DR VANDANA GAUR
ASSOCIATE PROFESSOR
DEPARTMENT OF PSYCHOLOGY
INTRODUCTION
• The reliability and validity of a test is important for data collection process.
• Before and after collecting the data , the researcher need to consider the
reliability and validity of their data.
• Reliability- Are we measuring accurately?
• Validity- Are we measuring the right things?
RELIABILITY
• It is the degree to which an assessment tool produces stable and
consistent results.
• The extent to which an experiment or measuring procedure yields the
same result on repeated trials
• A measure can be reliable without being valid but it cannot be valid
without being reliable.
TYPE OF RELIABILITY
• Test-retest Reliability
• Parallel form Reliability
• Split half or internal Consistency Reliability
• Rational equivalence Reliability
TEST-RETEST RELIABILITY
• Obtained by administering the same test twice over a period of time to a group
of individuals.
• The scores are tabulated and the correlation calculated
• Correlation is higher then there will be more reliability
PARALLEL FORM RELIABILITY
• Different version of an assessment tool to the same group of individuals
• Both version must contain same construct, skill ,knowledge etc.
• The number of items in both the forms should be the same.
• Distribution of the indexes of difficulty of the items in both should be similar
• Means and SD of both the forms should be equal or nearly so.
• Mode od administration and scoring of both the forms should be uniform
SPLIT HALF RELIABILITY
• “Splitting in half” all items of a test that are intended to probe the same area of
knowledge in order to form two “sets” of items
• Items are divided into two halves and then compared. Odd, even items, or 1-50
and 51-100 are two ways to split items.
• Product moment correlation is computed between two sets of scores.
• Spearman-Brown prophecy formula is used for estimating the reliability of the
whole test
FACTORS INFLUENCING RELIABIITY
Extrinsic Factors
• Time interval between testing: testing with shorter intervals resulting in the higher
reliability coefficient
• Guessing by the examinee
• Condition under which measure: Favorable conditions will give high reliability
coefficient
Intrinsic Factors
• Length of test: The longer the test, higher the reliability due to the fact that longer test
will provide more adequate sample of the behavior being measured
• Difficulty level of test: tests that are too easy or too difficult for the group members
taking it will tend to provide score of low reliability
• Discrimination value:When the test is composed of discriminating items,the items likely
to be high
• Homogeneity of items:More homogenous more reliabilty
• Indicate the degree to which an instrument measures what it is supposed to
measure
• The degree to which a study accurately assesses the specific concept or
parameter that the researcher is trying to measure
• Whether the researcher has measured what it set out to measure
• For a test to be reliable it also need to be valid
INTERNAL VALIDITY
• Content or curricular Validity
• Construct validity
• Criterion-related Validity
a- Predictive validity
b-Concurrent validity
EXTERNAL VALIDITY
The extent to which the results of a study are generalizable or transferable
CONTENT VALIDITY
• Involves assessing whether the questions asked and the measurement recorded
adequately represent the domains being studied
• Is the questionnaire comprehensive?
• Entire range or universe of the construct is measured
• Usually evaluated and scored by experts in the content area
CONSTRUCT VALIDITY
• It implies using the construct correctly
• The most complex type of validity, and involves relating an instrument for data
collection to a theoretical framework
• For example, a researcher inventing a new IQ test might spend a great deal of
time attempting to define intelligence in order to reach an acceptable level of
construct validity.
CRITERIAN –RELATED VALIDITY
• Used to product future or current performance
• Demonstrating the accuracy of a measure or procedure by comparing it with
another procedure which has already been demonstrated to be valid
• It correlates test results with another criterion of interest
FACTORS AFFECTING VALIDITY
• Range of ability-Like reliability, validity is also influenced by the range of ability of the
sample used. If the subjects have a very limited range of ability, the validiy coefficient
will be low
• Ambiguous direction: If the directions of the test are ambiguous ,the validity of the test
would be lowered
• Socio-cultural differences: Cultural differences among different societies are likely to
affect the validity of a test. However ,when a test is cross-cultural ,this factor does not
affect the validity of the test
• Addition of Inappropriate Items: Inappropriate items are likely to lower both reliability
and validity of the test
• Length of the test: Homogenous lengthening of the test not only increses the reliability
but also validity of the test
CONT…….
Environmental factors: Some environmental factors such as room temperature,
lightning and noise can influence the error rate
Research Factors: The researcher can influence the result of the study in many
ways
Instrumental factors: An inadequate sampling of items is the other common
source of instrumentation errors
Subject factors: Any changing physical, emotional or psychological state of the
subject could introduce error into measurement process
RELATIONSHIP BETWEEN RELIABILITY AND VALIDITY
Validity and reliability are closely related
• A test cannot be considered valid unless the measurement resulting from it are
reliable
• Likewise result from a test can be reliable and not necessarily valid
Reliability and Validity.pptx

Reliability and Validity.pptx

  • 1.
