RSE International Survey Analysis

This repository is used to analyse international surveys conducted by the Software Sustainability Institute.

We ran the first survey in 2016, which provided an insight into the demographics, job satisfaction, and practices of research software engineers (RSEs) in the UK. To support and broaden this work, the institute will conduct the survey at regular intervals and extend the geographical coverage to facilitate inter-country comparisons. The results of the surveys, anonymised and open licensed, will act as a a valuable resource to understand and improve the working conditions for RSEs.

In 2017 we also surveyed Canadian RSEs and we added four countries, Germany, Netherlands, South Africa and USA. Our thanks to our partners: Scott Henwood (Canada), Stephan Janosch and Martin Hammitzsch (Germany), Ben van Werkhoven and Tom Bakker (Netherlands), Anelda van der Walt (South Africa) and Daniel Katz and Sandra Gesing (USA).

Since 2018 we have worked differently and created a survey for all countries (rather than one survey for each one).

This site covers the 2022 survey results.

Setup

This repository is only for the survey analysis. Here's how to reproduce the analysis on your own computer. The following instructions only apply for the 2018 and 2022 surveys.

To reproduce the results on your machine, first clone the repository and setup the Python virtual environment:

git clone --recurse-submodules https://github.com/softwaresaved/international-survey-2022
cd international-survey-2022
python -m venv venv  # use python3 if your default python is still Python 2
source venv/bin/activate
python -m pip install -r requirements.txt

Use the following to set the environment variables on macOS and Linux. If you are on Windows, use the documentation linked here.

export RSE_SURVEY_YEAR=2022  # optional, this is detected automatically if in a folder like *-2018
export RSE_SURVEY_YEAR_PREV=2018  # optional, only needed if previous year != current year - 1
export RSE_SURVEY_FIGURE_TYPE=svg  # optional, set to svg or png to generate figures in that format
export RSE_SURVEY_FIGURE_DPI=300  # optional, set dpi for png or pdf output formats

Generating the full report

You can generate the full report (other than the country specific reports) by running:

python run.py

Country reports should be generated after the section specific reports (do not replace country with a specific country, this will generate all country reports):

python run.py country

If you'd like to view this report locally, you can do so via:

bundle install
bundle exec jekyll serve

Then you will be able to view the report by pointing a browser at http://127.0.0.1:4000/international-survey-2022/.

Generating individual sections of the report

To generate sections individually, first ensure that you have run the initialisation (which is the overview and sampling section) that is required by other sections. This can be done by passing init to run.py:

python run.py init

This should create a cache/processed_data.csv file within the docs folder. Once this is generated, you can run any of the sections in any order:

python run.py <section>

How it works

Generating the report

So for example to generate the current-employment section: python run.py current-employment

• Two types of reports are generated: section reports and country reports. Section reports are useful for comparing countries for a particular section of the survey, such as demographics or job satisfaction, whilst country reports give an overview of a country across the different sections of the survey. Both the section and country reports utilise templates from templates corresponding to each section (or from international survey lib submodule repo if they are not found). The country report template country_report.md is also found and used from one of these locations in the same way as for sections. The template file uses the Mustache templating language via the chevron module.
• Section reports are generated in _section
• Country reports are generated in _country
• Each table in the report has a corresponding CSV in csv
• Each figure in the report has a corresponding SVG in fig
• The source data used for the analysis is at data

Displaying the report as a website

• We use Jekyll and Github Pages to build the website hosted at https://softwaresaved.github.io/international-survey-2022.
• Stylesheets for the site are at _sass
• Configuration of the site is at _config.yml
• The hosted site can be viewed locally by using the command bundle exec jekyll serve in the docs folder and following the localhost URL. If this is the first time setting up Jekyll on your computer, ensure that you have Ruby and Bundle installed (gem install bundler).

Contributors

Here is a list of contributors for the 2022 version of the survey (last name alphabetic order). They are also mentioned in the .zenodo.json to be automatically added to the DOI.

• Radovan Bast, Norway/Nordics
• Alex Botzki, Belgium
• Jeff Carver, US
• Ian Cosden, US
• Steve Crouch, UK
• Florencia D’Andrea, Argentina
• Abhishek Dasgupta, UK
• William Godoy, US
• Alejandra Gonzalez-Beltran, Argentina
• Ulf Hamster, Germany
• Scott Henwood, Canada
• Simon Hettrick, UK
• Patric Holmvall, Sweden/Nordics
• Stephan Janosch, Germany
• Thibault Lestang, France
• Nick May, New Zealand
• Olivier Philippe, UK
• Johan Philips, Belgium
• Nooriyah Poonawala-Lohani, New Zealand
• Paul Richmond, UK
• Manodeep Sinha, Australia
• Florian Thiery, Germany
• Ben van Werkhoven, Netherlands
• Claire Wyatt, UK
• Qian Zhang, Canada

Licence

This repository contains code and public data. We have different licence for each

• The code is released under BSD 3-Clause License.
• The data stored in this repository is under the CC BY 2.5 SCOTLAND.

The repository is also archived on Zenodo: https://doi.org/10.5281/zenodo.6884882. If you want to cite this work and need a citation in a specific format, you can use the citation service on the zenodo.

Citations

The citation for the 2022 version is:

Simon Hettrick, Radovan Bast, Alex Botzki, Jeff Carver, Ian Cosden, Steve Crouch, Florencia D’Andrea, Abhishek Dasgupta, William Godoy, Alejandra Gonzalez-Beltran, Ulf Hamster, Scott Henwood, Patric Holmvall, Stephan Janosch, Thibault Lestang, Nick May, Olivier Philippe, Johan Philips, Nooriyah Poonawala-Lohani, Paul Richmond, Manodeep Sinha, Florian Thiery, Ben van Werkhoven, Claire Wyatt & Qian Zhang. "RSE Survey 2022", Pre-final release for 2022 results (Version 2022-v0.9.0). Zenodo DOI: https://doi.org/10.5281/zenodo.6884882. Check the repository's citation file.

The citation for the 2018 version is:

Olivier Philippe, Martin Hammitzsch, Stephan Janosch, Anelda van der Walt, Ben van Werkhoven, Simon Hettrick, … Manodeep Sinha. (2019, March 6). softwaresaved/international-survey: Public release for 2018 results (Version 2018-v.1.0.2). Zenodo. http://doi.org/10.5281/zenodo.2585783

Funders

The Software Sustainability Institute is supported by EPSRC grant EP/H043160/1 and EPSRC/ESRC/BBSRC grant EP/N006410/1, with additional project funding from Jisc and NERC. Collaboration between the universities of Edinburgh, Manchester, Oxford and Southampton.