This course is aimed at staff and UM students who want to use IBM SPSS to complement or enhance their statistical analysis, but who do not have the time to study the associated techniques from scratch. The course is intended for beginner IBM SPSS users.

The main aim is to teach participants basic statistics, and how IBM SPSS can help in analysing data efficiently. Course participants will learn how to input data into IBM SPSS, how to create variables, how to illustrate data graphically, how to get simple statistics such as mean, variance etc., and reliability analysis, which is essential to validate results obtained from a questionnaire. The only essential prerequisite is familiarity with MS Windows and the basic concepts of MS Excel.

The course consists of three sessions of 2 hours each.

Sampling techniques and sample size

• Random, systematic, stratified, cluster and multistage sampling
• Estimate confidence interval given sample size
• Estimate sample size given confidence interval

Basics of IBM SPSS

• Explain SPSS toolbars
• Generate categorical and continuous variables
• Enter data into SPSS
• Use of the Recode and the Select options
• Transform continuous to categorical variables
• Sorting data
• Analyse multiple responses by using the multiple response option
• Analyse multiple responses by stacking the data

Exploring IBM SPSS graphics

• Produce histograms to display distribution of continuous variables
• Generate bar graphs and pie charts for categorical variables
• Produce scatter plots to relate two continuous variables
• Generate clustered bar graphs to relate two categorical variables
• Produce box plots and line graphs

Descriptive statistics

• Define measures of location - mean, median and mode
• Define measures of dispersion - range, variance and standard deviation
• Define measure of symmetry - coefficient of skewness
• Define measure of peakedness - coefficient of kurtosis

Reliability analysis

• Inter-rater, Intra-rater and Test retest reliability
• The Kappa measure of agreement for nominal scaled responses
• Kendall tau measure of agreement for ordinal scaled responses
• Intraclass correlation measure of agreement for interval scaled responses
• Internal consistency using Cronbach's Alpha and Guttman split half coefficient

The course is intended to be a hands-on course, where a technique learnt will be applied to a case study. Further exercises will be available for the user who wants further practice.

Download the IBM SPSS: Introduction courseware [PDF] [Intranet]

This course is aimed at staff and UM students who already have a background of statistics and the use of IBM SPSS as a statistical tool. The main aim is to teach participants more advanced statistical analyses, and how IBM SPSS can help in analysing data efficiently. Course participants will learn how to perform simple and advanced tests on data using IBM SPSS. The only essential prerequisite is knowledge of the content covered in the IBM SPSS: Introduction course.

The course consists of three sessions of 2 hours each.

Aspects of hypothesis testing

• Null and alternative hypotheses
• Type I and Type II errors
• Level of significance and P-values

Testing the normality assumption

• P-P plots and Q-Q plots
• Kolmogorov Smirnov test

Parametric tests

• Pearson correlation
• One sample t-test
• Paired samples t-test
• Two independent samples t-test
• One-way ANOVA
• Two-way ANOVA

Non-parametric tests

• Spearman correlation
• Binomial (Sign) test
• Wilcoxon signed ranks test
• Mann Whitney test
• Kruskal Wallis test
• Chi-Square test
• Friedman

The course is intended to be a hands-on course, where a technique learnt will be applied to a case study. Further exercises will be available for the user who wants further practice.

Download the IBM SPSS: Intermediate courseware [PDF] [Intranet] 

This course is aimed at staff, postgraduate and doctoral students (Masters and Ph.D.), who already have a background of statistics and the use of IBM SPSS as a statistical tool, but need to use advanced statistical analyses that are required for postgraduate theses. The main aim is to teach participants about general linear models, generalized linear models, data reduction and clustering techniques. The only essential prerequisite is knowledge of the content covered in the IBM SPSS: Introduction and IBM SPSS: Intermediate courses.

The course consists of three sessions of 2 hours each.

General linear models

• Multiple regression models
• ANOVA regression models
• ANCOVA regression models

Generalized linear models

• Binary/Multinomial logistic regression models
• Log-linear models for count data

Data reduction technique

• Factor analysis by principal components

Clustering techniques

• Hierarchical methods - Ward method
• Non-hierarchical method - K-means
• Two step cluster
• Discriminant analysis

The course is intended to be a hands-on course, where a technique learnt will be applied to a case study. Further exercises will be available for the user who wants further practice.

Download the IBM SPSS: Advanced courseware [PDF] [Intranet]