Statistical Consulting

SPSS

SPSS is a computer application that provides statistical analysis of data. It allows for in-depth data access and preparation, analytical reporting, graphics and modelling.

More analytical power can be added by using optional SPSS modules.

Flinders University has a site licence for this product that allows for its unlimited use on computers owned by the University. There is also provision for home use by University staff and a graduate pack arrangement for use at home by postgraduate students.

The SPSS website is at www.spss.com.

Which SPSS products are we licenced for?

SPSS Base

Data access, manipulation, analysis, and presentation of your results. Includes extensive on line help and tutorial sessions. Base includes all the data management and graphing facilities and the following analysis procedures:

Reports
Frequencies
Descriptives
Explore
Crosstabs
Descriptive ratio statistics
Means
t tests
Oneway ANOVA
ANOVA - simple factorial
Correlate
Distances
Linear regression
PLUM regression
Curve estimation
Nonparametric tests
Multiple response
Factor analysis
TwoStep cluster analysis
Cluster
Quick cluster
Discriminant analysis
Multidimensional scaling

SPSS Regression Models

Predict behaviour or events when your data go beyond the assumptions of simpler regression techniques with the following analysis procedures:

Multinomial logistic regression
Binary logistic regression
Non-linear regression
Weighted least squares
Two-stage least squares
Probit regression

SPSS Advanced Models

Analyzing complex relationships with high end modeling procedures:

Generalized linear models
Generalized estimating equations
Mixed modelling
General linear models with random or mixed effects
Repeated measures ANOVA
Variance components analysis
Loglinear analysis
Hierarchical loglinear analysis
Generalized linear loglinear and logit modelling
Survival analysis
Kaplan-Meier analysis
Cox regression

SPSS Exact Tests

If you have a small number of cases, subset your data into fine breakdowns or have variables where 80 percent or more of the responses are in one category, then traditional tests could be incorrect. Exact test results are provided for many of the tests and procedures in Base.

SPSS Categories

Creates perceptual maps and predicts categories. Optimal scaling procedures free you from the restrictions of two-way tables, placing the relationships among your variables in a larger frame of reference. See a map of your data -- not just a statistical report. Includes the following analysis procedures:

Categorical regression
Correspondence analysis
Multiple correspondence analysis
Categorical principal components analysis
Nonlinear canonical correlation analysis
Multidimensional scaling
Preference scaling

SPSS Tables

Create more sophisticated presentation-ready tabular reports. Handle multiple-response questions and missing data. Calculate over 30 statistics in your tables automatically. Provides an intuitive graphical user interface with a preview pane that allows you to interact with your variables and see what your tables will look like before you click “OK”.

SPSS Trends

Improve forecasting with complete time-series analyses: multiple curve-fitting and smoothing models and methods for estimating autoregressive functions. Use powerful ARIMA modeling. Includes the following analysis procedures:

Create models – ARIMA and exponential smoothing models
Apply models – apply saved models to new or updated data
Seasonal decomposition
Spectral analysis

SPSS Missing Value Analysis

Helps you diagnose whether your missing data presents a serious problem, and then estimates the missing values using the data present for a valid conclusion. Includes the following analysis procedures:

Analyse patterns in missing data
Univariate, listwise and pairwise descriptive statistics
EM algorithm – estimation and imputation
Regression algorithm – estimation and imputation

SPSS Conjoint

Learn the relative importance of product attributes, and determine what the most preferred attribute levels are. Understand consumer preferences or ratings of existing or possible products in terms of product attributes and their levels. Includes the following analysis procedures:

Orthogonal design – generate and display
Conjoint analysis

AMOS

A powerful structural equation modeling and confirmatory factor analysis tool that specifies and fits your models easily and visually without using matrices. Unveil more critical information than with conventional methods such as regression or exploratory factor analysis.