Click on the following links. Please note these will open in a new window.
Identifying hypotheses and understanding the newly introduced dependent samples t test
Your book discussed the hypothesis for a dependent samples t test. In this activity, you will apply this information.
Answer the following questions:
- Under what conditions would you use a dependent samples t test?
- Go to this website: https://imgur.com/a/nRUn9
- Create a dependent samples t test hypothesis using the variables in the first graph (year and number of messages sent).
- Look at the second graph. How is the second graph similar to how a dependent samples t test is calculated?
- What makes a dependent samples t test different from a one-sample t test?
- In what way are these tests similar?
Sample size and calculation requirements of t tests
Your book discussed ideal sample size as well as how to calculate a t test by hand. In this activity, you will remind yourself how the hand calculation works and how the required calculation is related to your data points.
Go to this website: https://utopiayouarestandinginit.files.wordpress.com/2017/05/swedish-gender-wage-gap-by-percentile1.png
- How could you calculate a t test if you were interested in change over time?
- How could you calculate a t test if you were using a matched pairs design, using the variables already present? (i.e., gender).
- Let’s say that you had a sample of 100 volunteers (you did not have access to the database identified on the picture). You are looking for husband–wife pairs.
i.30 participants say they are not married.
ii.20 participants say they are in homosexual relationships.
iii.30 participants said their spouse did not want to participate.
iv.The remaining participants meet the eligibility requirements.
- Given how the dependent samples t test is calculated, how many participants do you have left?
- How does what your book says about a good enough sample for a dependent samples t test compare to what you have left?
Understanding conceptual differences between t tests: One sample versus dependent samples
So far your book has discussed two types of t tests: one sample and dependent samples. In this activity, you will practice developing hypotheses for each of those tests using the same variables.
First, in numerical terms what is the hypothesis for a:
- one-sample t test and
- dependent samples t test.
Second, use words to describe those hypotheses.
Third, go to this website: https://i.redd.it/hc8ep4vi0ezy.png
- Identify as many variables as you can from that graph. There are several. (Be sure to identify the variables, not the variable values--e.g., “seasons run” is a variable, and the values range from 8 to 30).
- Let’s say you were interested in a variable not on the graph--rating of the show.
- Using one of the variables identified above, create a repeated measures t test hypothesis with rating as the dependent variable.
- Using one of the variables identified above, create a matched pairs design with rating as the dependent variable.
- If you used seasons run, were there any that were not able to be included?
- Finally, create a one sample t test hypothesis using the variables on the graph.