to give 18 points to Member A of Group Z and 25 points (from another box) to Member B of Group W. If you decide to give 18 points to Member A of Group Z, then it means that you have also chosen 19 points for Member B of Group W. So please consider your choices carefully when you make them.
Now, each matrix page in the booklet contains different matrices, with different combinations of numbers in the boxes. So, as you go from one page to another, choose your boxes very carefully. Please note that you are never awarding points to yourself. We arranged the booklets so that your own individual identification letter never appears on the matrices in your booklet. Of course, we do not want you to give points to yourselves.
Regardless of your final choices, make sure that before each decision you carefully examine the two numbers contained in each box of the matrix. Once you make your decision, tick the box you chose and also write the numbers representing your choice in the spaces provided below each scale . . . (show this on chart). You may proceed now.
Two sets of statistical analyses are usually conducted to examine group pull scores on the Tajfel matrices: (a) matrix strategy analyses within each treatment condition and (b) matrix strategy analyses between treatment condition ( Sachdev & Bourhis, 1991).
(A) The within treatment condition analysis of the pull scores is of more immediate concern because this analysis is the one required to determine if pull scores obtained from subjects within each treatment condition are significantly different from zero use on the + 12 to - 12 matrix pull scales. It is clear that one must first determine if subjects actually used any of the six strategy pulls (P on FAV, MD on MIP + MJP, FAV on MJP, etc.) before determining whether differential use of the strategies were made by groups of subjects across treatment conditions within the study. The usual within treatment analysis is to apply a Wilcoxon Matched Pairs Test on the difference in scores between the opposed (O) and together (T) rank scores of each matrix type ( Turner, 1983a). In the example depicted in Table 9.3, one would test for the significance of group pull score of FAV on MJP by conducting a Wilcoxon Matched Pairs Test on the difference in score between the opposed and together rank score of Matrix Type A (O - T). In turn, the significance of group pull scores of the obverse pull, MJP on FAV, is determined by conducting a Wilcoxon Matched Pairs Test on the difference in scores between the (12 - opposed) and together rank scores of Matrix Type A: (12 - O) - T. Similar Wilcoxon Matched Pairs Tests are conducted to test the significance of the pull scores obtained from the remaining