    DR VANDANA GAUR ASSOCIATEPROFESSOR DEPARTMENT OF PSYCHOLOGY
  • 2.
    INTRODUCTION • The reliabilityand validity of a test is important for data collection process. • Before and after collecting the data , the researcher need to consider the reliability and validity of their data. • Reliability- Are we measuring accurately? • Validity- Are we measuring the right things?
  • 3.
    RELIABILITY • It isthe degree to which an assessment tool produces stable and consistent results. • The extent to which an experiment or measuring procedure yields the same result on repeated trials • A measure can be reliable without being valid but it cannot be valid without being reliable.
  • 4.
    TYPE OF RELIABILITY •Test-retest Reliability • Parallel form Reliability • Split half or internal Consistency Reliability • Rational equivalence Reliability
  • 5.
    TEST-RETEST RELIABILITY • Obtainedby administering the same test twice over a period of time to a group of individuals. • The scores are tabulated and the correlation calculated • Correlation is higher then there will be more reliability
  • 6.
    PARALLEL FORM RELIABILITY •Different version of an assessment tool to the same group of individuals • Both version must contain same construct, skill ,knowledge etc. • The number of items in both the forms should be the same. • Distribution of the indexes of difficulty of the items in both should be similar • Means and SD of both the forms should be equal or nearly so. • Mode od administration and scoring of both the forms should be uniform
  • 7.
    SPLIT HALF RELIABILITY •“Splitting in half” all items of a test that are intended to probe the same area of knowledge in order to form two “sets” of items • Items are divided into two halves and then compared. Odd, even items, or 1-50 and 51-100 are two ways to split items. • Product moment correlation is computed between two sets of scores. • Spearman-Brown prophecy formula is used for estimating the reliability of the whole test
  • 8.
    FACTORS INFLUENCING RELIABIITY ExtrinsicFactors • Time interval between testing: testing with shorter intervals resulting in the higher reliability coefficient • Guessing by the examinee • Condition under which measure: Favorable conditions will give high reliability coefficient Intrinsic Factors • Length of test: The longer the test, higher the reliability due to the fact that longer test will provide more adequate sample of the behavior being measured • Difficulty level of test: tests that are too easy or too difficult for the group members taking it will tend to provide score of low reliability • Discrimination value:When the test is composed of discriminating items,the items likely to be high • Homogeneity of items:More homogenous more reliabilty
  • 9.
    • Indicate thedegree to which an instrument measures what it is supposed to measure • The degree to which a study accurately assesses the specific concept or parameter that the researcher is trying to measure • Whether the researcher has measured what it set out to measure • For a test to be reliable it also need to be valid
  • 10.
    INTERNAL VALIDITY • Contentor curricular Validity • Construct validity • Criterion-related Validity a- Predictive validity b-Concurrent validity EXTERNAL VALIDITY The extent to which the results of a study are generalizable or transferable
  • 11.
    CONTENT VALIDITY • Involvesassessing whether the questions asked and the measurement recorded adequately represent the domains being studied • Is the questionnaire comprehensive? • Entire range or universe of the construct is measured • Usually evaluated and scored by experts in the content area
  • 12.
    CONSTRUCT VALIDITY • Itimplies using the construct correctly • The most complex type of validity, and involves relating an instrument for data collection to a theoretical framework • For example, a researcher inventing a new IQ test might spend a great deal of time attempting to define intelligence in order to reach an acceptable level of construct validity.
  • 13.
    CRITERIAN –RELATED VALIDITY •Used to product future or current performance • Demonstrating the accuracy of a measure or procedure by comparing it with another procedure which has already been demonstrated to be valid • It correlates test results with another criterion of interest
  • 14.
    FACTORS AFFECTING VALIDITY •Range of ability-Like reliability, validity is also influenced by the range of ability of the sample used. If the subjects have a very limited range of ability, the validiy coefficient will be low • Ambiguous direction: If the directions of the test are ambiguous ,the validity of the test would be lowered • Socio-cultural differences: Cultural differences among different societies are likely to affect the validity of a test. However ,when a test is cross-cultural ,this factor does not affect the validity of the test • Addition of Inappropriate Items: Inappropriate items are likely to lower both reliability and validity of the test • Length of the test: Homogenous lengthening of the test not only increses the reliability but also validity of the test
  • 15.
    CONT……. Environmental factors: Someenvironmental factors such as room temperature, lightning and noise can influence the error rate Research Factors: The researcher can influence the result of the study in many ways Instrumental factors: An inadequate sampling of items is the other common source of instrumentation errors Subject factors: Any changing physical, emotional or psychological state of the subject could introduce error into measurement process
  • 16.
    RELATIONSHIP BETWEEN RELIABILITYAND VALIDITY Validity and reliability are closely related • A test cannot be considered valid unless the measurement resulting from it are reliable • Likewise result from a test can be reliable and not necessarily valid