Flashcards
Research Basics - Measurement, Reliability and Validity
Cards in this deck (12)
- 1LEVELS OF MEASUREMENT1) Nominal - assignment of labels
2) Ordinal - assignment of values along some underlying dimension
3) Interval - assignment of values along some underlying dimension, with equal distances between points
4) Ratio - assignment of values along some underlying dimension, with equal distances between points, with a meaningful non-arbitrary zero - 2RELIABILITY- When a test measures the same thing more than once and results in the same outcome
- Reliability = True Score
True Score + Error Score
- Measured by Correlation Coefficient - 31) OBSERVED SCORE 2) TRUE SCORE 3) ERROR SCORE1) Score you record or observe
2) Perfect reflection of the true value of that variable, given no other internal/external influences
3) All the reasons the true and observed score differ - 41) METHOD ERROR 2) TRAIT ERROR1) Due to the testing situation
2) characteristics of the person taking the test - 5RELIABILITY COEFFICIENT- r = correlation
- The more similar the scores in terms of change from one time to another, the higher the correlation and the higher the reliability
- Range in value from -1.00 to +1.00.
A value of 1.00 would be perfect reliability. - 6TYPES OF RELIABILITY1) Test-Retest Reliability
2) Parallel forms of Reliability
3) Inter-rater Reliability
4) Internal Consistency - 7TEST-RETEST RELIABILITYA measure of how stable a test is over time
- 8PARALLEL FORMS RELIABILITYDifferent forms of the same test are given to the same group of participants, and then correlated with each other.
- 9INTER-RATER RELIABILITYA measure of the consistency from rater to rater
- 10INTERNAL CONSISTANCYExamines how unified the items are in a test.
- 11VALIDITYWhether the test or instrument actually measures what you need to have measured.
- 12TYPES OF VALIDITY1) Content Validity - the extent to which a test represents the universe of items from which it is drawn
2) Criterion Validity - How well a test estimates present performance (concurrent validity) or predicts future performance (predictive validity)
3) Construct Validity - The extent to which the results of a test are related to an underlying psychological construct.
Information
- Description
- Flashcards on Research Basics - Measurement, Reliability and Validity, created by Rachel Vince on 25/02/2017.
- Created by
- Rachel Vince
- Views
- 0
- Rating