Scribd
Upload
264 views

Validity and Reliability

This document discusses validity and reliability in measuring psychological constructs. It defines internal reliability as the consistency of items within a measure, which can be assessed through split-half and item analysis methods. External reliability refers to the stability of scores over time and contexts, and is evaluated using test-retest and replication techniques. The document also defines validity as the extent to which a measure accurately reflects the target construct. Validity can be examined through face validity, concurrent validity by comparing to other validated measures, and predictive validity by correlating with future outcomes.

Uploaded by

MissMumtaz
Copyright
© Attribution Non-Commercial (BY-NC)
Available Formats
Download as DOC, PDF, TXT or read online on Scribd
264 views

Validity and Reliability

This document discusses validity and reliability in measuring psychological constructs. It defines internal reliability as the consistency of items within a measure, which can be assessed through split-half and item analysis methods. External reliability refers to the stability of scores over time and contexts, and is evaluated using test-retest and replication techniques. The document also defines validity as the extent to which a measure accurately reflects the target construct. Validity can be examined through face validity, concurrent validity by comparing to other validated measures, and predictive validity by correlating with future outcomes.

Uploaded by

MissMumtaz
Copyright
© Attribution Non-Commercial (BY-NC)
Available Formats
Download as DOC, PDF, TXT or read online on Scribd
You are on page 1/ 2

Validity and reliability

What it is How it can be tested


Internal reliability
This concerns the extent to which a measure is consistent within itself. Eg. items on an IQ test must be measuring IQ not something else (eg. working memory, cultural knowledge). Techniques used to assess internal reliability: Reliability is the extent to which something is consistent / stable. This can be over time or contexts. It can refer to findings or results, or to the measure used to produce those results A reliable measure produces the same findings under the same circumstances.

i) Split half method the test is randomly divided into two so that
each is equivalent. Internal reliability can be assumed if participants scores are similar on both halves (i.e. they show a strong positive correlation)

ii) Item analysis Performance on each item is compared with the


overall score. A good positive correlation suggests high internal reliability

Reliabili ty

External reliability
This concerns the extent to which a measure varies from one time to another. This can be where a person is given a test on one occasion and again a month later. It has good external reliability if the scores are the same on both occasions. Similarly, where two researchers interview the same person and they agree on what they find, this also shows good reliability. It is usually assessed via a correlation between the two scores. The higher the correlation, the better the external reliability. Techniques used to assess external reliability:

a) Test retest the same person is tested twice over a period of


time. Good positive corrections between the two scores indicate high external reliability

b) Replication any research should produce similar findings is


repeated. Validity concerns the extent to which the research measures what it sets out to measure. Techniques used to assess validity include face validity, concurrent validity & predictive validity: i) Face validity - Simply eyeball the test. Does it look as if it measures what it says it measures? This is quite a basic test of validity, but a good starting point. ii) Concurrent validity - Results from a new test are compared with those from an older test that is known to have good validity. - If the tests yield similar effects the new test can be considered to have good validity iii) Predictive validity - A test has good predictive validity if it is able to predict performance on future tests (eg. GCSE scores to predict AS scores)

Validity

Regarding a measure or test of something, its usually about whether the test represents something meaningful.

Nafesa Mumtaz 2012

Nafesa Mumtaz 2012

You might also like