|
|
|
||||
|
|
|||||
| crosstabs | Crosstabulations |
| t-test | t-tests |
| glm | General linear models |
| regression | OLS regressions |
| pplot | Normal probability plot |
| logistic | Logistic regressions |
| npar | Non-parametric tests |
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_data\hs1.sav
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 boxThis 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
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.
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"
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
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
get file "c:\spss_data\hs1.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).
Choosing the Correct Statistical Test
in SPSS
Includes guidelines for choosing the correct non-parametric test
UCLA Researchers are invited to our Statistical Consulting Services
We recommend others to our list of Other Resources for Statistical Computing Help
These pages are Copyrighted (c) by UCLA Academic Technology Services