{{ message }}

Intersubject Variance using a 2 by 2\(m\) Crossover Design

Using the notation of Chow et al., \(\sigma^2_{BR}\) and \(\sigma^2_{BT}\) are the inter-subject variance for treatments \(R\) and \(T\) respectively. This calculator performs power and sample size calculatios for the two-sided hypothesis: \[ H_0: \frac{\sigma^2_{BT}}{\sigma^2_{BR}} = 1 \text{ versus } H_a: \frac{\sigma^2_{BT}}{\sigma^2_{BR}} \ne 1 \]

or the one-sided hypthesis:

\[ H_0: \frac{\sigma^2_{BT}}{\sigma^2_{BR}} \ge \delta \text{ versus } H_a: \frac{\sigma^2_{BT}}{\sigma^2_{BR}} \lt \delta \]

or the hypothesis of similarity:

\[ H_0: \frac{\sigma^2_{BT}}{\sigma^2_{BR}} \ge \delta \text{ or } \frac{\sigma^2_{BT}}{\sigma^2_{BR}} \le \frac{1}{\delta} \text{ versus } H_a: \frac{1}{\delta} \lt \frac{\sigma^2_{BT}}{\sigma^2_{BR}} \lt \delta \]

where \(\delta\) is the similarity limit using a parallel design with replicates. The model used to test this hypothesis is explained in more depth here.

Power Calculation Parameters

To input multiple values, seperate them by a comma.

{{ this.errors }}

Power Calculation Explanation

Solve For
The unknown you are interested in solving for.
N
The sample size used to test the hypothesis.
Alpha
The \(\alpha\) (Type I error rate) level of the hypothesis test.
Power
The power (1 - Type II error rate) of the hypothesis test.
Replications
The number of replications in the study (\(m\))
\(\sigma_{BT}\)
The intra-subject variability for the test treatment, \(T\)
\(\sigma_{WT}\)
The inter-subject variability for the test treatment, \(T\)
\(\sigma_{BR}\)
The intra-subject variability for the reference treatment, \(R\)
\(\sigma_{WR}\)
The inter-subject variability for the reference treatment, \(R\)
Similarity Limit \(\delta\)
The similarity limit against which the ratio of variabilities is tested.
\(\rho\)
An element of the covariance between \(S_{ijT}\) and \(S_{ijR}\), the random effects of the crossover model (details). If this is unknown, leave it 1.
Hypothesis
There are three types of hypotheses that can be tested: two-sided, one-sided and equivalence. Tests of equivalence must include a margin if the unknown and null means are equal. Read More

Calculation Results

No calculation has been generated yet.

Sign Up to Save Power and Sample Size Calculators!
N Replications Alpha Power \(\sigma_{BT}\) \(\sigma_{WT}\) \(\sigma_{BR}\) \(\sigma_{WR}\) \(\rho\) Similarity Limit
{{ val }}

Power Graph

There are no results to graph yet.

References

  • Chow, S., Shao, J., & Wang, H. (2003), Sample size calculations in clinical research, New York: Marcel Dekker.