Scribd
Upload
8 views

Reliability and Validity

The document discusses different types of reliability and validity in psychological testing. Reliability refers to consistency of test scores and is measured through test-retest reliability, alternate-form reliability, split-half reliability, and interrater reliability. Validity indicates how well a test measures the intended construct and includes content validity, face validity, construct validity, and criterion-related validity such as concurrent and predictive validity.

Uploaded by

Nishita Singh
Copyright
© © All Rights Reserved
We take content rights seriously. If you suspect this is your content, claim it here.
Available Formats
Download as PPTX, PDF, TXT or read online on Scribd
8 views

Reliability and Validity

The document discusses different types of reliability and validity in psychological testing. Reliability refers to consistency of test scores and is measured through test-retest reliability, alternate-form reliability, split-half reliability, and interrater reliability. Validity indicates how well a test measures the intended construct and includes content validity, face validity, construct validity, and criterion-related validity such as concurrent and predictive validity.

Uploaded by

Nishita Singh
Copyright
© © All Rights Reserved
We take content rights seriously. If you suspect this is your content, claim it here.
Available Formats
Download as PPTX, PDF, TXT or read online on Scribd
You are on page 1/ 21

PSYCHOLOGICAL TESTING: RELIABILITY & VALIDITY

Oindrila Mukherjee
RELIABILITY

What do you know / think about it?


CONCEPT

Reliability refers to the consistency of scores obtained by


the same persons when reexamined with the same test on
different occasions, or with different sets of equivalent
items, or under other variable examining conditions.

e.g. weighing machine


A reliable test should give similar results under different conditions (precision)

Does the measure give the same result each time it is administered on a respondent?

There are many types of reliability

Reliability is represented mathematically as a degree of consistency expressed in the form of a


correlation coefficient
TYPES OF RELIABILITY

Alternate- Internal
Test-Retest Split-Half Interrater
Form Consistency
TEST-RETEST RELIABILITY

Repeating the identical test on a second occasion

Correlating the scores of first test with second test

A minimal amount of variance is possible

AKA coefficient of stability

Correlation between scores for the same individuals at two different points in time

Temporal Stability
TEST-RETEST RELIABILITY:
DISADVANTAGES

Interval between test and person remembers how


Respondents are exposed
retest should be short s/he answered last time,
to the test items (practice,
especially when dealing answers the same the
recall)
with children next time

If the Interval is too


long construct (true
score) will change
ALTERNATE-FORM RELIABILITY

One way of avoiding the The same person is tested with


The correlation between the
difficulties encountered in one form of the test on the
scores obtained on the two
test-retest reliability is through first occasion and with
forms represents the reliability
the use of alternate forms of another, comparable form on
coefficient of the test
the test the second

Correlation between two


Aka Coefficient of
separate forms or scales
Equivalence / Parallel Form
developed to assess the same
Reliability
construct
ALTERNATE-FORM RELIABILITY

Occurs when an individual


participating in a research or The scores are then correlated to A given function is tested more
testing scenario is given two see if it is a reliable form of than once over time using
different versions of the same testing. equivalent forms of the test
test at different times

The letters ‘a’ and ‘e’ are used as


For e.g. you develop a test of In an alternate form of the same
they are similar to each other
memory where you ask test, you ask participants to
(vowels). The scores obtained
participants to recall words recall words beginning with the
for the two tests are correlated to
beginning with the letter ‘a’ letter ‘e’
find the reliability
SPLIT-HALF RELIABIITY

Single administration of one Divide the test into Administer both halves to
form of a test comparable halves (2 parts) the same person separately

Correlation between two


scores constructed as
Two scores are obtained for
Correlate these 2 scores (random) halves of a set of
each person
items on a scale {even &
odd items}
INTERRATER RELIABILITY

Aka scorer reliability

Some test scores have to be interpreted through judgement

For e.g. projective tests, creativity tests

2 or more administrators provide ratings / interpretations of each person

Correlation done between the ratings of the raters / scorers


INTERNAL CONSISTENCY RELIABILITY

how well are the


Correlation between
Aka inter-item items measuring
all the items on the
consistency what you want them
test
to measure

Precision of For e.g. I enjoy


For e.g. I hate ice
measurement of the eating eating ice
creams
question form creams
INTERNAL CONSISTENCY RELIABILITY

Kuder-Richardson

Alpha Coefficient

Omega Coefficient
VALIDITY

What the test How well does it


measures? measure?

Does the scale


For e.g. assess the
weighing construct /
machine variable it
intends to assess
TYPES OF VALIDITY

Construct Criterion-Related
Face Validity Content Validity
Validity Validity

Concurrent
Validity

Predictive
Validity
CONSTRUCT VALIDITY

Does your tool really measure the construct you want it to measure?
• e.g. tool on depression should not measure anxiety, mood, etc.

Tool should include indicators of the construct only


• Is based on existing and relevant knowledge about the construct

Mother of all other types of validity


• If content, face and concurrent validity are high, construct validity is also high
CONTENT VALIDITY

Does it cover a
Systematic examination representative sample of
Judgment by experts
of the test content the behavior domain to
be measured?

If any aspect is missing


Checks whether all
or if irrelevant
aspects of a construct are
information is added,
covered or not
content validity is low

e.g. depression scale (dimensions)


This type of validity focuses on the wording of the items of the test
FACE VALIDITY

What does the test Should not be


Does it ‘look’
superficially confused with
valid?
measure? content validity

Weakest form of
Informal and
validity but it is
subjective
important to

Focuses on the design/ layout/ spacing of the test


CRITERION-RELATED VALIDITY

Also called criterion validity

Indicates the effectiveness of a test in predicting an individual's behavior in specified situations

Performance or outcome of the test is measured against a criterion

A criterion is a measurement that is pre-established and considered to be valid

Correlation between the results of the developed test and the criterion measurement
CRITERION-RELATED VALIDITY

Types of Criterion Validity

Predictive Validity Concurrent Validity

Do test score correlate with a Do scores on this test correlate with


criterion obtained at a later scores on another test measuring the
point in time/ future same construct?

e.g. do depression scores e.g. correlating your depression


correlate with future depressive measure with Beck’s Depression
symptoms? Inventory
THANK YOU

You might also like

pFad - Phonifier reborn

Pfad - The Proxy pFad of © 2024 Garber Painting. All rights reserved.

Note: This service is not intended for secure transactions such as banking, social media, email, or purchasing. Use at your own risk. We assume no liability whatsoever for broken pages.


Alternative Proxies:

Alternative Proxy

pFad Proxy

pFad v3 Proxy

pFad v4 Proxy