Activity 5.4
Name: 5.4 SPSS Practice Activity: Calculating Power and t Values from Example Datasets and Interpreting Their Meaning
Type of activity: G*Power and SPSS Practice Activity
Activity description/instructions:
The following activity will present an opportunity to put into practice what you have learned and read about regarding power estimation and various types of t-tests available for use. The narratives supplied in these exercises will provide all the information you will need to determine a required sample size based on power and effect size, and what types of t-tests need to be executed in order to obtain valid results. Please read the following narrative and then use the Excel spreadsheet provided to import into SPSS to create a dataset that SPSS can use to perform the proper calculations.
Problem Narrative 5.4
Eagle Sleep Research, Inc. hypothesizes that people who are allowed to sleep for only four hours will score differently than people who are allowed to sleep for eight hours on a cognitive skills test. The researchers believe that thirty participants will be sufficient for their research given a moderate effect size and a desired power of 0.7. Therefore, they plan to recruit thirty participants into the sleep lab and randomly assigned to one of two groups. In one group (Group 1) participants sleep for eight hours and in the other group (Group 2) participants sleep for four hours. The next morning a cognitive skills test will be administered to all participants. Scores on the cognitive skills test range from 1-9 with high scores representing better performance. Before any data collection or participant recruitment is to begin, the researchers wish to check to see if thirty participants will be sufficient. They decide to use G-power to calculate their desired level of power using a two-tailed test, a moderate effect size of .5, and an alpha level of .05, and a desired power of 0.70.
Part I: Import Data into SPSS:
Use the spreadsheet you just downloaded “SleepResearch2” and import the data in this spreadsheet into SPSS for subsequent analyses. Name your SPSS data file “5.4 SPSS
SleepResearch2 Dataset” with the “.sav” extension. Your SPSS data file should be ready to use for analysis.
Part II: Short Answer Questions:
Create an MSWord document to answer the short answer questions from part II and III of this exercise. The name of the file will be “5.4 SPSS Exercise – Short Answer Questions”. Save this for subsequent upload as one of your deliverables for this exercise. Answer the following questions using the narrative provided above and G*Power:
1. Is the proposed sample size of 30 participants (15 per condition) sufficient to obtain the level of power desired?
2. What does it mean if your sample size is not sufficient to obtain the level of power desired?
3. Based on your calculations done in G*Power, what would you recommend to the lead researcher on this project with regard to the recommended sample size?
After you run your G*Power analysis and obtain your results, click on the “Protocol of Power Analyses” tab at the top of your G*Power window. A window will appear with your parameters and output. Select “Print” and print your G*Power output as a pdf file. Save this pdf file for upload as one of your deliverables for this exercise. Name your pdf file “5.4 GPower Protocol Output”.
Answer the following questions based on the narrative provided and add this to your short answer deliverable:
1. What kind of t-test will need to be performed (i.e., are these means and groups independent or related in any way) 2. Is this a one-tailed or two-tailed test
3. Calculate degrees of freedom.
4. What is the hypothesis?
5. Look up the value for t-critical in Table C.2, Appendix C in the text book at the p<.05 level for the type of test you identified earlier (one or two tailed?).
Part III: Running the Appropriate Analysis in SPSS:
Using the SleepResearch2 dataset, run the appropriate t-test in SPSS. Add the following to your list of short answer questions as information obtained from your SPSS output file:
1. What are your numerical results (t value, degrees of freedom, etc., written in APA format for results write up – t(xx) = x.xx, p = xx ).
2. What is your N for each group?
3. What are the means for each group?
4. What is the standard deviation for each group?
Save the SPSS output you just generated when running the analysis on your data as “5.4 SPSS SleepResearch2 Output” as a “.spv” file to upload as deliverable at the conclusion of this exercise. Part IV: Results Write Up:
Write up the results of your data analysis in standard APA format being sure to mention all relevant variables and findings in your write up. Make sure your format adheres to APA format and guidelines. Save the document as “5.4 SPSS Results Write Up”.
(If you have questions about what a complete results write up for a t-test looks like, either look up conference and journal articles that use t-tests as their statistical analysis or contact your instructor for specific guidance.)
Your deliverables for this SPSS exercise will be:
1. The G*Power output file “5.4 GPower Protocol Output”.
2. The IBM SPSS data file “5.4 SPSS SleepResearch2 Dataset”.
3. The IBM SPSS viewer file “5.4 SPSS SleepResearch2 Output”.
4. The Short Answer document “5.4 SPSS Exercise – Short Answer Questions”.
5. The APA formatted results write up “5.4 SPSS Results Write Up”.
Upload all five of these files as your submission for this exercise to be considered for full credit.
