ABSTRACT

All of the designs described in chapter 5 can be subjected to randomization tests within SPSS for Windows using the same general procedures. We suggest that you begin by copying all of the files in the SPSS subdirectory on the companion CD-ROM to a convenient directory. There are two kinds of files there: SPSS programs (or syntax documents; for the sake of consistency across the three packages, we refer to them elsewhere throughout the book as macros) with .sps extensions (design1.sps to design8.sps) and SPSS data files with .sav extensions (design1.sav to simple.sav). The next step, once this has been done, is to edit the data worksheet that corresponds to the design of interest so that it contains your own data, remaining in conformity with the worksheet specifications given in chapter 5. To edit one of the worksheets (e.g., design1.sav to analyze Design 1 data) use Open from the File menu to open it in SPSS and simply replace the numerical entries in all columns with your own values. Then, if you wish, use Save As from the File menu to save the current worksheet in the same directory, calling it owndes1 or something similar (the .sav extension will be added automatically). To run the program, with the worksheet open, open the program using Open from the File menu. You will need to scroll down the Types of File list to Syntax(*.sps) and then select the program you want by double-clicking on the file name in the list that appears. As the extensions are not displayed, the program names and data files may appear identical in the *.sav and *.sps lists, so you need to make sure that you have selected from the *.sps list. If you do not do that, you will probably get an error message telling you that the data file is already open. To run the pro-gram, select Run and then All from the Syntax Editor Window. After running, the results will appear in the Output Viewer window. These will comprise the value of the statistic used for the actual data, the count of arrangement statistics at least as large (in some cases, at least as small) as the statistic for the actual data, and the probability of obtaining a statistic as extreme as the one obtained. SPSS presents small probabilities in scientific notation.Thus, for example:

10**−2X 1.499250375

should be read as:

0.01499250375 by moving the decimal point two places to the left (indicated by the −2).