|Note: This manual is written with the assumption that you are using Windows 95 or 98, and that the standard type of computer for data analysis is an IBM or compatible. SPSS Inc. stopped updating SPSS for the Mac with Version 6.1, and many changes have taken place between Version 6.1 and Version 9.0. It appears that SPSS Inc.'s best suggestion is to combine a fast Mac (e.g., a G3 or G4) with Virtual PC software and the Windows version of SPSS.|
(If you are already familiar with Windows, skip to Starting SPSS for Windows)
Your Windows 95/98 opening screen should look something like Figure 1-1.
|For those of you just beginning to use Windows 95/98, the Windows program has its own, short tutorial. Unfortunately, if you are really a new user, a real person may need to help you get started using computer/Internet based instruction. The Windows program comes with a short tutorial. You need to start by going to the Start button, then go to "Help." Be sure the "Contents" tab is at the front by clicking on the tab, click on the top choice and work your way down until you feel comfortable. Alternatively, you can try the tutorial at http://www.windweaver.com/w95man.htm|
Several standard desktop icons, such as "My Computer" and "Recycle" will always appear on start up. Note the "Taskbar" along the bottom, with the "Start" button at the far left. If you own your own computer, you can do quite a bit of customizing of your desktop, choosing your favorite colors and scenes, rearranging and adding icons, moving the Taskbar to a different location, hiding it from view, and so on.
Although Windows makes use of the right-button key on the mouse, you will only use the left button for now, so when we say to "click" on the mouse button, it will always mean the left one.
|If mouse is being used by a left handed user these instructions may be reversed. We will assume a right handed user.|
A single click will often take you where you want to go, but if one click doesn't do anything, try a double click. (By the way, double clicking means to press the left mouse button twice in rapid succession. If nothing seems to happen, you probably need to double click more rapidly.) Now move your mouse so the pointer touches the "Start" button (don't click anything yet, just let the pointer rest somewhere on the button). Notice that a label appears in a little rectangle, just above the "Start" button, showing "Click here to begin" as shown in Figure 1-2.
These floating labels will usually appear in your active window whenever the mouse cursor moves over a button icon. They will then disappear after a few seconds, so pay attention! Now that you have found the Start button, click on it once and the Start Menu will appear, and your screen should look something like Figure 1-3.
This is the basic Start Menu, and it can also be customized by adding your most-often used programs to it. Without clicking anything yet, move the mouse pointer up and down the Start Menu. As you encounter each item, it will become highlighted. Nothing other than that will happen with the Help, Run, or Shut Down icons, but letting the pointer hover over Programs, Documents, Settings, or Find will cause other menus to slide out across your screen. The little triangles at the right side of the Start Menu are your clue here. If one is present, that means there are more menus for that item. If this is the first time you are using Windows, the Documents menu will say "empty." If you have used programs that saved files, up to 15 of the last ones you saved will show up on this menu. This is a quick way to find something you worked on recently, since one click will open that file and the applications program it belongs to (such as Word, SPSS, Excel, etc.).
For us now, the "Programs" icon is the important item on the list. Aim the mouse pointer to highlight that icon, and your screen should look somewhat like Figure 1-4, though with perhaps fewer or different items. Notice the little triangles next to most of the items: you guessed it, more sub-menus.
Move the mouse pointer around on those icons with a little triangle at the right (don't click anything yet) and watch what happens. When the pointer lands on a program icon, it becomes highlighted, and its sub-menu appears.
Starting SPSS for WindowsThe SPSS 9.0 for Windows icon should be on the Start Menu, but you might need to scroll down to see it. It should look something like Figure 1-5. Click on this, and wait while SPSS loads.
When SPSS has loaded you may get a menu that asks "What would you like to do?" (It looks like Figure 1-6.) For this Tutorial click "Cancel" to get rid of this.
Now look at your screen. The Taskbar is probably visible at the bottom of the screen. If it is not visible (making it "Autohide" is another option that can be set, but we will not go into that here), move your cursor so it is at the bottom of the screen and the Taskbar should show itself. It shows an "SPSS" button. Whatever programs you have open will show on the Taskbar, and the one you are currently using will be highlighted.
Next, observe the three small squares in the uppermost right-hand corner of the main SPSS window. The one furthest to the right, with an X in it, is used when you want to close any program you are using. Don't worry if you click on it accidentally, a dialog box will pop up asking if you want to save anything that changed since the last time you saved your work, before it actually lets you exit the program. You can save and quit, quit without saving, or change your mind about quitting by clicking "Cancel."
