Lecture 8 - Evaluating Questionnaires Flashcards
What does the observed score in measurement consist of?
Observed Score = True Score + Error.
What are the assumptions for psychological measurement?
- Psychological variables can be measured.
- Individual differences are meaningful and exist across people, groups, and time.
Define unidimensional and multidimensional scales.
Unidimensional Scale: Measures one construct/domain.
Multidimensional Scale: Includes multiple constructs/domains, often identified using factor analysis.
What is factor analysis used for in evaluating questionnaires?
To confirm whether a questionnaire measures one or multiple factors (Exploratory Factor Analysis for identifying factors, Confirmatory Factor Analysis for validating expected factors).
List types of validity in questionnaire evaluation.
Construct Validity
Face Validity
Content Validity
Convergent Validity
Discriminant Validity
Internal Validity
External Validity
What are the key types of reliability?
Inter-rater Reliability
Test-retest Reliability
Internal Consistency Reliability (including Cronbach’s Alpha).
What are the strengths of self-report measures?
Cost-effective
Quick data collection
No researcher presence needed
Easy to organize for analysis
Allows classification of individuals.
Name some biases associated with self-report methods.
Interviewer/Researcher Bias
Social Desirability Bias
Acquiescence Bias
Question Wording and Ordering Effects.
How can the quality of self-report data be maximized?
Avoid asking about distant events.
Ensure anonymity and transparency.
Use both positive and negative items with reverse scoring.
Include attention checks and criteria for case removal.
What is Cronbach’s Alpha?
A measure of internal consistency that calculates the average of all possible split-half reliabilities, indicating how strongly items in a scale correlate.
Why is piloting a newly developed measure important?
To identify errors, test feasibility, refine the questionnaire, and assess face validity through feedback.
How do positive and negative items in a questionnaire work?
Positive items align with the construct.
Negative items are reverse-coded to align scoring directionally with positive items.
This prevents response bias and enhances reliability.
What is test-retest reliability, and how is it measured?
Measures consistency of scores over time using Pearson’s Correlation or Intraclass Correlation Coefficient (ICC), with 0.75+ considered desirable and 0.90+ ideal for clinical measures.
