Exam 2: Psychological Assessment

Validity

Measuring what you think.

Reliability

How close to the true score are we likely to be.

Practicality

Does it make sense to apply it to this setting? Is it worth it? Related to utility.

Cross-Sectional Fairness

Is it accurate for someone in this group? Absence of test or assessor biases.

Correlation

Is the degree of the relationship between 2 variables.

No Correlation

No relationship; no pattern, random, no meaningful predictors.

Construct Validity

Asks if a quality is measurable and if it is an accurate measure.

Criterion Related Validity

Relationships between scores and other measures; do the scores predict performance on a criterion?

Concurrent Validity

Can predict score now; measures scores against established tests.

Predictive Validity

Can predict future scores or performance based on current measures.

Hit Rate

Hits/(hits + misses); measures prediction accuracy.

False Positives

Predict high/pass/trait but not accurate.

False Negatives

Predict absence of something but it's actually present.

Reliability Coefficient

A stat that quantifies reliability, ranging from 0 (not reliable) to 1 (reliable).

Test Retest

Correlates scores from the same test given at different times.

Alternate Forms

Same content tapped differently but equally; assess score correlation between versions.

Interscorer Reliability

Same people assessed by different administrators.

Split Half Reliability

Divide test into two halves and correlate scores.

Coefficient Alpha

Mean of all possible split halves.

Inter-Item Consistency

Degree of correlation among all items in a test.

Convergent Validity

Scores highly correlate with other tests as expected.

Discrimination Validity

Scores show little or no relationship with irrelevant measures.

Negative Correlation

An inverse relationship between two variables.

Positive Correlation

A direct relationship where higher values of one variable relate to higher values of another.

Strong Correlation

The closer to -1 or +1, the stronger the relationship, better prediction accuracy.

Weak Correlation

Scores closer to 0 indicate a weak relationship, making predictions less accurate.