The middle of the three small squares allows you to have the window you are working in fill up the whole screen, or to shrink it down to a smaller size. If the middle square shows two cascading rectangles in it, the window is already as big as it can get - clicking on this square will reduce the window in size. Try this now. In this shrunken window, the middle button now shows only one rectangle. Click on it to get back to the full screen view.
The last square, to the left of the other two, has what looks like a minus sign on it. Click this and watch what happens. Whoa, where did everything go? Look at your Taskbar. The button for "SPSS" is still there, but it is no longer highlighted. Click on it and see what happens. Aha. You have just learned how to minimize a window.
Leaving SPSSWe're not ready to actually use SPSS yet, so let's close it. There are at least four ways to do this. Move your mouse until the arrow is pointing at the word "File" in the upper-left hand corner of the screen and press the left mouse button once. A menu will appear. Move the arrow so it is pointing at the word "Exit" and press the left mouse button. This should close SPSS.
There is a second technique that can do the same thing. (Computers usually have more than one way to do everything.) Go back into SPSS and move your mouse until the arrow is pointing at the word "File" in the upper-left corner of the screen and press the left mouse button once, but this time, don't release the mouse button. Hold the mouse button down and move the mouse down until the word "Exit" is highlighted. Now release the mouse button and SPSS should close. This is called "click and drag" and is another way to use your mouse.
And now for a third way. Point your mouse at the SPSS icon in the upper-left corner of the screen. The icon will be just to the left of the words "Untitled: SPSS for Windows Data Editor." Move your mouse to the icon and double click on it. This has the same effect as the first two procedures; it closes SPSS.
And finally a fourth way. You already know the last way to close SPSS. Point your mouse at the X in the upper-right corner of the screen and click on it. SPSS will close.
Now you know how to move your mouse around and how to start and close SPSS. We'll show you more about Windows, but not much more. If you want to learn more about Windows, there are a lot of books available. The nice thing about Windows is that you don't have to know much to use it.
Looking at DataA data disk came with this tutorial if it was purchased new. Also, if you are in a computer lab, someone may have copied the data files on to your hard disk. For now, we'll assume you are using the data disk that came with the tutorial.
|SPSS works faster if the data is eventually copied onto your hard disk. If you don't know how to do this, ask someone like a lab technician, your instructor, or look at the manual that came with your computer.|
We need SPSS loaded into the computer, so start SPSS just as you did above. Now, put the data disk that came with this book in your A: drive.
|If you are using a drive other than A:, substitute the drive, and path for "A:" above.|
Point your mouse at the "Start" button on the task bar in the lower left-hand corner of the screen and press the left mouse button. Let your mouse rest on "Programs" and another menu should open. Point your mouse at "SPSS" and still another menu opens. Point your mouse at "SPSS 9.0 for Windows," click, and SPSS should start.
Your screen should look like Figure 1-6. (If you see a box asking "What do you want to do?," click on cancel to close this box.) At the very top of the screen, you'll see the words "SPSS Data Editor." Just below that line will be the menu bar with the following options: File, Edit, View, Data, Transform, Analyze, Graphs, Utilities, Windows, and Help. Point your mouse at "File" and press the left mouse button.
A box will open which is the File menu. Point your mouse at "Open" and press the left mouse button. (Also, you could have gotten to this point by clicking on the Open File icon just below "File" on the Menu bar.)
This opens a larger box called the Open File box. Here you need to tell SPSS where to find the data file to open. In the upper part of the box you'll see "Look in:." Click the down arrow on the "Look in" line and then click A:. Click on the file name, GSS98A, to highlight it and then click on "Open." In a few seconds, your data matrix will appear.
|There is also a portable version of this file, and all the others used in this Tutorial, which can be found on the Web at http://www.csubak.edu/ssric/|
A data matrix is a very important concept. The rows contain the cases and the columns contain the variables. (If you're familiar with spreadsheets, that's what this is.) Row 1 is case 1, row 2 is case 2, and so on. The top of each column contains the variable name. In this data set the variable names are abbreviations like ABANY and ABDEFECT. Unfortunately the abbreviations for the variable names do not tell you very much. We need some way to find out what these variables are. So try this. On the menu bar at the top of your screen, you'll see the word "Utilities." Point your mouse at "Utilities" and click the left button. This will open the Utilities menu. Point your mouse at "Variables" and click again. Your screen should look like Figure 1-7.
You'll see a list of all the variables in your data on the left side of the little window. (Also, see Appendix A for a list of variables.) Point your mouse at any of these variables and click. To the right of the variable list you'll see a short description of this variable. For example, point your mouse at the variable "ABANY" and press the left mouse button. This question asked if respondents thought that obtaining a legal abortion should be possible for a woman if she wants it for any reason. The possible answers are YES (value 1), NO (value 2), DK or don't know (value 8), NA or no answer (value 9), and NAP or not applicable (value 0). (Not applicable includes people who were not asked the question.) As you will see in Chapter 3, these values are very important!
