You are on page 1of 2

1

Introduction to SPSS
The Statistical Package for Social Scientists (SPSS) is computer software best suited to enter
[research] data and to conduct some analyses.
When you start an SPSS session, you come to the Data Editor window where you see at the
bottom the “Data View” and the “Variable View” windows.
 Data View: For in-putting data and to view all the data entered per variable (question
= columns) and per respondent (rows). It is also the space for conducting the analysis.
 Variable view: This is the starting point to enter data in SPSS. It is the space for
defining all the variables. The rows are the variables (questions), while the columns
are the attributes of the variables. When all the variables are defined as per below,
then get back to Data View to input the data questionnaire by questionnaire.
o Name: What the variable is all about. It is always written as one compound word
e.g. Age Group would be agegroup; education level would be edulevel
o Type: Mainly numeric usually for two expressions i.e. Yes or No; 1 or 2 etc...or
figures such as Income, Age and so on.
 Others e.g. comma, date (key to keep same format), dollar for Income,
etc...
 String = Answers where they can input text e.g. name of the author
o Width: Usually 8 is fine, except for String to have more e.g. 100 or whatever you
consider appropriate
o Decimals: If analyzing e.g. incomes and other numerical measures
o Labels: Explanation of the variable in detail
o Values: Defining the variables and always be as precise as possible. E.g 1=Single;
2=Married; 3=Divorced; 4=Co-habiting etc…
o Missing Values: This often arises because the respondent did not answer the
question. . It is often marked as system missing with a single period. The reasons
could be many, but a missing value could tell a story e.g. refer to the many taboo
subject in many cultures.
 It is advisable to always keep two sets of data. The first with all the
missing data.
 In another one, you can input the missing values black clicking on it and
inputting 999. It is only after inputting a value that you can label it e.g.
Value=999; Label=No Response.

1
2

 The above applies to both Numeric and String variables (e.g. Value=NR;
Label=No Response).
o Columns: Usually relevant for string variables. The default 8 is often fine.
o Measure:
 Scale (metric/continuous/quantitative data): This is when data values
indicate the order of values and/or the distance between values. These are
usually measures that are predictable e.g. how old are you? household size,
level of income. When age is not grouped e.g. 15, 16, 17 etc then it
becomes scale. But when age is grouped, then it becomes ordinal. You can
only do linear regression with Scale.
o Scale is opposed to categorical data, which has a limited number of distinct
values or categories (for example, gender or marital status).
 Nominal is one type of categorical data = e.g. color of shirt = Yellow,
Red, Green (none is inherently better). Cannot be ordered, no preference.
Sex (Male, Female), Yes/No is also nominal coz none better than others.
Types of media (Radio, TV, NPs).
 Ordinal = This expresses data that has meaningful order of categories.
There is a discernible measure e.g. age (when grouped), distance (5km is
shorter than 10km), height (10 meters is higher than 6 meters), etc…
 For categorical data, the most typical summary measure is the number or
percentage of cases in each category. The mode is the category with the
greatest number of cases. For ordinal data, the median (the value at which
half of the cases fall above and below) may also be a useful summary
measure if there is a large number of categories.

Analysing SPSS data

 Before analyzing data, always reflect about the research questions. It is key to choose
a good control group i.e. comparable group of respondents for pretesting.
 Review carefully the questionnaires and the code-book.

You might also like