Over the past few months I've been conducting a survey of registered Steam developers to get their opinions about how they really feel about Valve, called Operation Tell Valve All The Things.
PCGamer has a 90-day exclusive on the overview article which you can read here, but the 97-page report itself is fully public and is not under any publication restrictions. Feel free to share it and link to it yourself, just don't modify it.
This public version of the report is slightly different than what was sent to Valve. Chiefly, I've redacted everyone's personally identifying information and removed any responses from people who opted to exclude their data in whole or in part from the public. The vast majority of respondents (223 out of 232) shared their basic survey answers publicly, and a larger portion opted out of sharing their demographic data publicly.
The report itself with some additional context and commentary can also be found on PCGamer.
I've also embedded some of it here below:
My friend Jason O'Neil also created this visualization tool where you can compare how people voted by demographic information. (Note that in Survey 1 below the "community" field is mislabeled, but the rest of it should be correct, we'll try to fix that.)
- Dropdown or up/down on keyboard to change demographic highlighting. Note there's a bug where clicking the dropdown also advances the question. We'll fix it eventually.
- Click the arrows or left/right on keyboard to change questions.
- Mouseover the dots to see the details of each individual response
- "Show loud voices" scales responses in size proportional to the "importance" vote. "Very important" responses are larger than "Important" are larger than "Not important."
- Color coding generally follows: dark red (less fortune/advantage) to grey (middle) to dark blue (more fortune/advantage). So higher revenues or longer lifetime on steam is blue. White means "no demographic data."
The raw results are available in these spreadsheets if you want to check/reproduce my work: