Coding frequency and tools

Coding frequency

We asked respondents “In your current role, how often do you write code to complete your work objectives?”.

Coding frequency	Count
Never	6
Rarely	8
Sometimes	20
Regularly	35
All the time	61
^a Sample size = 130

What people are using code for

We asked respondents what data operations they carry out in their work, and whether they use code to do them. Please note we did not ask how much of each data operation is done with code, or how often.

Show chart Show table

Data operation	I do some or all of this by coding	I do this without coding
Analysis	116	8
Data cleaning	110	5
Data linking	97	6
Data transfer	63	15
Data visualisation	106	13
Machine learning	64	4
Modelling	80	5
^a Sample size = 130

Access to and knowledge of programming languages

Given a list of programming tools, we asked respondents to answer “Yes”, “No” or “Don’t know” for the following statements;

This tool is available to use for my work.
I know how to program with this tool to a level suitable for my work.

Access to tools does not necessarily refer to official policy. Some analysts may have access to tools others cannot access within the same organisation.

Please note that capability in programming languages is self-reported here and was not objectively defined or tested.

Access to coding tools

Show chart Show table

Programming language	Yes	Don't Know	No
C++ / C#	18	50	62
Java / Scala	28	52	50
Javascript / Typescript	43	41	46
Matlab	7	52	71
Python	109	11	10
R	125	2	3
SAS	58	31	41
SPSS	34	43	53
SQL	111	10	9
Stata	24	57	49
VBA	84	27	19
^a Sample size = 130

Knowledge of coding tools

Show chart Show table

Programming language	Yes	Don't Know	No
C++ / C#	17	3	110
Java / Scala	13	2	115
Javascript / Typescript	27	5	98
Matlab	35	4	91
Python	94	2	34
R	108	0	22
SAS	41	4	85
SPSS	33	4	93
SQL	110	0	20
Stata	16	4	110
VBA	47	7	76
^a Sample size = 130

Access and knowledge gaps

Using the data presented above we calculated the number of respondents with:

Access to tools they do not have the capability to use (access only),
Access to tools they are able to use (access and knowledge)
Or capability to use tools they cannot access at work (knowledge only)

Show chart Show table

Programming language	Access only	Access and knowledge	Knowledge only
C++ / C#	13	5	12
Java / Scala	22	6	7
Javascript / Typescript	23	20	7
Matlab	4	3	32
Python	25	84	10
R	19	106	2
SAS	28	30	11
SPSS	22	12	21
SQL	9	102	8
Stata	15	9	7
VBA	39	45	2
^a Sample size = 130

Coding capability

Change in coding ability during 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.

Show chart Show table

Ability Change	Count
Significantly worse	5
Slightly worse	13
No change	14
Slightly better	41
Significantly better	50
^a Sample size = 123

Where respondents first learned to code

Respondents with coding experience outside their current role were asked when they first learned to code. This output also includes analysts who code in their current role but reported no other coding experience.

These data only show where people first learned to code. They do not show all the settings in which they had learned to code, to what extent, or how long ago.

Show chart Show table

First coding experience	Count
In current role	9
In education	72
In private sector employment	5
In public sector employment	16
Self-taught	23
Other	4
^a Sample size = 129

Coding practices

We asked respondents who said they currently use code in their work how often they carry out various coding practices. For more information on the practices presented below, please read our guidance on Quality Assurance of Code for Analysis and Research.

General coding practices

Show chart Show table

	Percent
Question	I don't understand this question	Never	Rarely	Sometimes	Regularly	All the time
I use open source software when programming	4.0	1.6	3.2	14.5	20.2	56.5
My team open sources its code	4.8	21.8	21.0	25.0	15.3	12.1
I use a source code version control system e.g. Git	0.0	10.5	8.9	12.1	16.9	51.6
Code my team writes is reviewed by a colleague	0.8	2.4	7.3	16.1	28.2	45.2
I write repetitive elements in my code as functions	0.8	0.8	4.0	19.4	32.3	42.7
I unit test my code	5.6	12.9	24.2	21.8	18.5	16.9
I collect my code and supporting material into packages	1.6	33.9	16.1	25.0	14.5	8.9
I follow a standard directory structure when programming	10.5	9.7	13.7	15.3	29.0	21.8
I follow coding guidelines or style guides when programming	0.8	4.0	8.9	21.8	36.3	28.2
I write code to automatically quality assure data	1.6	12.9	18.5	28.2	25.0	13.7
My team applies the principles set out in the Aqua book when carrying out analysis as code	26.6	12.9	10.5	17.7	21.0	11.3
^a Sample size = 124

Documentation

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.

