R code and data for the .impact 2014 survey of effective altruists
Switch branches/tags
Nothing to show
Clone or download
Fetching latest commit…
Cannot retrieve the latest commit at this time.
Failed to load latest commit information.


The 2014 EA Survey

In May 2014, a team from .impact released a survey to better understand the EA community. Before this, we’ve mostly been using our first-hand, anecdotal impressions to describe the community. A few EA organizations have done surveys of their own supporters, but this particular survey supplies a more systematic understanding of EAs.

The full analysis was published on the EA forum on March 2015.

Replicating the Analysis

The core code for analysis is in imsurvey.R. It reads in data from imdata.csv and imdata-more-results.csv, which were two downloads of data collected from running the EA survey. It also reads data in from ace-results.csv, which was a specific survey run by Animal Charity Evaluators targeted at their audience, which included all of our questions along with some additional questions. All of that data is combined, throughly cleaned, and then analyzed.

The tech of the analysis relies heavily on the R packages magrittr, dplyr, and surveytools2.

To run the analysis, clone this repo, open up R within the repo folder, and then either source("imsurvey.R") or run individual parts of imsurvey.R.

Comparison with the LessWrong survey was done in lesswrong.R using the 2013 LessWrong survey data set, lwpublic2013.csv, for analysis. To run that analysis, source("lesswrong.R") within this repo.