WGN Decile & Quintile Analysis
JMJ
Updated 06 November, 2018
Preperation of Dataset
#setwd("~/Documents/Dropbox/Research/Eric")
setwd("~/Downloads")
combined2<-read.csv("~/Downloads/all DVs.csv", header=T, sep=",")
#removing empty rows
combined2<-subset(combined2, X.1 !="NA")
combined2 <- subset(combined2, cond != "<NA>")
combined2$bd<-combined2$beta*((1/(1+combined2$delta))^(2/52))*1.5
combined2$bd_not_centered<-combined2$bd
combined2$bd<-combined2$bd-mean(combined2$bd)
#OUTLIER ANALYSIS
hist(combined2$bd)
combined2$outliers[combined2$bd <= (mean(combined2$bd, na.rm = T)-(4*sd(combined2$bd, na.rm=T)))] <- 1
combined2$outliers[combined2$bd > (mean(combined2$bd, na.rm = T)-(4*sd(combined2$bd, na.rm=T)))] <- 0
table(combined2$outliers)##
## 0 1
## 1082 9combined2 <- subset(combined2, outliers == 0)
combined2$bd_ranked <- ((rank(combined2$bd))/nrow(combined2))*100
b1 <- c(0, 20, 40, 60, 80, 100)
b <- c(0, 10, 20, 30, 40, 50, 60, 70, 80, 90, 100)
combined2$quintiles <- .bincode(combined2$bd_ranked, breaks = b1, TRUE, TRUE)
combined2$deciles <- .bincode(combined2$bd_ranked, breaks = b, TRUE, TRUE)
library(ggplot2)
library(plyr)
cdata <- ddply(combined2, c("quintiles", "cond"), summarise,
N = length(SS),
mean = mean(SS),
sd = sd(SS),
ci = 1.96*(sd / sqrt(N)))
## this is with 95% confidence interval
ggplot(cdata, aes(x=quintiles, y=mean, colour=cond)) +
geom_errorbar(aes(ymin=mean-ci, ymax=mean+ci), width=.1) +
geom_line() +
geom_point()
## this is with +/- 1 standard deviation
cdata <- ddply(combined2, c("quintiles", "cond"), summarise,
N = length(SS),
mean = mean(SS),
sd = sd(SS),
se = sd / sqrt(N))
ggplot(cdata, aes(x=quintiles, y=mean, colour=cond)) +
geom_errorbar(aes(ymin=mean-se, ymax=mean+se), width=.1) +
geom_line() +
geom_point()
cdata1 <- ddply(combined2, c("deciles", "cond"), summarise,
N = length(SS),
mean = mean(SS),
sd = sd(SS),
se = sd / sqrt(N))
ggplot(cdata1, aes(x=deciles, y=mean, colour=cond)) +
geom_errorbar(aes(ymin=mean-se, ymax=mean+se), width=.1) +
geom_line() +
geom_point()
#small <- subset(combined2, select = c("cond","quintiles","deciles","SS","aware","directIE","indirectIE","ease","refdep","hong8","hong9","hong11","hong13","self","others","influence","interviewtime.x","believe.x","X.studies","gender","age","income","education","smoking"))
#write.csv(x = small, file = "WGN_complete_Dec7_2017.csv")Contrast Analysis
library(MASS)
combined2$cond_contrast <- as.character(combined2$cond)
combined2$cond_contrast[combined2$cond_contrast == "no"] <- 0
combined2$cond_contrast[combined2$cond_contrast == "SS"] <- -1
combined2$cond_contrast[combined2$cond_contrast == "LL"] <- 1
combined2$cond_contrast <- as.factor(combined2$cond_contrast)
combined2$quint_f <- as.factor(combined2$quintiles)
## Without Control Variables
summary(m1 <- aov(SS ~ cond_contrast * quint_f, combined2))## Df Sum Sq Mean Sq F value Pr(>F)
## cond_contrast 2 1.72 0.861 3.863 0.0213 *
## quint_f 4 27.66 6.916 31.017 <2e-16 ***
## cond_contrast:quint_f 8 2.93 0.366 1.640 0.1092
## Residuals 1067 237.92 0.223
## ---
## Signif. codes: 0 '***' 0.001 '**' 0.01 '*' 0.05 '.' 0.1 ' ' 1## With Control Variables
summary(m1 <- aov(SS ~ cond_contrast * quint_f + gender + age + income + education + X.studies, combined2))## Df Sum Sq Mean Sq F value Pr(>F)
## cond_contrast 2 1.19 0.597 2.666 0.0701 .
## quint_f 4 23.33 5.832 26.055 <2e-16 ***
## gender 2 0.25 0.126 0.562 0.5703
## age 1 0.17 0.165 0.738 0.3905
## income 13 2.48 0.191 0.853 0.6030
## education 8 1.79 0.223 0.997 0.4369
## X.studies 1 0.32 0.320 1.429 0.2322
## cond_contrast:quint_f 8 3.16 0.395 1.766 0.0802 .
## Residuals 869 194.53 0.224
## ---
## Signif. codes: 0 '***' 0.001 '**' 0.01 '*' 0.05 '.' 0.1 ' ' 1
## 173 observations deleted due to missingness## Analyses per quintile
summary(glm(SS ~ cond_contrast, family = binomial(link = "logit"), data = subset(combined2, quintiles == 1)))##
## Call:
## glm(formula = SS ~ cond_contrast, family = binomial(link = "logit"),
## data = subset(combined2, quintiles == 1))
##
## Deviance Residuals:
## Min 1Q Median 3Q Max
## -1.8867 -1.3308 0.6078 0.7502 1.0314
##
## Coefficients:
## Estimate Std. Error z value Pr(>|z|)
## (Intercept) 1.5950 0.2931 5.442 5.28e-08 ***
## cond_contrast0 -0.4711 0.4334 -1.087 0.277034
## cond_contrast1 -1.2414 0.3708 -3.348 0.000814 ***
## ---
## Signif. codes: 0 '***' 0.001 '**' 0.01 '*' 0.05 '.' 0.1 ' ' 1
##
## (Dispersion parameter for binomial family taken to be 1)
##
## Null deviance: 255.24 on 215 degrees of freedom
## Residual deviance: 242.82 on 213 degrees of freedom
## AIC: 248.82
##
## Number of Fisher Scoring iterations: 4#Here is the NO DEFAULT vs. LL contrast that Crystal requested.
LLcontrol<-subset(combined2, cond!="SS")
summary(glm(SS ~ cond_contrast, family = binomial(link = "logit"), data = subset(LLcontrol, quintiles == 1)))##
## Call:
## glm(formula = SS ~ cond_contrast, family = binomial(link = "logit"),
## data = subset(LLcontrol, quintiles == 1))
##
## Deviance Residuals:
## Min 1Q Median 3Q Max
## -1.6765 -1.3308 0.7502 1.0314 1.0314
##
## Coefficients:
## Estimate Std. Error z value Pr(>|z|)
## (Intercept) 1.1239 0.3193 3.520 0.000431 ***
## cond_contrast1 -0.7703 0.3918 -1.966 0.049292 *
## ---
## Signif. codes: 0 '***' 0.001 '**' 0.01 '*' 0.05 '.' 0.1 ' ' 1
##
## (Dispersion parameter for binomial family taken to be 1)
##
## Null deviance: 171.53 on 132 degrees of freedom
## Residual deviance: 167.49 on 131 degrees of freedom
## AIC: 171.49
##
## Number of Fisher Scoring iterations: 4summary(glm(SS ~ cond_contrast, family = binomial(link = "logit"), data = subset(combined2, quintiles == 2)))##
