Chapter 5 - Measurement Concepts Flashcards
True Score
Individuals actual/real score on a variable being measured.
Measurement Error
The Degree to which a measurement deviates from the true score value.
Shown in the greater variability of unreliable tests.
Pearson product-moment correlation coefficient
Symbolized as r, can range from 0.00 to +1.00 and 0.00 to -1.00
- 00 = no relation
- 1.00 & +1.00 = strong relationship
Test-Retest Reliability
Measuring the same individuals at two points in time
Example: IQ test at beginning and end of semester.
Alternate Forms Reliability
Two different forms of the same test to the same individuals at two points in time.
To avoid individuals remembering how they responded the first time.
Internal Consistency Reliability
The assessment of reliability using responses (data) collected at only one point in time with multiple measures.
Measures are reliable when multiple measures provide similar results.
Split-half reliability
Correlation of the total score on one half of the test compared to the total score on the other half of the same test.
Created randomly by dividing two parts.
Chronbach’s alpha
Provides the average of all possible split-half reliability coefficients.
Would want to do this on a computer.
Interrater reliability
The extent to which eaters agree in their observations.
It’s high when most of the observations result in the same judgement.
Construct validity
Concerns whether out methods of studying variables are accurate.
Face validity
Content of the measure appears to reflect the construct being measured.
Example: measure for depression including items like “I feel sad”
Content validity
Content of the measure is linked to the universe of content that defines the construct.
Predictive validity
Scores on the measure predict behavior on a criterion measured at a future time.
Example: GRE or SATs for college/graduate programs.
Concurrent validity
Scores on the measure are related to a criterion measured at the same time (concurrently).
Ex: If two groups were given measures and varied in predictable ways (those in therapy for anxiety scored higher than those with depression).
Convergent validity
Scores on the measure are related to other measures of the same construct.
Ex: if scores from the new measure were related to scores from other measures
Discrimination validity
Scores on the measure are not related to other measures that are theoretically different.
Reactivity
If awareness of being measured changes an individual’s behavior.
Ways to minimize this include allowing time for individuals to become used to the presence of the observer or recording equipment.
Nominal scales
Categories with no numeric scales. Impossible to define any quantities values.
Ex: Males/females
Ordinal scales
Rank ordering numeric values limited. Intervals between items is not known.
Ex: 2, 3, 4 - star restaurants
Interval scales
Numeric properties are literal. Assume equal interval between values. No true zero.
Ex: aptitude test score, intelligence, temperature.
Ratio
Zero indicates a sense of variable measured. Can form ratios.
Ex: reaction time, weight, age.
Reliability
The consistency or stability of a measure of behavior.
Most likely achieved when researchers use careful measurement procedures.
Increased by making multiple measures.
