SPSS Data Analysis

Drowning in Data

22 April 2014
nd

ROSLE MOHIDIN
Senior Lecturer
School Of Business & Economics, UMS

1
2
 Presentation Outline
 SPSS Environment -Review of SPSS Basics
– SPSS interface: data view and variable view
– How to enter data in SPSS
– How to clean and edit data
– How to transform variables
– How to sort and select cases
– How to get descriptive statistics
 Inferential Statistics in SPSS
– Independent t-test
– Regression

3
3
 Features of SPSS

 Originally developed for the people in Social

Science Areas, therefore, no heavy programming
background required
 Designed as User Friendly and has Pull Down
Menus to Execute Statistical Commands
 Ability to do Data Management & Manipulations
 Ability to Store Programs & Produce
Reports/Graphs

4
 SPSS Program Flow
Outside
Im
Data por
t ing
Source
SPSS Data
Data Modification/ Data Analysis
File Transformation
r y
Raw t E nt
ec
Data Dir
Pull-Down
Menu
OR
Syntax
Menu

(Data Steps) (Analysis Steps)

5
 An Example of Research use SPSS a
tool of Data analysis
 Youth Risk Behavior Surveillance System
(YRBSS, CDC)
– YRBSS monitors priority health-risk behaviors
and the prevalence of obesity and asthma
among youth and young adults.
– The target population is high school students
– Multiple health behaviors include drinking,
smoking, exercise, eating habits, etc.

6
6
 Data view
– The place to enter data
– Columns: variables
– Rows: records
 Variable view
– The place to enter variables
– List of all variables
– Characteristics of all variables

7
7
 You need a Questionnaire/code book/scoring
guide
 You give ID number for each case (NOT real
identification numbers of your subjects) if
you use paper survey.
 If you use online survey, you need
something to identify your cases.
 You also can use Excel to do data entry.

8
8
 Data View Window - Data Entry Site
(Columns=Variables, Rows=Cases)
Help Menu
Pull-down Menu bar Tool bar Information bar

Title bar

Variable
Names

Data View window Active cell Action bar9

 Variable View Window
Data Definition Site

64
Characters
Max, No Numeric,
space String, &
Length
Between Others
Beg letter, # of
Decimals Variable
@, #, or $ Description
Value
Code
Description Missing
Click here to see this view value
Description

10
 1. Click Variable View
2. Type variable name under
4. Description Name column (e.g. Q01).
2. Type of variable NOTE: Variable name can be 64
variable name
bytes long, and the first
3. Type:
character must be a letter or
numeric or
string one of the characters @, #, or
$.
3. Type: Numeric, string, etc.
4. Label: description of
variables.

1. Click this
Window

11
11
Based on your
code book!

12
12
 1. Two variables in the data set.
2. They are: Code and Q01.
3. Code is an ID variable, used to identify individual
case (NOT people’s real IDs).
4. Q01 is about participants’ ages: 1 = 12 years or
younger, 2 = 13 years, 3 = 14 years…

Under Data
View

13
13
 Save this
file as SPSS
data

14
14
 Cleaning the Data

 Key in values and labels for each variable

 Run frequency for each variable
 Check outputs to see if you have variables
with wrong values.
 Check missing values and Questionnaire if you
use surveys, and make sure they are real
missing.
 Sometimes, you need to recode string
variables into numeric variables
15
15
 Before we see
OK VS. Paste
buttons
Examples… <Output File>

1. OK - results/action
will be executed
16
Wrong
entries

17
17
 Descriptive statistics
– Purposes:
1.Find wrong entries
2.Have basic knowledge about the sample and
targeted variables in a study
3.Summarize data

Analyze Descriptive statistics

Frequency
18
18
 19
19
 20
20
 1. Skewness: a measure of the
asymmetry of a distribution.
The normal distribution is
symmetric and has a skewness
value of zero.
Positive skewness: a long right tail.
Negative skewness: a long left tail.
Departure from symmetry : a
skewness value more than twice
its standard error.
2. Kurtosis: A measure of the extent
to which observations cluster
around
a central point. For a normal
distribution, the value of the
Normal kurtosis
Curve
statistic is zero. Leptokurtic data
values are more peaked, whereas
platykurtic data values are flatter
and
more dispersed along the X axis. 21
21
 Example - School Data
Raw Data