## Call:
## glm(formula = SS ~ cond_contrast, family = binomial(link = "logit"),
## data = subset(combined2, quintiles == 2))
##
## Deviance Residuals:
## Min 1Q Median 3Q Max
## -1.4181 -1.1681 0.9544 0.9647 1.1868
##
## Coefficients:
## Estimate Std. Error z value Pr(>|z|)
## (Intercept) 0.5501 0.2293 2.399 0.0164 *
## cond_contrast0 -0.0268 0.3900 -0.069 0.9452
## cond_contrast1 -0.5720 0.3107 -1.841 0.0656 .
## ---
## Signif. codes: 0 '***' 0.001 '**' 0.01 '*' 0.05 '.' 0.1 ' ' 1
##
## (Dispersion parameter for binomial family taken to be 1)
##
## Null deviance: 294.68 on 215 degrees of freedom
## Residual deviance: 290.61 on 213 degrees of freedom
## AIC: 296.61
##
## Number of Fisher Scoring iterations: 4summary(glm(SS ~ cond_contrast, family = binomial(link = "logit"), data = subset(combined2, quintiles == 3)))##
## Call:
## glm(formula = SS ~ cond_contrast, family = binomial(link = "logit"),
## data = subset(combined2, quintiles == 3))
##
## Deviance Residuals:
## Min 1Q Median 3Q Max
## -1.239 -1.219 1.117 1.136 1.254
##
## Coefficients:
## Estimate Std. Error z value Pr(>|z|)
## (Intercept) 0.09764 0.22113 0.442 0.659
## cond_contrast0 0.04546 0.34741 0.131 0.896
## cond_contrast1 -0.27532 0.31612 -0.871 0.384
##
## (Dispersion parameter for binomial family taken to be 1)
##
## Null deviance: 300.82 on 216 degrees of freedom
## Residual deviance: 299.72 on 214 degrees of freedom
## AIC: 305.72
##
## Number of Fisher Scoring iterations: 3summary(glm(SS ~ cond_contrast, family = binomial(link = "logit"), data = subset(combined2, quintiles == 4)))##
## Call:
## glm(formula = SS ~ cond_contrast, family = binomial(link = "logit"),
## data = subset(combined2, quintiles == 4))
##
## Deviance Residuals:
## Min 1Q Median 3Q Max
## -0.9956 -0.9005 -0.9005 1.3708 1.4823
##
## Coefficients:
## Estimate Std. Error z value Pr(>|z|)
## (Intercept) -0.4439 0.2197 -2.02 0.0433 *
## cond_contrast0 -0.2492 0.3369 -0.74 0.4595
## cond_contrast1 -0.2492 0.3511 -0.71 0.4778
## ---
## Signif. codes: 0 '***' 0.001 '**' 0.01 '*' 0.05 '.' 0.1 ' ' 1
##
## (Dispersion parameter for binomial family taken to be 1)
##
## Null deviance: 281.39 on 215 degrees of freedom
## Residual deviance: 280.65 on 213 degrees of freedom
## AIC: 286.65
##
## Number of Fisher Scoring iterations: 4summary(glm(SS ~ cond_contrast, family = binomial(link = "logit"), data = subset(combined2, quintiles == 5)))##
## Call:
## glm(formula = SS ~ cond_contrast, family = binomial(link = "logit"),
## data = subset(combined2, quintiles == 5))
##
## Deviance Residuals:
## Min 1Q Median 3Q Max
## -0.9080 -0.7518 -0.7012 1.4732 1.7456
##
## Coefficients:
## Estimate Std. Error z value Pr(>|z|)
## (Intercept) -1.1192 0.2879 -3.887 0.000101 ***
## cond_contrast0 -0.1584 0.3976 -0.398 0.690325
## cond_contrast1 0.4463 0.3786 1.179 0.238449
## ---
## Signif. codes: 0 '***' 0.001 '**' 0.01 '*' 0.05 '.' 0.1 ' ' 1
##
## (Dispersion parameter for binomial family taken to be 1)
##
## Null deviance: 251.95 on 216 degrees of freedom
## Residual deviance: 249.00 on 214 degrees of freedom
## AIC: 255
##
## Number of Fisher Scoring iterations: 4## Dropping control condition
combined2$cond_contrast2 <- as.character(combined2$cond)
combined2$cond_contrast2[combined2$cond_contrast2 == "no"] <- NA
combined2$cond_contrast2[combined2$cond_contrast2 == "SS"] <- "SS"
combined2$cond_contrast2[combined2$cond_contrast2 == "LL"] <- "LL"
combined2$cond_contrast <- as.factor(combined2$cond_contrast)
summary(m1 <- aov(SS ~ cond_contrast2*quint_f, combined2))## Df Sum Sq Mean Sq F value Pr(>F)
## cond_contrast2 1 1.44 1.441 6.348 0.0120 *
## quint_f 4 16.51 4.127 18.178 2.78e-14 ***
## cond_contrast2:quint_f 4 2.27 0.567 2.498 0.0415 *
## Residuals 773 175.51 0.227
## ---
## Signif. codes: 0 '***' 0.001 '**' 0.01 '*' 0.05 '.' 0.1 ' ' 1
## 299 observations deleted due to missingnesssummary(m1 <- aov(SS ~ cond_contrast2*quint_f + gender + age + income + education + X.studies, combined2))## Df Sum Sq Mean Sq F value Pr(>F)
## cond_contrast2 1 0.67 0.6715 2.911 0.0885 .
## quint_f 4 12.17 3.0424 13.187 2.75e-10 ***
## gender 2 0.18 0.0906 0.393 0.6754
## age 1 0.24 0.2413 1.046 0.3069
## income 13 2.58 0.1985 0.860 0.5956
## education 8 1.34 0.1669 0.723 0.6710
## X.studies 1 0.48 0.4786 2.074 0.1503
## cond_contrast2:quint_f 4 2.17 0.5423 2.350 0.0531 .
## Residuals 578 133.35 0.2307
## ---
## Signif. codes: 0 '***' 0.001 '**' 0.01 '*' 0.05 '.' 0.1 ' ' 1
## 469 observations deleted due to missingnesssummary(glm(SS ~ cond_contrast2, family = binomial(link = "logit"), data = subset(combined2, quintiles == 1)))##
## Call:
## glm(formula = SS ~ cond_contrast2, family = binomial(link = "logit"),
## data = subset(combined2, quintiles == 1))
##
## Deviance Residuals:
## Min 1Q Median 3Q Max
## -1.8867 -1.3308 0.6078 1.0314 1.0314
##
## Coefficients:
## Estimate Std. Error z value Pr(>|z|)
## (Intercept) 0.3536 0.2271 1.557 0.119442
## cond_contrast2SS 1.2414 0.3708 3.348 0.000814 ***
## ---
## Signif. codes: 0 '***' 0.001 '**' 0.01 '*' 0.05 '.' 0.1 ' ' 1
##
## (Dispersion parameter for binomial family taken to be 1)
##
## Null deviance: 195.82 on 162 degrees of freedom
## Residual deviance: 183.77 on 161 degrees of freedom
## (53 observations deleted due to missingness)
## AIC: 187.77
##
## Number of Fisher Scoring iterations: 4summary(glm(SS ~ cond_contrast2, family = binomial(link = "logit"), data = subset(combined2, quintiles == 2)))##
## Call:
## glm(formula = SS ~ cond_contrast2, family = binomial(link = "logit"),
## data = subset(combined2, quintiles == 2))
##
## Deviance Residuals:
## Min 1Q Median 3Q Max
## -1.4181 -1.1681 0.9544 1.1868 1.1868
##
## Coefficients:
## Estimate Std. Error z value Pr(>|z|)
## (Intercept) -0.02198 0.20967 -0.105 0.9165
## cond_contrast2SS 0.57203 0.31069 1.841 0.0656 .
## ---
## Signif. codes: 0 '***' 0.001 '**' 0.01 '*' 0.05 '.' 0.1 ' ' 1
##
## (Dispersion parameter for binomial family taken to be 1)
##
## Null deviance: 237.27 on 172 degrees of freedom
## Residual deviance: 233.84 on 171 degrees of freedom
## (43 observations deleted due to missingness)
## AIC: 237.84
##
## Number of Fisher Scoring iterations: 4summary(glm(SS ~ cond_contrast2, family = binomial(link = "logit"), data = subset(combined2, quintiles == 3)))##
## Call:
## glm(formula = SS ~ cond_contrast2, family = binomial(link = "logit"),
## data = subset(combined2, quintiles == 3))
##
## Deviance Residuals:
## Min 1Q Median 3Q Max
## -1.219 -1.103 -1.103 1.136 1.254
##
## Coefficients:
## Estimate Std. Error z value Pr(>|z|)
## (Intercept) -0.1777 0.2259 -0.787 0.432
## cond_contrast2SS 0.2753 0.3161 0.871 0.384
##
## (Dispersion parameter for binomial family taken to be 1)
##
## Null deviance: 223.14 on 160 degrees of freedom
## Residual deviance: 222.38 on 159 degrees of freedom
## (56 observations deleted due to missingness)
## AIC: 226.38
##
## Number of Fisher Scoring iterations: 3summary(glm(SS ~ cond_contrast2, family = binomial(link = "logit"), data = subset(combined2, quintiles == 4)))##
## Call:
## glm(formula = SS ~ cond_contrast2, family = binomial(link = "logit"),
## data = subset(combined2, quintiles == 4))
##
## Deviance Residuals:
## Min 1Q Median 3Q Max
## -0.9956 -0.9956 -0.9005 1.3708 1.4823
##
## Coefficients:
## Estimate Std. Error z value Pr(>|z|)
## (Intercept) -0.6931 0.2739 -2.531 0.0114 *
## cond_contrast2SS 0.2492 0.3511 0.710 0.4778
## ---
## Signif. codes: 0 '***' 0.001 '**' 0.01 '*' 0.05 '.' 0.1 ' ' 1
##
## (Dispersion parameter for binomial family taken to be 1)
##
## Null deviance: 193.31 on 146 degrees of freedom
## Residual deviance: 192.81 on 145 degrees of freedom
## (69 observations deleted due to missingness)
## AIC: 196.81
##
## Number of Fisher Scoring iterations: 4summary(glm(SS ~ cond_contrast2, family = binomial(link = "logit"), data = subset(combined2, quintiles == 5)))##
