a set of procedures for describing, synthesizing, analyzing and interpreting quantitative data.
NOIR Scales-
emerge because different variables utilize varied ways of being measured, and thus, different kinds of data result.
different scales of measurement require specific statistics.
a statistic appropriate for a lower level of measurement may be applied to higher level data, since each subsequent scale has all characteristics of previous one.
The NOIR Scales
a.) Nominal-
lowest level of measurement, classifies persons or objects into mutually exclusive categories based upon shared characteristics.
categories may be true (persons/objects naturally fall into, such as sex, type of school), or artificial (operationally defined by the researcher, such as height, learning style).
not very precise, but sometimes necessary.
b.) Ordinal-
not only classifies, but also ranks objects of study in terms of degree of possession of characteristic of interest (e.g. most to least, highest to lowest).
amount of difference between ranks varies, so precision is not as much as desired.
c.) Interval-
has same characteristics as previous two, but rank intervals are equidistant.
type of scale utilized for intelligence and achievement tests.
problem is that minimum and maximum scores are arbitrarily assigned
(WAIS has FSIQ of 40 as lowest possible score, but does not indicate
absence of intelligence).
d.) Ratio-
highest, most precise level of measurement with all advantages of previous scales, but also has an absolute zero (e.g. no time, means just that).