2
Navigating SPSS

Chapter Menu

• Chapter Learning Objectives
• How the Program Works
• Managing Your SPSS Life
• Summary

Chapter Learning Objectives

1. Opening and using SPSS
2. Understand the three important file types
3. Managing your SPSS life
4. Importance of raw data files

How the Program Works

When you access the program (selecting the SPSS icon from your desktop or start menu) for the first time, a couple of different things can happen. First, a "Welcome to IBM SPSS Statistics" box might appear that basically asks "what would you like to do?" There is an option at the very bottom of this box that says "Don't show this dialog in the future" - go ahead and select that box (you don't actually want to see this box every time you open SPSS). The second thing that can happen (and in fact will happen when you close the "Welcome" window) is that an empty data file will open - this is the data file, one of three very important file types in SPSS.

Important File Types

There are three very important types of files used by SPSS. There is the aforementioned data file type (identified by a .sav file extension), a syntax file type (identified by a .sps file extension), and an output file type (identified by a .spv file extension).

Data Files

Data (.sav) files are where you store your numbers/data. Typically, you have one original data file for each major project that you are working on. Figure 2.1 shows what an empty .sav file looks like.

Notice that the empty data file looks similar to a spreadsheet, with a tool bar at the top filled with icons (including the typical "file", "edit", and "view" commands but also the nontypical "transform", "analyze", and "graphs" commands - the icons you see in Figure 2.1 may look a little different from SPSS version to a different SPSS version or from one computer to another). The bottom of the file has two spreadsheet tabs labeled "data view" (the tab currently shown in the screen shot above) and "variable view". We will explain the variable view tab later in this book, but for now you should recognize that the Data View tab visually reveals the data.

Figure 2.1 Empty .sav file.

Similar to Excel, data in this tab will be organized in rows and columns. Note that the columns (the top cells running from left to right) currently have the word "var" written in them. When you have data to enter in the .sav file, these top cells can be used to identify what the column data represent. As you record multiple observations (also called cases) of this variable, you will populate rows (identified by the numbers along the left side of the spread sheet above). When there is more than one case (rows) for each variable (columns), you will have a data matrix.

The matrix in Figure 2.2 contains four variables (Var1 to Var4) defining each column. For each variable, there are five rows, representing the five observations (aka cases) for each variable.

The .sav file technically contains everything a novice SPSS user needs to run analyses. 99% of the time, any command or manipulation you want to do can be completed using the dropdown options located in the commands at the top ribbon.

Although the .sav file can manage most of your statistical needs on its own, there is another important file that is crucial to maintaining and organizing your statistical work: the syntax file

Figure 2.2 Matrix with four variables.

Syntax Files

Syntax (.sps) files are where you write your statistical diary (where your data came from, what analyses you'd like to do, the date or any other notes you would like to leave for yourself later on). An empty (no diary entries/statistical commands) syntax file is presented in Figure 2.3.

Just as you put your numbers in your data file, you identify what you would like SPSS to do with the data via the Syntax file (.sps). Notice that: (1) the commands along the ribbon of the page are similar to the options in the .sav file and (2) there is a lot of blank white space instead of grids (you can actually write in this file like you would in a word processor). These two file types (data [.sav] and syntax [.sps]) are the two files that you, the SPSS user, works with. You have to enter information into both the .sav file and the .sps file for each project that you do.

Output Files

Output (.spv) files are different from data and syntax files. You do not create output file - output files are created by the SPSS program. If you have a syntax file that is full (writing in it instead of a big blank space like Figure 2.3) and you have a data file that has numbers in it (e.g. Figure 2.2, instead of the blank white cells as in Figure 2.1), the result of the interaction of your syntax commands and your data will produce an output (.spv) file. As with the data and syntax examples, a blank output file is presented in Figure 2.4.

You will almost never see empty output files like Figure 2.4. The reason for this is because output files are "born", they're not typically conjured up as this one has been. When you perform an analysis or a data manipulation - boom - an output file appears. An output file will be created the first time you "do something" to your data. As more "things" are done to the data, the result will be added to this output file. Output files are therefore like SPSS messengers - anytime something noteworthy is done to your data, the output file will pop up and report the event.

Figure 2.3 Syntax (.sps) file.

Figure 2.4 Empty output file.

The Output file has two "panels" (left and right). The right-hand panel presents all of the detailed output you request from your selected analyses. The left panel is merely an outline of the detailed information provided in the right-hand panel (this outline is sometimes useful to help you navigate through the detailed output on the right if you have a lot of analyses that have been performed).

The "Others"

There are other SPSS file types you will eventually encounter, but the big three that you should know are the data, syntax, and output files. You can pursue learning about SPSS's other capabilities after reading this book (if you decide that you like SPSS enough [or are going to use it enough] to learn more).

Managing Your SPSS Life

We've just presented the notion that SPSS has three important file types that you will be working with. If you're nervous about using new software (like many of us are), this fact can make learning this system feel even more daunting. Luckily, there are ways to manage how and where you save these files that will hopefully simplify the process (and declutter your desktop).

Before you start using SPSS, you will need to create a system for storing your syntax and data. You should start by choosing a location within your computer's file space that you can easily find. This file space may be somewhere in your personal file folder or on a designated external flash drive. It is also best to get in the habit of creating subfolders for each project or assignment. This will ensure that you can easily find the syntax file you need, open it up, and pick up right where you left off (even if it's been weeks or months since you last worked on the project). Figure 2.5 shows a flowchart of a system that works well if you are using this book to supplement a statistics course.

In the example above, the files are organized by class assignment folders. You can also organize them by topic or research area. Notice that within each HW folder, there is a separate space for the data and syntax files. If you've been paying attention, you might ask why there is not a place for the output file?

We have already mentioned that the syntax file is for writing and recording the manipulations and analysis that you will perform on your data file. The output file will pop up whenever you do something of note to the data. Because the output file is created by the interaction of the data file and the syntax file, there is not typically a need to save this file. Whenever you want the information again, you can simply rerun the syntax script on the original data file.

The Importance of Maintaining the Raw Data as an "Untouched" File

Maintaining the unadulterated status of your original (also called "raw") data file is key to using SPSS correctly, since this is what will enable you to redo and undo any errors that you make. In fact, early in your SPSS exploration, it is highly recommended that the raw data file be the only data file that you maintain for a given project.

Even the savviest of SPSS users are still human and make mistakes and errors that can compromise the data they are working with. Simple things such as filtering data on the wrong variable, accidently overwriting a...