Data Analysis Using SPSS

Module 1 - SPSS - Basic

Module 1.1 – Developing the familiarity with SPSS Processer

Entering data in SPSS editor

Solving the compatibility issues with different types of file

Inserting and defining variables and cases

Managing fonts and labels

Data screening and cleaning

Missing Value Analysis

Sorting

Transposing

Restructuring

Splitting and Merging

Compute and Recode functions.

Visual Binning and Optimal Binning

Research with SPSS (random number generation)

Module 1.2 – Working with descriptive statistics

Frequency tables

Using frequency tables for analyzing qualitative data Explore

Graphical representation of statistical data

histogram (simple vs. clustered)

boxplot

line charts

scatterplot (simple, grouped, matrix, drop-line)

P-P plots, Q-Q plots

Addressing conditionality’s and errors

Computing standard scores using SPSS

Reporting the descriptive output in APA

Module 1.3 – Hypothesis Testing

Sample and Population, concept of confidence interval

Testing normality assumption in SPSS

Testing for Skewness and Kurtosis

Kolmogorov–Smirnov test

Test for outliers Mahalanobis Test

Dealing with the non-normal data

testing for homoscedasticity (Levene’s test) and multicollinearity.

Module 1.4 – Testing the differences between group means

t – test (one sample, independent – sample, paired sample)

ANOVA-GLM 1 (one way)

Post-hoc analysis, reporting the output in APA format.

Module 1.5 – Correlational Analysis

Data entry for correlational analysis

Choice of a suitable correlational coefficient

non-parametric correlation (Kendall’s tau)

Parametric correlation (Pearson’s, Spearman’s)

Special correlation (Biserial, Point-biserial)

Partial and Distance Correlation

Module 1.6 – Regression

The method of Least Squares

Linear modeling

Assessing the goodness of fit

Simple regression

Multiple regression (sum of squares, R and R2 , hierarchical, step-wise)

Choosing a method based on your research objectives

checking the accuracy of regression model. Logistic regression

Reporting the output in APA format.

Module 1.7 – Non-parametric tests

When to use

Assumptions

Comparing two independent conditions (Wilcoxon rank-sum test, Mann- Whitney test)

Several independent groups (Kruskal- Wallis test)

Comparing two related conditions (Wilcoxon signed-rank test)

Several related groups (Friedman’s anova)

Post-hoc analysis in non- parametric analysis

Categorical testing: Pearson’s Chi-square test

Fisher’s exact test

Likelihood ratio

Yates’ correction

Log linear Analysis

Reporting the output in APA format.

Module 2 - SPSS - Advanced

Module 2.1 – General Linear Models (GLM 1 to 5)

Theoretical basis of GLM: Assumptions and practical considerations

Comparing several means

ANCOVA

Factorial Anova

Repeated Measure Anova

Mixed Design Anova

MANOVA

Module 2.2 – Factor Analysis

Theoretical foundations of factor analysis

Exploratory and Confirmatory factor analysis

testing data sufficiency for EFA and CFA

Principal component Analysis

Factor rotation

factor extraction

using factor analysis for test construction

Interpreting the SPSS output: KMO and Bartlett’s test

initial solutions

correlation matrix

anti-image

explaining the total variance

communalities

Eigen-values

scree plot

rotated component matrix

component transformation matrix

factor naming

Module 2.3 – Cluster Analysis

Basic concepts

purpose and uses

selecting distance measures

K-Means clustering and hierarchical clustering

combining the clusters

working with the SPSS output: agglomeration schedule

proximity matrix

cluster membership

icicle plot and dendograms

Transform values and transform measures

Module 2.4 – Profile Analysis

Basic concepts

Assumptions and Practical issues

Difference in levels

Parallelism and flatness in profiles

profile contrast (simple effect analysis and interaction contrast)

Doubly-Multivariate Design

Classifying profiles

Practical Demonstration using standard scales

Interpreting the SPSS output: test of between subject effects (Intercepts, groups, effects), test of within subject effects-sphericity

Greenhouse-Geisser, Hyunh-Feldt

test of within subject contrast- linear

quadratic

cubic

Module 2.5 – Discriminant Analysis

Derivation and test of discriminate function

types of discriminant function: direct

sequential

stepwise

Interpreting discriminant function: discriminant plots

structure matrix loading

Interpreting the SPSS output: Log determinants

Box test

Walks’ Lambda

canonical discriminant function coefficient

Function at group centroids, Classification results

Module 2.6 – Survival Analysis

Meaning and types: non-parametric

semi-parametric and parametric

Life tables

cumulative proportion surviving

Hazard and density functions

table of group differences

Using SPSS: procedures-Kaplan-Meier

Cox-Regression

Cox-Regression with time dependent covariate

survival function

Omnibus Test

Module 2.7 – Neural network Analysis

Understanding neural network structures

multilayer perceptron’s-uses in estimation of cost and time (predicted-by-observed charts

Residual-by-predicted charts)

Radial basis functions-ROC curve

cumulative gains and lift charts

Module 2.8 – Time Series Analysis

Using time series analysis for forecasting

detecting trends and patterns in data

Box-Jenkin (ARIMA) method – forecasting using moving averages method

forecasting using trend analysis

seasonal decomposition

spectral plots

running analysis

understanding and interpreting output