Ch 5

Validity Definition: Validity refers to the agreement between a test score or measure and the quality it is believed to measure. In simpler terms, validity assesses how accurately a test measures the concept it is intended to measure.
Historical Context: The definition of validity has undergone several revisions over time, leading to the current understanding that encompasses three types of validity evidence:
1. Construct-related
2. Criterion-related
3. Content-related
Types of Validity Evidence: Validity exists within overlapping categories, including but not limited to, face validity, predictive validity, and criterion validity.

Definition: Face validity refers to the extent to which a measure appears to assess what it is intended to measure based solely on appearance rather than statistical evidence.
Assessment: It involves subjective judgment—whether the items on a measure seem reasonably connected to the perceived purpose of the test. This type of validity does not rely on complex calculations but rather on the intuitive “feel” of the test.

Instructions: Participants provide feedback based on their feelings toward themselves, rating statements such as:
1. On the whole, I am satisfied with myself.
2. At times, I think I am no good at all.
3. I feel that I have a number of good qualities.
- The assessment questions are judged against the criteria of self-esteem to evaluate their relevance.

Logical Quality: Like face validity, content-related validity is judged based on logical consistency rather than statistical measures.
Construct Underrepresentation: Occurs when the test items do not adequately cover all aspects of the construct being assessed.
Construct-Irrelevant Variance: Refers to extraneous factors that may influence test outcomes other than the intended measurement.
Example: If students study specific topics and find their exams frequently diverge from those topics, this would indicate poor content validity.

Definition: Criterion validity refers to how well a test correlates with a specific criterion or outcome. This form of validity emphasizes the predictive relationship.
High Correlation: Criterion validity is typically measured through correlation coefficients, with higher correlations indicating stronger validity.
Example: A premarital test predicting marital success; although we cannot measure success at the time of testing, we can infer it based on past data.

Predictive Validity
- Definition: Predictive validity assesses how effectively a test predicts future events or behaviors. For example, does the SAT accurately predict future college success?
- Correlational Metric: If the SAT correlates highly with future college GPA, it demonstrates strong predictive validity.
Concurrent Validity
- Definition: Concurrent validity assesses the relationship between the test and criterion when both are measured simultaneously.
- Assessment Method: Developing a new depression scale and administering it alongside an existing scale allows for comparison and validation.

Definition: The validity coefficient represents the degree of correlation between a test and the related criterion, indicating the extent to which the test is valid for making statements about the criterion.
Acceptable Range: A validity coefficient generally ranges between 0.30 and 0.40 for adequacy, seldom exceeding 0.60.
Coefficient of Determination: It reflects the proportion of variance in the criterion that can be explained by the test score.

Definition: Construct-related evidence is critical in psychology due to the ambiguous nature of many psychological constructs (e.g., intelligence).
Measurement Challenge: Constructs often lack clear definitions, necessitating evidence assembly to clarify what a test measures.
- Convergent Validity: This occurs when two tests assess the same construct and show high correlation. It indicates that multiple measures are honing in on the same quality.
- Discriminant Validity: Conversely, it ensures that measures do not correlate with unrelated constructs, upholding the distinctness of unrelated qualities.

Interdependence: A test cannot be valid if it is not reliable; thus, reliability and validity are intrinsically linked concepts. A test’s performance must correlate positively with itself, distinct from any other variable.
Implications: Validity coefficients may exhibit variability impacted by reliability factors—constantly needing reassessment across different contexts and populations.

The study of validity emphasizes the need for rigorous evidence-grounded assessments in psychological testing. This chapter serves as a foundational overview of the crucial validity types, providing students and practitioners with guiding principles for evaluating psychological measures.