Results for the 2015 Scientific Python Survey
This repository is a companion to this blog post.
The results so far are contained in the
results.json file in JSON format.
This file will be updated as the results of the survey are processed and
anonymized. The file contains a list of entries, where each entry is a response in the survey. At the moment, this contains 781 responses (there were a few more responses, but these had issues and will be added manually very soon).
In version 1 (released 9th May 2015), each entry contains:
experience: how long the respondent has been using Python
fields: a list of fields that the respondent works in
installations: a list of dictionaries, where each dictionary corresponds to a single installation, which then gives the operating system and Python version (and in future, this will give the NumPy, SciPy, and Matplotlib versions). The list gives the primary installation first.
why_not_python3: a list of reasons as to why the respondent is not using Python 3 (if that is the case)
Note that for several of the fields, one of the options is 'Other'. Respondents were able to provide free-form text descriptions in these cases, but these data need to first be cleaned up and anonymized before they can be added here.
9 May 2015: Initial version
9 May 2015: Updated to include fields of research parsed from the 'Other' field
The following fields of research were not present from the original form but were parsed from the 'Other' field:
- Computer Science
- Machine Learning