HFIT-565

Assessment & Evaluation of Health Fitness Parameters

Fall Semester 2009

Dr. Marc Schaeffer

mschaef@american.edu

Examples for perfect correlation and using Excel to generate correlation coefficients

Examples of Plots of Perfect Correlation

• the first plot below is a perfect correlation with r = +1.000
• the second plot below is a perfect correlation with r = -1.000
• how do these plots differ?
• although it is possible to obtain experimental data that will graph with a perfect correlation, in practice it very, very rarely occurs

Using Excel to generate correlation coefficients

• We are using the data that look very similar, but there are three different iterations of the problem.
• Note that the scatter points mover closer and closer to the red line in each iteration.
• We should start by asking what is our experimental hypothesis.

On the average, I believe that people with larger feet are taller than people with smaller feet. If foot size (X) increases in size, then height (Y) also increases.

• Based on our experimental hypothesis, what is our null hypothesis?
• We can test our null hypothesis by performing a statistical test.
• Our test statistic will be a Pearson correlation coefficient r.
• It will be necessary to know how to find the df for these data.
• If our test statistic exceeds the (p < 0.05) criterion value for df = 3 (r = 0.878), we reject the null hypothesis.

back to Lecture 5

send email to Dr. Schaeffer

this page last modified by M Schaeffer

on September 17, 2009