Combined Stimulus Sets Meta-Awareness
Adam Barnas
Last compiled at 1:43 AM on November 05, 2020
1 Set up R environment.
library(tidyverse)
library(ggplot2)
library(ggpubr)
library(plyr)
library(magick)
library(png)
library(EBImage)
library(lme4)
library(lmerTest)2 Set the R working drectory to the main experiment directory.
setwd("/Users/adambarnas/Box/MetaAwareness/data/") 3 Read in data files.
Rensink_RTs_likelihood_no_NA <- read_csv("Rensink_RTs_likelihood_no_NA.csv")
Ma_RTs_likelihood_no_NA <- read_csv("Ma_RTs_likelihood_no_NA.csv")
Wolfe1_RTs_likelihood_no_NA <- read_csv("Wolfe1_RTs_likelihood_no_NA.csv")
Wolfe2_RTs_likelihood_no_NA <- read_csv("Wolfe2_RTs_likelihood_no_NA.csv")
tbl_all <- rbind(Rensink_RTs_likelihood_no_NA, Ma_RTs_likelihood_no_NA, Wolfe1_RTs_likelihood_no_NA, Wolfe2_RTs_likelihood_no_NA)4 Compute average likelihood rating.
tbl_all_subj_avg <- tbl_all %>%
group_by(workerId,image) %>%
dplyr::summarize(average = mean(likelihood_rating)) %>%
spread(image,average) %>%
mutate(subj_avg = rowMeans(.[-1], na.rm = TRUE))
mean(tbl_all_subj_avg$subj_avg)## [1] 3.1003485 Mixed effects model and correlation.
5.1 Raw RT
fit_raw <- lmer(detection_rt ~ likelihood_rating * change_type + (1 | workerId) + (1 | image) + (1 | stim_set), data=tbl_all)
summary(fit_raw)## Linear mixed model fit by REML. t-tests use Satterthwaite's method [
## lmerModLmerTest]
## Formula: detection_rt ~ likelihood_rating * change_type + (1 | workerId) +
## (1 | image) + (1 | stim_set)
## Data: tbl_all
##
## REML criterion at convergence: 23816.3
##
## Scaled residuals:
## Min 1Q Median 3Q Max
## -2.5417 -0.4219 -0.1208 0.1419 8.6021
##
## Random effects:
## Groups Name Variance Std.Dev.
## image (Intercept) 1.5986 1.2644
## workerId (Intercept) 8.4997 2.9154
## stim_set (Intercept) 0.2638 0.5136
## Residual 21.7899 4.6680
## Number of obs: 3923, groups: image, 480; workerId, 173; stim_set, 4
##
## Fixed effects:
## Estimate Std. Error df
## (Intercept) 12.10463 0.64645 11.87018
## likelihood_rating -0.52308 0.14306 2717.01985
## change_typedisappear -1.31017 0.62308 1151.03530
## change_typemovement -2.55755 0.90860 1657.93690
## change_typereplacement -1.39780 2.53863 2769.13562
## change_typesize -2.98798 1.23039 1373.57443
## likelihood_rating:change_typedisappear 0.11657 0.16280 2681.36634
## likelihood_rating:change_typemovement 0.35430 0.27413 3737.27688
## likelihood_rating:change_typereplacement 0.06829 0.71758 3656.85074
## likelihood_rating:change_typesize 0.32998 0.36705 2892.24576
## t value Pr(>|t|)
## (Intercept) 18.725 3.54e-10 ***
## likelihood_rating -3.656 0.000261 ***
## change_typedisappear -2.103 0.035706 *
## change_typemovement -2.815 0.004938 **
## change_typereplacement -0.551 0.581945
## change_typesize -2.428 0.015290 *
## likelihood_rating:change_typedisappear 0.716 0.474017
## likelihood_rating:change_typemovement 1.292 0.196284
## likelihood_rating:change_typereplacement 0.095 0.924186
## likelihood_rating:change_typesize 0.899 0.368720
## ---
## Signif. codes: 0 '***' 0.001 '**' 0.01 '*' 0.05 '.' 