Open source software packages like WordPress and Firefox have been major influences on modern computing. In statistics, open source software known as “R” has become the most popular environment for statistical computation and data analysis. A recent paper in PNAS studied the developers who make this software so successful to understand what motivates them and why they find contributing satisfying.
The study’s authors collected data from approximately a thousand R contributors who responded to a questionnaire distributed via e-mail. The respondents were asked about what drove them to participate in the project, with possible answers including taking pleasure in applying their skills and feeling a sense of responsibility toward the scientific community. They were also asked about extrinsic motivators, such as the potential that their work could help with academic advancement. Additionally, the surveys included questions about the characteristics of the software development work (e.g. repetitive, technical, social) and the demographics of participants.
The researchers hypothesized that developers of open source software enjoy being able to set their own pace and are motivated by altruism, rather than any aspirations of power associated with the work.
The survey data was analyzed using a regression analysis. Results showed that developers who experienced both internal and external motivations contributed more to the project. The researchers speculate that variability in the effectiveness of extrinsic motivators for the developers may be due to the academic nature of the project.
In terms of work characteristics, developers whose work had a social element were more likely to continue contributing to the project—for example, developers working on collaborative parts of the project were more likely to continue contributing than those who worked completely alone. Additionally, the data showed that open source projects tend to promote a high degree of social dependency among participants, as participants tend to rely heavily on each other to keep the work progressing.
As a result, there’s a bit of a feedback loop. The project itself made the developers socially dependent on each other, and that novel social network increased the likelihood that they would continue to work on the project. Conversely, when a contributor’s work was focused solely on software development, participants were less involved and less likely to engage in further development.
To ensure that there were no confounding factors, the researchers also analyzed the demographic characteristics of the software developers. They found that full-time employment (particularly in a field relevant to statistics) was associated with increased participation in R development. The authors think this suggests that participation in the project may be related to work that developers are doing for their employers. Finally, the data showed that academic background had no bearing on participation.
Though the authors think this is the first empirical study to examine the developers of the R statistical package, these findings probably confirm our expectations. It’s intuitive that people who participate in open source projects tend to do so because they get a sense of personal fulfillment from the project, enjoy the self-directed nature of the work, and find the social environment of the project to be enjoyable. After all, this work is uncompensated and cuts into leisure time.
These findings are in line with current thoughts on paid employment, which suggest that you can maximize workplace productivity by giving employees a positive environment and lots of agency. It’s always great to have solid data to support those ideas.
PNAS, 2015. DOI: 10.1073/pnas.1506047112 (About DOIs).
reader comments
60