Skip to main content

Table 4 Sample size data

From: Exploring the impact of public health teams on alcohol premises licensing in England and Scotland (ExILEnS): procotol for a mixed methods natural experiment evaluation

  Expected average effect size %/year (slope) Between-slope variance Residual variance model Number of areas in each group
From previous study of impact of licensing on alcohol-related crime [29]
 Crime ratesa −4.00% (−0.04) 0.003 0.03 29
−5.00% (−0.04) 0.003 0.03 19
−6.00% (−0.06) 0.003 0.03 13
From previous study of impact of licensing on alcohol-related hospital admissions [28]
 Rates −2.31% (−0.229) 0.110 0.011 34
Current study - minimum detectable effect size with proposed sample size
 20 areas per group −3.00% (−0.296) 0.110 0.011 20
  1. aEffect size is a range between 4 and 6% as estimated by d Vocht et al. (2016) using quadratic trends