## Call:
## glm(formula = SS ~ cond_contrast2, family = binomial(link = "logit"),
## data = subset(combined2, quintiles == 5))
##
## Deviance Residuals:
## Min 1Q Median 3Q Max
## -0.9080 -0.9080 -0.7518 1.4732 1.6744
##
## Coefficients:
## Estimate Std. Error z value Pr(>|z|)
## (Intercept) -0.6729 0.2458 -2.738 0.00618 **
## cond_contrast2SS -0.4463 0.3786 -1.179 0.23845
## ---
## Signif. codes: 0 '***' 0.001 '**' 0.01 '*' 0.05 '.' 0.1 ' ' 1
##
## (Dispersion parameter for binomial family taken to be 1)
##
## Null deviance: 168.62 on 138 degrees of freedom
## Residual deviance: 167.21 on 137 degrees of freedom
## (78 observations deleted due to missingness)
## AIC: 171.21
##
## Number of Fisher Scoring iterations: 4## If we want to report deciles...
summary(glm(SS ~ as.factor(cond), family = binomial(link = "logit"), data = subset(combined2, deciles == 1)))##
## Call:
## glm(formula = SS ~ as.factor(cond), family = binomial(link = "logit"),
## data = subset(combined2, deciles == 1))
##
## Deviance Residuals:
## Min 1Q Median 3Q Max
## -2.0393 0.0767 0.5350 0.5350 1.1127
##
## Coefficients:
## Estimate Std. Error z value Pr(>|z|)
## (Intercept) 0.1542 0.3212 0.480 0.63129
## as.factor(cond)no 1.7918 0.6958 2.575 0.01002 *
## as.factor(cond)SS 1.7177 0.5436 3.160 0.00158 **
## ---
## Signif. codes: 0 '***' 0.001 '**' 0.01 '*' 0.05 '.' 0.1 ' ' 1
##
## (Dispersion parameter for binomial family taken to be 1)
##
## Null deviance: 121.46 on 107 degrees of freedom
## Residual deviance: 107.26 on 105 degrees of freedom
## AIC: 113.26
##
## Number of Fisher Scoring iterations: 4summary(glm(SS ~ as.factor(cond), family = binomial(link = "logit"), data = subset(combined2, deciles == 2)))##
## Call:
## glm(formula = SS ~ as.factor(cond), family = binomial(link = "logit"),
## data = subset(combined2, deciles == 2))
##
## Deviance Residuals:
## Min 1Q Median 3Q Max
## -1.7653 -1.4181 0.6876 0.9196 0.9544
##
## Coefficients:
## Estimate Std. Error z value Pr(>|z|)
## (Intercept) 0.55005 0.32423 1.696 0.0898 .
## as.factor(cond)no 0.09181 0.50770 0.181 0.8565
## as.factor(cond)SS 0.77171 0.51328 1.503 0.1327
## ---
## Signif. codes: 0 '***' 0.001 '**' 0.01 '*' 0.05 '.' 0.1 ' ' 1
##
## (Dispersion parameter for binomial family taken to be 1)
##
## Null deviance: 132.95 on 107 degrees of freedom
## Residual deviance: 130.33 on 105 degrees of freedom
## AIC: 136.33
##
## Number of Fisher Scoring iterations: 4summary(glm(SS ~ as.factor(cond), family = binomial(link = "logit"), data = subset(combined2, deciles == 3)))##
## Call:
## glm(formula = SS ~ as.factor(cond), family = binomial(link = "logit"),
## data = subset(combined2, deciles == 3))
##
## Deviance Residuals:
## Min 1Q Median 3Q Max
## -1.5315 -1.0323 0.8607 0.9282 1.3298
##
## Coefficients:
## Estimate Std. Error z value Pr(>|z|)
## (Intercept) -0.3514 0.2994 -1.173 0.2406
## as.factor(cond)no 0.9704 0.5563 1.745 0.0811 .
## as.factor(cond)SS 1.1537 0.4484 2.573 0.0101 *
## ---
## Signif. codes: 0 '***' 0.001 '**' 0.01 '*' 0.05 '.' 0.1 ' ' 1
##
## (Dispersion parameter for binomial family taken to be 1)
##
## Null deviance: 147.90 on 107 degrees of freedom
## Residual deviance: 140.24 on 105 degrees of freedom
## AIC: 146.24
##
## Number of Fisher Scoring iterations: 4summary(glm(SS ~ as.factor(cond), family = binomial(link = "logit"), data = subset(combined2, deciles == 4)))##
## Call:
## glm(formula = SS ~ as.factor(cond), family = binomial(link = "logit"),
## data = subset(combined2, deciles == 4))
##
## Deviance Residuals:
## Min 1Q Median 3Q Max
## -1.3699 -1.3132 0.9964 1.0474 1.0520
##
## Coefficients:
## Estimate Std. Error z value Pr(>|z|)
## (Intercept) 0.31366 0.30182 1.039 0.299
## as.factor(cond)no 0.12818 0.52310 0.245 0.806
## as.factor(cond)SS -0.01138 0.43977 -0.026 0.979
##
## (Dispersion parameter for binomial family taken to be 1)
##
## Null deviance: 146.71 on 107 degrees of freedom
## Residual deviance: 146.63 on 105 degrees of freedom
## AIC: 152.63
##
## Number of Fisher Scoring iterations: 4summary(glm(SS ~ as.factor(cond), family = binomial(link = "logit"), data = subset(combined2, deciles == 5)))##
## Call:
## glm(formula = SS ~ as.factor(cond), family = binomial(link = "logit"),
## data = subset(combined2, deciles == 5))
##
## Deviance Residuals:
## Min 1Q Median 3Q Max
## -1.354 -1.145 1.011 1.081 1.210
##
## Coefficients:
## Estimate Std. Error z value Pr(>|z|)
## (Intercept) -0.07696 0.27756 -0.277 0.782
## as.factor(cond)no 0.48243 0.95407 0.506 0.613
## as.factor(cond)SS 0.30876 0.39370 0.784 0.433
##
## (Dispersion parameter for binomial family taken to be 1)
##
## Null deviance: 150.88 on 108 degrees of freedom
## Residual deviance: 150.13 on 106 degrees of freedom
## AIC: 156.13
##
## Number of Fisher Scoring iterations: 3summary(glm(SS ~ as.factor(cond), family = binomial(link = "logit"), data = subset(combined2, deciles == 6)))##
## Call:
## glm(formula = SS ~ as.factor(cond), family = binomial(link = "logit"),
## data = subset(combined2, deciles == 6))
##
## Deviance Residuals:
## Min 1Q Median 3Q Max
## -1.228 -1.121 -1.023 1.128 1.340
##
## Coefficients:
## Estimate Std. Error z value Pr(>|z|)
## (Intercept) -0.3747 0.3917 -0.957 0.339
## as.factor(cond)no 0.4925 0.4818 1.022 0.307
## as.factor(cond)SS 0.2412 0.5360 0.450 0.653
##
## (Dispersion parameter for binomial family taken to be 1)
##
## Null deviance: 149.57 on 107 degrees of freedom
## Residual deviance: 148.48 on 105 degrees of freedom
## AIC: 154.48
##
## Number of Fisher Scoring iterations: 4summary(glm(SS ~ as.factor(cond), family = binomial(link = "logit"), data = subset(combined2, deciles == 7)))##
