Data science education overview

More and more industry and government organizations recognize the value of data for decision-making and have set up general and specific data scientist roles to support data science and engineering, e.g., Chief Data Officer, Chief Analytics Officer, data modelers and data miners, in addition to data engineers and business analysts.

An increasing number of data science courses are available from research institutions and professional course providers. However, they look more like “old wine in new bottles”, i.e., a re-labeling and combination of existing subjects in statistics and IT. A fair summary of the templates for creating the current so-called data science courses is the hybridization of statistics, computing and informatics; or in a more specific sense, the combination of statistical analysis, data mining, machine learning, programming, and case study assignments.

With regard to course-offering disciplines, many more courses are available today in data science and analytics compared to five years ago when the first research degrees, Master of Analytics and PhD in Analytics, were established [6]. This is in addition to related courses offered in classic disciplines, in particular statistics, informatics and computing. As shown in references [7], [8], [9], [10], [11], the following is clear:

  • Core relevant disciplines offering data science-related courses include statistics and mathematics, IT (in particular, informatics and computing), business and management. Interestingly, more courses are offered in business and management than in science, engineering and technology (in particular, IT and statistics) [7]. As a result, such courses usually focus on offering a specific discipline-based body of knowledge.
  • An increasing number of initiatives have been created or are in development in seemingly ‘irrelevant’ disciplines, such as environment science, geographical science, physics, health and medical science, finance and economics, and even agricultural science.
  • Most courses are offered by a single faculty and in an individual discipline. Joint efforts across multiple relevant disciplines need to be made to create data science courses that go beyond a narrow focus and satisfy holistic requirements. In interdisciplinary terms, a simple solution is to create courses by mixing components offered in both statistics and IT/computing.

Data science courses are generally offered through the following channels:

  • On-campus courses: Most awarded courses and short courses are offered on campus, and are based on classic teaching/learning modes that include lectures, tutorials, and lab practice.
  • Off-campus courses: Online courses and in-house corporate training are the main types of off-campus courses, which are usually offered to professionals and people who are not seeking a degree qualification.
  • Mixed channel courses: More data science courses are offered jointly by on-campus and off-campus modes, and in other disciplines, than has previously been the case.

 

References:

[1] L. Cao. Data science: Profession and education.

[2] UTS, “Master of analytics by research and PhD thesis: Analytics, Advanced Analytics Institute, University of Technology Sydney,” 2011.

[Online]. Available: www.analytics.uts.edu.au

[3] Silk, “Data science university programs,” 2016. [Online]. Available: http://data-science-university-programs.silk.co/

[4] DSC, “College & university data science degrees,” 2016. [Online]. Available: http://datascience.community/colleges (accessed on 16 April 2016.)

[5] Github, “Data science colleges,” 2016, (retrieved on 4 April 2016). [Online]. Available: https://github.com/ryanswanstrom/awesomedatascience-colleges

[6] Classcentral, “Data science and big data — free online courses,” 2016. [Online]. Available: https://www.class-central.com/subject/data-science

[7] USDSC, “US degree programs in analytics and data science,” 2016. [Online]. Available: http://analytics.ncsu.edu/?page id=4184

Email:contacts@datasciences.org