Skip to main content

Table 5 Subscale descriptive statistics and reliability characteristics for modified 23-item knowledge and 23-item engagement scales

From: Validation and refinement of the Stakeholder-driven Community Diffusion Survey for childhood obesity prevention

  # items Mean (SD) α ICC WSCV (95% CI)
Analytic sample size, n - 164 164 14 14
Knowledge
1. Intervention factors 5 3.6 (0.8) 0.87 0.50 0.14 (0.10, 0.21)
2. Roles & resources 7 3.5 (0.7) 0.87 0.86 0.08 (0.05, 0.11)
3. Implementation & sustainability 6 3.5 (0.7) 0.86 0.78 0.09 (0.07, 0.14)
4. Problem 5 4.2 (0.6) 0.87 0.62 0.09 (0.06, 0.13)
Engagement
1. Dialogue & mutual learning 4 3.7 (0.9) 0.88 0.87 0.09 (0.06, 0.13)
2. Flexibility 4 3.9 (0.5) 0.75 0.87 0.07 (0.05, 0.10)
3. Influence & power 4 3.2 (0.9) 0.87 0.95 0.07 (0.05, 0.11)
4. Leadership & stewardship 8 3.8 (0.7) 0.90 0.96 0.03 (0.02, 0.05)
5. Trust & trustworthiness 3 4.0 (0.6) 0.80 0.70 0.08 (0.06, 0.12)
  1. CI confidence interval, ICC intraclass correlation coefficient, SD standard deviation, WSCV within-subject coefficient of variation. ICC estimates were based on multiple observers and two-way mixed-effects models and are commonly interpreted as < 0.40 “poor”; 0.40–0.59 “fair”; 0.60–0.74 “good”; 0.75–1.00 “excellent”. WSCV estimates are interpreted as the average percentage of variation between test and retest scores