1. When making a scatter diagram, it is easiest if you use graph paper
2. When changing the raw scores to Z scores, it is easiest (and you will make fewer errors) if you do all the Z scores for one variable and then all the Z scores for the other variable.
• Also, to make sure you have done it correctly, when you finish all the Z scores for a variable, add them up–they should add up to 0 (within rounding error).
3. Before calculating a correlation coefficient, estimate its strength and direction by drawing a scatter diagram.
• If the correlation coefficient you calculate is very different to what you expected, check all of your working (and don’t forget to double check that you plotted the scatter diagram correctly).
4. When calculating the cross-product of Z scores, pay careful attention to the sign of each Z score. As you know, a negative score multiplied by a negative score produces a positive score. errors in calculating the correlation coefficient are frequently made at this step, so be careful!
5. If you calculate a correlation coefficient to be larger than +1.00 or less than -1.00, you have made a mistake (so go back and check all of your calculations)!
6. When calculating the correlation coefficient, be sure to divide the sum of the cross-product of Z scores by the number of people in the study (N) and not by the total number of scores there are in the study (which will be double the number of people, since each person contributes two scores: one on each of the two variables).
• This is a common mistake, so be sure to avoid it.

Copyright © 1995-2008, Pearson Education, Inc., publishing as Pearson Prentice Hall Legal and Privacy Terms