**Introduction to SPSS**

**What is SPSS?**

Statistical Package for the Social Sciences, a fairly user friendly statistics
package that is appropriate for both novice and
sophisticated users. SPSS allows users to operate in a menu environment
or in a command (syntax) driven environment. It is available for the Mac
and Windows.

**SPSS Basics: Windows**

Available windows may vary from one version to another

Each has a toolbar that you can use instead of or along with the menus

*Data editor *
Where you open files; edit, select, or transform data, conduct statistical
analyses and produce graphs

*Output * Where you look at statistical results,
tables, charts

*Chart editor* Where you edit graphs

*Syntax editor* Where you can write and execute commands
in the SPSS command language

SPSS Basics: Menus

**Help menu**

If you have the SPSS tutorial installed, it is a very easy way to get started
with SPSS

*
Help topics*:

SPSS at a glance

Getting help

Tutorials

Data management

Statistical analysis

Graphical analysis

Interactive charts

Output management

Saving files

Printing files

Customizing SPSS

FAQ's

**Window menu**

Allows you to switch back and forth from the data
editor, ouput, syntax, and chart windows

I do most of my work in the data editor window

**File menu**

*New* for defining a new data file, and entering your own data (or new
syntax, output window)

*Open* existing data files in various formats; or SPSS output, syntax,
etc. files

*Read text data* open text data files and define variables

This is where you must know how your data file is
organized and formatted

Information is usually included in a codebook

**Using existing data files**

What type of file is it?

Mac or Windows?

Text, spreadsheet, SPSS, SPSS portable?

Is it only data, or are labels included?

What variables are where?

How are records formatted?

How many records per case?

Columns? Delimiters?

Decimals?

String vs. numeric variables?

How is missing data coded?

Value labels?

**Define variables**

*Variable View* in data editor

Assign variable names in an organized manner; keep them short

*Type* allows you to define string and numeric characteristics, others

*Width* total columns allowed

*Decimals*

*Label* allows you to give longer references to variables

Variable names are 8 characters or less and start with
a letter

Variable labels can be longer and can include spaces

*Values* assign strings to numbers

*Missing* define discrete or ranges of missing values for variables

*Columns* columns displayed

*Align*

*Measure*

**Edit menu**

This works very much like common spreadsheet editors

Cut, copy, paste, find

Options

*General*

I prefer to open a syntax window when the program
launches

Display order for variable lists

*Viewer/Draft viewer*

Fonts, page width and length

*Output labels*

Variable names and/or labels

*Chart*

Aspect ratio, frame, font

**File menu (cont.)**

*Save and Save as:* data can be saved in various formats

SPSS portable files are transportable across Mac, Unix, Windows platforms

SPSS files save variable definitions and formats with the data

Tab-delimited files save data, with or without variable names

value labels are not saved

Fixed ASCII writes variables to fixed columns

*Print*

*Quit*

**View menu**

Grid lines

Value labels

Fonts

Switch to variable view

**Utilities menu**

*Variables*

Shows the format and characteristics for variables

*File info*

Gives information about all the variables in a file

**Data menu**

*Insert variable* in spreadsheet

*Insert case* in spreadsheet

*Sort cases*

Ascending or descending

Multiple variables

*Split file*

Allows you to repeat analyses by groups

It is on until you turn it off

A sort must take place (default)

*Select cases*

Allows you to filter out a subset of cases

It is on until you turn it off

Filter rather than delete cases

Selecting conditionally: keypad, formulas, and functions

**Transform menu**

*Compute*

Allows you to construct a new variable

This can be done conditionally

*Recode*

Allows you assign new values to represent old values

Useful for collapsing interval variables to categorical variables

Be careful about recoding into the same variable

String variables can be recoded to numeric, and vice versa

Recodes can be done conditionally

*Count*

Gets the total number of values from a set across variables in a case

*Categorize variables*

Collapses a variable with many values to a specific number of categories

Done by percentiles

*Automatic recode*

Recodes string or numeric variables to consecutive integers

*Rank cases*

Different types of ranks

Different ways of dealing with ties

Can be done by subgroups

**Data Analysis: The Analyze Menu**

**Reports**

*Case summary*

Allows you to print values for selected cases and
variables

Can be done for subgroups

Can get you some descriptive stats

**Descriptive Statistics**

*Frequencies*

Gets a table of counts and percentages

Usually used for categorical variables and ordinal variables with not too many
categories

