Alonmarom/us-college-basketball-expenses

# US Collegiate Basketball Expenses Analysis – EDA Project This project explores historical athletic spending data from U.S. colleges and universities. The goal was to analyze spending trends and disparities in men's and women's basketball programs across different collegiate divisions. ## Dataset Overview **Source:** U.S. Department of Education / IPEDS / Kaggle **Size:** 132,327 entries (Filtered subset for Basketball: 8,443 rows). **Target Variables:** `exp_men` and `exp_women` (Basketball Expenses). ## Data Cleaning & Preparation - **Missing Values:** NaN values in all financial columns (e.g., expenses, revenue) were imputed with **zero (0)**. - **Filtering:** The dataset was strictly filtered to include only "Basketball" entries. - **Outliers:** **Retained** (Not removed), as extreme spending values belong to top-tier NCAA Division I programs and are critical for trend projection. ## Descriptive Statistics **Gender Disparity:** Average spending for Men's Basketball ($0.92M$) is more than double that of Women's Basketball ($0.43M$). **Skewness:** The Mean is significantly higher than the Median (e.g., Men's Median is $0.13M$), indicating that a few high-resource programs drive the overall average. **Correlation:** Extremely high correlation (**0.99**) between Expenses and Revenue, indicating high spending is tied directly to high revenue generation. ## Research Questions **Q1: What is the historical trend and projection for expenses?** → Strong, consistent **upward trend** for both genders (2003–2018), projecting continued spending escalation.  **Q2: Which divisional classification drives the highest expenses?** → **NCAA Division I-FBS** and **Division I without football** account for the highest expenditures and the largest gender spending gap.  **Q3: Is the gender spending gap consistent across divisions?** → Yes — the financial disparity (Men > Women) is **persistent across all divisions** (NCAA I, II, and III).  ## Conclusion The EDA confirms that **Institutional Classification** and **Revenue Generation** are the strongest factors influencing basketball spending. Future spending is projected to increase sharply, and the significant **gender disparity** observed across all divisions is expected to persist. ## 4. Video Presentation A short video presentation summarizing the EDA findings and conclusions: [Watch the full Project Video Summary on YouTube](https://youtu.be/BBmzhtDYkiw)# 美国大学篮球开支分析——探索性数据分析（Exploratory Data Analysis，EDA）项目 本项目针对美国高等院校的历史体育开支数据展开研究，旨在分析不同大学体育分区下男子与女子篮球项目的开支趋势与经费差异。 ## 数据集概览 **数据来源**：美国教育部 / 高等教育数据集成系统（Integrated Postsecondary Education Data System，IPEDS）/ Kaggle **数据规模**：原始数据集共132,327条记录，经筛选后的篮球项目子集包含8,443行数据。 **目标变量**：`exp_men`与`exp_women`，分别代表男子与女子篮球项目的开支。 ## 数据清洗与预处理 - **缺失值处理**：所有财务类列（如开支、营收）中的缺失值（NaN）均以**0填充**。 - **数据筛选**：严格筛选仅保留“篮球项目”相关记录。 - **异常值处理**：**保留异常值未予剔除**，因极端开支数值均来自顶级全国大学体育协会（National Collegiate Athletic Association，NCAA）一级联盟院校，对开支趋势预测至关重要。 ## 描述性统计分析 **性别差异**：男子篮球项目的平均开支达92万美元，约为女子篮球项目（43万美元）的两倍有余。 **偏态分布特征**：开支均值显著高于中位数（如男子篮球项目开支中位数为13万美元），表明少数高资源投入的项目拉高了整体平均开支水平。 **相关性分析**：开支与营收间存在极高的相关性（**0.99**），表明高开支与高营收直接挂钩。 ## 研究问题 **Q1：篮球开支的历史趋势与未来预测如何？** → 2003年至2018年间，男女篮球项目均呈现强劲且持续的**上升趋势**，预计未来开支仍将持续攀升。  **Q2：哪个体育分区的篮球开支最高？** → **NCAA一级联盟美式橄榄球碗赛分区（Division I-FBS）**与**无橄榄球项目的一级联盟**拥有最高的开支规模，同时性别开支差距也最为显著。  **Q3：性别开支差距在各体育分区中是否普遍存在？** → 是的——**男子篮球开支高于女子**的经费差异在所有分区（NCAA一级、二级、三级联盟）中均持续存在。  ## 研究结论 本次探索性数据分析证实，**院校体育分区等级**与**营收能力**是影响篮球项目开支的两大核心因素。未来篮球开支预计将大幅增长，且各分区普遍存在的显著性别开支差距或将持续存在。 ## 4. 视频演示文稿 一段总结本次EDA研究结果与结论的简短演示视频： [观看完整项目视频总结（YouTube链接）](https://youtu.be/BBmzhtDYkiw)