Why might the KW be more appropriate to use to analyze the relationship between these variables? Which test would your report in a research article and why?

The essay portion of the homework assignment to be submitted is:

1) In a brief paragraph, for each of the three parametric and one non-parametric analyses, write a brief synopsis of your results. Include the research hypothesis (not the null, as I want you to write it more like a research article in which you do not see a null stated, as the null is implied); tell me if the research hypothesis is supported by these data and the strength of the association. If the results are significant then you must go on to interpret the multiple comparison tests and tell the reader which groups differ and by how much. Lastly, identify the assumptions for ANOVA and if they were met in each analysis.

2) Look at the third ANOVA analysis {i.e., relationship between the hours of TV watched and general happiness} and then the Kruskal-Wallis for the same variables. Compare the results of these two tests.

Why might the KW be more appropriate to use to analyze the relationship between these variables? Which test would your report in a research article and why?