A NO-CODE SOLUTION FOR TEACHING STATISTICAL HYPOTHESIS TESTING FOR STUDENTS IN THE SOCIAL SCIENCES AND ECONOMICS

Keywords: RapidMiner, hypothesis testing, t-test, proportion test, ANOVA, no-code platform, Python macro integration

Teaching statistical hypothesis testing to students in the social sci-

ences and economics presents many challenges, as students in these fields

often lack programming skills, while advanced platforms such as Altair AI

Studio (RapidMiner) do not natively support common statistical testing

tools like t-tests, proportion tests, and ANOVA analysis. Our solution is

to build testing processes in Altair AI Studio (RapidMiner) that do not

require users to program. These processes primarily rely on using macro

settings to provide parameters and leverage internal Python execution

capabilities to handle the back-end. The developed processes include

one-sample t-tests, independent two-sample t-tests, paired two-sample t-

tests, one- and two-proportion tests, and ANOVA with post-hoc analysis.

Students only need to set input values via macros, eliminating the need

for programming and allowing them to focus on understanding core sta-

tistical concepts. A dataset, namely, Amazon service reviews, was used

to demonstrate the flexibility and applicability of the proposed solution.