# Tasks:

# Test your knowledge

## Note on Tasks

Before completing the Tasks, please work through each of the sub-pages for Tutorial 4 as these will help complete the task below.

For the task below we will be using the following dataset:

Skoczylis, Joshua, 2021, "Extremism, Life Experiences and the Internet", https://doi.org/10.7910/DVN/ICTI8T, Harvard Dataverse, Version 3.

## Task 1: Independent T-Test or non-parametric equivalent

Formulate a hypothesis and alternative hypothesis that allows you to test whether there is a difference in the Strain/Resilience Score between men and women.

Now answer the following:

Are the assumptions for parametric tests met?

Which test do you select and why?

Do you accept or reject the Null hypothesis?

What is your p-value and effect size?

Finally, generate a graph that illustrates your outcome (consider using JJStats Plot, Survey Plot or Flexplot)

## Task 2: ANOVAs or non-parametric equivalent

Formulate a hypothesis and alternative hypothesis that allows you to test what impact education has on a person's Racism Score.

Now answer the following:

Are the assumptions for parametric tests met?

Which test did you select and why?

Do you accept or reject the Null Hypothesis?

What is your p-value and effect size?

What does the post-hoc test tell you about the difference?

Now generate a graph to illustrate your output (consider using JJStats Plot, Survey Plot or Flexplot).

## Task 3: Factorical ANOVA

For the final task, formulate a hypothesis that allows you to test the impact of Gender and Religious Affiliation on a person's self-confidence score.

Note: Consider reducing the options for the religion variable by creating a new variable with fewer options.

Now answer the following:

Are the assumptions for parametric tests met? In this case, it is worth checking the Q/Q plot as the line only slightly diverges, continue to use a parametric test or use a Robust ANOVA.

Which test did you select and why?

Do you accept or reject the Null Hypothesis?

What is your p-value and effect size?

What does the post-hoc test tell you about the difference?

Does the interaction of Religious affiliation and Gender have an impact on a persons confidence

Now generate a graph to illustrate your output (consider using JJStats Plot, Survey Plot or Flexplot).