Scribd
Upload
95 views2 pages

Scales or Level of Measurements

There are four levels of measurement used in data analysis: nominal, ordinal, interval, and ratio. Nominal level data involves categorizing items with no meaningful ordering. Ordinal level indicates ordering but not magnitude of differences. Interval level shows exact differences between cases using constant units of measurement. Ratio level possesses all interval characteristics and ratios between values have meaning, with an absolute zero point.

Uploaded by

hytii games
Copyright
© © All Rights Reserved
We take content rights seriously. If you suspect this is your content, claim it here.
Available Formats
Download as DOCX, PDF, TXT or read online on Scribd
95 views2 pages

Scales or Level of Measurements

There are four levels of measurement used in data analysis: nominal, ordinal, interval, and ratio. Nominal level data involves categorizing items with no meaningful ordering. Ordinal level indicates ordering but not magnitude of differences. Interval level shows exact differences between cases using constant units of measurement. Ratio level possesses all interval characteristics and ratios between values have meaning, with an absolute zero point.

Uploaded by

hytii games
Copyright
© © All Rights Reserved
We take content rights seriously. If you suspect this is your content, claim it here.
Available Formats
Download as DOCX, PDF, TXT or read online on Scribd
You are on page 1/ 2

SCALES (or Levels) OF MEASUREMENT

Data can also be classified based on levels of measurements or scales of measurement. There are four
levels of measurement used in preparing data for analysis, namely nominal, ordinal, interval, and ratio
level. The nominal level data are of the lowest level, the most primitive or the most limited type of
measurement while the ratio levels are classified under the highest level.

NOMINAL SCALE

It is a measurement scale that involves the process of naming or labeling the items by placing cases into
categories and counting their frequency of occurrences. While the numbers indicate that the elements
are different, such difference is not according to order or magnitude. Each case must be placed in one
and only one category, but the categories must be equal with respect to some of their attributes or
properties. The categories be non- overlapping or mutually exclusive. There are no measurements and
no scales involved. Instead, these are just counts. Examples under nominal scale are gender (male or
female), political affiliation ( Team Unity, Genuine Opposition, KBL, Lakas CMD, Laban, Kampi, Liberal
Party), mode of adaptation ( conformity, innovation, ritualism, retreatism, and rebellion), time
orientation ( past, present, future), religion, region, civil status, names of companies, dichotomous
responses or preferences, car makers ( Toyota, Honda, Ford, Kia Pride, Hyundai, Volkswagen, BMW).
These data are not graded, ranked or scaled for qualities such as better or worse, higher or lower, more
or less. They are merely labeled with no meaningful ranking of the categories is applied. This indicates
that for the nominal level of measurement, there are no particular order for the groupings. The numbers
may not be added, subtracted, multiplied, or divided. Only the frequency and percentage of
observations falling to each category are usually computed. While we can also determine observations
falling into each category are usually computed. While we can also determine the mode under this scale
of measurement, we cannot do it for the mean and median.

Data under nominal level are often summarized in charts and bar graphs.

ORDINAL SCALE

It is a measurement scale that yields information about the ordering of categories. The magnitude of
numerical differences between and /or among cases are not determined though. The intervals between
the points or ranks in an ordinal level are not known. Therefore, it is not possible to assign scores to
cases located at points along the scale. Examples are ranking of honor students, assessment of levels of
job performance (poor, average, excellent), ranking of faculty members ( instructor, assistant professor,
associate professor, professor), hardness of material, IQ (low, average, high). In this scale, one case is
said to be greater than or less than or less than the other using a criterion rather than saying that it is
only equal or different from the others as what is meant in the nominal scale of measurement.

INTERVAL SCALE

It is a measurement scale that shows order of cases into categories considering and indicating the exact
differences between and among the cases. It uses constant units of measurement, for example, pesos,
centavos, Fahrenheit, Celsius, yards, feet, minutes, seconds, which yield equal intervals between points
on the scale. Calendar time is an interval variable with an arbitrary defined zero point. An interval
variable does not have a “true zero” point, but zero point may be arbitrarily defined for convenient
purposes only.

A temperature of 30 degrees Celsius in Manila cannot be compared to a 15 degrees Celsius in Baguio. It


does not make sense, therefore to talk of a temperature of 0 degrees Celsius indicating the absence of
heat or the absence of temperature in a particular place.

Scores on a SAT examination and scores in a history or a mathematics examinations are also examples of
interval scale measurement.

RATIO SCALE

It is a measurement that possesses all the characteristics of interval scale and for which the interval size
and the ratio of two values have meanings. In ratio scale, it is appropriate to speak of one number in
relation to another. Measurement of weights, heights, lengths and ages appropriately use the ratio
scale. Examples of comparisons of measurement such as, a tree 6 meters tall is twice as tall as the other
three - meters tall tree, a baby girl which weighs 10 lbs. is twice as heavy as a baby girl weighing 5 lbs.,
could mean that one variable value, or measurement, may be spoken of as double or triple the other
variable.

An absolute zero is always implied. Any number used represents a distance from a natural origin. One
object may be twice as long, three times as heavy or four times as numerous as the other object. The
essential difference between the ratio and an interval level variable is that measurements of the former
are made from a true zero point, whereas, the latter measurements are from arbitrarily defined zero
point of origin. Therefore, the ratio variable is formed directly from the variable values from which
meaningful interpretations are done.

You might also like