Load Libraries
library(dplyr)
library(papaja)
Load Data
# Data are in tab--delimited text files created from the original Excel File on OSF at: https://osf.io/pszjz/
s1 <- read.table("https://raw.githubusercontent.com/ndphillips/PsyKo-2017/master/data/jordan_s1.txt", sep = "\t", header = TRUE, stringsAsFactors = FALSE)
s2 <- read.table("https://raw.githubusercontent.com/ndphillips/PsyKo-2017/master/data/jordan_s2.txt", sep = "\t", header = TRUE, stringsAsFactors = FALSE)
Clean Data
s1 <- s1 %>% mutate(
likeall = (likework + likeromantic + likeacademic + likedrugs) / 4,
gender = ifelse(gender == 1, "male", "female"),
goodinfo = ifelse(goodinfo == 1, "Good", "None"),
condemn = ifelse(condemn == 1, "Target", "Other")) %>%
filter(percCorrect == 1 &
is.finite(age)) # As recommended by the main study author in an email
s2 <- s2 %>% mutate(
likeall = (likework + likeromantic + likeacademic + likedrugs) / 4,
gender = ifelse(gender == 1, "male", "female"),
direct = ifelse(direct == 1, "Direct", "Moral"),
signal = ifelse(signal == 1, "Target", "Other")) %>%
filter(percCorrect == 1)
Study 1
# Show structure of study 1 data
# Should have 619 participants
str(s1)
