2.2.5. Sample Size
The sample size for the study was calculated from an expected difference in change in insulin
sensitivity of 20% between the two groups at the end of the study, using the mean insulin sensitivity,
M, of a healthy Cameroonian population (M = 14.3 ± 2.2 mg/kg/min) [26]. Choosing α (two-sided) at
0.05 and a statistical power of 80%, the minimum sample size was 10 IR subjects per group. The
calculations are given below [27].
The standardized difference,
d = target difference/standard deviation
d = 2.86/2.2 = 1.3
For two equal groups, the sample size per group is given by the formula below [27]:
n = (2/d2) × Cp,power
where Cp,power is a constant defined by the value chosen for the power and p value, available in
statistical tables. With a power of 80% and a p value of 0.05, the constant is 7.9 [27].
Thus the sample size per group,
n = (2/1.32) × 7.9
n = 9.35 subjects
This number was rounded up to a minimum of 10 insulin-resistant subjects per group.
2.3. Statistical