ResearchEducation & Skills, Employment

Data Spotlight: Gender gaps across life

Feb 23, 2026
AIBM Team
Young male and female standing next to each other, couple breaking up, closeup

From birth to death, men and women experience different outcomes on key dimensions of human flourishing including health, education, and economic opportunity. Some of these differences are stark, others less so. Some favor women, others men.

Here we show gender gaps in selected measures at different points in the lifecycle (we recommend pressing the play button and watching all the way through):

 

Figure 1

American men do worse on measures like educational attainment, criminal status, and life expectancy. However, they have more favorable outcomes on measures like earnings and employment. For instance, about 79% of 25 year-old men were employed, compared to 72% of women; on the other hand, about 39% of 30-year old men have at least a bachelor’s degree, compared to 46% of women. The largest gaps are in incarceration, where men are far more likely to have ever been incarcerated. Outcome gaps related to natal and infant health are smaller in comparison.

Our analysis here is inspired by a similar representation of Danish data, undertaken by the Rockwool Foundation. The key points are simply that measures of gender equality should include boys and men, as an earlier paper by Richard V. Reeves and Allen Downey shows. Likewise efforts to promote gender equality should look both ways.

Data note

Data included in this visualization come from a number of sources:

  • Apgar (age 0): National Center for Health Statistics Linked Birth and Infant Data (2022 cohort). The Apgar score, which is measured at 1 and 5 minutes after birth, measures a number of characteristics, such as heart rate and reflexes, and synthesizes each individual measure into a score on a scale of 1 to 10.
  • NICU (age 0): National Center for Health Statistics Linked Birth and Infant Data (2022 cohort); this measure represents if an infant was admitted into a neonatal intensive care unit within the first 6 months of life.
  • NAEP Math (age 10): National Assessment for Educational Progress 4th Grade Math Assessment (2024); age 10 inferred from the test-takers being in 4th grade
  • NAEP Reading (age 10): National Assessment for Educational Progress 4th Grade Reading Assessment (2024); age 10 inferred from the test-takers being in 4th grade
  • SAT Math (age 16-18): SAT Suite of Assessments (2025); age 16-18 inferred from average ages when students take the SAT
  • SAT ERW (age 16-18): SAT Suite of Assessments (2025); age 16-18 inferred from average ages when students take the SAT
  • High School Graduation (age 18): Averaged Freshman Graduation Rate (AFGR) via National Center for Education Statistics (2023); population counts also collected from the NCES; the AFGR takes cross-sections of an incoming freshman class, and compares that population to the graduating population four years later; due to the nature of this measurement, the AFGR is sensitive to factors such as students transferring schools and students graduating shorter or longer than four years. While this makes the AFGR less robust, national level comparisons between the AFGR and the more robust ACGR track very closely together.
  • Employment (age 25): Current Population Survey (CPS) via IPUMS (2025)
  • Bachelor’s Degree (age 30): Current Population Survey (CPS) via IPUMS (2025)
  • Depression (age 37): CDC Behavioral Risk Factor Surveillance System (BRFSS) (2024); this measure represents whether Americans aged 35 to 39 years-old were “(ever told) ([they] had) a depressive disorder (including depression, major depression, dysthymia, or minor depression)”
  • Incarceration (age 39): National Longitudinal Survey; incarceration rates are calculated as the share of 39 year-old individuals that were ever incarcerated in their life.
  • Income (age 40): Current Population Survey (CPS) via IPUMS (2025)
  • Life Expectancy (Age 65): United States Mortality Tables via Human Mortality Database

To allow for visualization on the same figure and accommodate data limitations, each measure was standardized as follows:

  • For continuous variables (e.g., income), the standardized score is calculated as the difference between the male mean and the overall sample mean, divided by the overall sample standard deviation..
  • For binary variables (e.g., employment), the standardized score is calculated as the log-odds ratio between the male prevalence and the overall prevalence.

Bars show men’s deviation from the overall sample mean (rather than the direct male–female difference). In approximately balanced samples for continuous variables, this corresponds to roughly half the gender gap; if one gender is overrepresented (e.g., births are ~51% male), the overall mean shifts slightly toward that group, though this effect is minimal in the datasets used here.

 

In order to place all measures on the same scale, we standardize the outcomes rather than using the raw measures; for instance, while the difference between average SAT math scores between men and women are 515 and 500, respectively, the effect size for men is about 0.06. This standardized effect size allows us to plot employment rates alongside test scores for instance. See the data note above for a more detailed explanation of our methodology.

Naturally, this visualization will never be able to tell the entire story, as a myriad of other outcomes, like parent and marital status, could be shown. Despite this limitation, the outcomes shown here still cover a wide range of important milestones and indicators across age, and they highlight places where we can help boys and girls, and men and women alike.

The AIBM team thanks Allen Downey for his analysis on this piece.

Related Research

Male medical professional talks with elderly patient in wheelchair
Education & Skills, Employment

Healthcare is the new working-class jobs engine. Can it become a career pathway for men?

While healthcare has emerged as the dominant replacement sector for manufacturing in the U.S. since the early 1980s, much of its growth is concentrated in roles that many men have not historically viewed as natural occupational pathways.

Father holds baby as they both smile
Education & Skills, Employment

Dads are taking more parental leave than the headlines suggest

As state paid family leave programs have expanded rapidly across the United States, Anders Knospe and Richard Reeves find that fathers' leave-taking has grown dramatically, and that the gap between fathers and mothers is narrower than standard measures suggest.

In a bustling city street, a man in a hat holds a cardboard sign reading Need work
Education & Skills, Employment

What’s really going on with men and work? Seven facts

This brief provides seven key facts about men and work by mapping who is affected, tracing how trends have shifted over time, and confronting the questions we may still not have good answers to.

Back to All