debruine

library(faux)

within <- list(
  pet = c("dog" = "Has a dog", "cat" = "Has a cat"),
  time = c("day" = "Tested during the day", "night" = "Tested at night")
)
wcells <- faux:::cell_combos(within)

between <- list(
  country = c("UK" = "United Kingdom", "NL" = "Nederland"),

debruine

library(faux)
library(tidyverse)

inatt <- function(within, between, n, mu, sd, r, prop_bad = 0) {
  good_n <- n * (1-prop_bad)
  bad_n <- n * prop_bad
  good_dat <- sim_design(within, between, n = good_n, mu, sd, r)
  if (bad_n > 0) {
    # should r be 0 for inattentive responders?
    bad_dat <- sim_design(within, between, n = bad_n, unlist(mu)%>% mean(), sd, r) %>%

debruine

#devtools::install_github("debruine/faux")
library(faux)

# make unique pairs of level names to ask for cors
unique_pairs <- function(v) {
  if (is.numeric(v)) { v <- LETTERS[1:v] }
  
  expand.grid(a = v, b = v) %>% 
    filter(a != b) %>% t() %>% 
    tibble::as_tibble() %>% 

debruine

#' Check readline input
#'
#' @param prompt the prompt for readline
#' @param type what type of check to perform, one of c("numeric", "character", "length", "minlength", "maxlength", "exact", "grep")
#' @param compare the comparator for exact and (min|max)length types
#' @param ... other arguments to pass to grep
#'
#' @return the validated result of readline
#' @export
#'

debruine

library(faux)
library(tidyverse)

data <- sim_design(within = list(T = c("T1", "T2", "T3", "T4", "T5")),
                   id = "ID")

processed_data <- data %>%
  gather(T, raw, T1:T5) %>% # put in long format
  group_by(ID) %>% # start doing calculations within groups
  mutate(Mean_T = mean(raw),

debruine

#devtools::install_github("debruine/faux")
library(faux)
library(afex)
library(tidyverse)

## Null data (no effect)

des_null <- check_design(list(task = c("test", "white", "asian")), 
                         n = 400, 
                         mu = c(100, 100, 90), 

debruine

#devtools::install_github("debruine/faux")
library(faux)

# compare the correlations between simulated normally distributed data and the likert-scale version
check_likert <- function(n = 100, r = 0.5, 
                         prob = c(.05, .1, .2, .3, .2, .1, .05)) {
  df <- faux::rnorm_multi(n, 2, 0, 1, r) %>%
    dplyr::mutate(
      L1 = norm2likert(X1, prob, 0, 1),
      L2 = norm2likert(X2, prob, 0, 1)

debruine

---
title: 'Example'
output: html_document
---



```{r, message=FALSE}
# load required packages
library("lme4")        # model specification / estimation

debruine

library(faux)
library(tidyverse)

sub_sd <- 1 # subject intercept SD
stim_sd <- 1 # stimulus intercept SD
err_sd <- 2 # residual error SD
grand.i <- 600 # grand interecpt (mean RT)
ord_eff <- 0.5 # coefficient of order effect
sch_eff <- 0.5 # coefficient of schizotypy effect
ord_sch_ixn <- 0.5 # coefficient of order*schizotypy interaction

debruine

a <- rep(c(1, 2), each = 4)
b <- c(1,2,3,4,5,6,7,8)
frame <- data.frame(a,b)

library(dplyr)
library(tidyr)

x <- frame %>%
  group_by(a) %>%
  summarise(Mean = mean(b),