Responding to CARS is voluntary. The results presented here are from a self-selecting sample of government analysts. Because respondents are self-selecting, the results we present reflect the views of the analysts who participated.
For more detail, see the data collection page.
We asked respondents if they had any coding experience, inside or outside of work. Of 1300 respondents, 92.9% had coding experience. We asked all respondents with coding experience “In your current role, how often do you write code to complete your work objectives?”
To see how this compares to other years, see the data collection page
| In your current role, how often do you write code to complete your work objectives? | Percentage |
|---|---|
| Never | 9.4% |
| Rarely | 12.7% |
| Sometimes | 17.7% |
| Regularly | 36.1% |
| Always | 24.1% |
| Sample size = 1208 |
Given a list of programming tools, we asked all respondents with coding experience if the tool was available to use for their work. Note this includes respondents who do not code in their current role.
Access to tools does not necessarily refer to official policy. Some analysts may have access to tools others cannot access within the same organisation.
| Programming tool | Yes | No | Don't know |
|---|---|---|---|
| Python | 76.1% | 11.5% | 12.4% |
| R | 92.9% | 4.5% | 2.6% |
| SQL | 71.9% | 7% | 21.1% |
| Matlab | 4.6% | 40.5% | 54.9% |
| SAS | 22.4% | 31.7% | 45.9% |
| SPSS | 23.7% | 30.4% | 45.9% |
| Stata | 15.8% | 30.7% | 53.5% |
| VBA | 43% | 16% | 41% |
| Sample size = 1208 |
Given the same list of programming tools, the same respondents were asked if they knew how to program with the tool to a level suitable for their work, answering “Yes”, “No” or “Not required for my work”.
Please note that capability in programming languages is self-reported here and was not objectively defined or tested. The statement “not required for my work” was similarly not defined.
| Programming tool | Yes | No | Not required for my work |
|---|---|---|---|
| Python | 50.3% | 34% | 15.6% |
| R | 80.1% | 13.2% | 6.6% |
| SQL | 62.3% | 22.1% | 15.6% |
| Matlab | 11.8% | 45.9% | 42.3% |
| SAS | 19.7% | 43.7% | 36.6% |
| SPSS | 26.7% | 36.6% | 36.8% |
| Stata | 11.7% | 49.4% | 38.9% |
| VBA | 21.8% | 44.5% | 33.8% |
| Sample size = 1208 |
The proportion of respondents who report having the capability to use R and Python, is shown alongside the proportion who are able to use SAS, SPSS or Stata, for the past four years of the survey.
| Programming language type | Year | Know how to programme with these tools (percent) | Lower confidence limit (percent) | Upper confidence limit (percent) |
|---|---|---|---|---|
| 2021 | Open Source | 77% | 0.7413160 | 0.7958949 |
| 2022 | Open Source | 80.3% | 0.7802509 | 0.8231395 |
| 2023 | Open Source | 72.6% | 0.7005951 | 0.7491138 |
| 2024 | Open Source | 80.9% | 0.7869734 | 0.8296660 |
| 2021 | Proprietary | 60.1% | 0.5687348 | 0.6321733 |
| 2022 | Proprietary | 56.3% | 0.5359005 | 0.5893030 |
| 2023 | Proprietary | 36.1% | 0.3351431 | 0.3873442 |
| 2024 | Proprietary | 40.7% | 0.3805305 | 0.4338641 |
Differences in preferred languages may lead to silos between analytical professions. Here we show the percentage of respondents reporting capability in different tools, within the different analytical professions.
Please note that respondents might be members of more than one profession, and may report capability in more than one tool.
| Profession | Python | R | Matlab | SAS | SPSS | SQL | Stata | VBA |
|---|---|---|---|---|---|---|---|---|
| Data Engineers | 84.2% | 78.9% | 21.1% | 15.8% | 31.6% | 89.5% | 5.3% | 31.6% |
| Data Scientists | 88.5% | 90.4% | 19.9% | 14.7% | 13.5% | 84% | 7.1% | 21.8% |
| Government Digital and Data | 67.2% | 65.5% | 15.5% | 29.3% | 13.8% | 75.9% | 5.2% | 24.1% |
| Government Economic Service | 39.2% | 79.4% | 3.9% | 11.8% | 9.8% | 38.2% | 36.3% | 11.8% |
| Government Geography Profession | 80% | 80% | 5% | 10% | 15% | 80% | 5% | 30% |
| Government Operational Research Service | 65.6% | 87% | 22.9% | 22.9% | 8.4% | 74% | 7.6% | 33.6% |
| Government Science & Engineering | 64.9% | 83.8% | 27% | 16.2% | 13.5% | 64.9% | 8.1% | 18.9% |
| Government Social Research | 25.2% | 75.7% | 3.5% | 9.6% | 53.9% | 30.4% | 13.9% | 3.5% |
| Government Statistician Group | 43.7% | 86.8% | 11% | 28.5% | 35.9% | 64.8% | 9.8% | 20% |
We asked respondents to answer “Yes”, “No” or “Don’t know” for the following questions:
Please note these outputs include people with coding experience who do not code in their current role.
