SuAVE proved to be a useful tool for teaching undergraduate research methods in sociology, as many students came to statistics classes with little math background and a lot of anxiety. SuAVE’s user-friendly, visual approach to statistical analysis enabled students to explore simple social science questions with survey data in an intuitive way, without having to master first the mathematical tools of statistics. Its unique capacity to animate transitions between the big picture and individual cases with great ease was critical to explaining statistical approaches to students, especially in presenting the relationship between small N, case based, idiographic, comparative arguments, and large N, variable based, nomothetic explanations.

The visual approach implemented in SuAVE allows students to scrutinize cases by looking up all relevant information about a specific case. This proves to be helpful in three ways:

  • it allows the student to look more closely at unusual values in univariate distributions, to find out whether these are true outliers or data errors.
  • it lets the student examine outliers in bivariate distributions of correlated indicators of the same concept to better understand the ability of individual indicators to capture underlying concepts.
  • using deviant case analysis to discover causal explanations with SuAVE, students learn that, rather than be treated as random noise, exceptions to patterns can teach us a lot about the social world.

Here are selected student comments from the first two classes taught by Prof. Akos Rona-Tas in 2015 and 2016 (SOCIOLOGY 103M “Computer Applications to Data Management in Sociology”):

  • I really liked used SuAVE. I think the visualization portion of it was extremely useful and I learned a lot from using it. In addition, it was very user friendly and easy to navigate.
  • It was cool to see how you can find specific information about one respondent. The way to categorize respondents and understand odd findings was interesting because you would sometimes expect the opposite.
  • I really enjoyed SuAVE, it made the analyzing procedure easier by allowing us to look at multiple variables at once.

For more information see the “Using SuAVE as a teaching tool” Powerpoint presentation by Prof. Rona-Tas.

Additional features of SuAVE, such as the ability to launch various survey analysis, image processing, machine learning, and text analytics codes implemented in Jupyter notebooks, integration with LimeSurvey, and the annotation and view sharing functionality, have let to additional uses in the classroom. Here are a few resources and examples: