In a Repeated Measures study, one sample experiences multiple treatments. Joe use the Excel Tool Pak to perform the appropriate ANOVA for the hypothesis tests and then shows you how to use Excel's built-in statistical functions to understand the results. You'll then have an in-depth understanding of this important technique.
- [Voiceover] Now I'm gonna tell you about a very popular way of doing research. It's called a repeated measures design. In this type of study, you use just one sample. Unlike other research designs, everyone in the sample goes through all the conditions of the study. Why would you use this design? Repeated measures requires fewer individuals than other designs, where each condition gets its own unique sample. In this kind of study, each person experiences the conditions in a different random order, or you could also use this design when you wanna track someone's progress in learning a skill, either a physical skill or a cognitive one.
You measure them repeatedly; here's an example. Researchers at the University of Brighton in England were interested in soccer players' progress as they repeatedly completed a sprint test. You can read their study in the European Journal of Applied Physiology for the year 2000. The test consisted of six sprints of about 34 meters. Halfway through each sprint, they had to turn left or right and then back around to the sprint path. A signal just before the start of each sprint indicated the direction of the turn. After they completed a sprint, they jogged back to the starting point to start the next sprint.
Would their times improve because they learned how to take the turns? Now, most studies don't provide the raw data, but this one is an exception. In the data, columns B through G represent the six sprints. Each row, two through eight, represents the data for one athlete. The dependent variable is the time, in seconds, to complete each sprint. So this is a test of the null hypothesis that the six sprint means are equal, against the alternative hypothesis that they're not. Excel doesn't have a data analysis tool called Anova: Repeated Measures, but Anova: Two-Factor Without Replication gets the job done.
It's two factor because Excel considers the row variable to be a factor, even though Athlete is not part of our null hypothesis. So I'll select that tool from the ToolPak and in the Input Range box, I'll enter A1 through G8. You have to include the column with the ID info for the row variable. Check the box next to Labels and New Worksheet Ply and call it Repeated Measures ANOVA, then click on OK, and there's a result.
I'll expand the columns, and here they are. And you know in the table, we can replace rows with Athlete and columns with Sprint. The high value for F for Sprint in cell E23 and the low p-value in F23 tell us that with degrees of freedom equal five and 30, we can reject the null hypothesis. The other F ratio doesn't really mean anything to us, as Athlete had nothing to do with our null hypothesis. Its high value just tells us that athletes are different from one another, which is hardly news, so we can delete the contents of E22 through G22.
So we learned how to test the overall differences in athletes' repeated sprint times, and repeated sprint time can be an important factor in a variety of sports. More generally, you can now apply analysis of variance to the very popular and powerful repeated measures design.
He explains how to organize and present data and how to draw conclusions using Excel's functions, charts, and 3D maps and the Solver and Analysis ToolPak add-ons. Learn to calculate mean, variance, standard deviation, and correlation; visualize sampling distributions; and test differences with analysis of variance (ANOVA). Then find out how to use linear, multiple, and nonlinear regression testing to analyze relationships between variables and make predictions. Joseph also shows how to perform advanced correlations, variable frequency testing, and simulations.
By the end of this course, you should have the foundational knowledge you need to take other statistics-related courses and perform basic analysis in the workplace.
- Identify the type of function that returns more than one value.
- Identify the type of function that returns values for the standard normal distribution.
- Name the different amounts of a factor in the analysis of varianc e.
- Explain how to use the Tool Pak Regression tool for exponential regression.
- Define negative correlation.
- Recall the term used to describe a simulation that includes randomness.