## Call:
## glm(formula = SS ~ as.factor(cond), family = binomial(link = "logit"),
## data = subset(combined2, deciles == 7))
##
## Deviance Residuals:
## Min 1Q Median 3Q Max
## -1.2163 -0.9508 -0.8826 1.1390 1.5043
##
## Coefficients:
## Estimate Std. Error z value Pr(>|z|)
## (Intercept) -0.7419 0.3842 -1.931 0.0535 .
## as.factor(cond)no 0.1823 0.5278 0.345 0.7298
## as.factor(cond)SS 0.8329 0.4886 1.705 0.0882 .
## ---
## Signif. codes: 0 '***' 0.001 '**' 0.01 '*' 0.05 '.' 0.1 ' ' 1
##
## (Dispersion parameter for binomial family taken to be 1)
##
## Null deviance: 146.71 on 107 degrees of freedom
## Residual deviance: 143.15 on 105 degrees of freedom
## AIC: 149.15
##
## Number of Fisher Scoring iterations: 4summary(glm(SS ~ as.factor(cond), family = binomial(link = "logit"), data = subset(combined2, deciles == 8)))##
## Call:
## glm(formula = SS ~ as.factor(cond), family = binomial(link = "logit"),
## data = subset(combined2, deciles == 8))
##
## Deviance Residuals:
## Min 1Q Median 3Q Max
## -0.9196 -0.8540 -0.7687 1.4592 1.6512
##
## Coefficients:
## Estimate Std. Error z value Pr(>|z|)
## (Intercept) -0.6419 0.3907 -1.643 0.100
## as.factor(cond)no -0.1791 0.5325 -0.336 0.737
## as.factor(cond)SS -0.4260 0.5242 -0.813 0.416
##
## (Dispersion parameter for binomial family taken to be 1)
##
## Null deviance: 131.26 on 107 degrees of freedom
## Residual deviance: 130.58 on 105 degrees of freedom
## AIC: 136.58
##
## Number of Fisher Scoring iterations: 4summary(glm(SS ~ as.factor(cond), family = binomial(link = "logit"), data = subset(combined2, deciles == 9)))##
## Call:
## glm(formula = SS ~ as.factor(cond), family = binomial(link = "logit"),
## data = subset(combined2, deciles == 9))
##
## Deviance Residuals:
## Min 1Q Median 3Q Max
## -0.8620 -0.7585 -0.7276 1.5297 1.7080
##
## Coefficients:
## Estimate Std. Error z value Pr(>|z|)
## (Intercept) -1.09861 0.38490 -2.854 0.00431 **
## as.factor(cond)no -0.09531 0.52768 -0.181 0.85667
## as.factor(cond)SS 0.30010 0.55611 0.540 0.58944
## ---
## Signif. codes: 0 '***' 0.001 '**' 0.01 '*' 0.05 '.' 0.1 ' ' 1
##
## (Dispersion parameter for binomial family taken to be 1)
##
## Null deviance: 123.61 on 107 degrees of freedom
## Residual deviance: 123.05 on 105 degrees of freedom
## AIC: 129.05
##
## Number of Fisher Scoring iterations: 4summary(glm(SS ~ as.factor(cond), family = binomial(link = "logit"), data = subset(combined2, deciles == 10)))##
## Call:
## glm(formula = SS ~ as.factor(cond), family = binomial(link = "logit"),
## data = subset(combined2, deciles == 10))
##
## Deviance Residuals:
## Min 1Q Median 3Q Max
## -1.0455 -0.6681 -0.6576 1.3153 1.8098
##
## Coefficients:
## Estimate Std. Error z value Pr(>|z|)
## (Intercept) -0.3185 0.3286 -0.969 0.3324
## as.factor(cond)no -1.0678 0.5353 -1.995 0.0461 *
## as.factor(cond)SS -1.1029 0.5341 -2.065 0.0389 *
## ---
## Signif. codes: 0 '***' 0.001 '**' 0.01 '*' 0.05 '.' 0.1 ' ' 1
##
## (Dispersion parameter for binomial family taken to be 1)
##
## Null deviance: 128.27 on 108 degrees of freedom
## Residual deviance: 122.22 on 106 degrees of freedom
## AIC: 128.22
##
## Number of Fisher Scoring iterations: 4Moderator Analyses
summary(m1 <- aov(SS ~ cond_contrast * quint_f, combined2))## Df Sum Sq Mean Sq F value Pr(>F)
## cond_contrast 2 1.72 0.861 3.863 0.0213 *
## quint_f 4 27.66 6.916 31.017 <2e-16 ***
## cond_contrast:quint_f 8 2.93 0.366 1.640 0.1092
## Residuals 1067 237.92 0.223
## ---
## Signif. codes: 0 '***' 0.001 '**' 0.01 '*' 0.05 '.' 0.1 ' ' 1summary(m1 <- aov(SS ~ cond_contrast * quint_f * aware, combined2))## Df Sum Sq Mean Sq F value Pr(>F)
## cond_contrast 2 1.20 0.599 2.858 0.0582 .
## quint_f 4 22.43 5.607 26.747 <2e-16 ***
## aware 1 0.05 0.047 0.226 0.6343
## cond_contrast:quint_f 8 2.12 0.264 1.261 0.2612
## cond_contrast:aware 2 0.26 0.131 0.626 0.5350
## quint_f:aware 4 0.44 0.111 0.530 0.7137
## cond_contrast:quint_f:aware 6 2.45 0.408 1.948 0.0712 .
## Residuals 578 121.17 0.210
## ---
## Signif. codes: 0 '***' 0.001 '**' 0.01 '*' 0.05 '.' 0.1 ' ' 1
## 476 observations deleted due to missingnesssummary(m1 <- aov(SS ~ cond_contrast * quint_f * directIE, combined2))## Df Sum Sq Mean Sq F value Pr(>F)
## cond_contrast 2 1.20 0.599 2.959 0.05266 .
## quint_f 4 22.43 5.607 27.693 < 2e-16 ***
## directIE 1 1.28 1.279 6.317 0.01223 *
## cond_contrast:quint_f 8 2.16 0.271 1.337 0.22237
## cond_contrast:directIE 2 1.49 0.746 3.682 0.02576 *
## quint_f:directIE 4 2.93 0.732 3.615 0.00638 **
## cond_contrast:quint_f:directIE 8 2.00 0.250 1.237 0.27482
## Residuals 576 116.62 0.202
## ---
## Signif. codes: 0 '***' 0.001 '**' 0.01 '*' 0.05 '.' 0.1 ' ' 1
## 476 observations deleted due to missingnesssummary(m1 <- aov(SS ~ cond_contrast * quint_f * indirectIE, combined2))## Df Sum Sq Mean Sq F value Pr(>F)
## cond_contrast 2 1.20 0.599 2.828 0.0599 .
## quint_f 4 22.43 5.607 26.470 <2e-16 ***
## indirectIE 1 0.75 0.753 3.554 0.0599 .
## cond_contrast:quint_f 8 2.07 0.259 1.222 0.2834
## cond_contrast:indirectIE 2 0.45 0.226 1.068 0.3442
## quint_f:indirectIE 4 0.30 0.076 0.356 0.8396
## cond_contrast:quint_f:indirectIE 8 0.90 0.113 0.532 0.8326
## Residuals 576 122.01 0.212
## ---
## Signif. codes: 0 '***' 0.001 '**' 0.01 '*' 0.05 '.' 0.1 ' ' 1
## 476 observations deleted due to missingnesssummary(m1 <- aov(SS ~ cond_contrast * quint_f * ease, combined2))## Df Sum Sq Mean Sq F value Pr(>F)
## cond_contrast 2 1.20 0.599 2.925 0.0544 .
## quint_f 4 22.43 5.607 27.378 < 2e-16 ***
## ease 1 4.51 4.511 22.026 3.36e-06 ***
## cond_contrast:quint_f 8 1.78 0.223 1.087 0.3702
## cond_contrast:ease 2 0.75 0.374 1.826 0.1621
## quint_f:ease 4 0.41 0.104 0.506 0.7315
## cond_contrast:quint_f:ease 8 1.07 0.134 0.652 0.7338
## Residuals 576 117.96 0.205
## ---
## Signif. codes: 0 '***' 0.001 '**' 0.01 '*' 0.05 '.' 0.1 ' ' 1
## 476 observations deleted due to missingnesssummary(m1 <- aov(SS ~ cond_contrast * quint_f * refdep, combined2))## Df Sum Sq Mean Sq F value Pr(>F)
## cond_contrast 2 1.20 0.599 3.055 0.0479 *
## quint_f 4 22.43 5.607 28.588 < 2e-16 ***
## refdep 1 3.45 3.450 17.590 3.17e-05 ***
## cond_contrast:quint_f 8 1.94 0.243 1.237 0.2747
## cond_contrast:refdep 2 4.63 2.313 11.793 9.56e-06 ***
## quint_f:refdep 4 1.61 0.403 2.055 0.0853 .
## cond_contrast:quint_f:refdep 8 1.89 0.236 1.202 0.2951
## Residuals 576 112.97 0.196
## ---
## Signif. codes: 0 '***' 0.001 '**' 0.01 '*' 0.05 '.' 0.1 ' ' 1
## 476 observations deleted due to missingnesssummary(m1 <- aov(SS ~ cond_contrast * quint_f + aware, combined2))## Df Sum Sq Mean Sq F value Pr(>F)
## cond_contrast 2 1.20 0.599 2.843 0.059 .
## quint_f 4 22.43 5.607 26.608 <2e-16 ***
## aware 1 0.05 0.047 0.225 0.635
## cond_contrast:quint_f 8 2.12 0.264 1.255 0.265
## Residuals 590 124.32 0.211
## ---
## Signif. codes: 0 '***' 0.001 '**' 0.01 '*' 0.05 '.' 0.1 ' ' 1
## 476 observations deleted due to missingnesssummary(m1 <- aov(SS ~ cond_contrast * quint_f + directIE, combined2))## Df Sum Sq Mean Sq F value Pr(>F)
