Stack Overflow Developer Survey 2026
AI coding agents: Want to use in the next 6 months by employment status
In the I prefer not to say group, 30% selected “Emdash”; in the Employed group, 2% did. Of the 33 answers tested, “Emdash” differs most by employment status, a small association: χ²(5, N = 7648) = 289.45, Holm-adjusted p < .001, V = .19. Respondents could select more than one answer, so the percentages add to more than 100%. These are unweighted respondents, not population estimates.
How results are checked
Named 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 employment status, 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: Want to use in the next 6 months | Employed (A) | I prefer not to say (B) | Independent contractor, freelancer, or self-employed (C) | Not employed (D) | Retired (E) | Student (F) | All |
|---|---|---|---|---|---|---|---|
| Claude Code | 67.6%E3,671 | 60.5%46 | 65.8%713 | 61.5%251 | 54.3%63 | 61.8%329 | 66.3%5,073 |
| GitHub Copilot | 31.1%1,690 | 44.7%C34 | 28.8%312 | 35.0%143 | 32.8%38 | 35.2%187 | 31.4%2,404 |
| OpenAI Codex | 25.8%1,401 | 42.1%A32 | 33.8%A366 | 34.8%A142 | 25.9%30 | 34.0%A181 | 28.1%2,152 |
| Cursor | 17.3%938 | 32.9%AE25 | 20.6%223 | 29.7%ACEf121 | 12.1%14 | 21.6%115 | 18.8%1,436 |
| Open Code | 17.6%954 | 36.8%ACEF28 | 19.7%214 | 27.9%ACE114 | 10.3%12 | 21.4%e114 | 18.8%1,436 |
| Google Antigravity | 11.7%636 | 30.3%A23 | 18.5%A200 | 25.5%AC104 | 14.7%17 | 24.6%Ac131 | 14.5%1,111 |
| Gemini Code Assist | 9.1%495 | 30.3%ACF23 | 14.0%A152 | 18.9%A77 | 25.0%AC29 | 15.4%A82 | 11.2%858 |
| Pi | 9.1%495 | 28.9%ACEF22 | 14.0%A152 | 17.2%A70 | 7.8%9 | 12.4%66 | 10.6%814 |
| JetBrains AI | 8.9%483 | 27.6%ACEF21 | 12.9%A140 | 16.9%A69 | 8.6%10 | 12.8%A68 | 10.3%791 |
| Mistral Vibe | 5.7%307 | 22.4%AEFc17 | 11.2%A121 | 13.5%A55 | 6.9%8 | 10.0%A53 | 7.3%561 |
| Windsurf | 3.7%201 | 18.4%ACE14 | 7.6%A82 | 15.0%ACef61 | 5.2%6 | 9.0%A48 | 5.4%412 |
| Warp | 3.2%173 | 22.4%ACEF17 | 7.7%A83 | 12.0%A49 | 6.0%7 | 7.7%A41 | 4.8%370 |
| Devin | 3.3%177 | 19.7%ACFe15 | 6.9%A75 | 11.8%AC48 | 6.0%7 | 8.5%A45 | 4.8%367 |
| Amazon Q Developer | 3.1%167 | 23.7%ACEF18 | 7.6%A82 | 12.3%Ac50 | 4.3%5 | 7.9%A42 | 4.8%364 |
| Kilo Code | 3.1%170 | 22.4%ACEFd17 | 6.7%A73 | 10.8%A44 | 4.3%5 | 6.6%A35 | 4.5%344 |
| Google Jules | 2.4%130 | 22.4%ACEF17 | 7.3%A79 | 13.0%AC53 | 6.9%A8 | 9.6%A51 | 4.4%338 |
| Cline | 2.9%155 | 21.1%ACEF16 | 7.0%A76 | 12.0%ACF49 | 5.2%6 | 5.8%A31 | 4.4%333 |
| Continue | 2.4%129 | 25.0%ACDEF19 | 6.1%A66 | 11.3%AC46 | 7.8%A9 | 6.8%A36 | 4.0%305 |
| Replit Agent | 2.0%108 | 18.4%ACF14 | 5.6%A61 | 12.0%ACf49 | 6.0%A7 | 6.8%A36 | 3.6%275 |
| Aider | 2.0%111 | 18.4%ACEF14 | 6.2%A67 | 11.0%ACF45 | 5.2%6 | 5.6%A30 | 3.6%273 |
| Emdash | 1.8%96 | 30.3%ACDEF23 | 6.1%A66 | 9.3%A38 | 3.4%4 | 5.5%A29 | 3.3%256 |
| BLACKBOX AI | 1.7%92 | 21.1%ACEFd16 | 5.5%A60 | 10.0%AC41 | 5.2%a6 | 6.8%A36 | 3.3%251 |
| Codium (Qodo) | 1.8%99 | 21.1%ACEFd16 | 5.3%A57 | 10.0%AC41 | 4.3%5 | 5.8%A31 | 3.3%249 |
| Supermaven | 1.7%90 | 21.1%ACEFd16 | 5.6%A61 | 9.6%A39 | 4.3%5 | 5.5%A29 | 3.1%240 |
| Augment Code | 1.5%83 | 22.4%ACDEF17 | 5.4%A59 | 10.0%ACf41 | 4.3%5 | 5.3%A28 | 3.0%233 |