0.1 ' ' 1
##
## Correlation of Fixed Effects:
## (Intr) lklhd_ chng_typd chng_typm chng_typr chng_typs
## liklhd_rtng -0.749
## chng_typdsp -0.709 0.735
## chng_typmvm -0.505 0.548 0.490
## chng_typrpl -0.187 0.200 0.180 0.160
## chang_typsz -0.375 0.401 0.360 0.340 0.132
## lklhd_rtng:chng_typd 0.636 -0.840 -0.873 -0.464 -0.169 -0.339
## lklhd_rtng:chng_typm 0.370 -0.499 -0.376 -0.845 -0.108 -0.239
## lklhd_rtng:chng_typr 0.152 -0.201 -0.151 -0.118 -0.891 -0.090
## lklhd_rtng:chng_typs 0.284 -0.376 -0.283 -0.239 -0.082 -0.845
## lklhd_rtng:chng_typd lklhd_rtng:chng_typm
## liklhd_rtng
## chng_typdsp
## chng_typmvm
## chng_typrpl
## chang_typsz
## lklhd_rtng:chng_typd
## lklhd_rtng:chng_typm 0.432
## lklhd_rtng:chng_typr 0.171 0.110
## lklhd_rtng:chng_typs 0.321 0.227
## lklhd_rtng:chng_typr
## liklhd_rtng
## chng_typdsp
## chng_typmvm
## chng_typrpl
## chang_typsz
## lklhd_rtng:chng_typd
## lklhd_rtng:chng_typm
## lklhd_rtng:chng_typr
## lklhd_rtng:chng_typs 0.085corr <- tbl_all %>%
group_by(image) %>%
dplyr::summarize(detection_rt = mean(detection_rt), likelihood_rating = mean(likelihood_rating), change_type = unique(change_type))
corr %>%
ggscatter(y = "detection_rt", x = "likelihood_rating", ylab = "Raw Change Detection RT (sec)", xlab = "Likelihood of Detecting Change", title = "N = 173", add = "reg.line", cor.coef = TRUE, cor.coeff.args = list(method = "pearson", label.x = 1, label.y = 37.5), conf.int = TRUE, xlim = c(1, 5), ylim = c(5, 40))
corr %>%
ggscatter(y = "detection_rt", x = "likelihood_rating", color = "change_type", palette = c("#0d2240", "#00a8e1", "#f7a800", "#E31818", "#dfdddc"), ylab = "Raw Change Detection RT (sec)", xlab = "Likelihood of Detecting Change", title = "N = 173", add = "reg.line", conf.int = TRUE, xlim = c(1, 5), ylim = c(5, 40)) + stat_cor(aes(color = change_type), label.x = c(1, 1, 2.5, 2.5, 4), label.y = c(37.5, 35, 37.5, 35, 37.5), method = "pearson")
Drop image-116 from Wolfe1.
tbl_all_no_outlier <- tbl_all %>%
filter(image != "image-116")
fit_raw_no_outlier <- lmer(detection_rt ~ likelihood_rating * change_type + (1 | workerId) + (1 | image), data=tbl_all_no_outlier)
summary(fit_raw_no_outlier)## Linear mixed model fit by REML. t-tests use Satterthwaite's method [
## lmerModLmerTest]
## Formula: detection_rt ~ likelihood_rating * change_type + (1 | workerId) +
## (1 | image)
## Data: tbl_all_no_outlier
##
## REML criterion at convergence: 23750.4
##
## Scaled residuals:
## Min 1Q Median 3Q Max
## -2.5600 -0.4240 -0.1206 0.1410 8.6445
##
## Random effects:
## Groups Name Variance Std.Dev.
## image (Intercept) 1.448 1.203
## workerId (Intercept) 8.474 2.911
## Residual 21.575 4.645
## Number of obs: 3921, groups: image, 479; workerId, 173
##
## Fixed effects:
## Estimate Std. Error df t value
## (Intercept) 12.3927 0.5733 1612.2338 21.617
## likelihood_rating -0.5941 0.1416 2788.3993 -4.197
## change_typedisappear -1.4225 0.6084 1721.5165 -2.338
## change_typemovement -2.8247 0.8932 2272.0900 -3.162
## change_typereplacement -1.6872 2.5110 2900.2981 -0.672
## change_typesize -3.2660 1.2091 1666.2308 -2.701
## likelihood_rating:change_typedisappear 0.1779 0.1611 2696.9948 1.104
## likelihood_rating:change_typemovement 0.4168 0.2720 3728.1867 1.532
## likelihood_rating:change_typereplacement 0.1452 0.7120 3648.9687 0.204
## likelihood_rating:change_typesize 0.4024 0.3635 2882.1582 1.107
## Pr(>|t|)
## (Intercept) < 2e-16 ***
## likelihood_rating 2.79e-05 ***
## change_typedisappear 0.01949 *
## change_typemovement 0.00159 **
## change_typereplacement 0.50169
## change_typesize 0.00698 **
## likelihood_rating:change_typedisappear 0.26965
## likelihood_rating:change_typemovement 0.12552
## likelihood_rating:change_typereplacement 0.83839
## likelihood_rating:change_typesize 0.26833
## ---
## Signif. codes: 0 '***' 0.001 '**' 0.01 '*' 0.05 '.' 