SPSS Results Interpretation: Competency Domains and Partial η²

SPSS Results Interpretation: Competency Domains

• Scientific Methods

  • Domain: Scientific Methods
  • Comparison: Clinical PhD > Clinical PsyD
  • Partial η²:  = ^{2}_{p} = 0.13
  • Interpretation: About 13% of the variance in ratings is explained by the predictor (after controlling for other variables). Described as a nearly large effect in the context of typical benchmarks.

• Life Sciences (Biology)

  • Domain: Life Sciences (Biology)
  • Comparison: Clinical PhD > Counseling PhD
  • Partial η²: ^{2}_{p} = 0.10
  • Interpretation: About 10% of the variance in ratings is explained by the predictor (after controlling for other variables). Considered a moderate effect.

• Psychopathology (Abnormal)

  • Domain: Psychopathology (Abnormal)
  • Comparison: Clinical PhD > Counseling PhD
  • Partial η²: ^{2}_{p} = 0.13
  • Interpretation: About 13% of the variance in ratings is explained by the predictor (after controlling for other variables). Considered a nearly large effect.

Partial η² Cheat Sheet

• Partial η² (eta squared) definition: tells you the proportion of variance in the dependent variable that is explained by your independent variable, after controlling for other variables in the model.

• Guidelines (Cohen, 1988; general benchmarks):

  • Small effect:  ≈ 0.01 (≈ 1% of variance explained)
  • Medium effect:  ≈ 0.06 (≈ 6% of variance explained)
  • Large effect:  ≈ 0.14 (≈ 14% of variance explained)

• Interpretation tip:

  • Use p-values to tell you IF a difference is statistically significant.
  • Use partial η² to tell you HOW BIG that difference is in practical terms.

• Examples from results above:

  • η²_p = 0.10 → 10% variance explained, moderate effect (Life Sciences domain).
  • η²_p = 0.13 → 13% variance explained, nearly large effect (Scientific Methods, Psychopathology).

Formula and Conceptual Notes

• Conceptual definition: Partial η² represents the proportion of total variance in the dependent variable that is attributable to a specific effect, after accounting for other effects in the model.

• Formula (definition):
  \eta^{2}{p} = \frac{SS{effect}}{SS{effect} + SS{error}}

  • Where:
  • SS_{effect} = Sum of squares for the effect of interest
  • SS_{error} = Sum of squares for error (residual) after accounting for other effects

• Practical interpretation:

  • It is a measure of effect size in ANOVA-like analyses that controls for other predictors.
  • Different from partial omega squared or between-study measures; ensure you specify the model structure when reporting.

Worked Examples from the Transcript

• Life Sciences (Biology):

  • Partial η²: ^{2}_{p} = 0.10
  • Interpretation: 10% of variance in ratings explained by the predictor; considered a moderate effect.

• Scientific Methods:

  • Partial η²: ^{2}_{p} = 0.13
  • Interpretation: 13% of variance explained by the predictor; described as nearly large effect.

• Psychopathology (Abnormal):

  • Partial η²: ^{2}_{p} = 0.13
  • Interpretation: 13% of variance explained by the predictor; described as nearly large effect.

Practical Implications and Reporting Guidance

• When reporting results, present both statistical significance (p-value) and practical significance (partial η²).

  • Example phrasing: "Clinical PhD significantly differs from Counseling PhD on the competency domain, F(df1, df2) = value, p = value, η²_p = 0.13." (replace with actual F and p values as available)

• Context matters:

  • A small p-value with a small η²_p may indicate a statistically reliable but practically modest difference.
  • A larger η²_p (e.g., ≥ 0.10) suggests a meaningful difference in real-world terms, especially in applied settings like clinical domains.

• Reporting conventions to adopt:

  • Always specify the domain, the cohorts/comparisons, and the effect size with its interpretation.
  • Use the benchmark guidelines to label the magnitude, and cite that they come from Cohen (1988) for clarity.

• Real-world relevance:

  • The results suggest that degree of competency differences between PhD and PsyD/Counseling pathways vary by domain, with life sciences showing a moderate effect and scientific methods/psychopathology showing a nearly large effect, which may inform training emphasis and program evaluation.

Quick Reference Summary

• Partial η² tells the practical size of an effect after controlling for other variables.
• Benchmarks: ^{2}_{p} \approx 0.01\ (small),\ 0.06\ (medium),\ 0.14\ (large)
• Report both p-values and η²_p to convey statistical and practical significance.
• Example results:
  • Life Sciences: ^{2}_{p} = 0.10 → moderate effect.
  • Scientific Methods: ^{2}_{p} = 0.13 → nearly large effect.
  • Psychopathology: ^{2}_{p} = 0.13 → nearly large effect.