How do we measure how cognitively demanding a job is? The traditional approach has relied on occupational analysts rating jobs against standardised frameworks. Zisman and Ganzach (2023) propose a simpler alternative: look at the average intelligence of the people actually doing the job. They call this occupational intelligence, and argue it is a valid and practical measure of occupational complexity.

What They Found

Examining data across a wide range of occupations, the authors found that occupational intelligence correlated strongly with established analyst-based measures of job complexity, specifically the DOT and O*NET frameworks (Zisman & Ganzach, 2023). In other words, jobs that analysts rate as cognitively demanding tend to be staffed by people with higher average intelligence, and vice versa. The measure held up well as a proxy for complexity.

Importantly, the relationship with occupational pay was considerably weaker (Zisman & Ganzach, 2023). This is a meaningful distinction: a job being cognitively complex and a job being well-paid are related but separate things, and conflating them produces a muddier picture of what occupational complexity actually means.

The study also replicated a well-established finding: the interaction between individual intelligence and job complexity in predicting pay. People with higher intelligence earn more, but this effect is amplified in more complex jobs. Crucially, occupational intelligence produced the same interaction pattern as the traditional analyst-based measures, further supporting its validity (Zisman & Ganzach, 2023).

Why It Matters

Analyst-based measures of occupational complexity are resource-intensive to produce and periodically go out of date. Occupational intelligence offers a data-driven alternative that can be derived from large existing datasets and updated as workforce composition changes. For researchers studying the relationship between cognitive ability, work, and economic outcomes, it is a potentially valuable tool, though the authors note that variability within occupations and the US-specific nature of the samples are limitations worth keeping in mind (Zisman & Ganzach, 2023).

Zisman, C., & Ganzach, Y. (2023). Occupational intelligence as a measure of occupational complexity. Personality and Individual Differences, 203, 112005. https://doi.org/10.1016/j.paid.2022.112005