INTRODUCTION
TO SPSS
WISDOM MGOMEZULU
mgomezuluwisdom@[Link]
Outline
• Introduction to SPSS
• Temperate designing for data entry
• Data entry
• Data cleaning
• Basic statistical analysis such as descriptive, hypothesis tests, modelling, etc
Make sure you have SPSS installed in your computer
Introduction to SPSS
SPSS –Statistical Package for Social Scientists
o Computer program/software for carrying out statistical analyses
o Very powerful and user-friendly
Nature of SPSS: three main windows
Data editor window is used for entering, displaying and editing data.
- shows data in two forms
• Data view
• Variable view
Output viewer for displaying results on an analysis or general sequence
of commands.
Syntax editor window is used for typing and running SPSS commands.
Introduction to SPSS
*You must save the data editor window and output viewer window
separately.
Make sure to save both if you want to save your changes in data or
analysis.*
Introduction to SPSS ct’d
Data view:
o Rows are cases
o Columns are variables
Variable view:
o Rows define the variables
• Name, Type, Width, Decimals, Label, Missing, etc.
Scale – age, weight, income
Nominal – categories that cannot be ranked (ID number)
Ordinal – categories that can be ranked (level of
satisfaction)
Opening SPSS
SPSS GUI-Variable view
Data view
SPSS MENUS
Gives you access to various functions in SPSS
File menu -lets u open, save, print, export data, close and access
recently used files
Edit-cut, copy, etc
View-to customize SPSS desktop: hide/show tool bar, status bar, etc
Data menu-data manipulation: sort cases, merge files, split file, etc
Transform-compute new variables & make changes to existing ones
Analyze-lets us perform all the statistical analysis
Graph- to make plots from the data
Utilities-Gives information about the variables and files
Add-ons-tells us other prog of SPSS family
Windows and help
Types of variables &
measurement
Types of variables
Quantitative -numeric in nature
Qualitative-non numeric
Levels of measurements
Nominal-categorical e.g. gender, Religion, race, occupation, etc
Ordinal : This is data ordered sequence. For example,
agreement (strongly agree, disagree) satisfaction (total
satisfaction, satisfied, dissatisfied), etc
Interval or Ratio: -numbers go from low to high in equal
intervals. Height and weight- examples.
Designing Template
Variable view
Rows define the variables
Name, Type, Width, Decimals, Label, values, Missing, Columns, Align, Measure, etc
Defining and naming variables
Data can be of various types:
Numeric*-variable whose values are numbers
U cannot perform any statistical analysis if u define it as string
String*-text format
Date
Comma
Currency, etc
Naming variables
Avoid use of space-unrecognised character
Designing Template
Variable names are usually short forms e.g. hhage for “age of the household
head”, etc
You can use under-score e.g. hh_edulevl for “education level of the household
head”
A blank space and special SPSS characters like, +, /, -, *, !, ’, and ? can not be
used
A variable name cannot be repeated for different columns
Note: there is no defined rules for naming variables-depends on the
comfortability of the researcher.
However, conventional short forms/symbols shud be considered
Designing Template
Width-specifies the length of the variable in characters
Decimals-affect the display of the data in data view. Take note of sensitive
variables
Label-details or description of the variable name. e.g. hh_id variable as
“household identification number”
Values-assigning the values or codes for the variable
For example:
Gender: 1=male, 0=female
Marital status: 1=married, 2=single, 3=divorced, 4=widow(er), etc
Occupation: 1=farming, 2=employment, 3=busines, etc
Designing Template
Assigning values
Click on the cell in the values column. A dialogue box will appear
Gender for example: enter 1 for male in the box labeled “value” and
specify its name (male) in the box labeled “value label”
This will activate the “add button” click the add button and repeat
the steps to specify for female
The buttons “change” and “remove” are used when u have made
mistakes
Note: The variable gender is qualitative but bcoz u have assigned the
Value labels-dialogue box
Designing template
Missing values-to specify missing values when
coding data
Specify certain values to variables for which the
respondent gave no responses
Usually out-of-range values: 999, 9999, 99, 888, etc
Why missing values?
Measure-to specify whether your variable is scale,
nominal or ordinal
Defining missing values
Multiple response questions/variables
Multiple response questions/variables
Define each response as a separate variable
Maintain the value labels for each variable
Examples: World Vision questionnaire
B11-Livelihood sources
B13
Ranking livelihood sources*
Open ended questions (un coded questions)
You can code them after data collection*
Or type them as text
A poorly designed template will give you problems!
DATA ENTRY
The data view is the mode that you enter your data
Rows are cases
The facilitator to demonstrate
Columns are variables
Missing values*
Multiple response questions*
Hands on using questionnaire
Data cleaning/management
Saving SPSS data file
Click on File> Click Save As and give a file name>
Click Save
DATA ANALYSIS-HANDS ON
Opening an existing SPSS data file
Click on File> Click Open> Click Data> Click File Name> Click
Open
Summarizing data: Descriptive analysis
Measures of central tendency: mean, median, mode-explore*
Frequencies-categorical
Tables
Comparing means-(quantitative variables)
One sample t-test
Independent sample t-test
Chi-square test (categorical variables)
Data Analysis
Custom tables
Modeling
Reading SPSS output and interpretation
Descriptive analysis
Data Analysis-descriptive
Data Analysis
Multiple response analysis
Analyse-multiple response-define variable sets
Categories, dichotomies, variable name, label, close
Analyse immediately before you close
Cross tabulation
To describe the relationship between variables-nominal
Analyse-descriptives-crosstabulation
Chi-square analysis and interpretation
Advanced data handling in SPSS
Sorting cases
Finding values, variables, cases
Merging files
Splitting files
Selecting cases
Recording values
Transforming variables
Computing new variables
Statistical Modelling
Linear regression, logit and probit models
Analyse-regression-linear/probit/logit
Dependent variables, independent variables
Diagnostic tests
T-TEST
There are tree types of T-test
1. Independent-samples t test (two-sample t test).
2. Paired-samples t test (dependent t test).
3. One-sample t test.
Independent-samples t test (two-sample t test).
Compares the means of one variable for
two groups of cases.
Descriptive statistics for each group and
Levene’s test for equality of variances are
provided, as well as both equal- and
unequal-variance t - values and a 95%-
confidence interval for the difference in
means.
Paired-samples t test (dependent t test).
Compares the means of two variables for a
single group.
This test is also for matched pairs or case-
control study designs.
The output includes descriptive statistics for
the test variables, the correlation between
them, descriptive statistics for the paired
differences, the t test, and a 95%-confidence
interval.
One-sample t test.
Compares the mean of one variable with a
known or hypothesized value.
Descriptive statistics for the test variables
are displayed along with the t test. A 95%-
confidence interval for the difference
between the mean of the test variable and
the hypothesized test value is part of the
default output.
IT GETS BETTER WITH
PRACTICE (WIZ, 2018)
THANK YOU