## cond_contrast 2 1.20 0.599 2.873 0.0573 .
## quint_f 4 22.43 5.607 26.886 <2e-16 ***
## directIE 1 1.28 1.279 6.133 0.0135 *
## cond_contrast:quint_f 8 2.16 0.271 1.298 0.2418
## Residuals 590 123.04 0.209
## ---
## Signif. codes: 0 '***' 0.001 '**' 0.01 '*' 0.05 '.' 0.1 ' ' 1
## 476 observations deleted due to missingnesssummary(m1 <- aov(SS ~ cond_contrast * quint_f + indirectIE, combined2))## Df Sum Sq Mean Sq F value Pr(>F)
## cond_contrast 2 1.20 0.599 2.858 0.0582 .
## quint_f 4 22.43 5.607 26.751 <2e-16 ***
## indirectIE 1 0.75 0.753 3.591 0.0586 .
## cond_contrast:quint_f 8 2.07 0.259 1.235 0.2759
## Residuals 590 123.66 0.210
## ---
## Signif. codes: 0 '***' 0.001 '**' 0.01 '*' 0.05 '.' 0.1 ' ' 1
## 476 observations deleted due to missingnesssummary(m1 <- aov(SS ~ cond_contrast * quint_f + ease, combined2))## Df Sum Sq Mean Sq F value Pr(>F)
## cond_contrast 2 1.20 0.599 2.941 0.0536 .
## quint_f 4 22.43 5.607 27.523 < 2e-16 ***
## ease 1 4.51 4.511 22.143 3.16e-06 ***
## cond_contrast:quint_f 8 1.78 0.223 1.093 0.3661
## Residuals 590 120.19 0.204
## ---
## Signif. codes: 0 '***' 0.001 '**' 0.01 '*' 0.05 '.' 0.1 ' ' 1
## 476 observations deleted due to missingnesssummary(m1 <- aov(SS ~ cond_contrast * quint_f + refdep, combined2))## Df Sum Sq Mean Sq F value Pr(>F)
## cond_contrast 2 1.20 0.599 2.919 0.0548 .
## quint_f 4 22.43 5.607 27.318 < 2e-16 ***
## refdep 1 3.45 3.450 16.809 4.71e-05 ***
## cond_contrast:quint_f 8 1.94 0.243 1.182 0.3074
## Residuals 590 121.10 0.205
## ---
## Signif. codes: 0 '***' 0.001 '**' 0.01 '*' 0.05 '.' 0.1 ' ' 1
## 476 observations deleted due to missingnessCreating attention check variable to use for excluding people who failed.
#recoding "reader" into attention check failure/success
combined2$attn[combined2$reader == "reader" | combined2$reader == "Reader"|combined2$reader == "READER"|combined2$reader== " reader"|combined2$reader== "reader " | combined2$reader== "readers" ]<-1
combined2$attn[combined2$reader != "reader" & combined2$reader != "Reader"& combined2$reader != "READER" & combined2$reader!= " reader"&combined2$reader!= "reader " & combined2$reader!= "readers" ]<-0
table(combined2$attn)##
## 0 1
## 277 805pass<-subset(combined2, attn==1)Looking at ease, implied endorsement, reference dependence, and belief about the default’s effects (although only half the sample has this data)