Additional statistics like percentiles can be obtained

Most summary statistics like means are better obtained from other procedures

The format subcommand can be used to order the table by counts or values

Graphs are better obtained from graphing procedures

The format subcommand can be used to order the table by counts or values

*Crosstabs*

Gets contingency tables (two-way tables)

Used for summarizing two categorical or ordinal variables at once

Cell information can include observed and expected frequencies, row and column
%

Layers can give you tables by subgroups (i.e., three-way tables)

*Descriptives*

Computes summary statistics like mean, standard deviation, min, maximum

Options lets you select the statistics you want and order the display of
results

This procedure can also save standardized values (z-scores) to the data set

*Explore*

Computes summary statistics and produces simple graphs and plots

As the name says, it is useful for exploring data

Stem and leaf plots and boxplots can be used to identify outliers

Analyses can be conducted by groups to explore differences

*Correlate*

Bivariate

Computes correlations between two ordinal or interval variables

Will produce correlation matrices

One can specify listwise or pairwise treatment of missing values

**Graphing**

**Bar charts**: summarize variables by groups and subgroups

*Simple* displays summaries by groups

Display frequencies or percents for groups [Ex: N for HSP]

Display summary statistics for a variable by groups [Ex: Mean(RDG) by HSP]

Statistical displays for a set of variables [Ex: Medians (RDG WRTG MATH)]

*Cluster* displays summaries by groups and sub-groups (clusters) [Ex: %'s
for HSP SEX]

An elaboration of simple bar charts

Bars for "missing" can be discarded in the options menu

Can use pairwise or listwise deletion

*Histograms*: summarize an interval variable [Ex: SCI with normal curve]

A normal curve can be superimposed

**Line charts:** summarize by groups

*Simple* is like a simple bar chart, midpoints of bar tops are connected
[Ex: Mean(RDG--CIV)]

*Multiple* is analogous to cluster bar charts

Can get separate lines for groups or variables

Series--transpose data switches groups with variables [Ex: Mean(RDG--CIV) by
SES]

**Scatterplots:** show relationships between variables

*Simple* shows the relationship between two variables [Ex: RDG w/ WRTG by
SEX]

Axes are usually interval variables

Markers can be used for separate groups

Options can fit regression lines to the total group or subgroups

Sunflowers are for marking # of hidden cases (can't be done for subgroups)

**Some other charts**

Pie, area, high-low, boxplots, control, time series

**Editing and Saving Output**

**Output Viewer**

Click on tree in left sends you to a table on right

Double-click on table to get in edit mode

(or go to Edit menu and open/edit SPSS Pivot Table Object)

Double-click on cell you wish to change

Somewhat limited editing options for tables

Objects can be cut and pasted to other applications

**File Menu**

Tables can be saved as SPSS output files (.spo)

They can be exported as HTML or text files

**Editing charts**

Done in the chart editor (or in chart windows for other versions of SPSS)

Double-click on chart to open it in chart editor

Most changes are made by drilling down the chart, highlighting, and selecting
from menus

Depending on the type of chart being edited, different characteristics can be
changed, including:

Axis scaling and ticks

Number of intervals

Interpolation, fitting lines, reference
line

Labels, text styles, and fonts

Titles, footnotes, annotation, and legends

Frames

Fill patterns, spacing, and colors for
bars

Bar style

Marker colors, styles, and sizes

Line styles and colors

Swapping axes

Transposing data

**Saving charts**

Save as SPSS output (.spo)

Export as a jpeg, pict, png, tiff, or bmp (not SPSS files)

Cut and paste to another application