Interpreting and Using Statistics in Psychological Research. Andrew N. Christopher

statistics and statistical tools throughout this class. As I said in the previous chapter, above all else, I want you to understand the information that statistics provides you. Toward that end, as we move along, there will be some opportunities to do some calculations, as doing calculations will help you understand what information various statistics are giving you. In addition, it is important to note that most researchers do not calculate statistics themselves; rather, they have software that helps them do this task. One particularly popular software program for statistical analyses is the Statistical Package for the Social Sciences (SPSS). It can do a lot of great statistical work for us. That’s a good thing; however, we must understand what exactly it is that SPSS is doing for us. The software can “spit out” a lot of statistical information that we need to be able to interpret correctly, or else we will misuse the statistical tools we are learning about in this class. Each chapter, with the exception of the last one, will contain at least one opportunity to use this software.

      Let us start to familiarize ourselves with SPSS right now. It would be helpful if, as you read this section, you have access to a computer with SPSS loaded on it. That way, you can work along in SPSS as you read this information. If you cannot access SPSS right now, that’s fine. There will be screenshots, as there will be in all subsequent SPSS sections of this book, to help you see what’s going on as you use this software.

      First, let’s open SPSS. To do so, click on the name of the program from the Start menu as we do for any Microsoft Windows–based program (Microsoft Corporation, Redmond, WA). Here is the dialogue box you will see:

      Simply click “Cancel” because we want to create a new spreadsheet.

      SPSS has two main screens that we will navigate: Data View and Variable View. In the lower left-hand corner of the screen, you will see two tabs, one for Data View (which is the one SPSS normally opens into) and one for Variable View. The Data View is where we enter data into SPSS; the Variable View is where we manage our data and is the place we should start when using SPSS. So at this point, select Variable View and let’s set up our first SPSS spreadsheet.

      Variable View

      What we will do is set up a spreadsheet in Variable View, and then we will get some practice entering data into Data View. We will use some of the variables measured in Terrell et al.’s (2008) research, specifically those in Table 2.4. Let’s make sense of the Variable View of SPSS. There are 11 columns in Variable View; some are vitally important, whereas others are much less important. We will discuss them in the order in which they appear in SPSS.

      1 Name. We must give each variable in our study a name, one that allows us to know what the variable is. You may know what each variable is now, but if you exit your SPSS spreadsheet and come back to it later, there is always the chance you will forget such details. So, name each variable as precisely as possible.In our example, we can name our first variable Sex. Easy enough.We can name our second variable Class_Standing. Notice that rather than a space between Class and Standing, there is an underscore. We cannot have spaces between words when naming variables.Let’s call our third variable Ethnicity.Go ahead and provide names for the variables of GPA, Height, and Weight.For the tendency to act aggressively, notice we have four items (Aggressive1, Aggressive2, Aggressive3, and Aggressive4). Each one at this point is a separate variable, so each needs its own name.Once you have provided a name to each of our 10 variables, here is what the Variable View will look like:Now let’s move on and consider more of the Variable View.

      2 Type. For the first variable of Sex, move your cursor to its Type box. This is where we specify the type of variable it is. Click on the right side of the Type box for Sex. Here is what you will see:Almost always we will use numeric data, which is the default type of variable that SPSS provides. There will be some exceptions, and when they arise, we will deal them at that time. As you can see on the right side of the Variable Type dialogue box, you can format the width and number of decimal places for that variable as well. The width of eight characters and two decimal places are the SPSS defaults.

      3 Width. Of course, you can set the width here, as well as in the previous dialogue box. To do so here, just click on the Width box for the variable you want to adjust the width for, and then click the “up” or “down” arrow to set your desired width.

      4 Decimals. Similar to setting the width, you can click on the Decimals box for the variable you want to adjust the number of decimal places for; then you can click the arrows accordingly. In this example, only the variable of GPA needs to have decimals as GPAs can take on at least two decimal places. However, even if a variable cannot take on any decimal value, it is fine to leave the default of two decimal places intact.

      5 Label. Here, you can give a variable a more precise name that will appear on the statistical output we generate. It can be the same as the name designation we gave that variable previously, but it does not have to be. In fact, it is easier to be more descriptive with labels because you can leave spaces between words here. For example, for the variable named Class_Standing, we can give it a label such as Class Standing of the Participant. When we ask SPSS to provide us with statistical information, it will use the labels we provide in making its output for us to interpret.

      6 Values. SPSS needs numbers to do its work for us. For nominal variables, we need to provide SPSS with numbers. For example, SPSS won’t understand input such as male, female, or transgendered. However, it does understand 1, 2, and 3. So, in the Values box, this is where we assign numbers to certain data. For the variable of Sex, click on the right side of its Values box. Here is what you will see:For Value, enter a 1. For Label, enter male. Then click on Add. When we begin entering our data in SPSS, for a male participant, we will enter a 1 for the Sex variable. Now for Value, enter a 2. For Label, enter female. Then click on Add. Finally, for Value, enter a 3. For Label, enter transgendered and then click on OK.We have two other variables for which we need to provide values to SPSS, specifically, Class_Standing and Ethnicity. For Class_Standing, let’s give a value of 1 for first-year; a value of 2 for sophomore; a value of 3 for junior; and a value of 4 for senior. For Ethnicity, let’s give a value of 1 to African American; a value of 2 to Caucasian, and so on, with a value of 6 to other.As these are all nominal (categorical) data, remember that the number value we give each category means nothing in terms of relative standing of the categories. For instance, for Sex, a 2 for female does not mean women are twice as much of anything as men, who are coded as a 1.We do not need to enter a value for scale variables (in this case, for GPA, Height, Weight, or the four aggression items) because such measurements are numeric to start with. We can simply enter the raw data that each respondent provided us. However, you can always do so; it will never hurt to include a value for each and every variable in your SPSS spreadsheets.

      7 Missing. Sometimes a researcher cannot collect data on every variable from every participant in a study. In such cases, this column allows us to specify how to tell SPSS when there is missing data for a variable. For our purposes in this class, we will use examples in which all participants supplied data for each variable we are dealing with. However, should you collect data in your class, here is how you can handle any missing data you might have. Click on the right side of the Missing column for one of the variables. Select Discrete Missing Values, and enter a number that makes no sense given how the variable was measured. For instance, a grade-point average can be from 0 to 4.00. So if a GPA is missing for a participant, we can just enter it as 99999. Clearly, that value is nonsense given how GPA is measured, and now SPSS knows to treat it as missing data.

      8 Columns. If you’ve changed the width of a variable in the third column, you will need to change this number as well to match the width.

      9 Align. Here we tell SPSS how we want our data to appear, either right-justified, left-justified, or centered. For our statistical purposes, it makes no difference what you select here. I typically use the right-justified default.

      10 Measure. We