Deciding on the right method of analyzing empirical research data can be difficult - especially when it comes to choosing specific methods of inferential statistics, where there is often not one right way, but a variety of valid options (e.g., using an ANOVA or its non-parametric alternative for data that is not perfectly normally distributed). Novice researchers not only lack the experience to know when a particular hypothesis test is appropriate, but struggle to find suitable literature to familiarize themselves (i.e., literature that is not too superficial, yet comprehensible).

With the present article we provide a remedy following the didactic method of scaffolding: We present a systematization of the most elementary inferential statistical methods, namely hypothesis tests for group differences with one independent variable and one (quasi-)metric dependent variable. We start by explaining basic terms (e.g., independent variable) and then give step-by-step instructions for choosing a proper hypothesis test based on data properties, implementing it from scratch, and reporting or interpreting its results. With these practical cookbook-like guidelines, this article serves as a concise starting point for young researchers entering the field of empirical research, as a valuable resource for their instructors, and as a basis for automating statistical procedures in a software system.

Keywords: Empirical research, data analysis, inferential statistics, guidelines.