MBA Admission Equity Audit

Analysis of MBA admissions equity and inclusion using synthetic data – evaluates gender and racial disparities with insights from Human Capital Theory, Social Reproduction, and SDG 4.

---

📘 README

Equity and Inclusion Analysis Using Synthetic Data

This project examines MBA admission outcomes through the lens of equity and inclusion, analyzing whether applicants are treated fairly across gender and race, and exploring possible social reproduction patterns within institutional selection.

⚠️ Important: Dataset is synthetic and does not represent real applicants.

---

1️⃣ Project Context and Theoretical Framework

Objective

• Evaluate fairness in MBA admissions by gender and race
• Identify potential structural barriers or biases in selection

Theoretical Framework

• Human Capital Theory (Becker, 1964): Skills, knowledge, and qualifications (GPA, GMAT) increase productivity and admission chances
• Social Reproduction Theory (Bourdieu, 1970): Institutions can reproduce social inequalities, favoring candidates whose background matches system expectations
• SDG 4 – Quality Education: Ensures equitable access to quality education. Disparities in admission conflict with this goal

---

2️⃣ Data Source and License

• Origin: Synthetic data inspired by aggregated statistics from Wharton Class of 2025
• Platform: Kaggle - MBA Admission Dataset
• License: Apache 2.0
• Usability Score: 10/10

Contains no real individual information; purely illustrative

---

3️⃣ Data Preparation and Cleaning

1. Imported dataset into Python
2. Cleaning and preprocessing:
   • race = null → replaced with International
   • admission = null → interpreted as Deny
3. Imported into Power BI via Python script
4. Final validation and transformations in Power Query before report building

---

4️⃣ Dataset Structure

Column	Description
application_id	Unique identifier
gender	Male / Female
international	TRUE / FALSE
gpa	GPA on 4.0 scale
major	Business / STEM / Humanities
race	White / Black / Asian / Hispanic / Other / International
gmat	GMAT score (out of 800)
work_exp	Years of work experience
work_industry	Consulting, Finance, Technology, etc.
admission	Admit / Waitlist / Deny

---

5️⃣ Power BI Report Structure

Page 1 – Applicant Landscape

• Gender and race distribution
• Major and industry distribution
• Academic background (GPA, GMAT)
• Key finding: 63.66% of applicants were male at entry

Page 2 – Institutional Selection Logic

• Distribution of GMAT, GPA, and work experience
• Analysis of selection thresholds
• Checks whether selection is merit-based

Page 3 – Equity Audit: Admission Outcomes & Social Reproduction

Gender Outcomes

• Admission: 50% Male / 50% Female
• Despite male overrepresentation at entry (63.66%), outcomes achieve parity ✅

Racial Outcomes

Race	Entry %	Admitted %	Δ	GMAT avg (Admitted)
International	29.74%	30.89%	↑ +1.15	689.89
White	23.51%	27.11%	↑ +3.60	692.42
Asian	18.52%	21.11%	↑ +2.59	694.11
Black	14.79%	8.89%	↓ -5.90	699.00
Hispanic	9.62%	6.89%	↓ -2.73	699.52
Other	3.83%	5.11%	↑ +1.28	685.87

Analysis:

• Black & Hispanic applicants have the highest GMAT averages, yet their representation declines at admission
• White, Asian, International, and Other increase or remain stable
• High entry performance does not guarantee equity at exit → social reproduction effect

---

6️⃣ Key Metrics: Average GMAT (Admitted)

Race	GMAT avg (Admitted)	Δ vs Overall Avg (692.73)
Hispanic	699.52	+6.79
Black	699.00	+6.27
Asian	694.11	+1.38
White	692.42	-0.31
International	689.89	-2.84
Other	685.87	-6.86

Insight:

• Hispanic and Black applicants must exceed the overall average to be admitted
• International and Other admitted below average
• Suggests structural barriers for certain groups

---

7️⃣ Entry vs Exit Percentages

Race	Entry %	Admitted %	Change	Trend
International	29.74%	30.89%	↑ +1.15	✅
White	23.51%	27.11%	↑ +3.60	✅
Asian	18.52%	21.11%	↑ +2.59	✅
Black	14.79%	8.89%	↓ -5.90	❌
Hispanic	9.62%	6.89%	↓ -2.73	❌
Other	3.83%	5.11%	↑ +1.28	✅

Interpretation:

• Black and Hispanic representation declines sharply despite highest GMAT averages
• Shows some groups face higher thresholds to compete

---

8️⃣ Limitations

• Dataset is synthetic
• Results do not reflect real institutions
• Findings are illustrative and educational only

---

9️⃣ Conclusion

• Gender: Outcomes achieve parity, correcting male overrepresentation
• Race: Disparities persist despite similar academic profiles
• Selection is not perfectly neutral; structural barriers exist
• Highlights the need for multi-dimensional evaluation of admissions processes

Even with high merit indicators, some groups face disadvantages → relevance of Human Capital Theory, Social Reproduction Theory, and SDG 4

---

✅ Executive Insights

1. Entry-level distribution: relatively balanced
2. Exit-level outcomes: some groups increase, others decrease
3. High academic performers (GMAT) are not necessarily the most represented at admission

Reveals partial inequity and social reproduction in MBA admissions