 |
Introduction to Research Methods in Political Science: |
GETTING STARTED WITH SPSS
This tool covers a variety of
SPSS features with which you will need to become familiar.
To open one of the SPSS system files (those with names ending in ".sav") that accompany POWERMUTT, open your browser, go to http://www.csupomona.edu/~jlkorey/POWERMUTT/downloads.html, click on the file you want, and save it. If SPSS is loaded on your computer, you may be able to open a system file in SPSS by double-clicking on it after you have downloaded and saved it.
Alternatively, you can open SPSS
first, then open a system file within it. If there is a shortcut to SPSS on
your desktop, double-click it. If not, click on “Start” and “All
Programs.” Find SPSS and open it. There are two ways that you can
then open a file within SPSS.
1) You may see a dialog box
asking “What would you like to do?” If you wish to use this box to open a
file, either double click on the file under “Open an existing data source,” or
"Open another type of file." Click on "OK." This will open
another dialog box that you can use to select the file you want.
2) If the initial dialog box does
not appear, or if you do not wish to use it (and click on “cancel”), go to the
menu bar and click on “File,” “Open,” and “Data.”
You should now see a dialog box
called “Open File.” All of the data files used in this project are SPSS
system files. You can also import files from a variety of other formats,
such as Microsoft Excel. Choose the appropriate type of file from “Files
of type:” Use “Look in:” to browse your directories for the file you wish
to open, and either double-click on the file or single-click and then click
“Open.”
Viewing and Editing a Data File
There are two “views” within
SPSS.
“Data View” provides a
spreadsheet in which each row represents a case and each column a
variable. Though you normally will not wish to do so, you can
change the value of a variable for a given case simply by typing in the new
value. (More systematic ways to modify the values of variables are
explained in the tools describing “Recode” and “Compute.”)
“Variable View” provides more
detailed information about each variable in your file, and also allows you to
make a number of changes in the ways these variables are described.
In some cases (“Name” and
“Label”) you can make changes simply by typing them in. “Name,” not
surprisingly, refers to the variable name. In general, variable names may
be up to 64 characters long, though in early versions of SPSS the limit was 8
characters. The first character should be a letter. Remaining
characters may consist of letters, numbers, or some special characters.
Do not include blanks within a name. “Label” is means of providing a more
detailed description of a variable, and is very useful in making output easier
to read. A label can include blanks and special characters.
For “Width,” “Decimals,” and
“Columns”, you can make changes by clicking near the right side of the field,
and then clicking the up or down arrow to select the number you want.
“Align” and “Measure” can be
changed by clicking on the right side of the field and then making your choice
from the drop-down menu that will appear.
Finally, when you click on the
right side of a field under “Type,” “Values,” or “Missing,” a dialog box will
open up for your use in making changes.
In most cases, the “type” of
variable will be numeric. You may also have some variables of another
type, most often “string” (i.e., text) variables consisting of letters as well
as numbers (for example, in a file containing data about different countries,
the names of the countries would constitute a string variable). Most
variables are numeric, since most statistical procedures require data that have
been coded numerically.
Just as variable names can be
labeled to make output easier to read, so too can specific values of a variable
be given labels. To add or edit value labels, click on the right side of
the field under “Values.” In the resulting dialog box, enter the value,
then the value label, then click on “Add.” To remove a value label,
select the label, then click on “Remove.”
Specifying missing data is
especially important, since SPSS procedures provide you with different ways to
treat such data. For each variable, you may specify up to three discrete
values as missing, or may specify a range of values and, if you wish, one
discrete value as missing. In addition, any blanks in your data will
automatically be considered as “system missing” or “sysmis.” The
“Missing” feature is especially useful even if the dataset you are using
already comes with missing values specified. Whether a particular value
should be treated as missing may depend on the specific research question you
are asking. For example, if a respondent in a survey identifies with the
Libertarian Party, that information is not missing, but you may choose to treat
it as such if you are only interested in comparing Democrats and Republicans.
Note: If you are using survey
data, you may need to weight cases by a
variable included in the dataset to correct for known over or under
representation of some categories of respondents.
Changing the Way
Information About Variables is Displayed
In the menu bar, click on “Edit,”
on “Options...,” and on the “General” tab. For now at least, you can
accept most of the default options, but you should decide how you want
variables shown when you are carrying out SPSS procedures. Under
“Variable Lists,” choose either “Display Names” or “Display Labels.”
Also, decide whether you prefer to have variables listed alphabetically or in
the order in which they appear in the file (as shown in the “Variable View.”)
It will also probably make your
output easier to understand if, while still in “Options,” you click on the
“Output” tab and select the following from the four drop-down menus that
appear:
1.
Names and Labels
2. Values and Labels
3. Names and Labels
4. Values and Labels
Obtaining Information
About a Variable
From the menu bar, click on
“Utilities.” Clicking on “Variables…” will allow you to scroll through
the variables in the dataset, examining them one at a time, while clicking on
“File Info” will provide similar information for all of the variables in the
dataset.
Analyzing Data and Preparing Charts
The most important parts of SPSS,
the parts that provide you with result, are found under the headings of
"Analyze" and "Graphs." Specific tools are provided
for each of the procedures used in POWERMUTT. See the "Tools"
section of the Table of Contents
for specifics. A couple of general notes:
SPSS often produces voluminous output. Printing it all is unaesthetic, uneconomical, and evil. Keep the good stuff to print out or to copy and paste into another application such as Microsoft Word.
Select the first portion of
output you wish to preserve by clicking on it. Hold down the control
(Ctrl) key and select additional portions.
To print the portions you have
selected, click on “File” (in the menu bar), and on “Print...” Under
“Print range,” make sure that “Selection” is selected. Click on
“OK.” Note: before printing your output, it is a good idea to first click
on “File” and on “Print Preview” You may wish to then click on
“File” and “Page Setup…” and select “Landscape” rather than “Portrait” for the
printing of your output.
You can also copy and paste
output. After selecting the portions of your output that you wish to
preserve, click on “Edit” (in the tool bar) and on “Copy.” Then open your
other application, click on “Edit” and on “Paste.”
If you make any changes to your file and wish to save the revised version, click on “File” and “Save” to replace the original file. To save the new version of the file under a different name (recommended if there is any chance that you might later want to return to the original version), click on “File,” and “Save as….” This will open up a dialog box called “: Save Data As….” If you wish to delete some variables, click on “Variables” and deselect the variables you do not wish to retain. The default option is to save the file as an SPSS system file (with a “.sav” extension). If you wish to save the file is some other format, click on the arrow to the right of the “Save as file type:” window, and choose the desired format. Finally, browse until you find the directory in which you wish to store your file, type in a file name in the “File name:” box, and click on “Save.”
There are a number of places you
can go to get help in using SPSS:
,
give it a “high four” with your mouse. A pop up window will display
the instructions used to produce that result.
Last updated
May 1, 2013 .
© 2003---2013 John L. Korey. Licensed under a 
Creative Commons Attribution-Noncommercial-Share Alike 3.0 Unported License.