## 'data.frame': 619 obs. of 51 variables:
## $ id : int 1 3 4 5 8 9 11 13 15 16 ...
## $ goodinfo : chr "Good" "Good" "None" "None" ...
## $ condemn : chr "Target" "Target" "Target" "Other" ...
## $ qAnswered : int 16 16 16 16 16 16 16 16 16 16 ...
## $ totalCorrect : int 16 16 16 16 16 16 16 16 16 16 ...
## $ percCorrect : num 1 1 1 1 1 1 1 1 1 1 ...
## $ ansall : int 1 1 1 1 1 1 1 1 1 1 ...
## $ dowork : int 4 2 4 3 2 2 1 4 1 1 ...
## $ doromantic : int 2 1 5 5 2 2 1 4 1 1 ...
## $ doacademic : int 2 5 4 3 3 2 2 4 1 1 ...
## $ dodrugs : int 2 3 3 2 3 2 1 5 2 1 ...
## $ trustswork : int 5 5 4 5 6 6 7 5 6 7 ...
## $ trustsromantic : int 3 6 5 4 5 5 7 5 7 7 ...
## $ trustsacademic : int 5 5 4 5 3 5 5 6 7 7 ...
## $ trustsdrugs : int 5 5 4 5 5 5 7 5 6 7 ...
## $ trustgwork : int 4 5 4 5 5 5 7 6 6 6 ...
## $ trustgromantic : int 4 5 4 5 5 5 6 5 6 7 ...
## $ trustgacademic : int 5 3 4 5 3 5 6 5 6 6 ...
## $ trustgdrugs : int 5 3 4 5 5 5 5 4 6 5 ...
## $ likework : int 5 6 4 4 5 5 6 5 6 7 ...
## $ likeromantic : int 3 7 3 5 5 5 7 5 6 7 ...
## $ likeacademic : int 5 6 4 5 4 5 3 6 6 5 ...
## $ likedrugs : int 5 4 4 4 4 5 6 4 6 5 ...
## $ compq1work : int 2 2 1 1 2 1 2 2 1 2 ...
## $ compq1romantic : int 2 2 1 1 2 1 2 2 1 2 ...
## $ compq1academic : int 2 2 1 1 2 1 2 2 1 2 ...
## $ compq1drugs : int 2 2 1 1 2 1 2 2 1 2 ...
## $ compq2work : int 1 1 1 1 1 1 1 1 1 1 ...
## $ compq2romantic : int 1 1 1 1 1 1 1 1 1 1 ...
## $ compq2academic : int 1 1 1 1 1 1 1 1 1 1 ...
## $ compq2drugs : int 1 1 1 1 1 1 1 1 1 1 ...
## $ compq3work : int 3 3 3 3 3 3 3 3 3 3 ...
## $ compq3romantic : int 3 3 3 3 3 3 3 3 3 3 ...
## $ compq3academic : int 3 3 3 3 3 3 3 3 3 3 ...
## $ compq3drugs : int 3 3 3 3 3 3 3 3 3 3 ...
## $ compq4work : int 1 1 1 2 2 1 1 1 1 1 ...
## $ compq4romantic : int 1 1 1 2 2 1 1 1 1 1 ...
## $ compq4academic : int 1 1 1 2 2 1 1 1 1 1 ...
## $ compq4drugs : int 1 1 1 2 2 1 1 1 1 1 ...
## $ sumdo : num 5.5 5.25 4 4.75 5.5 6 6.75 3.75 6.75 7 ...
## $ sumtrusts : num 4.5 5.25 4.25 4.75 4.75 5.25 6.5 5.25 6.5 7 ...
## $ sumtrustg : num 4.5 4 4 5 4.5 5 6 5 6 6 ...
## $ sumlike : num 4.5 5.75 3.75 4.5 4.5 5 5.5 5 6 6 ...
## $ sumtotal : num 4.75 5.06 4 4.75 4.81 ...
## $ age : int 47 35 38 24 29 34 23 28 24 24 ...
## $ gender : chr "male" "male" "male" "female" ...
## $ edu : int 6 5 4 5 6 4 6 4 6 5 ...
## $ income : int 1 7 7 3 4 5 7 4 4 4 ...
## $ generaltrust : int 5 6 1 5 6 5 5 4 6 6 ...
## $ previousexperience: int 3 2 3 2 4 3 1 1 1 1 ...
## $ likeall : num 4.5 5.75 3.75 4.5 4.5 5 5.5 5 6 6 ...
Barplot
papaja::apa_barplot(data = s1,
dv = "likeall",
factors = c("goodinfo", "condemn"),
id = "id", args_legend = list(x = "topleft"), ylim = c(1, 7))
ANOVA
# Conduct an ANOVA on likeall as a function of goodinfo and condemn
# According to the paper on p. 359
# Information main effect, F(1, 165) = 137.93, p < .001
# Condemn main effect, F(1, 165) = 13.20, p < .001
# Interaction: F(1, 165) = 8.51, p = .004
s1.aov <- aov(likeall ~ goodinfo * condemn,
data = s1)
summary(s1.aov)
## Df Sum Sq Mean Sq F value Pr(>F)
## goodinfo 1 69.4 69.44 88.850 <2e-16 ***
## condemn 1 1.5 1.47 1.886 0.1701
## goodinfo:condemn 1 2.5 2.46 3.151 0.0764 .
## Residuals 614 479.8 0.78
## ---
## Signif. codes: 0 '***' 0.001 '**' 0.01 '*' 0.05 '.' 0.1 ' ' 1
## 1 observation deleted due to missingness
Study 2
# Show structure of study 2 data
# Should have 803 participants
str(s2)
## 'data.frame': 803 obs. of 38 variables:
## $ id : int 1 2 3 4 5 6 7 8 9 10 ...
## $ direct : chr "Moral" "Moral" "Direct" "Moral" ...
## $ signal : chr "Other" "Target" "Other" "Other" ...
## $ qAnswered : int 4 4 4 4 4 4 4 4 4 4 ...
## $ totalCorrect : int 4 4 4 4 4 4 4 4 4 4 ...
## $ percCorrect : num 1 1 1 1 1 1 1 1 1 1 ...
## $ dowork : int 4 2 4 4 4 5 4 2 4 4 ...
## $ doromantic : int 4 2 3 4 4 3 4 3 3 3 ...
## $ doacademic : int 4 2 4 4 1 3 2 6 5 3 ...
## $ dodrugs : int 4 2 5 4 3 3 4 4 3 5 ...
## $ trustswork : int 4 6 4 4 5 5 2 4 5 4 ...
## $ trustsromantic : int 3 6 4 4 3 5 4 4 5 2 ...
## $ trustsacademic : int 4 6 4 4 4 5 2 4 4 5 ...
## $ trustsdrugs : int 4 6 3 4 4 5 4 4 4 5 ...
## $ trustgwork : int 4 6 4 4 4 5 2 4 5 4 ...
## $ trustgromantic : int 4 6 4 4 2 5 4 4 4 3 ...
## $ trustgacademic : int 4 6 4 4 4 5 2 4 4 5 ...
## $ trustgdrugs : int 4 6 4 4 3 5 4 4 3 6 ...
## $ likework : int 5 6 4 4 5 5 2 4 5 4 ...
## $ likeromantic : int 3 6 4 4 4 5 4 4 4 5 ...
## $ likeacademic : int 4 6 4 4 4 5 2 4 5 2 ...
## $ likedrugs : int 4 6 4 4 4 5 4 4 3 6 ...
## $ compq1work : int 2 1 2 2 1 1 2 1 2 2 ...
## $ compq1romantic : int 2 1 2 2 1 1 2 1 2 2 ...
## $ compq1academic : int 2 1 2 2 1 1 2 1 2 2 ...
## $ compq1drugs : int 2 1 2 2 1 1 2 1 2 2 ...
## $ sumdo : num 4 6 4 4 5 4.5 4.5 4.25 4.25 4.25 ...
## $ sumtrusts : num 3.75 6 3.75 4 4 5 3 4 4.5 4 ...
## $ sumtrustg : num 4 6 4 4 3.25 5 3 4 4 4.5 ...
## $ sumlike : num 4 6 4 4 4.25 5 3 4 4.25 4.25 ...
## $ sumtotal : num 3.94 6 3.94 4 4.12 ...
## $ age : num 27 28 27 25 30 24 24 37 22 22 ...
## $ gender : chr "male" "female" "male" "male" ...
## $ edu : int 5 5 5 5 4 4 5 4 5 5 ...
## $ income : int 5 5 9 6 5 3 6 2 4 7 ...
## $ generaltrust : int 3 6 3 5 3 4 1 2 5 5 ...
## $ previousexperience: int 2 1 1 1 1 1 3 3 1 1 ...
## $ likeall : num 4 6 4 4 4.25 5 3 4 4.25 4.25 ...
Barplot
papaja::apa_barplot(data = s2,
dv = "likeall",
factors = c("signal", "direct"),
id = "id", args_legend = list(x = "topleft"), ylim = c(1, 7))
ANOVA
# Conduct an ANOVA on likeall as a function of goodinfo and condemn
# According to paper on page 361:
# Signal type, F(1, 799) = 9.92, p = .002
# Signaler: F(1,799) = 198.62, p < .001
# Interaction: F(1, 799) = 14.01, p < .001
s2.aov <- aov(likeall ~ direct * signal,
data = s2)
summary(s2.aov)
## Df Sum Sq Mean Sq F value Pr(>F)
## direct 1 1.2 1.23 2.065 0.1511
## signal 1 79.5 79.46 133.892 <2e-16 ***
## direct:signal 1 2.4 2.43 4.097 0.0433 *
## Residuals 799 474.2 0.59
## ---
## Signif. codes: 0 '***' 0.001 '**' 0.01 '*' 0.05 '.' 0.1 ' ' 1