| Is Git available to use for your work? | Percentage |
|---|---|
| Yes | 82.5% |
| No | 6% |
| Don't know | 11.5% |
| Sample size = 1208 |
| Do you know how to use Git for your work? | Percentage |
|---|---|
| Yes | 66.4% |
| No | 23.8% |
| Not required for my work | 9.8% |
| Sample size = 1208 |
Respondents with coding experience were asked where they first learned to code.
These data only show where people first learned to code. They do not show all the settings in which they learned to code, to what extent, or how long ago.
| Where did you first learn to code? | Percentage |
|---|---|
| Self-taught | 25.1% |
| Primary/secondary education | 5.9% |
| Higher Education | 43.2% |
| Current role | 8.7% |
| Previous public sector employment | 13.2% |
| Previous private sector employment | 2.2% |
| Other | 1.7% |
| Sample size = 1208 |
We asked respondents with coding experience how many years of experience they had in a coding role, excluding any years in education.
This data includes any experience in other roles and sectors, but does not define the type of experience.
| How many years of experience do you have in a coding role? | Percentage |
|---|---|
| None | 4.7% |
| Less than 1 year | 12.2% |
| 1 - 3 years | 25.7% |
| 3 - 5 years | 18.5% |
| Over 5 years | 38.9% |
| Sample size = 1208 |
We asked “Has your coding ability changed during your current role?”
This question was only asked of respondents with coding experience outside of their current role. This means analysts who first learned to code in their current role are not included in the data.
| How has your coding ability changed during your current role? | Percentage |
|---|---|
| It has become significantly better | 34.9% |
| It has become slightly better | 32.7% |
| It has stayed the same | 16.7% |
| It has become slightly worse | 10% |
| It has become significantly worse | 5.7% |
| Sample size = 1103 |
We asked all respondents who use code in their role, it they use AI tools when coding at work.
The term ‘AI’ was not defined in this question.
| Do you use any AI tools when coding at work? | Percentage |
|---|---|
| Yes | 39.4% |
| No | 60.6% |
| Sample size = 1095 |
We asked all respondents who use AI tools when coding at work, what they use these tools for.
| AI use | Yes | No |
|---|---|---|
| Converting code to a new language | 29.5% | 70.5% |
| Debugging code | 67.1% | 32.9% |
| Documenting code | 20% | 80% |
| Generating synthetic data | 12.1% | 87.9% |
| Getting answers to a coding question | 78.7% | 21.3% |
| Improving efficiency of code | 45.2% | 54.8% |
| Other | 0% | 100% |
| Testing code | 14.2% | 85.8% |
| Writing code | 68.2% | 31.8% |
| Sample size = 431 |
We asked all respondents who use AI tools when coding at work, how much they trust the accuracy and reliability of the outputs.
| Do you trust the outputs from AI tools when coding at work? | Percentage |
|---|---|
| Strongly trust | 5.8% |
| Somewhat trust | 56.4% |
| Neutral/not sure | 19.3% |
| Somewhat distrust | 16.2% |
| Strongly distrust | 2.3% |
| Sample size = 431 |
RAP refers to the use of good software engineering practices to make analysis pipelines more reproducible. This approach aims to use automation to improve the quality and efficiency of analytical processes.
The following links contain more resources on RAP:
We asked all respondents if they had heard anything about reproducible analytical pipelines (RAPs).
| Year | Heard of RAP (percent) | Lower confidence limit | Upper confidence limit |
|---|---|---|---|
| 2021 | 75.8% | 0.7276143 | 0.7867370 |
| 2022 | 82.1% | 0.7982960 | 0.8423761 |
| 2023 | 88.2% | 0.8615990 | 0.8993581 |
| 2024 | 86.2% | 0.8404089 | 0.8812603 |
We asked respondents who had heard of RAP whether they agreed with a series of statements.
| Statement | Strongly Disagree (%) | Disagree (%) | Neutral (%) | Agree (%) | Strongly Agree (%) |
|---|---|---|---|---|---|
| I think it is important to implement RAP in my work | 1.6% | 3.9% | 11.6% | 33.3% | 49.5% |
| I have the resources I need to help me implement RAP in my work | 5.7% | 14.5% | 23.2% | 31.5% | 25.1% |
| I am currently implementing RAP in my work | 10% | 14.2% | 17.6% | 29% | 29.1% |
| My department values RAP | 2.8% | 6.3% | 21.9% | 36.7% | 32.3% |
| Sample size = 1068 |
We asked respondents who reported writing code at work about the good practices they apply when writing code at work. These questions cover many of the coding practices recommended in the quality assurance of code for analysis and research guidance, as well as the minimum RAP standards set by the cross-government RAP champions network.
Coding practices have been classified as either ‘Basic’ or ‘Advanced’. Basic practices are those that make up the minimum RAP standards, while Advanced practices help improve reproducibility. The percentage of respondents who reported applying these practices either ‘Regularly’ or ‘All the time’ is shown below.
Open sourcing was defined as ‘making code freely available to be modified and redistributed’
| RAP component | Type | Percentage of analysts who code in their work |
|---|---|---|
| Use open source software | Basic | 79.9% |
| Work to quality standards | Basic | 69% |
| Peer review | Basic | 58.7% |
| Version control | Basic | 56.5% |
| Create project README | Basic | 43.6% |
| Open source own code | Basic | 15.5% |
| Manually test code | Advanced | 79.3% |
| Use control flow | Advanced | 58.4% |
| Follow code style guidelines | Advanced | 56.9% |
| Write functions | Advanced | 56.2% |
| Use configuration files | Advanced | 53.9% |
| Document dependencies | Advanced | 34.6% |
| Document functions | Advanced | 31.8% |
| Write automated tests | Advanced | 16.6% |
| Sample size = 1095 |
We asked respondents who reported writing code at work how frequently they apply good coding practices when writing code at work.
| Statement | I don't understand this question (%) | Never (%) | Rarely (%) | Sometimes (%) | Regularly (%) | Always (%) |
|---|---|---|---|---|---|---|
| Have code reviewed | 0.1% | 6.3% | 11.5% | 23.4% | 27.7% | 31.1% |
| Manually test code | 0.7% | 3.7% | 4.1% | 12.2% | 32% | 47.3% |
| Separate settings from code | 6.6% | 9% | 9.8% | 20.7% | 27.2% | 26.7% |
| Use a standard code style | 9.9% | 13.6% | 5.7% | 14% | 28.8% | 28.1% |
| Use control flow | 2.5% | 8.7% | 10.2% | 20.3% | 29.6% | 28.8% |
| Use open source software | 0.7% | 4.9% | 6.4% | 8% | 23.7% | 56.3% |
| Write automated tests | 10% | 35.6% | 20.4% | 17.4% | 10.4% | 6.2% |
| Write functions | 1.1% | 7.2% | 9.9% | 25.7% | 26.8% | 29.4% |
| Sample size = 1095 |
We asked respondents who reported writing code at work how frequently they write different forms of documentation when programming in their current role.
Embedded documentation is one of the components which make up a RAP minimum viable product. Documentation is important to help others be clear on how to use the product and what the code is intended to do.
| Statement | I don't understand this question (%) | Never (%) | Rarely (%) | Sometimes (%) | Regularly (%) | Always (%) |
|---|---|---|---|---|---|---|
| Create README files | 2.3% | 20.3% | 13.5% | 20.4% | 20.7% | 22.8% |
| Document dependencies | 4.3% | 28.1% | 14.5% | 18.4% | 17.7% | 16.9% |
| Document functions | 12.1% | 33.1% | 11.5% | 11.5% | 16.3% | 15.4% |
| Document manual QA steps | 3.5% | 14% | 11.9% | 26.3% | 27.2% | 17.2% |
| Document pipeline design | 9.3% | 23.3% | 15.4% | 23% | 19.3% | 9.7% |
| Sample size = 1095 |
We asked respondents who reported writing code at work how frequently they use good working practices in the coding projects they work on.
| Statement | I don't understand this question (%) | Never (%) | Rarely (%) | Sometimes (%) | Regularly (%) | Always (%) |
|---|---|---|---|---|---|---|
| Have a succession plan | 0.8% | 5.6% | 4.9% | 13.3% | 37% | 38.4% |
| Publish code in the open | 5% | 43.8% | 18.8% | 16.8% | 10.3% | 5.2% |
| Understand project aims | 0.1% | 1.6% | 1% | 10.4% | 40.5% | 46.4% |
| Understand project roles | 0.7% | 3% | 4.5% | 16.3% | 39.1% | 36.3% |
| Use version control | 2.7% | 20.1% | 8.5% | 12.1% | 22.1% | 34.4% |
| Work to quality standards | 4.1% | 5.6% | 5.5% | 15.8% | 37.4% | 31.7% |
| Sample size = 1095 |