Correlation & RegressionJeff Sinn, Winthrop University, SPSS Guide – Correlation & Regression
SPSS Guide: Correlation & Regression
Once the data are entered, go to
Analyze, Correlation, Bivariate
to get this dialogue box.Move the variables (quantitativeonly) that you wish to correlateinto the variables box and hit
1.775**.368-.637*.003.239.02612121212.775**1.585*-.360.003.046.25112121212.368.585*1.055.239.046.86612121212-.637*-.360.0551.026.251.86612121212Pearson CorrelationSig. (2-tailed)NPearson CorrelationSig. (2-tailed)NPearson CorrelationSig. (2-tailed)NPearson CorrelationSig. (2-tailed)NtestscrstudyingSATabsencestestscrstudyingSATabsencesCorrelation is significant at the 0.01 level (2-tailed).**.Correlation is significant at the 0.05 level (2-tailed).*.
This is a correlation matrix. It gives resultsfor
In this case, the p-value is below themagic .05 so we
the Ho.[We think absences really docorrelate negatively with test score].
In this case, the p-value is NOTbelow the magic .05 so we
the Ho. [We are NOTconfident that there is a correlationbetween SAT and test score].
Every r value (a sample statistic) strives to represent
(The actual correlation value in the population).When r gets bigger, we get more confident that therereally is a correlation. We know one of two thingsmust be true.H
= 0 [There is NO actual correlation]
0 [This is a correlation]
KEY POINT: If p (the middle number) drops below .05, we REJECT the Ho. This makes us happy. Wewant to reject the null hypothesis because it means we have evidence that we found a true relationship.We explain a finding as follows:
The [research] hypothesis was supported. Absences correlate significantlywith Test Score, r (10) = -.637, p
Note: More on this later.
Degrees of freedom (df) = n-2.