Skip to content
Open navigation
OpenOpen workspace

Stack Overflow Developer Survey 2026

AI coding agents: Used in past year and want to again by gender

In the Non-binary group, 17% selected “Pi”; in the Woman group, 3% did. Of the 9 answers tested, “Pi” differs most by gender, a negligible association: χ²(4, N = 5839) = 46.31, Holm-adjusted p < .001, V = .09. Respondents could select more than one answer, so the percentages add to more than 100%. These are unweighted respondents, not population estimates.

Published by crosstabs.com · Last updated ·
How results are checkedNamed statistical review pending. Selected automated reference cases are published on the validation page.

The table

The table

Question asked: Have you used any of the following coding agents or assistants in the past year? Do you want to use any of the following in the next 6 months? Respondents could select more than one answer. Shown by gender, as the percentage of each group's respondents who selected each answer, with the number of respondents beneath. A respondent is counted in every row they selected, so a column adds to more than 100%.

AI coding agents: Used in past year and want to again by gender: percentage of each group's respondents who selected each answer, and respondent counts
AI coding agents: Used in past year and want to againMan (A)Non-binary (B)Prefer not to say (C)Prefer to self-describe: (D)Woman (E)All
Claude Code63.1%3,33759.4%3859.9%9455.9%1966.7%19663.1%3,684
GitHub Copilot34.0%1,79835.9%2329.3%4626.5%935.4%10433.9%1,980
OpenAI Codex21.3%1,12920.3%1323.6%3735.3%e1216.0%4721.2%1,238
Open Code12.7%67010.9%718.5%e2920.6%79.5%2812.7%741
Cursor12.2%6446.3%47.0%1117.6%611.2%3312.0%698
Google Antigravity8.5%4486.3%410.2%1614.7%57.1%218.5%494
Gemini Code Assist6.3%3356.3%46.4%108.8%36.8%206.4%372
JetBrains AI5.7%3037.8%56.4%1011.8%47.5%225.9%344
Pi3.8%19917.2%AE117.6%1214.7%AE52.7%84.0%235
Mistral Vibe2.2%1161.6%11.3%28.8%a32.4%72.2%129
Warp2.1%1093.1%21.3%28.8%a31.7%52.1%121
Windsurf1.8%944.7%31.3%25.9%22.0%61.8%107
Amazon Q Developer1.5%781.6%11.3%25.9%24.1%A121.6%95
Devin1.5%801.6%10.6%15.9%21.4%41.5%88
Kilo Code1.4%763.1%20.0%08.8%AE31.4%41.5%85
Cline1.3%691.6%11.3%28.8%A32.0%61.4%81
Continue1.4%731.6%10.0%08.8%AE31.4%41.4%81
Google Jules0.9%481.6%10.6%18.8%ACE31.0%31.0%56
Replit Agent0.9%471.6%11.3%25.9%A21.0%30.9%55
BLACKBOX AI0.5%291.6%10.6%18.8%AC32.0%A60.7%40
Aider0.5%271.6%11.3%211.8%ACE41.0%30.6%37
Supermaven0.6%311.6%10.0%05.9%Ae20.7%20.6%36
Codium (Qodo)0.5%271.6%10.0%011.8%AE40.7%20.6%34
Trae0.5%271.6%10.0%05.9%Ae20.7%20.5%32
Augment Code0.5%243.1%A20.0%05.9%A21.0%30.5%31
Respondents5,29064157342945,839

A letter beside a percentage names the columns it is significantly higher than (uppercase 95%, lowercase 90%; pairwise column tests, Bonferroni-corrected). 25,064 of 30,903 respondents have no answer to one of the two questions (not asked or not answered) and are left out. The table shows 25 of the 33 answers: the most selected. The CSV and JSON downloads have every answer.

Analysis

Results

Chi-square test of “Pi”

χ² = 46.31

df = 4, Holm-adjusted p < .001

Effect size

Cramér's V = 0.089

a negligible association

The strongest answer, APA-style: χ²(4, N = 5839) = 46.31, Holm-adjusted p < .001, V = .09. Each answer is tested on its own, as the respondents who selected it against those who did not, across the 5 groups (N = 5,839). The p-values are Holm-adjusted for the 9 answers tested; 1 of them differs by gender at the 0.05 level. 24 answers are not tested: too few respondents selected them for the test to be reliable (an expected count below 1, or more than 20% of the expected counts below 5).

