Skip to main content

Table 2 Sample size calculations

From: The Active for Life Year 5 (AFLY5) school based cluster randomised controlled trial: study protocol for a randomized controlled trial

Outcome

ICC (95%CI)

Alpha

(two-sided)

Power

Minimal effect detectablea

Total number schools (pupils) required

Time spent in MVPA per dayb

0.07

(0.00, 0.21)

5%

80%

0.25 SD

51 (1274)

Time spent sedentary per dayb

0.02

(0.00, 0.11)

5%

80%

0.25 SD

51 (1272)

Fruit and vegetable portions per dayc

0.04

(0.00, 0.09)

5%

85%

0.25 SD

51 (1268)

Secondary outcomes

Self-report time spent screen-viewing

0.00

(0.00, 0.03)

1%

90%

0.25 SD

40 (1000)

Snacks portions per dayc

0.03

(0.00, 0.07)

1%

90%

0.30 SD

50 (1254)

High fat food portions per dayc

0.05

(0.00, 0.10)

1%

85%

0.30 SD

51 (1274)

High energy drink portions per dayc

0.02

(0.00, 0.05)

1%

90%

0.30 SD

43 (1070)

Mean BMI

0.00

(0.00, 0.02)

1%

90%

0.30 SD

32 (794)

Mean Waist

0.05

(0.03, 0.20)

1%

80%

0.30 SD

51 (1274)

% overweight or obese (by BMI)

0.03

(0.01, 0.05)

1%

55%

20%d

51 (1274)

% centrally obese (by Waist)

0.06

(0.02, 0.09)

1%

60%

20%d

51 (1274)

% overweight or obese (by BMI)

0.03

(0.01, 0.05)

1%

80%

30%e

51 (1274)

% centrally obese (by Waist)

0.06

(0.02, 0.09)

1%

82%

30%e

51 (1274)

  1. a Difference in mean continuously measured outcomes and relative % difference for binary outcomes; b Assessed by accelerometer; c For these outcomes the minimal effect detectable is calculated using SD differences on the log-scale; d Relative difference (i.e. ability to detect an odds ratio of at least 0.8 or 1.2; e Relative difference of larger magnitude (i.e. ability to detect an odds ratio of at least 0.7 or 1.3)
  2. SD: standard deviation; Calculations assume similar numbers of schools randomised to control or intervention and ~ 25 pupils per school; In all calculations we have used the ICC upper 95% CI value