Lakens

library(metafor)

g <- escalc(
  measure = "SMD",
  n1i = 43, # sample size in group 1 is 50
  m1i = 1.3, # observed mean in group 1 is 5.6
  sd1i = 1, # observed standard deviation in group 1 is 1.2
  n2i = 43, # sample size in group 2 is 53
  m2i = 1, # observed mean in group 1 is 4.9
  sd2i = 1

Lakens

library(segHT)
library(rpact)

looks <- 3
n_seg <- 50
alpha_level <- 0.05 
true_d <- 0.5 # can not enter 0, segmented_hyp_test_outcomes gives error

###############################
# Segmented procedure ----

Lakens

---
title: "Power Analysis for Interactions"
author: "Daniel Lakens"
date: "28-3-2020"
output:
  html_document: default
  pdf_document: default
  word_document: default
editor_options:
  chunk_output_type: console

Lakens

# Scienceverse Sim

# install scienceverse
# devtools::install_github("scienceverse/scienceverse")

library(scienceverse)
library(faux)

set.seed(2) # set.seed(2) is a random draw where H1 is corroborated.

Lakens

# Scienceverse Sim

# install scienceverse
# devtools::install_github("scienceverse/scienceverse")

library(scienceverse)
library(faux)

set.seed(2) # set.seed(2) is a random draw where H1 is corroborated.

Lakens

n<-2000 #total number of datapoints (per condition) you are willing to collect after initial 10

D<-0.0 #True effect size (Keep SD below to 1, otherwise, this is just mean dif, not d)
SD<-1 #Set True standard deviation.

p<-numeric(n) #store p-values
x<-numeric(n) #store x-values
y<-numeric(n) #store y-values

n<-n+10 #script calculates p-values after 10 people in each condition, so add 10 to number of datapoints

Lakens

---
title: "Observed Alpha Levels"
output:
  word_document: default
  html_document:
    df_print: paged
editor_options:
  chunk_output_type: console
---

Lakens

# Code by Chelsea Parlett, small additions by Daniel Lakens
library(pwr)
library(ggplot2)

#set up vector of effect sizes
es <- seq(0.01,1, length = 1000)
#specify power for test 
pow <- 0.8

#calculate ns (sample size for 80% power in two-sided test)

Lakens

#Install pwr package if needed
if(!require(pwr)){install.packages('pwr')}
library(pwr)

alpha_level = 0.05 #set alpha level
n = 100 #set number of observations
st_dev = 1 #set true standard deviation
SESOI <- 0.5 #set smallest effect size of interest (raw mean difference)

# calculate lower and upper critical values c_l and c_u

Lakens

optimal_alpha <- function(power_function, costT1T2 = 1, prior_H1H0 = 1, error = "minimal") {
  #Define the function to be minimized
  f = function(x, power_function, costT1T2 = 1, prior_H1H0 = 1, error = "minimal") {
    y <- 1 - eval(parse(text=paste(power_function)))
    print(c(x, y, x+y)) #optional: print alpha, beta, and objective
    if(error == "balance"){
      max((costT1T2*x - prior_H1H0*y)/(prior_H1H0+1), (prior_H1H0*y - costT1T2*x)/(prior_H1H0+1))
    } else if (error == "minimal"){
      2*(costT1T2*x + prior_H1H0*y)/(prior_H1H0+1)
    }