Premium
Comparing four contemporary statistical software tools for introductory data science and statistics in the social sciences
Author(s) -
Abbasnasab Sardareh Sedigheh,
Brown Gavin T. L.,
Denny Paul
Publication year - 2021
Publication title -
teaching statistics
Language(s) - English
Resource type - Journals
SCImago Journal Rank - 0.425
H-Index - 13
eISSN - 1467-9639
pISSN - 0141-982X
DOI - 10.1111/test.12274
Subject(s) - computer science , software , statistical software , statistical analysis , data science , visualization , computational statistics , data visualization , perspective (graphical) , interface (matter) , usability , statistics education , human–computer interaction , mathematics education , statistics , artificial intelligence , machine learning , psychology , mathematics , programming language , bubble , maximum bubble pressure method , parallel computing
Research students in social science disciplines frequently struggle to master statistical analysis. A contributing factor may be the statistical software that is used, as the design of such software may not address the needs of non‐statisticians or non‐computer programming students. Hence, decisions about which statistical software tools are most suitable for such end‐users need to be made at the introductory level. This paper first identifies key human‐computer interaction (HCI) factors that may directly influence students' statistical analysis performance. Factors include technical properties such as user interface design, statistical features available, visualization, data handling, preparation, and manipulation, and usage properties such as speed/number of steps, ease of command/use, and efficiency. Four popular software systems (ie, SPSS, R within RStudio Desktop, R Commander & jamovi) were evaluated. Findings suggest that HCI usage factors from an interaction perspective are likely to be especially important for students gaining an introductory knowledge of statistics.