SPSS Class Notes Analyzing Data

---

1.0  SPSS commands used in this unit

crosstabs	Crosstabulations
t-test	t-tests
glm	General linear models
regression	OLS regressions
pplot	Normal probability plot
logistic	Logistic regressions
npar	Non-parametric tests

2.0 Demonstration and explanation

For this section we will be using the hs1.sav data set that we worked with in previous sections.

• File
   Open
    Data
     select C:spss_datahs1.sav

2.1  Chi-square

The chi-square test is used to determine if there is a relationship between two categorical variables.

• Analyze
   Descriptive Statistics
    Crosstabs...
     select prgtype for the rows and ses for the columns
      click on "Statistics"
       check the chi-square box

2.2  t-tests

This is the one-sample t-test, testing whether the sample of writing scores was drawn from a population with a mean of 50.

• Analyze
   Compare Means
    One Sample t-test
     select write and compare it to 50

This is the two-sample independent t-test with separate (unequal) variances.

• Analyze
   Compare Means
    Independent Samples t-test
     select write as the dependent variable and female as the independent 
     variable

This is the paired t-test, testing whether or not the mean of write equals the mean of science.

• Analyze
   Compare Means
    Paired Samples t-test
     select write and science

2.2  ANOVA

In this example the glm command is used to perform a one-way analysis of variance (ANOVA).

• Analyze
   General Linear Models
    Univariate
     select write as the dependent variable and prog as the fixed factor

In this example the glm command is used to perform a two-way analysis of variance (ANOVA).  The plot option creates plots of the means, which can be a great visual aid to understanding the data.

• Analyze
   General Linear Models
    Univariate
     select write as the dependent variable and prog and ses as fixed factors
      Plots
       select prog to be the X axis and ses to be the separate lines
        Add
         Continue

The Tukey test is used to test all the pair-wise comparisons of the levels of prog.

• Repeat the above analysis (dialogue recall)
   Post Hoc
    select prog and choose Tukey test

Here the glm command performs an analysis of covariance (ANCOVA). Note that the results are exactly the same as in the regression where write and science are regressed on math.

• Analyze
   General Linear Models
    Univariate
     select math as the dependent variable and science and write as covariates
      model 
       select custom
        choose main effect in the build terms field and select every variable in 
        the Factors & Covariates field and move them to the Model field.

2.3  Regression

This is plain old OLS regression.

• Analyze
   Regression
    Linear
     select math as the dependent variable and write and science as independent 
     variables

It is often very useful to look at the standardized residual versus standardized predicted plot in order to look for outliers and to check for homogeneity of variance.  The ideal situation is to see no observations beyond the reference lines, which means that there are no outliers.  Also, we would like the points on the plot to be distributed randomly, which means that all the systematic variance has been explained by the model.

• Analyze
   Regression
    Linear
     select math as the dependent variable and female, write and socst
     as independent variables    Save
      Click on Unstarndarized residuals
       Plots...
        select Zresid for the Y axis and ZPred for the X axis
         Continue
          OK
           Double click on the plot
            Options
             Y Axis Reference Line
              add a line at Y = -2.5
               Apply
                add a line at Y = 2.5
                 Apply

The P-P plots command produces a normal probability plot.  It is a method of testing if the residuals from the regression are normally distributed.

• Analyze
   Descriptives
    P-P plots
     select res_1 and the test distribution to be "normal"

The Q-Q plots produces a normal quantile plot. It is another method for testing if the residuals are normally distributed. The normal quantile plot is more sensitive to deviances from normality in the tails of the distribution, whereas the normal probability plot is more sensitive to deviances near the mean of the distribution.

• Analyze
   Descriptives
    Q-Q plots
     Select res_1 and the test distribution to be "normal"

2.4  Logistic regression

Logistic regression requires a dependent variable that is dichotomous (i.e., has only two values).  As we do not have such a variable in our data set, we will create one called honcomp (honors composition). This is purely for illustrative purposes only!

• Transform
   Compute
    select honcomp for the "target variable" and for numeric expression enter 
    "write >= 60".

• Analyze
   Regression
    Binary Logistic
     select honcomp as the dependent variable, and select read and socst as 
     covariates

2.5  Non-parametric tests

The binomial test is the nonparametric analog of the single-sample two-sided t-test.

• Analyze
   Nonparametric Tests
    Binomial 
     select write and define the cut point to be 50

The signrank test is the nonparametric analog of the paired t-test.

• Analyze
   Nonparametric Tests
    2 Related Samples 
     select write and read as the test pair list and select Wilcoxon 
     as the test type

The Mann Whitney U test is the nonparametric analog of the independent two-sample t-test.

• Analyze
   Nonparametric Tests
    2 Independent Samples 
     select write as the test variable list, 
     select female as the group variable
      click on Define Groups and enter 0 and 1
       Continue
        select Mann Whitney U as the test type

The Kruskal Wallis test is the nonparametric analog of the one-way ANOVA.

• Analyze
   Nonparametric Tests
    K Independent Samples 
     select write as the test variable list and select prog as the group variable
      click on Define Range and enter 1 for the Minimum and 3 for the Maximum
       Continue

3.0 Syntax version

get file "c:spss_datahs1.sav".

* chi-square test.
crosstabs
  /tables prgtype by ses
  /statistic = chisq.

* t-tests.
t-test
  /testval=50
  /variables=write.

t-test
  groups=female(0 1)
  /variables=write.

t-test
  pairs= write with science (paired).

* anova.
glm
  write  by prog
  /design = prog.

glm
  write  by prog ses
  /design = prog, ses, prog*ses
  /plot = profile(prog*ses).

glm
  write  by prog ses
  /design = prog, ses, prog*ses
  /posthoc = prog(tukey).
  
* ancova.
glm
 math with science write
 /design= science write.

* regression.
regression
  /dependent math
  /method=enter write science.

regression
  /dependent math
  /method=enter socst write ses
  /scatterplot=(*zresid ,*zpred ).

*residual plots.
pplot
  /variables=res_1
  /type=p-p
  /dist=normal.

pplot
  /variables=res_1
  /type=q-q
  /dist=normal.

* creating a dichotomous variable.
compute honcomp = (write > 60).
execute.

* logistic regression.
logistic regression var=honcomp
  /method=enter read socst.

* non-parametric tests.

* binomial test.
npar test
  /binomial (.50)= write (50).

* sign test.
npar test
  /sign= read with write (paired).

*signrank test.
npar tests
  /m-w= write by female(1 0).

* kruskal-wallis test.
npar tests
  /k-w=write by prog(1 3).

4.0 For more information

• Choosing the Correct Statistical Test in SPSS
  Includes guidelines for choosing the correct non-parametric test

• SPSS Frequently Asked Questions
  Covers many different topics including: ANOVA, Generalized Linear Models (GLM) and linear regression
• SPSS Regression Webbook
  Includes such topics as diagnostics, categorical predictors, testing interactions and testing contrasts
• SPSS Data Analysis Examples
  Includes examples of common data analysis techniques

• SPSS Annotated Output


Includes annotated output for descriptive statistics, correlation, regression and logistic regression

• SPSS Library
  Topics in ANOVA and other subjects