#WITH CONTROL CONDITION
summary(lm(ease ~ cond*bd, combined2))##
## Call:
## lm(formula = ease ~ cond * bd, data = combined2)
##
## Residuals:
## Min 1Q Median 3Q Max
## -2.1339 -1.0458 -0.3833 0.9678 3.3121
##
## Coefficients:
## Estimate Std. Error t value Pr(>|t|)
## (Intercept) 2.06195 0.09489 21.731 <2e-16 ***
## condno 0.18080 0.13489 1.340 0.181
## condSS 0.07132 0.13072 0.546 0.586
## bd -0.70460 0.45220 -1.558 0.120
## condno:bd -0.61714 0.68110 -0.906 0.365
## condSS:bd -0.53135 0.60439 -0.879 0.380
## ---
## Signif. codes: 0 '***' 0.001 '**' 0.01 '*' 0.05 '.' 0.1 ' ' 1
##
## Residual standard error: 1.318 on 600 degrees of freedom
## (476 observations deleted due to missingness)
## Multiple R-squared: 0.03194, Adjusted R-squared: 0.02387
## F-statistic: 3.959 on 5 and 600 DF, p-value: 0.001533summary(lm(refdep ~ cond*bd, combined2))##
## Call:
## lm(formula = refdep ~ cond * bd, data = combined2)
##
## Residuals:
## Min 1Q Median 3Q Max
## -2.1082 -0.7918 -0.5459 0.4564 3.4353
##
## Coefficients:
## Estimate Std. Error t value Pr(>|t|)
## (Intercept) 1.86504 0.08434 22.113 <2e-16 ***
## condno -0.23905 0.11990 -1.994 0.0466 *
## condSS 0.06748 0.11619 0.581 0.5616
## bd 0.41307 0.40195 1.028 0.3045
## condno:bd -1.12903 0.60541 -1.865 0.0627 .
## condSS:bd -2.22311 0.53722 -4.138 4e-05 ***
## ---
## Signif. codes: 0 '***' 0.001 '**' 0.01 '*' 0.05 '.' 0.1 ' ' 1
##
## Residual standard error: 1.172 on 600 degrees of freedom
## (476 observations deleted due to missingness)
## Multiple R-squared: 0.06058, Adjusted R-squared: 0.05275
## F-statistic: 7.739 on 5 and 600 DF, p-value: 4.56e-07summary(lm(directIE ~ cond*bd, combined2))##
## Call:
## lm(formula = directIE ~ cond * bd, data = combined2)
##
## Residuals:
## Min 1Q Median 3Q Max
## -0.8643 -0.5696 -0.3877 0.3976 3.6452
##
## Coefficients:
## Estimate Std. Error t value Pr(>|t|)
## (Intercept) 1.63078 0.06389 25.526 <2e-16 ***
## condno -0.19000 0.09082 -2.092 0.0369 *
## condSS -0.08694 0.08801 -0.988 0.3236
## bd -0.12989 0.30447 -0.427 0.6698
## condno:bd -0.52075 0.45859 -1.136 0.2566
## condSS:bd -0.13559 0.40694 -0.333 0.7391
## ---
## Signif. codes: 0 '***' 0.001 '**' 0.01 '*' 0.05 '.' 0.1 ' ' 1
##
## Residual standard error: 0.8876 on 600 degrees of freedom
## (476 observations deleted due to missingness)
## Multiple R-squared: 0.01711, Adjusted R-squared: 0.008921
## F-statistic: 2.089 on 5 and 600 DF, p-value: 0.06511summary(lm(indirectIE ~ cond*bd, combined2))##
## Call:
## lm(formula = indirectIE ~ cond * bd, data = combined2)
##
## Residuals:
## Min 1Q Median 3Q Max
## -0.9902 -0.6504 -0.5308 0.3852 3.3670
##
## Coefficients:
## Estimate Std. Error t value Pr(>|t|)
## (Intercept) 1.61746 0.07043 22.965 <2e-16 ***
## condno -0.02724 0.10012 -0.272 0.786
## condSS 0.08916 0.09703 0.919 0.358
## bd 0.07434 0.33566 0.221 0.825
## condno:bd -0.78401 0.50557 -1.551 0.121
## condSS:bd -0.34350 0.44863 -0.766 0.444
## ---
## Signif. codes: 0 '***' 0.001 '**' 0.01 '*' 0.05 '.' 0.1 ' ' 1
##
## Residual standard error: 0.9786 on 600 degrees of freedom
## (476 observations deleted due to missingness)
## Multiple R-squared: 0.01097, Adjusted R-squared: 0.002729
## F-statistic: 1.331 on 5 and 600 DF, p-value: 0.2492summary(lm(self ~ cond*bd, combined2))##
## Call:
## lm(formula = self ~ cond * bd, data = combined2)
##
## Residuals:
## Min 1Q Median 3Q Max
## -0.7872 -0.4159 -0.2480 -0.1982 3.7596
##
## Coefficients:
## Estimate Std. Error t value Pr(>|t|)
## (Intercept) 1.48091 0.05511 26.872 < 2e-16 ***
## condno -0.26908 0.07834 -3.435 0.000634 ***
## condSS -0.11851 0.07592 -1.561 0.119062
## bd -0.45309 0.26264 -1.725 0.085025 .
## condno:bd 0.35456 0.39559 0.896 0.370462
## condSS:bd 0.15655 0.35104 0.446 0.655780
## ---
## Signif. codes: 0 '***' 0.001 '**' 0.01 '*' 0.05 '.' 0.1 ' ' 1
##
## Residual standard error: 0.7657 on 600 degrees of freedom
## (476 observations deleted due to missingness)
## Multiple R-squared: 0.02698, Adjusted R-squared: 0.01887
## F-statistic: 3.327 on 5 and 600 DF, p-value: 0.005673summary(lm(others ~ cond*bd, combined2))##
## Call:
## lm(formula = others ~ cond * bd, data = combined2)
##
## Residuals:
## Min 1Q Median 3Q Max
## -1.8221 -0.6651 0.2957 0.4157 2.4744
##
## Coefficients:
## Estimate Std. Error t value Pr(>|t|)
## (Intercept) 2.65317 0.06405 41.425 <2e-16 ***
## condno -0.22370 0.09105 -2.457 0.0143 *
## condSS 0.04918 0.08823 0.557 0.5774
## bd -0.24983 0.30524 -0.818 0.4134
## condno:bd 1.37959 0.45974 3.001 0.0028 **
## condSS:bd 0.27369 0.40796 0.671 0.5026
## ---
## Signif. codes: 0 '***' 0.001 '**' 0.01 '*' 0.05 '.' 0.1 ' ' 1
##
## Residual standard error: 0.8899 on 600 degrees of freedom
## (476 observations deleted due to missingness)
## Multiple R-squared: 0.03055, Adjusted R-squared: 0.02247
## F-statistic: 3.781 on 5 and 600 DF, p-value: 0.002224summary(lm(influence ~ cond*bd, combined2))##
## Call:
## lm(formula = influence ~ cond * bd, data = combined2)
##
## Residuals:
## Min 1Q Median 3Q Max
## -0.8884 -0.4544 -0.3038 0.4390 3.6042
##
## Coefficients:
## Estimate Std. Error t value Pr(>|t|)
## (Intercept) 1.54408 0.05747 26.866 < 2e-16 ***
## condno -0.22923 0.08170 -2.806 0.00518 **
## condSS -0.15422 0.07918 -1.948 0.05191 .
## bd -0.19252 0.27390 -0.703 0.48240
## condno:bd 0.05451 0.41255 0.132 0.89493
## condSS:bd -0.44831 0.36608 -1.225 0.22120
## ---
## Signif. codes: 0 '***' 0.001 '**' 0.01 '*' 0.05 '.' 0.1 ' ' 1
##
## Residual standard error: 0.7985 on 600 degrees of freedom
## (476 observations deleted due to missingness)
## Multiple R-squared: 0.02617, Adjusted R-squared: 0.01806
## F-statistic: 3.225 on 5 and 600 DF, p-value: 0.006979#WITHOUT CONTROL CONDITION
nocontrol<-subset(combined2, cond!="no")
