Validity is one of two criteria
(the other being reliability) by which researchers judge
their results or measurement tools. A valid result is one that accurately measures what it
claims to be measuring. Using shoe size as a measurement of intelligence is not a valid
measure of intelligence. It lacks face validity since it is not obvious that it is
measuring what it claims to measure.
One test of validity might be
the extent to which your measurements allow you to make predictions about future
behaviour. If your measurement of intelligence does not predict how people perform in
examinations then perhaps it is not a valid measurement of intelligence. The concept of
validity refers to the extent to which the data we collect gives a true measurement of
"social reality" (what is "really happening" in society).
The concepts of reliability and
validity very important in sociological research:
If data is reliable but not
valid, then it may have limited use. We can make general statements about the world, but
such statements may not actually apply to any particular social group.
If data is valid, but not
reliable, we may not be able to use it to make general statements about the world (for
example, we may be able to understand something about one particular group of people that
may not necessarily apply to all people in that group).
Concurrent validity is a
parameter used in sociology, psychology, and other psychometric or behavioral sciences.
Concurrent validity is demonstrated where a test correlates well with a measure that has
previously been validated. The two measures may be for the same construct, or for
different, but presumably related, constructs.
In social science and
psychometrics, construct validity refers to whether a scale measures the unobservable
social construct (such as "fluid intelligence") that it purports to measure. The
unobservable idea of a unidimensional easier-to-harder dimension must be
"constructed" in the words of human language and graphics. Nomological validity
is a form of construct validity. It is the degree to which a construct behaves as it
should within a system of related constructs called a nomological set.
On the Validity of
Official Statistics A Comparative Study of White, Black, and Japanese High-School Boys -
William J. Chambliss, Univ. of California, Richard H. Nagasawa, Portland State College,
Journal of Research in Crime and Delinquency
A continuing debate in the sociology of deviance is the degree to which official
statistics are representative of the distribution and rate of deviance in the society.
Some argue that official statistics are totally meaningless as indices of deviance rates;
others maintain that, although official statistics are distorted, they nonetheless
indicate important trends and distributions of real rates of deviance. Official and
unofficial delinquency rates of black, white, and Japanese youths in a large metropolitan
area were analyzed. Our conclusion is that official statistics are so misleading that they
are virtually useless as indicators of actual deviance in the population. It is suggested
that the visibility of the offenses, the bias of the policing agencies, and the demeanor
of the youth account for the rate and distribution of delinquency among the three groups
and that official rates are a complete distortion of the actual incidence.
Critical Sociology of Science and Scientific Validity - Sal Restivo,
Rensselaer Polytechnic Institute, Julia Loughlin, Syracuse University, Science
Communication, Vol. 8, No. 3, 486-508 (1987)
Critical sociology of science, in conjunction with an emancipatory epistemology, focuses
attention on the social and especially value aspects of validity and objectivity. This is
the foundation for our arguments concerning the relevance of the sociology of science for
explaining and understanding science, and for fashioning science policy. After outlining
the basic features of critical sociology of science and emancipatory epistemology, we
explore their implications for the problem of validity in applied social science research.
Drawing on research on the use of illegal drugs, we argue that Campbell's suggestions
regarding how to define research problems and treat qualitative methods, as well as other
aspects of his "sociology of scientific validity, " may improve validity in the
sense of achieving consensus and vigor within specialties. More generally, however, his
program may inhibit progress in developing accurate, objective knowledge of science and
Validity - Toward an Interdisciplinary Science Studies
Robert A. Neimeyer, William R. Shadish, JR, Memphis State University
Science Communication, Vol. 8, No. 3, 463-485 (1987)
In his lead article for this series, Campbell discusses the role of a "disputatious
and mutually reinforcing community of truth seekers" in promoting a sociology of
scientific validity. Situating his work in the broader context of the sociology of
science, we first raise a number of questions about the conceptual and empirical warrant
for his recommendations, local factors that might affect their implementation, and
temporal and "level of analysis "issues which could influence the optimal form
that a scientific community might take. In general, however, these considerations do not
vitiate Campbell's prescriptions so much as point up the need for conceptually informed
empirical research on the outcome of varymg social structures for science. We then survey
some of the potential contributions to science studies that are being made by
psychologists of diverse specialties, ranging from a relatively mdividualistic focus on
creativity and cognitive processes to a broader view of scientific behavior in a social
Validity of Self-Reports
About Quality of Life Among Patients With Schizophrenia
Nasreen Khatri, M.Sc., David M. Romney, Ph.D. and Guy Pelletier, Ph.D.
Psychiatr Serv 52:534-535, April 2001
Lehman's Quality of Life Interview was administered to 22 patients with schizophrenia and
their proxies and to 15 patients with cancer and their proxies. The results indicated that
there was a discrepancy between responses on global objective and subjective measures for
patients with schizophrenia but not for patients with cancer. A discrepancy was also found
for the proxies of the patients with schizophrenia but not for the proxies of the patients