51 KB

README.md

File metadata and controls

80 lines (55 loc) · 5.51 KB

DATA ANALYSIS (ADVANCED STATISTICS) PROJECT 2 WITH SPSS

Author: Patrick Chimadzuma

📋 Project Overview

This repository showcases a comprehensive advanced statistics portfolio, completing a full academic project for EDF 652: Educational Statistics. The analysis demonstrates mastery of a wide range of statistical techniques, from foundational descriptive statistics to complex inferential models, using IBM SPSS for data analysis and interpretation.

📁 Repository Contents

This portfolio contains the following key deliverables:

  • EDUCATIONAL STATISTICS PROJECT SOLUTIONS.pdf: The complete solution set featuring:
    • Full IBM SPSS output tables for all advanced analyses.
    • Detailed step-by-step interpretations and conclusions for each statistical test.
    • Answers to all 8 complex project questions.
  • EDUCATIONAL STATISTICS PROJECT QUESTIONS.pdf: The original project instructions and question paper, providing full context for the analyses performed.
  • DRUG_TRIAL.SAV DATA
  • HEAD TEACHER_STD7_ENGLISH (1).SAV DATA
  • WEIGHT_GENDER).SAV DATA

🔬 Summary of Advanced Analyses Performed

This project demonstrates advanced competency in statistical analysis, including:

  • Descriptive Statistics & Estimation: Calculation of weighted means from grouped data and point/interval estimation for population parameters.
  • Inferential Statistics:
    • T-Tests: Independent samples t-test to compare group means (Physics vs. Chemistry majors).
    • ANOVA: One-Way Analysis of Variance to test for differences in means across multiple groups (e.g., Divisions), including post-hoc analyses.
    • ANCOVA: Analysis of Covariance to compare group means on a dependent variable after controlling for a covariate (creativity).
    • Multiple Linear Regression: Building a regression model to predict a dependent variable using multiple predictors, including interpretation of R, R², coefficients, and significance.
  • Non-Parametric Statistics:
    • Mann-Whitney U Test: Used for comparing two independent groups when the assumption of normality is violated.
    • Kruskal-Wallis Test: The non-parametric equivalent of ANOVA for comparing three or more independent groups.
  • Normal Distribution Applications: Practical application of z-scores and the empirical rule to solve real-world problems within a normal distribution.

🧠 Key Findings

The project yielded several robust conclusions from diverse datasets:

  • Group Comparisons: Successfully identified significant differences in student performance across different educational divisions using ANOVA.
  • Predictive Modeling: Built a significant regression model, identifying Age of learner, Gender, and Division as statistically significant predictors of academic performance.
  • Covariate Control: Utilized ANCOVA to control for creativity, revealing true group differences in problem-solving skills that were not apparent in a standard ANOVA.
  • Non-Parametric Results: Applied appropriate non-parametric tests (Mann-Whitney U, Kruskal-Wallis) to analyze non-normally distributed data, finding significant differences in weight by gender and in statistics scores across drug trial groups.

🔧 How to Navigate This Repository

For recruiters and reviewers, the provided PDFs are the primary artifacts:

  1. For the Full Analysis: Open EDUCATIONAL STATISTICS PROJECT SOLUTIONS.pdf. This document is a complete report, showcasing not only the statistical output but, most importantly, the ability to derive meaning and business/educational insights from complex data.
  2. For the Project Scope: Open EDUCATIONAL STATISTICS PROJECT QUESTIONS.pdf to understand the original, complex requirements that the analysis successfully addressed.

ℹ️ Note on Reproducibility

  • The analytical outputs are presented in their final form within the solution PDF, which includes all necessary SPSS output tables.
  • The original syntax files (.sps) are not included in this repository to ensure privacy and confidentiality, adhering to academic and ethical standards.
  • The comprehensive explanations and statistical conclusions contained in the SOLUTIONS.pdf fully document the methodology, results, and professional interpretation.

🛠️ Skills Demonstrated

This project provides concrete evidence of advanced, industry-relevant skills:

  • Statistical Software Proficiency: Expert use of IBM SPSS for complex data analysis.
  • Statistical Methodology: Application of both parametric (t-test, ANOVA, ANCOVA, Regression) and non-parametric (Mann-Whitney U, Kruskal-Wallis) techniques.
  • Data Interpretation & Storytelling: Advanced ability to interpret statistical output and translate complex results into clear, actionable conclusions.
  • Experimental Design & Hypothesis Testing: Strong understanding of designing statistical tests and validating assumptions.
  • Technical Reporting: Professional presentation of sophisticated analytical results suitable for an academic or business intelligence context.

👨‍💻 Author

Patrick Chimadzuma

  • Master of Education Candidate

📄 License

This project is for academic and professional portfolio purposes. The data is used under educational fair use and has been anonymized to protect privacy.

This project represents my original work for EDF 652: Educational Statistics and serves as a demonstration of my advanced analytical capabilities.