Scribd
Upload
100 views4 pages

Scales of Measurement: Dr. Michael Passer Psychology 209 University of Washington 1

The document discusses the four scales of measurement: nominal, ordinal, interval, and ratio. It provides examples for each scale and describes their key properties. Nominal scales involve qualitative categories without mathematical relationships between values. Ordinal scales represent ordered categories where differences between values are quantitative but intervals may not be equal. Interval scales have equal intervals but no true zero point, so ratios cannot be formed. Ratio scales have a true zero point so ratios can be formed.

Uploaded by

jymrprzbtst
Copyright
© © All Rights Reserved
We take content rights seriously. If you suspect this is your content, claim it here.
Available Formats
Download as PDF, TXT or read online on Scribd
100 views4 pages

Scales of Measurement: Dr. Michael Passer Psychology 209 University of Washington 1

The document discusses the four scales of measurement: nominal, ordinal, interval, and ratio. It provides examples for each scale and describes their key properties. Nominal scales involve qualitative categories without mathematical relationships between values. Ordinal scales represent ordered categories where differences between values are quantitative but intervals may not be equal. Interval scales have equal intervals but no true zero point, so ratios cannot be formed. Ratio scales have a true zero point so ratios can be formed.

Uploaded by

jymrprzbtst
Copyright
© © All Rights Reserved
We take content rights seriously. If you suspect this is your content, claim it here.
Available Formats
Download as PDF, TXT or read online on Scribd
You are on page 1/ 4

Dr.

Michael Passer Psychology 209 University of Washington 1

Scales of Measurement

The concept of scales of measurement refers to rules for


assigning scale values to measurements.

Nominal
A scale of measurement in which the scale values represent
categories that only differ from one another qualitatively (i.e.,
differ in “type” rather than in “amount”). In other words, the
different values of a variable are categories that only
represent different “types” of something. Variables measured
using a nominal scale are also known as "qualitative"
variables.

Examples are ethnic group membership, your college major,


religious affiliation, eye color, political affiliation.

Numbers are used merely to represent unique categories


and have no mathematical properties:

Example: Political affiliation might be coded on a questionnaire in


any of the following ways. The numbers are arbitrary and each
method is as good as the others.

Option 1 Option 2 Option 3


1 = Republican 1 = Democrat 1 = Independent
2 = Democrat 2 = Republican 2 = Republican
3 = Independent 3 = Independent 3 = Democrat
4 = Other 4 = Other 4 = Other
Dr. Michael Passer Psychology 209 University of Washington 2

Ordinal:
A scale of measurement in which the scale values represent
categories that differ quantitatively in terms of their order,
but in which the intervals between numbers (i.e., between
categories) cannot be assumed to be equal.
st nd rd
Examples are 1 2 and 3 place finishers in a race, birth order,
rank order of your 5 favorite movies from all time, rank your top 3
friends, your class rank.

In a race, we cannot assume that the difference in time between the


first and second place finishers is the same as the difference in
time between the second and third place finishers. The numbers
only tell us about the order in which they crossed the finish line.

Likewise, when you rank order the names of your three favorite
friends, the numbers 1, 2, and 3 only tell us who you like most,
second most, and third most. Perhaps you like your top ranked
friend a lot more than your second ranked friend. Perhaps you like
your top two friends almost equally. If we measure friendship on
an ordinal scale by obtaining rankings, we cannot make
assumptions about distances between the ranks.
Dr. Michael Passer Psychology 209 University of Washington 3

Interval
A scale of measurement in which the distance between any
two adjacent scores is the same as the distance between
any other two adjacent scores. However, there is no “true” or
“natural” zero point and therefore meaningful ratios cannot
be formed. In short, numbers are spread across equal
intervals without a natural zero point.

An example is temperature, measured either in degrees Celsius or


degrees Fahrenheit. The intervals are equidistant (i.e., a 1 degree
increase from 15 to 16 degrees represents the same amount of
increase in temperature as does 30 to 31 degrees, but 30 degrees is
not “twice as hot” as 15 degrees).

NOTE: Many variables measured by psychologists are often


analyzed statistically as if they were measured on an interval scale.
Common examples include IQ scores, personality test scores,
aptitude test scores, and scores from rating scales and attitude
scales [e.g., “To what extent do your disagree or agree with the
following statement: Congress is doing a good job: -2 (Strongly
Disagree) -1 (Disagree) 0 (Neutral) +1 (Agree) +2 (Agree
Strongly)”]. There is considerable disagreement, however, as to
whether the scale of measurement for these variables truly is an
interval scale, or better represents an ordinal scale. For example,
does the 5-point difference in IQ scores of 110 versus 105 represent
the same amount of difference in intelligence as that represented by
IQ scores of 105 versus 100? In measuring attitudes, does the 1-
point difference between “Strongly Agree” and “Agree” represent the
same amount of change in the strength of an attitude as the 1-point
difference between “Agree” and “Neutral”?

One thing is for sure, an IQ score of 130 or an attitude rating of +2


(Strongly Agree) cannot be interpreted as meaning that someone has
twice as much intelligence or twice as positive an attitude as
someone who has an IQ score of 65 or an attitude rating of +1
(Agree).
Dr. Michael Passer Psychology 209 University of Washington 4

Ratio
With ratio scales, the scale values are numbers that
represent equal distances in some attribute, and there also
is an absolute zero point. Thus, meaningful ratios can be
formed.

Examples are length, height, weight, time, number of errors made


performing a task, and number of tickets sold.

If IQ scores were ratio (they are not), you could say that
someone with a 130 IQ was twice as smart as someone with
a 65 IQ (a ratio of 2 to 1), but we cannot make this claim.
However, for height measured in inches, which is a ratio
scale, you may properly say that somebody 60 inches tall is
twice as tall as somebody 30 inches tall.

You might also like