Does a College Degree Predict Job Performance?

In a study of 10,765 agents hired, education level had a weak association with the production milestone. Rates moved only slightly across education groups and their confidence intervals overlapped. The study did not support using a degree as a precise forecast of who would produce.

Source: "Decision Traces," Saad Bin Shafiq, Nodes, 2026, education analysis on the subset with available education data. Read it on arXiv.

What the data showed

Across agents with evaluable tenure and available education data, production rates moved slightly by education level, but the confidence intervals overlapped. The data could not confidently rank one education level above another. See the approved model comparison.

A degree is a credential, not a forecast

The reason mirrors why resume keywords fail. A degree describes something a candidate did in the past. It does not measure how they will perform in a specific role with specific demands. It can carry a weak signal about persistence or aptitude, but it is a noisy proxy, and screening hard on it removes capable people for a small and uncertain gain. See why keywords fail.

The wider evidence on degree inflation

This is consistent with research outside the study. The Harvard Business School and Accenture report Dismissed by Degrees, drawn from 26 million job postings and 600 employers, found that employers generally rate degreed and non-degreed workers in the same role as nearly equally productive, while a bachelor's degree increasingly gets used as a rough proxy for skills it does not actually measure. Requiring a degree where the role does not need one screens out capable candidates and shrinks the talent pool.

What this does not say

This is one carrier and one role, insurance sales. The effect was small, not zero, and the direction did favor more education. The takeaway is narrow: a degree is a weak screening signal for production, and a team that screens hard on it should check that choice against its own outcomes rather than assume it.

Frequently asked questions

Does a college degree predict job performance? In this study of 10,765 agents hired, only weakly. Production rates rose slightly with education, but the differences fell within overlapping confidence intervals.

Should employers require a degree? It depends on whether the role genuinely needs one. Research on degree inflation finds that requiring a degree for jobs that did not previously need one excludes capable candidates for little measured gain.

What predicts performance better than a degree? In this dataset, personality and behavioral assessment were far stronger signals than education or resume keywords.

Is this specific to insurance? The data is from insurance agent hiring. The pattern, that credentials are weak proxies for production, is consistent with broader research, though the specific numbers are carrier-specific.