ISA SPSS summary
Levels of measurement and measures of central tendency (mean, median, mode)
The mean is only useful for interval and ratio variables (however, there is a lot of discussion on
whether or not the mean of an ordinal variable might be useful as well, Erasmus thinks it is not
useful); the median for ordinal, interval and ratio variables, and the mode for all variable levels.
Interval variable examples (meaningful ranking, distances are equal but no absolute zero point):
Fahrenheit, Celsius, pH.
Note that Kelvin is a ratio variable, as 0.0 means ‘no heat’. For ratio, a 0 point means the complete
absence of something: e.g. height, age, weight, etc.
When working with ratio variables, but not interval variables, you can look at the ratio of two
measurements. E.g. 4 grams is twice as much as 2 grams (ratio), but 100C is not twice as much as 50C
(interval).
Subjects may be asked to rate their level of pain, how much they like a consumer product, their
attitudes about capital punishment, their confidence in an answer to a test question. Typically these
ratings are made on a 5-point or a 7-point scale. Another example is movie ratings (ranging from 1 to
5 stars). These scales are ordinal scales since there is no assurance that a given difference represents
the same thing across the range of the scale. For example, there is no way to be sure that a
treatment that reduces pain from a rated pain level of 3 to a rated pain level of 2 represents the
same level of relief as a treatment that reduces pain from a rated pain level of 7 to a rated pain level
of 6. This is also the case for scales containing terms such as ‘very satisfied’, ‘neutral’, ‘very
dissatisfied’. Other examples of ordinal variables are:
1. Ranking movies in a top 50
Note that jersey numbers are considered to be a nominal variable, as these numbers serve as
alternate forms of names for the players. Ranking makes no sense.
Emory University wants to know which dormitories the students prefer. The administration counts
the number of applications for each dorm. Administrators assign a rank to each dorm based on the
number of applications received. The answer to this question is ordinal, not ratio!
Altering the dataset
• Select cases (Data – Select Cases)
Sometimes you only want to look at a select part of your data. For example, in a European survey you
might only want to look at the Brits, French, Germans, etc instead of the entirety of Europe.
Data > Select Cases
If condition is satisfied.. Select corresponding variable and number (e.g. country = 3) > Continue >
OK.
Tip! Right click on variable > Variable information to see the corresponding labels.
Tip! Do not forget to switch off the filter when you are done with Select Cases, as SPSS will not do
this automatically. Data > Select Cases > All cases. You will also see if the filter is on by looking at the
bottom right of your screen (it will literally say Filter ON).
• Recode into different variable (Transform – Recode into Different Variable)
Recoding into a different variable transforms an original variable into a new variable. That is, the
changes do not overwrite the original variable; they are instead applied to a copy of the original
, variable under a new name. You always want to keep the original variable, so do not choose Recode
into same variable!
Recoding variables allows for between-group comparisons. E.g., from your age variable you can
create groups labelled 18 or younger, between 19 and 28, between 29 and 38, etc.
E.g., families with children, you want to create groups of 0-2 children, 3-5, 6 or more…
In the recode into different variables screen, you select the variable that you want to divide in
groups, and use the arrow to put it in the middle. Give your new variable a new name (can be
anything) with the Output Variable part, and press Change so that the question mark becomes the
new name.
Then select Old and New Values…
At new value, enter 1, and select the range of the cases that should fall within group 1. Do the same
for all the other groups. Note that in the example above, 20 belongs to the second group, not the
first! 50 also belongs to the second group, not the third! When done, press Continue.
Your new variable will appear at the bottom of Variable View and at the right of Data view. Now you
can change its properties. Don’t forget to add values so that you can read your results.
Levels of measurement and measures of central tendency (mean, median, mode)
The mean is only useful for interval and ratio variables (however, there is a lot of discussion on
whether or not the mean of an ordinal variable might be useful as well, Erasmus thinks it is not
useful); the median for ordinal, interval and ratio variables, and the mode for all variable levels.
Interval variable examples (meaningful ranking, distances are equal but no absolute zero point):
Fahrenheit, Celsius, pH.
Note that Kelvin is a ratio variable, as 0.0 means ‘no heat’. For ratio, a 0 point means the complete
absence of something: e.g. height, age, weight, etc.
When working with ratio variables, but not interval variables, you can look at the ratio of two
measurements. E.g. 4 grams is twice as much as 2 grams (ratio), but 100C is not twice as much as 50C
(interval).
Subjects may be asked to rate their level of pain, how much they like a consumer product, their
attitudes about capital punishment, their confidence in an answer to a test question. Typically these
ratings are made on a 5-point or a 7-point scale. Another example is movie ratings (ranging from 1 to
5 stars). These scales are ordinal scales since there is no assurance that a given difference represents
the same thing across the range of the scale. For example, there is no way to be sure that a
treatment that reduces pain from a rated pain level of 3 to a rated pain level of 2 represents the
same level of relief as a treatment that reduces pain from a rated pain level of 7 to a rated pain level
of 6. This is also the case for scales containing terms such as ‘very satisfied’, ‘neutral’, ‘very
dissatisfied’. Other examples of ordinal variables are:
1. Ranking movies in a top 50
Note that jersey numbers are considered to be a nominal variable, as these numbers serve as
alternate forms of names for the players. Ranking makes no sense.
Emory University wants to know which dormitories the students prefer. The administration counts
the number of applications for each dorm. Administrators assign a rank to each dorm based on the
number of applications received. The answer to this question is ordinal, not ratio!
Altering the dataset
• Select cases (Data – Select Cases)
Sometimes you only want to look at a select part of your data. For example, in a European survey you
might only want to look at the Brits, French, Germans, etc instead of the entirety of Europe.
Data > Select Cases
If condition is satisfied.. Select corresponding variable and number (e.g. country = 3) > Continue >
OK.
Tip! Right click on variable > Variable information to see the corresponding labels.
Tip! Do not forget to switch off the filter when you are done with Select Cases, as SPSS will not do
this automatically. Data > Select Cases > All cases. You will also see if the filter is on by looking at the
bottom right of your screen (it will literally say Filter ON).
• Recode into different variable (Transform – Recode into Different Variable)
Recoding into a different variable transforms an original variable into a new variable. That is, the
changes do not overwrite the original variable; they are instead applied to a copy of the original
, variable under a new name. You always want to keep the original variable, so do not choose Recode
into same variable!
Recoding variables allows for between-group comparisons. E.g., from your age variable you can
create groups labelled 18 or younger, between 19 and 28, between 29 and 38, etc.
E.g., families with children, you want to create groups of 0-2 children, 3-5, 6 or more…
In the recode into different variables screen, you select the variable that you want to divide in
groups, and use the arrow to put it in the middle. Give your new variable a new name (can be
anything) with the Output Variable part, and press Change so that the question mark becomes the
new name.
Then select Old and New Values…
At new value, enter 1, and select the range of the cases that should fall within group 1. Do the same
for all the other groups. Note that in the example above, 20 belongs to the second group, not the
first! 50 also belongs to the second group, not the third! When done, press Continue.
Your new variable will appear at the bottom of Variable View and at the right of Data view. Now you
can change its properties. Don’t forget to add values so that you can read your results.
