Predicting College Student Placement

Do academic performance, cognitive ability, internship experience, and communication skills predict whether a student gets placed in a job? This project fits a logistic regression model to a dataset of 10,000 college students to find out.

All Four Predictors vs. Placement

Each panel splits one predictor into five equal-width ranges across its actual scale (e.g. CGPA from its minimum to maximum, IQ from its minimum to maximum) and plots the real placement rate observed within each range. Internship Experience shows the two actual groups, No and Yes. CGPA and Communication Skills climb sharply and steadily across their range; IQ climbs more modestly; Internship Experience barely moves at all.

Overall Placement Rate

Only 16.6% of students in this dataset were placed, so the model works against a heavily unbalanced baseline.

Summary

Using a dataset of 10,000 college students, this project models job placement based on academic performance, cognitive ability, internship experience, and communication skills. CGPA stands out as the strongest predictor, with IQ and communication skills also playing a meaningful role. Internship experience, by contrast, adds little once the other factors are known. Full model details and performance metrics are on the Model page.