Scribd
Upload
20 views

Nasarawa State University Keffi Psychometrics

Uploaded by

safiyalabaran.sl
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
20 views

Nasarawa State University Keffi Psychometrics

Uploaded by

safiyalabaran.sl
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/ 7

NASARAWA STATE UNIVERSITY KEFFI

SCHOOL OF POSTGRADUATE STUDIES


FACULTY OF SOCIAL SCIENCES
INDUSTRIAL/ORGANIZATIONAL PSYCHOLOGY

SAFIYA LABARAN SALIHU AMS23P1O0003 PSYCHOMETRICS PSY864


ASS.PROF EMMANUEL ALHASSAN
TOPIC:
CLASSICAL TEST THEORY [CTT] vs ITEM RESPONSE THEORY [IRT]

What are Psychometric Tests? Psychometrics is the scientific study of


personality traits and inferred action through the use of
mathematical measurements.
Psychometric tests are specialized assessments created by scientists
to analyse people according to certain criteria. They are built to be
scientific and systematic tools to test a person’s ability to do a job.
Some tests are given to participants to measure their personality.
Psychometric tests are built upon one of two psychometric theories.
Classical test theory states that all testing experiences include some
elements that will induce errors. Thus, by administrators should
control for these errors by adjusting the resulting score to gain a
more accurate assessment. Item response theory is a theory that
individual test questions are the most important aspects of test. Each
item is chosen for specific purposes. Answer to individual items is an
assessment tool place the participant on a continuum for various
traits. Conglomerations of answers to multiple items can give the
most refined and accurate picture.
Classical Test Theory
What is the classical test theory [CTT] of psychometric tool
administration? Tests built upon the premise of classical test theory
calculate the measure and measure reliability based on an equation
that factors in error. There are three main aspects of CTT-founded
evaluations:
 Test score: This is also referred to as the observed score of a
participant. It is the raw data that results from a participant’s
answers. It may seem to be untrained mind that the initial
score of a participant is completely accurate. However, external
factors and other issues can cause errors in the data
 Error: Any false answer that ass potentially caused by an
external force or an internal bias.
 True test score: This is the result of adjustments performed to
the observed score or test score to compensate for errors.

Researchers attempt to make classical theory tests so accurate


that no errors are present. These tests are analysed in terms of
how reliable they are. The reliability of a classical test is
determined by subtracting the percentage of potential errors
from the total possible score. However, it is being
supplemented and replaced by item response theory. These
test are considered reliable automatically without any fear of
errors corrupting the results.

Item Response Theory


The item response theory [IRT] is a theory that focuses on an
individual’s responses to discrete questions. Each question lends
insight onto a person’s position on one or more spectrum of
personality traits. The main focus of IRT tests is the performance on
an exam made up of many individual items. One score is not given
for the entire test. Instead, the responder is evaluated based on
spectrums. This shows researchers their subjects strengths and
weaknesses rather than giving one score for the entire test. Many
major tests use the IRT approach because it allows for the creation of
large test banks.
The item response theory is also called the latent response or latent
trait theory. IRT analyses aspects of an individual’s personality of
lifestyle of which they may not be cognizant.
Example of item response testing include:
 Law school Admissions Test [LSAT]: This test is used for all
potential law school students throughout the united states.
Test creators formed an extremely large bank of questions
from which the questions for individual participants are
drawn. Not only those this promote randomization and
uniqueness for the students, but it also tests students on a
wide variety of characteristics and abilities.

 Scholastic Aptitude Test [SAT]: This test is used to determine if


a high school students is ready to enter undergraduate
programs at universities or colleges.

 Graduate Record Examinations [GRE]: This examination is


meant to determine whether or not a student graduating from
an undergraduate college can enter a graduate programme
 such as a master’s or a doctorate programme
Differences between CTT and IRT
There are multiple important differences between the CTT and IRT
test theories.
 Classical test theory examines the test as a whole. Item
response theory examines peoples’ responses to individual
questions.

 Classical test theory automatically assumes the presence of


errors in participants’ responses. Item response theory analyzes
the answer to the question thoroughly and does not attribute
any bias to the answers.

 Classical test theory is limited in its scope whereas item


response theory can be used with large numbers of answers.

 Classical test theory was created before item response theory

 Classical test theory examines only one metric whereas item


response theory can examine multiple metrics.

Lesson Summary
Psychometrics is the scientific measurements of some aspect
of an individual’s capabilities or knowledge
Psychometric tests are assets and tools that have been
deliberately created by scientists to examine the responses of
their subjects. There are two main theories regarding
psychometric test creation and administration.
Classical test theory states that the responses of a person to a
test should be considered in aggregate. Additionally,
practitioners who adhere to classical test theory states that all
people have some biases or other factors that result in errors in
the final score they receive. There is no way to escape these
errors. Therefore, a person’s psychometric test score, or
observed score, should be adjusted to compensate. The
observed score is a grade that a person receives. An error is
caused by problems in a person’s method of answering, biases,
mistakes of the test administrator, or other factors. The true
score of the test is the result of the adjustments performed to
the test score. Ideally, the test will be refined so as few errors
as possible can affect the results. The true score represents the
most probable score that a participant would have received if
there were no errors at all.

The second main psychometric test type is item response


theory. This theory is also called latent response or latent
response or latent trait theory. The main idea of item response
theory is that each individual item or answer in a test is
associated with personality traits or abilities. Thus, an answer
given to a questions demonstrate where a person is on one or
more spectrums. With every question answered, the
representation of people on spectrums becomes more and
more precise. Item response theory tests utilize a large number
of questions to make representations of respondents as
accurate as possible. Thus, test creators utilize immense
question banks. The LSAT, or Law School Admissions Test, was
created using item response theory. CRT looks at an entire test.
The IRT examines specific parts of tests for certain traits.

References:

Galton, F. [1897]. Psychometric experiments. Brain: A Journal of


Neurology, 11,149-162
Jones, L.V., & Thissen,D [.2007]. A history and overview. In C.R. Rao and
S. Sinharay [eds]. Handbook of statistics, 26: Psychometrics [pp.1-27].
Stout,W. [2002]. Psychometrics: From practice to theory and back.
Psychometricka , 67[4], 485-518.

Groenen, P.J., & Andries van der Ark, L. [2006]. Statistica Neerlandica,
60 [2], 135-144.

Borsboom,D. [2006]. The attack of the paychometricians. Psychometrika,


71[3] 425-440

Guilford, J.P. [1936[. Psychometric Methods. New York,NY: McGraw-Hill


Book company.

You might also like