AI coding agents: Used in past year and want to again by gender: the test of each answer, selected against not selected, by gender
AnswerSelectedχ² (df = 4)Adjusted pCramér's V
Claude Code63.1%3.451.0000.024
GitHub Copilot33.9%2.741.0000.022
OpenAI Codex21.2%9.45.4000.040
Open Code12.7%9.49.4000.040
Cursor12.0%7.07.7940.035
Google Antigravity8.5%3.381.0000.024
Gemini Code Assist6.4%0.451.0000.009
JetBrains AI5.9%4.201.0000.027
Pi4.0%46.31< .0010.089
Mistral Vibe2.2%Not tested: expected counts too small
Warp2.1%Not tested: expected counts too small
Windsurf1.8%Not tested: expected counts too small
Amazon Q Developer1.6%Not tested: expected counts too small
Devin1.5%Not tested: expected counts too small
Kilo Code1.5%Not tested: expected counts too small
Cline1.4%Not tested: expected counts too small
Continue1.4%Not tested: expected counts too small
Google Jules1.0%Not tested: expected counts too small
Replit Agent0.9%Not tested: expected counts too small
BLACKBOX AI0.7%Not tested: expected counts too small
Aider0.6%Not tested: expected counts too small
Supermaven0.6%Not tested: expected counts too small
Codium (Qodo)0.6%Not tested: expected counts too small
Trae0.5%Not tested: expected counts too small
Augment Code0.5%Not tested: expected counts too small

How to read it

No single chi-square test covers this table. A respondent appears in every row they selected, so the rows are not separate groups of people, and a test of the whole table would count the same person several times. Each answer is tested on its own instead: did the share who selected it differ between the groups? Testing 9 answers gives 9 chances of a false alarm, so the p-values are Holm-adjusted to keep the chance of any false alarm in this table at 0.05.

For “Pi” the test rejects independence at the 0.05 level (Holm-adjusted p < .001): differences this large between the groups are unlikely if selecting it did not depend on gender. Cramér's V of 0.089 is in the negligible range, which is trivially weak in practical terms, even where a sample this large can detect it.

With 5,839 respondents, even small differences reach statistical significance. Read the percentages and the effect size before the p-value.

Limits

  • The survey has no weights. These tables are unweighted counts of the people who answered, so they describe the respondents, not developers in general.
  • The test treats respondents as a simple random sample. They are people who chose to answer a survey Stack Overflow promoted, not a random sample of developers, so a result holds for these respondents and may not hold beyond them.
  • The adjustment covers the answers of this question by this group only, not the other tables in the library. The headline answer is the strongest of those tested, so its effect size is larger than a typical answer's.
  • A difference between groups is not evidence that group membership causes the answer.

Next steps

Work with this table

Open the test of “Pi” in the calculator

The link fills in the respondents who selected this answer and those who did not, by group. The whole table does not open there: its rows overlap, and the calculator's test needs rows that are separate people.

Open this question in the full file

The workspace opens all 30,903 respondents in your browser with this question selected. The survey has no weights, so the table there is unweighted too. Change the group, add a filter or export from there.

Open in the workspace →

Run this on your own data — free, no signup

Upload a CSV, Excel, SPSS, Stata, SAS, R, Parquet or JSON file. Everything runs in your browser; your file never leaves your device.

Open the workspace →

The same question by other groups

Other questions that differ by gender

Source

Stack Overflow. 2026 Developer Survey, https://survey.stackoverflow.co/2026, licensed under the ODbL 1.0. Tabulated by crosstabs.com. Variables AICodeAgentAdmired (AI coding agents: Used in the past year AND Want to use in the next 6 months) and Gender. Stack Overflow licenses the survey data under the Open Database License (ODbL) 1.0 and its contents under the Database Contents License (DbCL) 1.0: credit Stack Overflow, keep the licence notice, and share any database made from the data under the ODbL. These tables and their downloads are offered on those terms.

  1. [1]Stack Overflow: 2026 Developer Survey public results (survey_results_public.csv)survey.stackoverflow.co
  2. [2]Stack Overflow: Open Database License (ODbL) 1.0opendatacommons.org