Write a paper where you demonstrate that you can use statistics to describe data and to test hypotheses. You can use any dataset. You can also use any statistics program you’re comfortable with, though obviously JASP will be the choice for most, or all, of you.
To get started, find an interesting dataset and pick two variables that you think might be causally related, so that you have a hypothesis: X > Y. Then think of a third variable that you might need to control for to isolate the true effect of X on Y. In this paper you won’t be evaluated on the logic behind the hypothesis (why/how X might affect Y) or the quality of your measures, so don’t worry too much about those. However, in writing the paper, you will need to make sure you explain the hypothesis and measures in a few sentences, so I can follow what you’re presenting.
Your first job is to learn something about the individual variables. Then, run an appropriate test to determine if X seems to have an effect on Y. If so, how strong is the relationship, and is it positive or negative? Third, rerun the test with the third variable controlled for in an appropriate way and determine what happens to the relationship between X and Y. Then all you have to do is write up a report of what you saw and found in clear language, including whatever tables and charts you think might be helpful in helping me understand what you found. In the report, you should be sure to discuss both substantive significance (real-world importance) and statistical significance and should do your best to describe everything in language that could be understood by someone who hasn’t taken a statistics class. Don’t hide behind jargon!
