Skip to content
Open navigation
OpenOpen workspace

Stack Overflow Developer Survey 2026

AI coding agents: Used in the past year by education

In the Bachelor's degree (B.A., B.S., B.Eng., etc.) group, 71% selected “Claude Code”; in the Primary/elementary school group, 42% did. Of the 28 answers tested, “Claude Code” differs most by education, a small association: χ²(7, N = 11722) = 138.98, Holm-adjusted p < .001, V = .11. 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 education, 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 the past year by education: percentage of each group's respondents who selected each answer, and respondent counts
AI coding agents: Used in the past yearPrimary/elementary school (A)Secondary school (e.g. American high school, German Realschule or Gymnasium, etc.) (B)Some college/university study without earning a degree (C)Associate degree (A.A., A.S., etc.) (D)Bachelor's degree (B.A., B.S., B.Eng., etc.) (E)Master's degree (M.A., M.S., M.Eng., MBA, etc.) (F)Professional degree (JD, MD, Ph.D, Ed.D, etc.) (G)Other, please specify (H)All
Claude Code42.4%6455.3%41068.6%AB97067.6%AB25271.0%ABGH3,39370.3%ABGh2,36764.1%AB45959.9%A10668.4%8,021
GitHub Copilot63.6%9660.0%44559.8%84662.7%23462.1%2,96661.6%2,07659.2%42455.9%9961.3%7,186
OpenAI Codex34.4%5230.6%22732.0%45333.8%12631.2%1,48930.0%1,01129.2%20929.4%5230.9%3,619
Cursor21.9%3322.2%16527.0%G38224.7%9228.5%BG1,36426.5%G89220.5%14726.6%4726.6%3,122
Google Antigravity21.9%3317.4%12918.4%F26017.2%6417.5%F83614.6%49315.4%11015.8%2816.7%1,953
Gemini Code Assist21.9%f3317.5%13018.8%EF26616.1%6014.3%68513.4%45316.8%12023.2%EF4115.3%1,788
Open Code19.2%2918.2%EF13517.2%EF24315.3%5713.5%64412.4%41715.1%10815.8%2814.2%1,661
JetBrains AI10.6%1615.6%11614.1%20017.7%e6612.3%58913.9%46911.5%8216.4%2913.4%1,567
Windsurf10.6%168.0%597.2%1027.2%277.4%F3565.5%1864.6%339.0%166.8%795
Warp11.3%FG176.2%465.7%815.4%205.8%2764.9%1653.5%255.6%105.5%640
Cline9.3%144.0%304.6%656.2%234.6%2184.6%1544.3%318.5%154.7%550
Pi10.6%EFG164.9%365.9%e844.0%154.1%1944.5%1513.9%285.6%104.6%534
Amazon Q Developer9.9%BEFG153.4%255.0%704.6%174.1%1943.7%1252.4%176.8%g124.1%475
Continue10.6%CEFbd164.3%324.5%633.8%143.6%1743.5%1194.7%347.3%134.0%465
Kilo Code7.9%EFG124.0%304.2%595.1%192.9%1383.1%1042.2%167.3%EGf133.3%391
Mistral Vibe10.6%BCDEF163.5%264.1%E583.5%132.3%1113.6%E1224.6%E336.8%E123.3%391
Devin7.3%FG113.9%293.7%f533.8%143.3%f1562.2%732.1%154.5%83.1%359
Google Jules8.6%EFGc133.5%F263.6%Fe514.3%F162.2%1061.7%572.4%1711.3%BCEFGd202.6%306
Replit Agent7.9%EFG123.8%F283.7%Fe533.5%F132.3%F1091.2%421.7%128.5%EFGc152.4%284
Tabnine5.3%F83.0%223.2%F451.6%62.1%981.6%541.5%116.2%EFGd112.2%255
Aider6.6%EF103.4%E252.8%391.9%71.7%801.9%653.1%223.4%62.2%254
Codium (Qodo)4.6%EF72.3%172.5%362.4%91.4%681.4%471.8%134.5%EF81.7%205
BLACKBOX AI7.9%CDEFG123.8%EFg282.1%F302.1%81.6%F780.7%241.3%96.2%CEFG111.7%200
Supermaven6.6%CEFGb102.2%F161.8%F252.1%F81.3%F620.6%191.1%85.1%EFG91.3%157
Trae6.6%BCEFG101.9%F141.8%F252.4%F91.4%F650.5%160.8%64.5%EFG81.3%153
Respondents1517421,4143734,7803,36971617711,722

A letter beside a percentage names the columns it is significantly higher than (uppercase 95%, lowercase 90%; pairwise column tests, Bonferroni-corrected). 19,181 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 “Claude Code”

χ² = 138.98

df = 7, Holm-adjusted p < .001

Effect size

Cramér's V = 0.109

a small association

The strongest answer, APA-style: χ²(7, N = 11722) = 138.98, Holm-adjusted p < .001, V = .11. Each answer is tested on its own, as the respondents who selected it against those who did not, across the 8 groups (N = 11,722). The p-values are Holm-adjusted for the 28 answers tested; 25 of them differ by education at the 0.05 level. 5 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 the past year by education: the test of each answer, selected against not selected, by education
AnswerSelectedχ² (df = 7)Adjusted pCramér's V
Claude Code68.4%138.98< .0010.109
GitHub Copilot61.3%7.23.8110.025
OpenAI Codex30.9%5.79.8110.022
Cursor26.6%32.49< .0010.053
Google Antigravity16.7%19.61.0320.041
Gemini Code Assist15.3%43.58< .0010.061
Open Code14.2%35.67< .0010.055
JetBrains AI13.4%20.12.0320.041
Windsurf6.8%24.22.0080.045
Warp5.5%19.19.0320.040
Cline4.7%15.79.0810.037
Pi4.6%23.23.0110.045
Amazon Q Developer4.1%27.09.0030.048
Continue4.0%28.04.0030.049
Kilo Code3.3%32.90< .0010.053
Mistral Vibe3.3%53.54< .0010.068
Devin3.1%26.97.0030.048
Google Jules2.6%100.16< .0010.092
Replit Agent2.4%86.74< .0010.086
Tabnine2.2%36.88< .0010.056
Aider2.2%32.08< .0010.052
Codium (Qodo)1.7%28.12.0030.049
BLACKBOX AI1.7%98.17< .0010.092
Supermaven1.3%73.89< .0010.079
Trae1.3%74.55< .0010.080

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 28 answers gives 28 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 “Claude Code” 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 education. Cramér's V of 0.109 is in the small range, which is real but modest: knowing someone's group tells you only a little about their answer.

With 11,722 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 “Claude Code” 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 education

Source

Stack Overflow. 2026 Developer Survey, https://survey.stackoverflow.co/2026, licensed under the ODbL 1.0. Tabulated by crosstabs.com. Variables AICodeAgentHaveWorkedWith (AI coding agents: Used in the past year) and Education. 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