CSAM Measurement & Treatment Outcomes 4 — Questions and Answers
Question 1: A treatment program for domestic violence offenders reports anger reduction on self-report measures but collateral partner reports show no change. This discrepancy most likely reflects:
- Partners are unreliable informants in high-conflict relationships
- Possible response bias on self-report due to legal or program incentives (Correct answer)
- Regression to the mean affecting partner perception
- Instrument sensitivity differences between scales
Correct answer: Possible response bias on self-report due to legal or program incentives
Mandated or legally involved clients may underreport anger on self-administered measures to appear compliant, while actual behavior observed by partners remains unchanged.
Question 2: Which statistical approach is most appropriate for analyzing repeated anger measurements (baseline, mid-treatment, post-treatment, follow-up) in a group outcome study?
- Independent samples t-test at each time point
- Mixed-model repeated measures ANOVA or multilevel modeling (Correct answer)
- Pearson correlation between first and last measurement
- Chi-square test of proportions improved
Correct answer: Mixed-model repeated measures ANOVA or multilevel modeling
Mixed-model repeated measures ANOVA or multilevel modeling accounts for within-person correlation across measurement occasions and handles missing data more appropriately.
Question 3: When selecting an anger outcome measure for use in a correctional setting, the most critical psychometric property to verify is:
- Face validity so items seem relevant to inmates
- Construct validity with demonstrated sensitivity to treatment-induced change (Correct answer)
- Test-retest reliability over a 1-year interval
- Factor structure consistent with a two-factor anger model
Correct answer: Construct validity with demonstrated sensitivity to treatment-induced change
Treatment sensitivity (responsiveness) ensures the instrument can detect the changes expected from intervention, which is essential for evaluating program effectiveness.
Question 4: The concept of 'anger management' treatment success should ideally be defined by:
- Elimination of all angry emotional experiences
- Reduction in anger frequency, intensity, and duration plus improved regulation skills (Correct answer)
- Zero incidents of verbal or physical aggression
- Client self-report of feeling 'in control' at program completion
Correct answer: Reduction in anger frequency, intensity, and duration plus improved regulation skills
Clinically, success means anger is less frequent, less intense, shorter in duration, and better regulated—not that anger is eliminated, which is neither realistic nor healthy.
Question 5: A meta-analysis of anger management programs calculates an overall effect size of d = 0.70. According to Cohen's conventions, this indicates a:
- Small effect with limited practical significance
- Medium-to-large effect suggesting meaningful treatment impact (Correct answer)
- Large effect equivalent to complete anger resolution
- Negligible effect not worth clinical investment
Correct answer: Medium-to-large effect suggesting meaningful treatment impact
Cohen's d = 0.70 falls in the medium-to-large range (medium ≈ 0.50, large ≈ 0.80), indicating practically meaningful treatment gains.
Question 6: A CSAM practitioner is evaluating whether an anger management group reduced recidivism in a court-ordered sample. The most appropriate outcome measure for this goal is:
- Post-group anger questionnaire scores
- Official arrest or re-offense records over a 12-month follow-up period (Correct answer)
- Therapist-rated anger severity at termination
- Self-reported satisfaction with the program
Correct answer: Official arrest or re-offense records over a 12-month follow-up period
For criminal justice populations, objective recidivism data from official records provides the most externally valid measure of behavioral change.
Question 7: Which of the following represents a process measure rather than an outcome measure in anger management treatment evaluation?
- 6-month follow-up trait anger score
- Collateral report of aggressive incidents at discharge
- Session-by-session therapeutic alliance ratings (Correct answer)
- Pre-post change on the Novaco Anger Scale
Correct answer: Session-by-session therapeutic alliance ratings
Process measures assess what happens during treatment (e.g., alliance, skill acquisition rate), while outcome measures assess the end result of treatment.
CSAM Measurement & Treatment Outcomes 4
A treatment program for domestic violence offenders reports anger reduction on self-report measures but collateral partner reports show no change.
This discrepancy most likely reflects: