Repeating the

software survey

 

By Simon Hettrick

 5 March 2020 - Advisory Board   @sjh5000  ORCID: 0000-0002-6809-5195

Software survey 2014

Use

software

Fundamental to

results

Develop own code

69%

92%

56%

Not without problems

  1. Select 15 Russell Group universities

  2. Identify website structure for their faculties/colleges/schools

  3. Write scraper

  4. Collect all emails addresses

  5. Select 1000 email addresses from each university

  6. Send 15,000 emails

It was worth it

The new survey

210,000

academic staff
 

150,000

of whom conduct research

 

384

survey responses needed

(confidence level of 95% with margin of error of 5%)

3000

research staff
 

3000

research postgraduates

28

RSE Groups
 

88,000

research staff

They get:

 

 

 

 

 

If they:

  • CC-By the anonymised data and store it openly
  • Promise not to adapt the current questions
  • Very valuable data on software use
  • Survey
  • Analysis code
  • Ethics paperwork
  • Email text
  • Report template

Southampton

 These slides: http://bit.ly/2U8CxAy

"Vital"

"Not important"

"Professional"

"Beginner"

Studies

  • National software survey:
    https://www.software.ac.uk/blog/2017-09-06-journey-reproducibility-excel-pandas

  • Southampton software survey:
    https://github.com/Southampton-RSG/soton_software_survey_analysis_2019/blob/master/report/Research%20software%20at%20the%20University%20of%20Southampton.pdf

 

Links

  1. UKRI roadmap: http://bit.ly/UKRIRoadmap2019

  2. Southampton results: https://github.com/Southampton-RSG/soton_software_survey_analysis_2019

Software survey 2020

By Simon Hettrick

Software survey 2020

Presented at the Software Sustainability Institute Advisory Board meeting 5 March 2020.

  • 893