Pearson r Correlation in SPSS - Part 2



5
158433

How to calculate the correlation coefficient in SPSS is covered in this video. The correlation is also tested for significance and a scatterplot is constructed. For more SPSS help: http://tinyurl.com/nywhxw3 Correlation Pearson's r Correlation Correlation video in SPSS Video Transcript: To run the correlation we go to Analyze, and then Correlate, and then Bivariate. And bivariate if you think about it, 'bi' just means two and variate it is just another name for a variable so this is the correlation we use when we have two variables, like we do here. So select Bivariate and then here notice hours media and college GPA are on the left-hand side here of our Bivariate correlations box. You should see on your screen hours media is already selected, so go ahead and click the right arrow button to move it to the Variables box. And then select college GPA and click the right arrow button once again to move it over as well. Now notice here under Correlation Coefficients that Pearson is chosen by default. That's Pearson's r, that's what we want to run here in this analysis. If you want to run something like say Spearman's you could check that box, but what we want to run is the Pearson correlation, which is the most commonly used correlation coefficient by far. So that looks good Pearson is selected. Everything else looks great let's go ahead and click OK. And when we click OK, the correlation analysis is presented in our output here. And the way you read this here is you want to see where hours media and college GPA meet or intersect. So these two variables if you see they meet right here in this box here and it's also shown here once again as it's redundant. So either box is fine to look at. Notice that our correlation where it's shown here is Pearson correlation is negative, so it is negative, .66 if I round to two decimal places and then the p-value is .001. So using an alpha of .05 and a two-tailed test where a two-tailed test allows for a positive correlation or a negative correlation, we can use the following decision rule as we have for all of our tests: if p is less than or equal to .05, the test is significant which here would indicate that there is a significant relationship between hours of media watched and college GPA. Whereas if p is greater than .05, the test is not significant, indicating that there is not a significant relationship between hours of media watched and college GPA. Our p-value in our output is once again .001. So applying the p value to our decision rule, we can see that .001 is in fact less than .05, which indicates the test is significant. In other words in our example there is a significant relationship between hours of media watched and college GPA. We could write our results as follows: There is a significant negative relationship between hours of media watched and college GPA. And then we have r, which is for Pearson’s r, we have 18 degrees of freedom. Now the degrees of freedom interestingly are not reported in this table here instead we just get an N. But the degrees of freedom for Pearson's r are equal to the total number of people in your study, or your sample size, minus two. So N - 2. But we have 20 people, N - 2 or 20-2 is 18, so that's where I got this here. Next we have our correlation coefficient of negative .66 which you can see in the output here when we round to two decimal places. And then we have a p-value of .001, which is shown in our table here. Notice as well in the output here that SPSS uses asterisks to indicate if the test is significant and they might be a little hard to see but there are two asterisks here. And notice the note down here where the two asterisks are showing and it says correlation is significant at the .01 level, two tailed. Now if you saw one asterisk showing instead of two, this would read correlation is significant at the .05 level, two tailed. And it's just the way SPSS outputs it by default to make it a little more user-friendly. We could turn that off if we wanted to in the Bivariate correlations box but we just left it as is. YouTube Channel: https://www.youtube.com/user/statisticsinstructor Channel Description: For step by step help with statistics, with a focus on SPSS (with Excel statistics videos as well). Subscribe today! YouTube Channel: https://www.youtube.com/user/statisticsinstructor

Published by: Quantitative Specialists Published at: 10 years ago Category: آموزشی