Now you know how to get a preexisting data file in SPSS and how to find out what the variables are in the file. We will tell you more about this later, but here we just want to give you a brief introduction to SPSS for Windows.
A Brief Tour Through SPSSNow that you have the file opened, let's look at some things you can do with SPSS. You're already familiar with the variable ABANY. Let's find out what percent of people surveyed thought it ought to be legal for a woman to have an abortion for any reason. (If you have the Variables window open showing the variable labels and values, point your mouse at the close button and click it.) On the menu bar you will see "Analyze." Point your mouse at "Analyze" and click it. A box opens that looks like Figure 1-8.
This lists the statistical procedures in SPSS. We want to use "Descriptive Statistics" so point your mouse at "Descriptive Statistics." This opens another box listing the statistical procedures you can use to summarize your data. Point your mouse at "Frequencies" and click it. This opens the Frequencies box. Since ABANY is the first variable in the data, it's already highlighted. Point your mouse at the right arrow next to the list of variables and click it. The label ABANY will move to the box called Variable(s). This is how you select variables. Point your mouse at "OK" and click it. In a few seconds, a new screen should appear that looks like Figure 1-9. We are now in a different part of SPSS for Windows, called the Output Window. This is where the results, or output, are displayed.
|Instead of seeing a list of variables, you may see a list of variable labels. You can change this so SPSS displays the list of variables. To do this, click on Edit in the menu bar, then click on Options and on the General tab. Look for Variable Lists in the General tab and click on Display names. Finally, click on OK. This change will not take effect until the next data set is opened. You can reopen the data set you are using by clicking on File in the menu bar and then on New and then on Data and then select the GSS98A file. However, if you do this, remember that you will lose any changes (e.g., recodes) you have made to the data file (unless you save the file first). We remind you of this again, along with some other options that you can control, later in Chapter 4.|
The Output Window is divided into two vertical frames or panes. The left-hand frame contains a summary of the output or information that SPSS gives you. This information is in outline form and can be used to select what you want to view. Simply click on the information you want to look at and that information will appear in the right-hand pane. You can also collapse and expand the outline by clicking on the plus and minus signs in the left-hand pane. The plus sign indicates that the information is collapsed (or hidden) and the minus sign indicates that it is expanded (or shown). You can use the scroll bars on each pane to scroll through the Output Window.
On the right side, the frequency distribution for ABANY is divided into four parts: (1) the title, (2) notes on the table (there aren't any for this table), (3) statistics (a summary of the number of missing and valid observations), and (4) the actual table showing the frequency distribution. Click on "Statistics" in the left-hand pane and you will see that there were 1778 valid and 1054 missing cases. Click on "abortion--for any reason" and you will see the frequency distribution. In Figure 1-9, you can see that 728 people said yes, 1050 said no, 98 said they didn't know, 6 didn't answer the question, and 950 were coded not applicable. (These 950 respondents were not asked this question. In survey research it's very common to ask some, but not all, of the respondents a particular question. In this case, only about 1800 of the approximately 2800 respondents were asked this question.) Of those who had an opinion, we want to know what percent of the respondents said yes or no, so we should look at the Valid Percents in the table. About 41% of the respondents who had an opinion thought it should be legal, while 59% thought it should be not be legal.
It would be interesting to know if men or women were more likely to favor allowing a legal abortion when the woman wants it for any reason. We're going to use a cross tabulation(usually called a crosstab in SPSS) to determine this. Point your mouse at "Analyze" and press the left mouse button. Then point your mouse at "Descriptive Statistics" and finally, point your mouse at "Crosstabs" and press the mouse button. Your screen should look like Figure 1-10.
The list of variables in your data set is on the left of the screen. We want to move the variable ABANY into the box next to the list of variables where it says "Rows." Click on the variable "ABANY" which will highlight it.
Now click on the arrow pointing to the right which is next to the Rows box. Notice that this moves ABANY into the Rows box. We also need to move the variable SEX into the Columns box. You will have to use the scroll bar in the box containing the list of variables to find this variable. (You can also click anywhere in this box and then type the letter "S" to move to the first variable starting with the letter S.) Point your mouse at the down arrow next to the list of variables and click. If you keep pressing the mouse button, the list of variables will move down and eventually you will see the variable SEX. Highlight it and click on the arrow pointing to the right which is next to the Columns box. This moves SEX into the Columns box. Now your screen should look like Figure 1-11.
