What is the main focus of non-parametric statistics?

The main focus of non-parametric statistics is indeed on observations that involve nominal or ordinal data. Non-parametric methods do not assume a specific distribution for the data, making them particularly useful when dealing with data that does not meet parametric assumptions, such as normality. This approach is valuable for analyzing ranked data or categorical data, which are common in social science research, including psychology.

Non-parametric statistical tests, like the Wilcoxon signed-rank test or the Kruskal-Wallis test, allow researchers to analyze data without the constraints of assumptions required for parametric tests. As a result, these methods are adaptable for a wider variety of research scenarios, where traditional parametric tests might be inappropriate due to the nature of the data.

The other options highlight concepts that are more aligned with parametric statistics. For instance, testing based on population means and providing estimates for population parameters are hallmarks of parametric methods, which rely on certain assumptions about the underlying population distribution. Additionally, the focus on large sample sizes tends to relate more closely to the strengths of parametric tests, which often require sufficient sample size to meet their assumptions, rather than the objectives of non-parametric statistics. Thus, the correct choice is clearly linked to the foundational characteristic