Skip to main content

Table 1 Observations and parameters by group and actual treatment for the 2-stage randomised design

From: Estimation of treatment preference effects in clinical trials when some participants are indifferent to treatment choice

Actual treatment

 

Choice arm (Total sample size = m)

Random arm (Total sample size = n)

  

Choose A

Choose B

Undecided (and randomised)

A

Sample size

m 1

—

m 3 /2

n 1

Observations

x 11 ,…x 1,m1

 

v 11 ,…v 1,m3/2

y 11 ,… y 1,n1

Mean

μ 11

(μ 12 )

μ 13

μ 1

SD

σ 11

 

σ 13

σ 1

B

Sample size

—

m 2

m 3 /2

n 2

Observation

 

x 21 ,…x 2,m2

v 21 ,…v 2,m3/2

y 21 ,…y 2,n2

Mean

(μ 21 )

μ 22

μ 23

μ 2

SD

 

σ 22

σ 23

σ 2

  1. μ12 and μ21 are not observable