Show chart Show table

	Percent
Question	I don't understand this question	Never	Rarely	Sometimes	Regularly	All the time
Code comments	0.8	0.8	3.2	8.9	29.0	57.3
Documentation for each function or class	3.2	12.1	6.5	21.0	32.3	25.0
README files	0.8	15.3	12.9	10.5	30.6	29.8
Desk notes	19.4	21.8	11.3	22.6	18.5	6.5
Analytical Quality Assurance (AQA) logs	13.7	19.4	21.0	29.0	12.1	4.8
Data or assumptions registers	12.9	34.7	17.7	12.9	16.9	4.8
Flow charts	1.6	22.6	24.2	32.3	16.1	3.2
^a Sample size = 124

Dependency management

Respondents who currently use code in their work were asked whether they use any tools for dependency management. We provided examples of tools that may be used for dependency management:

Requirements files, e.g. python requirements.txt or R DESCRIPTION files
Virtual environments (e.g. venv or renv) or virtual machines
Containers e.g. Docker

Show chart Show table

Use dependency management software	Count
Yes	68
No	39
I don't know what dependency management is	17
^a Sample size = 124

Reproducible workflow packages

Respondents were asked whether they use continuous integration technologies. As above, respondents were provided with examples of what those might be:

GitHub actions
Jenkins
Travis

Show chart Show table

Use reproducible workflow packages	Count
Yes	16
No	89
I don't know what reproducible workflows are	19
^a Sample size = 124

RAP knowledge and opinions

We asked respondents about their knowledge of and opinions on reproducible analytical pipelines (RAP). RAP refers to the use of practices from software engineering to make analysis more reproducible. These practices build on the advantages of writing analysis as code by ensuring increased quality, trust, efficiency, business continuity and knowledge management. The RAP champions are a network of analysts across government who promote and support RAP development in their departments.

Knowledge of RAP

We asked our respondents whether they had heard of RAP and what their knowledge is of their own department RAP champion.

Show chart Show table

RAP champion knowledge	Count
Have not heard of RAP	17
Heard of RAP, have not heard of RAP champions	19
Heard of RAP, does not know department champion	31
Heard of RAP champions, no champion in department	5
Knows department RAP champion	44
^a Sample size = 130

Show chart Show table

Line manage anyone who writes codes	count
Yes	55
No	27
I don't line manage anyone	48
^a Sample size = 130

Opinions on RAP

We asked respondents who had heard of RAP whether they agreed with a series of statements.

Show chart Show table

	Percent
Question	Strongly disagree	Disagree	Neutral	Agree	Strongly agree
I feel confident implementing RAP in my work	3.5	9.7	16.8	38.1	31.9
I feel supported to implement RAP in my work	2.7	10.6	19.5	38.1	29.2
I know where to find resources to help me implement RAP	1.8	15.0	15.0	38.1	30.1
I understand what the key components of the RAP methodology are	2.7	8.8	15.0	38.1	35.4
I think it is important to implement RAP in my work	3.5	0.9	5.3	40.7	49.6
I and/or my team are currently implementing RAP	6.2	13.3	20.4	26.5	33.6
I or my team are planning on implementing RAP in the next 12 months	5.3	11.5	23.0	27.4	32.7
^a Sample size = 113

RAP scores

RAP component scores “regularly” or “all the time” to the relevant questions. For documentation, this includes both code comments and README files. For the continuous integration and dependency management components we only collected “yes”, “no” or “I don’t understand the question” responses. As such, we gave “yes” responses a score of 1. The sum total of each respondent’s scores is presented here as “RAP scores”.

A score of one for each RAP component is derived where respondents answered “regularly” or “all the time” to the relevant questions. For documentation, this includes both code comments and README files. For the continuous integration and dependency management components we only collected “yes”, “no” or “I don’t understand the question” responses. As such, we gave “yes” responses a score of 1. The sum total of each respondent’s scores is presented here as “RAP scores”. “Basic components” are the components which make up the RAP MVP. “Advanced components” are components that help improve reproducibility, but were are considered part of the minimum standard.

RAP components

Show chart Show table

Component	Type	Count
AQUA book guidance	Basic	40
Documentation	Basic	68
Peer review	Basic	91
Team open source code	Basic	34
Use open source software	Basic	95
Version control	Basic	85
Code packages	Advanced	29
Continuous integration	Advanced	47
Dependency management	Advanced	68
Follow code style guidelines	Advanced	80
Function documentation	Advanced	71
Functions	Advanced	93
Unit testing	Advanced	44
^a Sample size = 124

Basic RAP scores

Show chart Show table

Basic RAP score	Count
0	5
1	21
2	9
3	24
4	31
5	26
6	8
^a Sample size = 124

Advanced RAP scores

Show chart Show table

Advanced RAP score	Count
0	10
1	18
2	16
3	20
4	21
5	11
6	13
7	15
^a Sample size = 124

Profession profile: Government Data Scientist

Coding frequency and tools