Raw numbers by themselves are seldom useful. Most people understand percentages better. To get SPSS to compute percentages, point your mouse at the button labeled "Cells" at the bottom of the screen and click on it. This will open the Crosstabs: Cell Display box. Find the box called "Column" percentages and click on this box. This will place an X in this box and your screen should look like Figure 1-12.
Now click on "Continue" and you will be back to the Crosstabs box. To tell SPSS to run the Crosstabs procedure, click on "OK." After a few seconds your screen should look like Figure 1-13. Use the scroll bar to look at all the information that SPSS gives you in the Output Window.
Figure 1-13 shows the results, or "output." It shows, for example, that males and females differ very little in their opinions about a woman obtaining a legal abortion for any reason. Forty-two percent of the males and 40 percent of the females approve of a woman obtaining an abortion for any reason. The difference between these two percentages is so small that it could easily be a chance or random difference.
You can also examine other items in the survey to compare men and women. Who has more education? Is the average age at birth of first child younger for women than for men? Comparing means will answer these questions. Click on "Analyze," point your mouse to "Compare Means," and then click on "Means." Your screen should look like Figure 1-14.
Now put age at birth of first child (AGEKDBRN) and years of school completed (EDUC) in the Dependent List box and SEX in the Independent List box. By now you have a good idea how to do this. Highlight "AGEKDBRN" in the list of variables on the left of the screen by pointing your mouse at it and clicking. Then click on the arrow next to the Dependent List box. Do the same for "EDUC." Now highlight "SEX" and click on the arrow next to the Independent List box. This should move AGEKDBRN and EDUC into the Dependent List box and SEX into the Independent List box and your screen should look like Figure 1-15. Then click on "OK" and the output should look like Figure 1-16.
Women had their first child at an average age of 22.52 years, while the average for men is 25.21 years, a difference of more than two years. Now look at the mean years of school completed for men and women. There isn't much difference (about 0.19 of a year) between men and women.
Another way of examining relationships is to look at Pearson Correlation Coefficients. One could hypothesize that the respondents' education is correlated with the educational achievements of their parents. The Pearson Correlation Coefficient will tell us the strength of the linear relationship between father's education, mother's education, and the respondent's education. The closer the correlation is to 1, the stronger the relationship, and the closer it is to 0, the weaker the relationship.
Point your mouse at "Analyze" and press the mouse button. Now point your mouse at "Correlate" and then click on "Bivariate." Your screen should look like Figure 1-17.
Now move the following three variables into the Variables box: EDUC, MAEDUC, and PAEDUC. These variables refer to the number of years of school completed by the respondent and the respondent's parents. Highlight each of these variables and press the arrow next to the Variables box. The screen should look like Figure 1-18.
Click on "OK" and the correlations will appear in your output box. The output should look like Figure 1-19.
The strongest correlation is between father's and mother's education. As we predicted, there is also a fairly strong correlation between respondent's education and parent's education.
We can also look at a scatterplot showing the relationship between father's education and the respondent's education. Click on "Graphs" in the menu bar and then click on "Scatter." This will open the Scatterplot box. Click on "Simple" and then on "Define." This will open the Simple Scatterplot box. Scroll down the list of variables on the left until you see "EDUC" and click on it to highlight it. Then click on the arrow to the left of the Y Axis box to move EDUC into this box. Scroll down this same list until you find "PAEDUC" and click on it. Then click on the arrow to the left of the X Axis box to move PAEDUC into it. Your screen should look like Figure 1-20.
Now click on "OK" and an output box should open. Your screen should look like Figure 1-21. Each dot in the scatterplot represents a case in your data set. In general, the higher the education of the father, the higher the education of the child. However, it is far from a perfect relationship. Many fathers with high education have children with less education and many fathers with low education have children with more education.
|Depending on the number of cases in the scatterplot, a dot may represent more than one case.|
Overview of ChaptersChapter 2 will acquaint you with how to enter new data into SPSS for Windows using the Data Editor. Chapter 3 explains how to take your data, or that collected by someone else, and modify it in a way that makes it easier to understand. Chapter 4 starts the sections where you really get to see the results of your work. In Chapter 4 you will learn how to look at each variable, one at a time. We call this univariate analysis. Chapters 5 through 7 will teach you how to look at two variables at a time, or what we call bivariate analysis. Chapter 5 shows you how to cross tabulate two variables. Chapter 6 shows you alternative ways of comparing more than one variable at a time, and Chapter 7 will teach you how to do this using linear regression procedures. Finally, Chapter 8 shows you how to explore relationships among sets of variables using multivariate cross tabulations and multiple regression.