summary(lm(ease ~ cond*bd, nocontrol))##
## Call:
## lm(formula = ease ~ cond * bd, data = nocontrol)
##
## Residuals:
## Min 1Q Median 3Q Max
## -2.1339 -1.0043 -0.4741 0.9714 3.3121
##
## Coefficients:
## Estimate Std. Error t value Pr(>|t|)
## (Intercept) 2.06195 0.09515 21.670 <2e-16 ***
## condSS 0.07132 0.13108 0.544 0.587
## bd -0.70460 0.45347 -1.554 0.121
## condSS:bd -0.53135 0.60608 -0.877 0.381
## ---
## Signif. codes: 0 '***' 0.001 '**' 0.01 '*' 0.05 '.' 0.1 ' ' 1
##
## Residual standard error: 1.322 on 405 degrees of freedom
## (374 observations deleted due to missingness)
## Multiple R-squared: 0.0294, Adjusted R-squared: 0.02221
## F-statistic: 4.089 on 3 and 405 DF, p-value: 0.007051summary(lm(refdep ~ cond*bd, nocontrol))##
## Call:
## lm(formula = refdep ~ cond * bd, data = nocontrol)
##
## Residuals:
## Min 1Q Median 3Q Max
## -2.1082 -0.8981 -0.6879 0.4558 3.3049
##
## Coefficients:
## Estimate Std. Error t value Pr(>|t|)
## (Intercept) 1.86504 0.09104 20.486 < 2e-16 ***
## condSS 0.06748 0.12542 0.538 0.590875
## bd 0.41307 0.43388 0.952 0.341643
## condSS:bd -2.22311 0.57990 -3.834 0.000146 ***
## ---
## Signif. codes: 0 '***' 0.001 '**' 0.01 '*' 0.05 '.' 0.1 ' ' 1
##
## Residual standard error: 1.265 on 405 degrees of freedom
## (374 observations deleted due to missingness)
## Multiple R-squared: 0.0544, Adjusted R-squared: 0.04739
## F-statistic: 7.766 on 3 and 405 DF, p-value: 4.709e-05summary(lm(directIE ~ cond*bd, nocontrol))##
## Call:
## lm(formula = directIE ~ cond * bd, data = nocontrol)
##
## Residuals:
## Min 1Q Median 3Q Max
## -0.7566 -0.6134 -0.5202 0.3979 3.4922
##
## Coefficients:
## Estimate Std. Error t value Pr(>|t|)
## (Intercept) 1.63078 0.06494 25.112 <2e-16 ***
## condSS -0.08694 0.08946 -0.972 0.332
## bd -0.12989 0.30949 -0.420 0.675
## condSS:bd -0.13559 0.41364 -0.328 0.743
## ---
## Signif. codes: 0 '***' 0.001 '**' 0.01 '*' 0.05 '.' 0.1 ' ' 1
##
## Residual standard error: 0.9023 on 405 degrees of freedom
## (374 observations deleted due to missingness)
## Multiple R-squared: 0.004939, Adjusted R-squared: -0.002432
## F-statistic: 0.6701 on 3 and 405 DF, p-value: 0.5708summary(lm(indirectIE ~ cond*bd, nocontrol))##
## Call:
## lm(formula = indirectIE ~ cond * bd, data = nocontrol)
##
## Residuals:
## Min 1Q Median 3Q Max
## -0.9223 -0.6711 -0.6149 0.3761 3.3670
##
## Coefficients:
## Estimate Std. Error t value Pr(>|t|)
## (Intercept) 1.61746 0.07237 22.349 <2e-16 ***
## condSS 0.08916 0.09970 0.894 0.372
## bd 0.07434 0.34490 0.216 0.829
## condSS:bd -0.34350 0.46098 -0.745 0.457
## ---
## Signif. codes: 0 '***' 0.001 '**' 0.01 '*' 0.05 '.' 0.1 ' ' 1
##
## Residual standard error: 1.006 on 405 degrees of freedom
## (374 observations deleted due to missingness)
## Multiple R-squared: 0.003958, Adjusted R-squared: -0.00342
## F-statistic: 0.5365 on 3 and 405 DF, p-value: 0.6575summary(lm(self ~ cond*bd, nocontrol))##
## Call:
## lm(formula = self ~ cond * bd, data = nocontrol)
##
## Residuals:
## Min 1Q Median 3Q Max
## -0.7872 -0.4409 -0.3647 0.3988 3.7596
##
## Coefficients:
## Estimate Std. Error t value Pr(>|t|)
## (Intercept) 1.48091 0.06020 24.598 <2e-16 ***
## condSS -0.11851 0.08294 -1.429 0.154
## bd -0.45309 0.28692 -1.579 0.115
## condSS:bd 0.15655 0.38348 0.408 0.683
## ---
## Signif. codes: 0 '***' 0.001 '**' 0.01 '*' 0.05 '.' 0.1 ' ' 1
##
## Residual standard error: 0.8365 on 405 degrees of freedom
## (374 observations deleted due to missingness)
## Multiple R-squared: 0.01379, Adjusted R-squared: 0.006482
## F-statistic: 1.887 on 3 and 405 DF, p-value: 0.1311summary(lm(others ~ cond*bd, nocontrol))##
## Call:
## lm(formula = others ~ cond * bd, data = nocontrol)
##
## Residuals:
## Min 1Q Median 3Q Max
## -1.8221 -0.6960 0.2956 0.3677 2.4039
##
## Coefficients:
## Estimate Std. Error t value Pr(>|t|)
## (Intercept) 2.65317 0.06383 41.563 <2e-16 ***
## condSS 0.04918 0.08794 0.559 0.576
## bd -0.24983 0.30422 -0.821 0.412
## condSS:bd 0.27369 0.40660 0.673 0.501
## ---
## Signif. codes: 0 '***' 0.001 '**' 0.01 '*' 0.05 '.' 0.1 ' ' 1
##
## Residual standard error: 0.8869 on 405 degrees of freedom
## (374 observations deleted due to missingness)
## Multiple R-squared: 0.002572, Adjusted R-squared: -0.004817
## F-statistic: 0.3481 on 3 and 405 DF, p-value: 0.7906summary(lm(influence ~ cond*bd, nocontrol))##
## Call:
## lm(formula = influence ~ cond * bd, data = nocontrol)
##
## Residuals:
## Min 1Q Median 3Q Max
## -0.8884 -0.5194 -0.3366 0.4486 3.6042
##
## Coefficients:
## Estimate Std. Error t value Pr(>|t|)
## (Intercept) 1.54408 0.06097 25.326 <2e-16 ***
## condSS -0.15422 0.08399 -1.836 0.0671 .
## bd -0.19252 0.29055 -0.663 0.5080
## condSS:bd -0.44831 0.38834 -1.154 0.2490
## ---
## Signif. codes: 0 '***' 0.001 '**' 0.01 '*' 0.05 '.' 0.1 ' ' 1
##
## Residual standard error: 0.8471 on 405 degrees of freedom
## (374 observations deleted due to missingness)
## Multiple R-squared: 0.02373, Adjusted R-squared: 0.0165
## F-statistic: 3.282 on 3 and 405 DF, p-value: 0.02091#ALSO CONFIRMING THAT PEOPLE IN THE CONTROL CONDITION ANSWERED THESE QUESTIONS, AND THE REACTANCE QUESTIONS
control<-subset(combined2, cond=="no")
summary(control$hong11)## Min. 1st Qu. Median Mean 3rd Qu. Max. NA's
## 1.000 3.000 3.000 3.239 4.000 5.000 102summary(control$refdep)## Min. 1st Qu. Median Mean 3rd Qu. Max. NA's
## 1.000 1.000 1.000 1.599 2.000 5.000 102summary(control$ease)## Min. 1st Qu. Median Mean 3rd Qu. Max. NA's
## 1.000 1.000 2.000 2.193 3.000 5.000 102combined2$reactance <- (combined2$hong8+combined2$hong9+combined2$hong11+combined2$hong13)/4
summary(combined2$reactance)## Min. 1st Qu. Median Mean 3rd Qu. Max. NA's
## 1.000 1.750 2.000 2.091 2.500 4.250 476Demographic analyses
table(combined2$education)##
##
## 0
## 9th, 10th or 11th grade
## 5
## Associate degree (for example: AA, AS)
## 113
## Bachelor's degree (for example: BA, AB, BS)
## 422
## Doctorate degree (for example: PhD, EdD)
## 12
## High school diploma or the equivalent (for example: GED)
## 97
## Master's degree (for example: MA, MS, MEng, MEd, MSW, MBA)
## 92
## Professional degree (for example: MD, DDS, DVM, LLB, JD)
## 14
## Some college, and currently enrolled
## 98
## Some college, but not currently enrolled
## 229combined2$edu[combined2$education=="9th, 10th or 11th grade"]<-1## Warning in `[<-.factor`(`*tmp*`, combined2$education == "9th, 10th or 11th
## grade", : invalid factor level, NA generatedcombined2$edu[combined2$education=="High school diploma or the equivalent (for example: GED)"]<-2## Warning in `[<-.factor`(`*tmp*`, combined2$education == "High school
## diploma or the equivalent (for example: GED)", : invalid factor level, NA
## generatedcombined2$edu[combined2$education=="Associate degree (for example: AA, AS)"]<-3## Warning in `[<-.factor`(`*tmp*`, combined2$education == "Associate degree
## (for example: AA, AS)", : invalid factor level, NA generatedcombined2$edu[combined2$education=="Some college, but not currently enrolled"]<-4## Warning in `[<-.factor`(`*tmp*`, combined2$education == "Some college, but
## not currently enrolled", : invalid factor level, NA generatedcombined2$edu[combined2$education=="Some college, and currently enrolled"]<-5## Warning in `[<-.factor`(`*tmp*`, combined2$education == "Some college, and
## currently enrolled", : invalid factor level, NA generatedcombined2$edu[combined2$education=="Bachelor's degree (for example: BA, AB, BS)"]<-6## Warning in `[<-.factor`(`*tmp*`, combined2$education == "Bachelor's degree
## (for example: BA, AB, BS)", : invalid factor level, NA generatedcombined2$edu[combined2$education=="Professional degree (for example: MD, DDS, DVM, LLB, JD)"]<-6## Warning in `[<-.factor`(`*tmp*`, combined2$education == "Professional
## degree (for example: MD, DDS, DVM, LLB, JD)", : invalid factor level, NA
## generatedcombined2$edu[combined2$education=="Master's degree (for example: MA, MS, MEng, MEd, MSW, MBA)"]<-7## Warning in `[<-.factor`(`*tmp*`, combined2$education == "Master's degree
## (for example: MA, MS, MEng, MEd, MSW, MBA)", : invalid factor level, NA
## generatedcombined2$edu[combined2$education=="Doctorate degree (for example: PhD, EdD)"]<-8## Warning in `[<-.factor`(`*tmp*`, combined2$education == "Doctorate degree
## (for example: PhD, EdD)", : invalid factor level, NA generatedcombined2$edu<-as.numeric(combined2$edu)
combined2$edu[combined2$education=="9th, 10th or 11th grade"]<-1
combined2$edu[combined2$education=="High school diploma or the equivalent (for example: GED)"]<-2
combined2$edu[combined2$education=="Associate degree (for example: AA, AS)"]<-3
combined2$edu[combined2$education=="Some college, but not currently enrolled"]<-4
combined2$edu[combined2$education=="Some college, and currently enrolled"]<-5
combined2$edu[combined2$education=="Bachelor's degree (for example: BA, AB, BS)"]<-6
combined2$edu[combined2$education=="Professional degree (for example: MD, DDS, DVM, LLB, JD)"]<-6
combined2$edu[combined2$education=="Master's degree (for example: MA, MS, MEng, MEd, MSW, MBA)"]<-7
combined2$edu[combined2$education=="Doctorate degree (for example: PhD, EdD)"]<-8
table(combined2$edu)##
## 1 2 3 4 5 6 7 8
## 5 97 113 229 98 436 92 12#centering
summary(combined2$edu)## Min. 1st Qu. Median Mean 3rd Qu. Max.
## 1.000 4.000 5.000 4.898 6.000 8.000combined2$eduCentered<-combined2$edu - 4.889
summary(combined2$eduCentered)## Min. 1st Qu. Median Mean 3rd Qu. Max.
## -3.889000 -0.889000 0.111000 0.009336 1.111000 3.111000#re-coding income as numeric
table(combined2$income)##
## $10,000 to $19,999 $100,000 to $124,999
## 0 105 69
## $125,000 to $149,999 $150,000 to $199,999 $20,000 to $29,999
## 36 24 133
## $200,000 or more $30,000 to $39,999 $40,000 to $49,999
## 12 145 132
## $50,000 to $59,999 $60,000 to $69,999 $70,000 to $79,999
## 102 85 65
## $80,000 to $89,999 $90,000 to $99,999 Less than $10,000
## 52 41 81combined2$inc[combined2$income=="Less than $10,000"]<-1## Warning in `[<-.factor`(`*tmp*`, combined2$income == "Less than $10,000", :
## invalid factor level, NA generatedcombined2$inc[combined2$income=="$10,000 to $19,999"]<-2## Warning in `[<-.factor`(`*tmp*`, combined2$income == "$10,000 to
## $19,999", : invalid factor level, NA generatedcombined2$inc[combined2$income=="$20,000 to $29,999"]<-3## Warning in `[<-.factor`(`*tmp*`, combined2$income == "$20,000 to
## $29,999", : invalid factor level, NA generatedcombined2$inc[combined2$income=="$30,000 to $39,999"]<-4## Warning in `[<-.factor`(`*tmp*`, combined2$income == "$30,000 to
## $39,999", : invalid factor level, NA generatedcombined2$inc[combined2$income=="$40,000 to $49,999"]<-5## Warning in `[<-.factor`(`*tmp*`, combined2$income == "$40,000 to
## $49,999", : invalid factor level, NA generatedcombined2$inc[combined2$income=="$50,000 to $59,999"]<-6## Warning in `[<-.factor`(`*tmp*`, combined2$income == "$50,000 to
## $59,999", : invalid factor level, NA generatedcombined2$inc[combined2$income=="$60,000 to $69,999"]<-7## Warning in `[<-.factor`(`*tmp*`, combined2$income == "$60,000 to
## $69,999", : invalid factor level, NA generatedcombined2$inc[combined2$income=="$70,000 to $79,999"]<-8## Warning in `[<-.factor`(`*tmp*`, combined2$income == "$70,000 to
## $79,999", : invalid factor level, NA generatedcombined2$inc[combined2$income=="$80,000 to $89,999"]<-9## Warning in `[<-.factor`(`*tmp*`, combined2$income == "$80,000 to
## $89,999", : invalid factor level, NA generatedcombined2$inc[combined2$income=="$90,000 to $99,999"]<-10## Warning in `[<-.factor`(`*tmp*`, combined2$income == "$90,000 to
## $99,999", : invalid factor level, NA generatedcombined2$inc[combined2$income=="$100,000 to $124,999"]<-11## Warning in `[<-.factor`(`*tmp*`, combined2$income == "$100,000 to
## $124,999", : invalid factor level, NA generatedcombined2$inc[combined2$income=="$150,000 to $199,999"]<-12## Warning in `[<-.factor`(`*tmp*`, combined2$income == "$150,000 to
## $199,999", : invalid factor level, NA generatedcombined2$inc[combined2$income=="200,000 or more"]<-13## Warning in `[<-.factor`(`*tmp*`, combined2$income == "200,000 or more", :
## invalid factor level, NA generatedcombined2$inc<-as.numeric(combined2$inc)
combined2$inc[combined2$income=="Less than $10,000"]<-1
combined2$inc[combined2$income=="$10,000 to $19,999"]<-2
combined2$inc[combined2$income=="$20,000 to $29,999"]<-3
combined2$inc[combined2$income=="$30,000 to $39,999"]<-4
combined2$inc[combined2$income=="$40,000 to $49,999"]<-5
combined2$inc[combined2$income=="$50,000 to $59,999"]<-6
combined2$inc[combined2$income=="$60,000 to $69,999"]<-7
combined2$inc[combined2$income=="$70,000 to $79,999"]<-8
combined2$inc[combined2$income=="$80,000 to $89,999"]<-9
combined2$inc[combined2$income=="$90,000 to $99,999"]<-10
combined2$inc[combined2$income=="$100,000 to $124,999"]<-11
combined2$inc[combined2$income=="$150,000 to $199,999"]<-12
combined2$inc[combined2$income=="200,000 or more"]<-13
table(combined2$inc)##
## 1 2 3 4 5 6 7 8 9 10 11 12
## 81 105 133 181 132 102 97 65 52 41 69 24#centering
combined2$incCentered<-combined2$inc - mean(combined2$inc)
summary(combined2$incCentered)## Min. 1st Qu. Median Mean 3rd Qu. Max.