| Respondents | 5,432 | 76 | 1,084 | 408 | 116 | 532 | 7,648 |
A letter beside a percentage names the columns it is significantly higher than (uppercase 95%, lowercase 90%; pairwise column tests, Bonferroni-corrected). 23,255 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 “Emdash”
χ² = 289.45
df = 5, Holm-adjusted p < .001
Effect size
Cramér's V = 0.195
a small association
The strongest answer, APA-style: χ²(5, N = 7648) = 289.45, Holm-adjusted p < .001, V = .19. Each answer is tested on its own, as the respondents who selected it against those who did not, across the 6 groups (N = 7,648). The p-values are Holm-adjusted for the 33 answers tested; 33 of them differ by employment status at the 0.05 level.
| Answer | Selected | χ² (df = 5) | Adjusted p | Cramér's V |
|---|---|---|---|---|
| Claude Code | 66.3% | 21.63 | .001 | 0.053 |
| GitHub Copilot | 31.4% | 16.02 | .007 | 0.046 |
| OpenAI Codex | 28.1% | 57.46 | < .001 | 0.087 |
| Cursor | 18.8% | 58.24 | < .001 | 0.087 |
| Open Code | 18.8% | 52.51 | < .001 | 0.083 |
| Google Antigravity | 14.5% | 146.53 | < .001 | 0.138 |
| Gemini Code Assist | 11.2% | 115.93 | < .001 | 0.123 |
| Pi | 10.6% | 74.12 | < .001 | 0.098 |
| JetBrains AI | 10.3% | 67.34 | < .001 | 0.094 |
| Mistral Vibe | 7.3% | 99.38 | < .001 | 0.114 |
| Windsurf | 5.4% | 152.77 | < .001 | 0.141 |
| Warp | 4.8% | 157.14 | < .001 | 0.143 |
| Devin | 4.8% | 135.32 | < .001 | 0.133 |
| Amazon Q Developer | 4.8% | 175.05 | < .001 | 0.151 |
| Kilo Code | 4.5% | 135.71 | < .001 | 0.133 |
| Google Jules | 4.4% | 238.13 | < .001 | 0.176 |
| Cline | 4.4% | 159.01 | < .001 | 0.144 |
| Continue | 4.0% | 208.66 | < .001 | 0.165 |
| Replit Agent | 3.6% | 202.34 | < .001 | 0.163 |
| Aider | 3.6% | 180.37 | < .001 | 0.154 |
| Emdash | 3.3% | 289.45 | < .001 | 0.195 |
| BLACKBOX AI | 3.3% | 216.64 | < .001 | 0.168 |
| Codium (Qodo) | 3.3% | 197.00 | < .001 | 0.160 |
| Supermaven | 3.1% | 206.78 | < .001 | 0.164 |
| Augment Code | 3.0% | 236.75 | < .001 | 0.176 |
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 33 answers gives 33 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 “Emdash” 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 employment status. Cramér's V of 0.195 is in the small range, which is real but modest: knowing someone's group tells you only a little about their answer.
With 7,648 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 “Emdash” 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 employment status
- Other work, school or life responsibilities by employment status
- Role in choosing software tools by employment status
- How they learned to code in the past year by employment status
- Activities in a typical work week by employment status
- How they learned to code for AI in the past year by employment status
- Involvement in building internal tools by employment status
Source
Stack Overflow. 2026 Developer Survey, https://survey.stackoverflow.co/2026, licensed under the ODbL 1.0. Tabulated by crosstabs.com. Variables AICodeAgentWantToUseNext6Months (AI coding agents: Want to use in the next 6 months) and Employment. 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.