Lecture 3: Reliability and Validity - PowerPoint PPT Presentation

1 / 7
About This Presentation
Title:

Lecture 3: Reliability and Validity

Description:

Measuring something in a consistent and stable manner ... consistency: the consistency of performance among the items that compose a scale ... – PowerPoint PPT presentation

Number of Views:95
Avg rating:3.0/5.0
Slides: 8
Provided by: hyoso
Category:

less

Transcript and Presenter's Notes

Title: Lecture 3: Reliability and Validity


1
Lecture 3Reliability and Validity
  • Reliability
  • Validity

2
Question How reliable?
  • Measuring something in a consistent and stable
    manner
  • A measure is reliable if it consistently gives
    the same answer
  • X T E.
  • - X Total score (Observed measurement
    score)
  • T True score Systemic score Reliable
  • E Random error
  • Reliability The ratio of T to X
  • 0 lt Reliability coefficient lt 1.0

3
  • Reliability consists of 3 different dimensions
  • - stability the consistency of a result or of a
    measure at different points in time.
  • Ex) test-retest
  • internal consistency the consistency of
    performance among the items that compose a scale
  • Ex) split-half technique, Cronbachs alpha
  • equivalency the consistency between two parallel
    forms of a test
  • Ex) intercoder reliability

4
Question How valid?
  • Measuring what researchers intend to measure
  • Measurement Equations (V Valid I Invalid).
  • X T E.
  • T V I.
  • X V I E.

5
Types of Validity
  • Judgment-based Face validity (whether it
    measures what it appears to measure -gt very
    subjective, not very informative)
  • Criterion-based
  • - Predictive validity (measurement instrument
    against some future outcome)
  • - Concurrent validity (measurement instrument
    against some present criterion)
  • Theory-based Construct validity (whether the
    measurement is logically related to other
    concepts in the framework)

6
  • Internal Validity
  • Threats to internal validity (i.e., artifacts)
    history, maturation, testing, instrumentation,
    statistical regression, experimental
    morality(text, pp. 27-32)
  • External Validity how well the results of a
    study can be generalized across populations,
    settings, and time
  • A study that lacks external validity cannot be
    projected to other situations it is valid only
    for the sample tested.

7
Relationship between Reliability and Validity
  • One can assume reliability if validity can be
    established.
  • If a measurement is unreliable, the measurement
    cannot be valid.
  • Reliability is necessary but not sufficient for
    validity.
Write a Comment
User Comments (0)
About PowerShow.com