In the recent years academia has been making efforts to adjust to the growing demand for data science and data scientists. These include academic journals dedicated to data science and big data, as well as scientific meetings and conferences. But despite these signs of interest in data science, it is still difficult to consider data science as an academic discipline.
Many universities have recently started to offer academic degrees in data science, and data science programs start to appear in universities and community colleges. The programs might have some large variation in curriculum, and unlike some other degree programs, a degree in data science might not provide the full information about the skills and knowledge of the person who earned it. Therefore, when hiring a data scientist one might also need to carefully review the curriculum of the specific program, as just having a data science degree might not tell the full story about the candidate.
The differences in curriculum can be attributed to the fact that data science is a new field, but can also be the result of the absence of data science departments in academia. In some cases a statistics department can change its name to “statistics and data science”, but normally the data science part is still secondary to the statistics portion. In the absence of data science departments, data science programs can be housed in the statistics department, business school, computer science department, or a program that combines several departments. The combination of departments affects the curriculum, leading to substantial differences between the different data science programs.
In addition to the absence of data science departments, academia is also missing something much more important in turning data science into an academic discipline – data scientists. Academic researchers related to data science can be broadly separated into two groups: The first is scientists involved with the data-enabled research. These researchers normally have a primary field, and they use data and computational tools to make discoveries in their domain. The second group is the methodology scientists, who focus on the development of new techniques that can be used for data science. These can include machine learning, statistics, deep learning, VLDB, HPC, and more, but these fields are not data science, and in fact have existed for many years before the recent rise of data science. Pure data scientists whose mission is to turn data, any data, into scientific discoveries in multiple fields are still rare in academia. In the absence of data science departments, hiring data scientists has the immediate obstacle of departmental affiliation and hiring committees. Many of the data scientists in academia are in fact researchers and educators who were hired for other purposes, and changed the focus of their work into data science. It can be expected that as the discipline grows, data science departments will start to form, and data scientists will become more prevalent in academia. Until then, data science cannot be considered an academic discipline.