 Subject 1
– Subject # (1)
– Female (1)
– Intensive (1)
– Reading (90)
– Math (67)

 Subject 2
– Subject # (2)
– Female (1)
– Moderate (2)
– Reading (72)
– Math (46)

 Subject 3
– Subject # (3)
– Male (0)
– Basic (3)
– Reading (41)
– Math (73)

22
 School Data
Variable View

Variable View Activated

23
 School Data
Completed Dataset – Data View

24
 School Data
Completed Dataset – Variable View

25
 Click to Obtain
Data File Information

26
Variable Information

27
Value Code Information

28
Basic Statistical Methods

 Independent t-test
 Multiple
Regression

29
 Independent t-test
– Is there a significant difference between 2
groups?

Assumptions 1. Normality 2. Variance 3.

Equality Independence

# of Variables Characteristics School Data

N=100
Dependent = 1 Continuous Math Score
Range of 0-100
Independent =1 Categorical Gender
2-levels
30
 How to calculate t-value?

Mean Difference
t-value= Group Variability

31
t-test
Medium
Variability

High
Variability

Low
Variability

32
 Independent t-test

1. Go to Analyze.

2. Choose
Compare Means.

3. Choose
Independent
Samples t Test.

33
 t-test

1. Choose Dependent
& Independent Variables.

34
 Independent Variable
Descriptives &
Analysis

Dependent Variable

Variance Equality Test t - statistics t= Mean Diff

Std. Error Diff

t= Z1 – Z2 = 63.20 – 54.10 = 9.093 = 3.295

SD12 + SD22 (13.914)2 +(13.064)2 2.760
N1 N2
41 59
35
Conclusion &
Chart

 There is a
significant
difference in
math ability
between
males and
females.

36
 Multiple Regression
– Which IVs can predict the DV and to estimate the effects of
these variables on DV?

Assumptions 1. Normality 2. Variance 3. 4. Linear

Equality Independen Relationship
ce

# of Variables Characteristics Health Survey

Data
N=100
Dependent =1 Continuous LDL Value
0-200
Independent > 1 Continuous or HT, WT, BMI, &
Dichotomous (0 Exercise
or 1) Variables 37
 Multiple Regression Diagram

HT
DV

WT

LDL
IV
BMI

Exercise

All 4 IVs are predicting LDL

38
Health Survey Data of N=100

39
 Multiple Regression

1.Choose Regression

2. Choose Linear Regression

40
 2. Choose Statistics you need.

1. Choose DV, IV, & Method.

3. Choose Residual Plots.

41
 Descriptives
& Correlation
Tables

Descriptive
Stats.

Correlation
Coefficients &
corresponding
p-values.

42
 R2=how much of the variability in the outcome is accounted
Main Analysis for by the predictors (regression sum of squared/total sum of squares)

Adj. R Sq=Adj for the # of

R=r between pred and Parameters in the model
observ value of the DV

Global test to
see if any
coefficient is
different from
“0” Partial/Part Tolerance
t & Sig=IV Correlations &VIF
B=Reg Coefficient predictability

Beta=Stdized. Reg
Coefficient.
Something is Wrong
if Beta >1!!

43
 Residual
Analysis

Residual Normality Linearity and

Equal Variance & residual independence

44
 Conclusion Multiple Regression
 IVs explain about
40% of the variability
of LDL level.
 The significant
predictors of LDL
were BMI and Hrs of
Exercise.
 The collinearity
statistics didn’t show
exceptionally large
multicollinearity
among predictors.
 Assumptions of
residual normality
and equal variance
were met. 45
 Key Concepts
 Statistical Models depend on the theory
and data. Choose your model wisely to
see if it can answer your research
questions.
 Check Assumptions. Model conclusions
may not be valid unless the assumptions
were met. If not, use appropriate
corrections, do data transformations, or
even use other statistical methods.

46
 Conclusions
 Statistical judgments come into
our daily lives. Statistics are
more than mathematical
calculations or scientific
research, but they are the way
of logical thinking…
Thank you
47