BAPSY Psychological Research and Statistics 5 — Questions and Answers
Question 1: Which scale of measurement has an absolute zero point that means the complete absence of the attribute?
- Nominal
- Ordinal
- Interval
- Ratio (Correct answer)
Correct answer: Ratio
Ratio scales (e.g., height, reaction time) have a true zero, allowing meaningful ratio comparisons such as 'twice as long.'
Question 2: A researcher uses a pretest-posttest control group design. Compared to a one-group pretest-posttest design, this controls for:
- Experimenter bias only
- History and maturation threats to internal validity (Correct answer)
- Demand characteristics and social desirability
- All possible confounds simultaneously
Correct answer: History and maturation threats to internal validity
Adding a control group that also experiences the passage of time and outside events controls for history and maturation, which affect both groups equally.
Question 3: Construct validity of a psychological measure refers to:
- Whether the measure predicts a relevant future criterion
- The degree to which the measure actually assesses the theoretical construct it purports to measure (Correct answer)
- Whether different raters score the measure consistently
- How stable scores are across repeated administrations
Correct answer: The degree to which the measure actually assesses the theoretical construct it purports to measure
Construct validity is the extent to which a test measures the psychological construct (e.g., anxiety, intelligence) it was designed to capture.
Question 4: When comparing three or more group means, why is ANOVA preferred over multiple t-tests?
- ANOVA is more sensitive to small effect sizes
- Multiple t-tests are computationally impossible for three groups
- Multiple t-tests inflate the familywise Type I error rate (Correct answer)
- ANOVA can handle non-normal distributions better
Correct answer: Multiple t-tests inflate the familywise Type I error rate
Running multiple t-tests increases the probability of at least one false positive; ANOVA controls familywise error at the nominal alpha level.
Question 5: In observational research, reactivity refers to the problem that:
- Observers may rate behaviors differently over time
- Participants change their behavior when they know they are being observed (Correct answer)
- The dependent variable reacts to changes in the independent variable
- Correlational data cannot establish cause and effect
Correct answer: Participants change their behavior when they know they are being observed
Reactivity (also called the Hawthorne effect) occurs when awareness of being observed alters participants' natural behavior.
Question 6: A standardized score (z-score) of +2.0 means the raw score is:
- Two points above the mean
- Two standard deviations above the mean (Correct answer)
- In the 2nd percentile
- Twice the value of the mean
Correct answer: Two standard deviations above the mean
A z-score expresses distance from the mean in standard deviation units; z = +2.0 indicates the score is two SDs above the mean.
Question 7: External validity is MOST threatened in a study that uses:
- A large random sample from the target population
- Standardized procedures and controlled laboratory conditions
- A convenience sample of undergraduate psychology students (Correct answer)
- Multiple dependent variable measures
Correct answer: A convenience sample of undergraduate psychology students
WEIRD (Western, Educated, Industrialized, Rich, Democratic) convenience samples limit the generalizability of findings to broader populations.
BAPSY Psychological Research and Statistics 5
Which scale of measurement has an absolute zero point that means the complete absence of the attribute?