SURVEY RESULTS

Before launching the Data Science Academy, we have run a survey to understand the needs of the research community for further education in data science. All respondents have voluntarily taken the survey that was advertised on social media and a computer calculations mailing list. Below, we report the most important findings.  Would you have any questions, contact us at hello@dasc.academy

Demographics

Most of the respondents are from physical sciences. This is because we contacted many networks of researchers in our field. Despite not all areas are equally represented, we have found that the most results are similar for all disciplines. The only outliers are the programming skills and  math background, which are higher for physical sciences. As for the career level of the respondents, it approximately split in 4 equal parts betweem MSc students, PhD students, postdocs, and more established researchers.

Preferred learning style

Approximately 50% of the respondents would like to improve their data science skills and they would do that in courses of several weeks. Almost all of them would relocate to attend the course. All respondents are quite open to the way they’d work on a project.

Experience in programming and data science

As mentioned above, these results are skewed by the high preponderance of respondents from the physical sciences. It gives us a snapshot about the diversity of backgrounds in programming and in math. Notably, only 0.5% do not have programming experience, while the rest is confident either with low level programming languages or with high level packages where scripting is required.  

In addition, most people have a sound background in statistics, and many have at least a good understanding of calculus and linear algebra. This means that a course in data science would be easily accessible for researchers from all disciplines, with physical scientists having a slight advantage compared to the others. We would strongly recommend those with little programming and/or math skills, to first become familiar with some of the concepts and methods from these disciplines which would allow them to learn data science.