## -4.3697 -2.3697 -0.3697 0.0000 1.6303 6.6303summary(aov(SS~cond + eduCentered + eduCentered*cond, data=combined2))## Df Sum Sq Mean Sq F value Pr(>F)
## cond 2 1.72 0.861 3.526 0.0298 *
## eduCentered 1 4.76 4.758 19.478 1.12e-05 ***
## cond:eduCentered 2 0.94 0.468 1.917 0.1475
## Residuals 1076 262.82 0.244
## ---
## Signif. codes: 0 '***' 0.001 '**' 0.01 '*' 0.05 '.' 0.1 ' ' 1summary(aov(SS~cond + incCentered + incCentered*cond, data=combined2))## Df Sum Sq Mean Sq F value Pr(>F)
## cond 2 1.72 0.8613 3.484 0.03102 *
## incCentered 1 2.17 2.1704 8.781 0.00311 **
## cond:incCentered 2 0.38 0.1896 0.767 0.46459
## Residuals 1076 265.96 0.2472
## ---
## Signif. codes: 0 '***' 0.001 '**' 0.01 '*' 0.05 '.' 0.1 ' ' 1summary(aov(SS~cond + gender + gender*cond, data=combined2))## Df Sum Sq Mean Sq F value Pr(>F)
## cond 2 1.22 0.6118 2.447 0.0871 .
## gender 2 0.39 0.1943 0.777 0.4600
## cond:gender 3 0.38 0.1271 0.508 0.6766
## Residuals 904 225.97 0.2500
## ---
## Signif. codes: 0 '***' 0.001 '**' 0.01 '*' 0.05 '.' 0.1 ' ' 1
## 170 observations deleted due to missingnesssummary(aov(SS~cond + age + age*cond, data=combined2))## Df Sum Sq Mean Sq F value Pr(>F)
## cond 2 1.72 0.8613 3.455 0.0319 *
## age 1 0.22 0.2165 0.868 0.3516
## cond:age 2 0.07 0.0374 0.150 0.8606
## Residuals 1076 268.22 0.2493
## ---
## Signif. codes: 0 '***' 0.001 '**' 0.01 '*' 0.05 '.' 0.1 ' ' 1summary(aov(SS ~ cond*age*bd, data=combined2))## Df Sum Sq Mean Sq F value Pr(>F)
## cond 2 1.72 0.861 3.799 0.0227 *
## age 1 0.22 0.216 0.955 0.3287
## bd 1 19.72 19.723 86.994 < 2e-16 ***
## cond:age 2 0.01 0.004 0.017 0.9835
## cond:bd 2 5.11 2.553 11.261 1.45e-05 ***
## age:bd 1 0.24 0.244 1.077 0.2995
## cond:age:bd 2 0.63 0.315 1.388 0.2499
## Residuals 1070 242.58 0.227
## ---
## Signif. codes: 0 '***' 0.001 '**' 0.01 '*' 0.05 '.' 0.1 ' ' 1summary(aov(SS ~ cond*eduCentered*bd, data=combined2))## Df Sum Sq Mean Sq F value Pr(>F)
## cond 2 1.72 0.861 3.825 0.0221 *
## eduCentered 1 4.76 4.758 21.129 4.81e-06 ***
## bd 1 17.17 17.172 76.264 < 2e-16 ***
## cond:eduCentered 2 0.91 0.453 2.013 0.1341
## cond:bd 2 4.47 2.236 9.931 5.33e-05 ***
## eduCentered:bd 1 0.22 0.224 0.994 0.3191
## cond:eduCentered:bd 2 0.05 0.027 0.119 0.8882
## Residuals 1070 240.93 0.225
## ---
## Signif. codes: 0 '***' 0.001 '**' 0.01 '*' 0.05 '.' 0.1 ' ' 1summary(aov(SS ~ cond*incCentered*bd, data=combined2))## Df Sum Sq Mean Sq F value Pr(>F)
## cond 2 1.72 0.861 3.814 0.02236 *
## incCentered 1 2.17 2.170 9.612 0.00198 **
## bd 1 18.74 18.736 82.972 < 2e-16 ***
## cond:incCentered 2 0.30 0.151 0.667 0.51349
## cond:bd 2 5.16 2.581 11.429 1.23e-05 ***
## incCentered:bd 1 0.18 0.178 0.789 0.37451
## cond:incCentered:bd 2 0.34 0.171 0.757 0.46929
## Residuals 1070 241.62 0.226
## ---
## Signif. codes: 0 '***' 0.001 '**' 0.01 '*' 0.05 '.' 0.1 ' ' 1summary(aov(SS ~ cond*gender*bd, data=combined2))## Df Sum Sq Mean Sq F value Pr(>F)
## cond 2 1.22 0.612 2.704 0.0675 .
## gender 2 0.39 0.194 0.859 0.4241
## bd 1 16.01 16.009 70.753 < 2e-16 ***
## cond:gender 3 0.36 0.119 0.524 0.6656
## cond:bd 2 6.14 3.070 13.566 1.57e-06 ***
## gender:bd 2 0.55 0.275 1.216 0.2968
## cond:gender:bd 2 0.33 0.165 0.730 0.4824
## Residuals 897 202.96 0.226
## ---
## Signif. codes: 0 '***' 0.001 '**' 0.01 '*' 0.05 '.' 0.1 ' ' 1
## 170 observations deleted due to missingnessFIGURES
combined2$cond2[combined2$cond=="no"]<-"Control"
combined2$cond2[combined2$cond=="LL"]<-"LL"
combined2$cond2[combined2$cond=="SS"]<-"SS"
SScontrol<-subset(combined2, cond2!="LL")
f1<-(glm(SS~cond2*bd_not_centered, data=SScontrol, family="binomial"))
library(visreg)
col<-c("black", "blue")
col2<-c("gray80", "cornflowerblue")
col3<-adjustcolor(col2, alpha=.7)
visreg(f1, "bd_not_centered", by="cond2", line=list(col=col), fill=list(col=col3), overlay=TRUE, partial=FALSE, scale="response", xlab="Subjective Present Value of LL", ylab="Percent Choosing SS")
LLcontrol<-subset(combined2, cond2!="SS")
f2<-(glm(SS~cond2*bd_not_centered, data=LLcontrol,family="binomial"))
col<-c("black", "red")
col2<-c( "gray80", "indianred1" )
col3<-adjustcolor(col2, alpha=.45)
visreg(f2, "bd_not_centered", by="cond2", line=list(col=col), fill=list(col=col3), overlay=TRUE, partial=FALSE, xlab="Subjective present value of LL", scale="response", ylab="P (choosing SS)")
#REACTANCE GRAPHS AS REQUESTED, ALTHOUGH THERE IS ONLY REACTANCE DATA FOR ABOUT HALF THE SAMPLE
f3<-(lm(reactance~cond2*bd_not_centered, data=combined2))
col<-c("black", "red", "blue")
col2<-c( "gray80", "indianred1" , "cornflowerblue")
col3<-adjustcolor(col2, alpha=.45)
visreg(f3, "bd_not_centered", by="cond2", line=list(col=col), fill=list(col=col3), overlay=TRUE, partial=FALSE, xlab="Subjective present value of LL", scale="response", ylab="Reactance")
cdata <- ddply(combined2, c("quintiles", "cond"), summarise,
N = length(reactance),
mean = mean(reactance, na.rm = T),
sd = sd(reactance, na.rm = T),
se = sd / sqrt(N))
## this is with 95% confidence interval
ggplot(cdata, aes(x=quintiles, y=mean, colour=cond)) +
geom_errorbar(aes(ymin=mean-se, ymax=mean+se), width=.1) +
geom_line() +
geom_point()