Student Outcomes: from College to Career

This dataset looks at how college students move from graduation to their first jobs, things like degree type, GPA, salary, and early employment outcomes. I wanted to get a quick sense of what patterns show up across campuses and whether factors like age or major make any real difference once students hit the workforce.

Nothing too complicated, just straightforward exploration to see what stands out and what doesn’t.

library(readr)
library(readxl)
library(googlesheets4)
library(dplyr)

univ <- read.csv("C:/R/RProjects/guilzee.github.io/posts/StudentOutcomes/Input/synthetic_illinois_university_student_outcomes.csv")

Alright, diving into this dataset of Illinois university student outcomes. First thing, libraries loaded, CSV pulled in, we’re ready to roll :)

Before anything fancy, I like to get a feel for what I’m working with.

When these students started

range(univ$start_year)

[1] 2016 2020

So I checked the range of start years—2016 to 2020. That tells me we’re looking at roughly one full college cycle.

Then I looked at how many different campuses we’re dealing with.

Where did they study?

campusct <- univ %>%
  count(univ$campus_location) %>%
  nrow()
print(campusct)

[1] 3

Three total, which makes sense for a state university system.

Next up, total students across those years

How many students are we talking about?

studentct <- univ %>%
  count(univ$student_id) %>%
  nrow()
print(studentct)

[1] 5000

five thousand!

How long did it take them to finish their degree?

univ$program_duration <- univ$graduation_year - univ$start_year

mean(univ$program_duration)

[1] 4

On average, students took four years to finish, which tracks with the standard timeline. No major surprises there.

Age at graduation

range(univ$age_at_graduation)

[1] 22 29

Then I checked age at graduation, and the range came out between 22 and 29. That spread suggests a mix of traditional and slightly older students—maybe transfers or people coming back to school later.

I wondered if age tied into what degrees people chose. I bucketed the ages and grouped them by degree program.

Does age influence degree choice?

univ$agebucket <- cut(
  univ$age_at_graduation,
  breaks = c(0, 22, 25, 30, 40, 100),   # bins are (0,22], (22,25], (25,30], ...
  labels = c("≤22", "23–25", "26–30", "31–40", "40+"),
  right = TRUE,
  include.lowest = TRUE
)


group <- univ %>%
  select(degree_program, agebucket) %>%
  group_by(degree_program, agebucket) %>%
  summarise(count = n()) %>%
  arrange(degree_program, agebucket)

age_vs_degreeprogram <- group %>%
  arrange(desc(count))

Turns out, not really. The top programs stayed consistent across age groups: Business, STEM, and Health Sciences. So age didn’t seem to drive what major someone picked.

After that, I wanted to see if age showed up somewhere else, like in jobs or pay after graduation

What about age and career outcomes?

age_vs_salaryandjob <- univ %>%
  group_by(agebucket) %>%
  summarize(
    avgsalary = mean(salary_first, na.rm = TRUE),
    jobrate = mean(employment_status_6mo == "Employed", na.rm = TRUE)
  )

I calculated the average salary and employment rate by age group. Again, not much variation. Whether someone graduated at 22 or 29, their early career outcomes looked pretty similar.