0.1 ' ' 1
##
## Correlation of Fixed Effects:
## (Intr) lklhd_ chng_typd chng_typm chng_typr chng_typs
## liklhd_rtng -0.824
## chng_typdsp -0.797 0.744
## chng_typmvm -0.544 0.547 0.494
## chng_typrpl -0.200 0.198 0.180 0.155
## chang_typsz -0.403 0.400 0.363 0.334 0.127
## lklhd_rtng:chng_typd 0.704 -0.841 -0.881 -0.464 -0.168 -0.340
## lklhd_rtng:chng_typm 0.408 -0.498 -0.379 -0.850 -0.107 -0.238
## lklhd_rtng:chng_typr 0.167 -0.201 -0.152 -0.117 -0.894 -0.090
## lklhd_rtng:chng_typs 0.312 -0.376 -0.286 -0.239 -0.081 -0.850
## lklhd_rtng:chng_typd lklhd_rtng:chng_typm
## liklhd_rtng
## chng_typdsp
## chng_typmvm
## chng_typrpl
## chang_typsz
## lklhd_rtng:chng_typd
## lklhd_rtng:chng_typm 0.431
## lklhd_rtng:chng_typr 0.171 0.109
## lklhd_rtng:chng_typs 0.321 0.226
## lklhd_rtng:chng_typr
## liklhd_rtng
## chng_typdsp
## chng_typmvm
## chng_typrpl
## chang_typsz
## lklhd_rtng:chng_typd
## lklhd_rtng:chng_typm
## lklhd_rtng:chng_typr
## lklhd_rtng:chng_typs 0.085corr %>%
filter(image!="image-116") %>%
ggscatter(y = "detection_rt", x = "likelihood_rating", ylab = "Raw Change Detection RT (sec)", xlab = "Likelihood of Detecting Change", title = "N = 173", add = "reg.line", cor.coef = TRUE, cor.coeff.args = list(method = "pearson", label.x = 1, label.y = 25), conf.int = TRUE, xlim = c(1, 5), ylim = c(5, 30))
corr %>%
filter(image!="image-116") %>%
ggscatter(y = "detection_rt", x = "likelihood_rating", color = "change_type", palette = c("#0d2240", "#00a8e1", "#f7a800", "#E31818", "#dfdddc"), ylab = "Raw Change Detection RT (sec)", xlab = "Likelihood of Detecting Change", title = "N = 173", add = "reg.line", conf.int = TRUE, xlim = c(1, 5), ylim = c(5, 30)) + stat_cor(aes(color = change_type), label.x = c(1, 1, 2.5, 2.5, 4), label.y = c(27.5, 25, 27.5, 25, 27.5), method = "pearson")
5.2 Log RT
fit_log <- lmer(log10(detection_rt) ~ likelihood_rating * change_type + (1 | workerId) + (1 | image) + (1 | stim_set), data=tbl_all)
summary(fit_log)## Linear mixed model fit by REML. t-tests use Satterthwaite's method [
## lmerModLmerTest]
## Formula: log10(detection_rt) ~ likelihood_rating * change_type + (1 |
## workerId) + (1 | image) + (1 | stim_set)
## Data: tbl_all
##
## REML criterion at convergence: -4197.3
##
## Scaled residuals:
## Min 1Q Median 3Q Max
## -2.7630 -0.5472 -0.1255 0.3285 5.1747
##
## Random effects:
## Groups Name Variance Std.Dev.
## image (Intercept) 0.0015873 0.03984
## workerId (Intercept) 0.0123117 0.11096
## stim_set (Intercept) 0.0005797 0.02408
## Residual 0.0162902 0.12763
## Number of obs: 3923, groups: image, 480; workerId, 173; stim_set, 4
##
## Fixed effects:
## Estimate Std. Error df
## (Intercept) 1.010e+00 2.169e-02 6.701e+00
## likelihood_rating -1.602e-02 4.004e-03 2.993e+03
## change_typedisappear -3.489e-02 1.766e-02 1.481e+03
## change_typemovement -6.969e-02 2.563e-02 1.938e+03
## change_typereplacement -2.251e-02 7.078e-02 2.732e+03
## change_typesize -8.286e-02 3.481e-02 1.522e+03
## likelihood_rating:change_typedisappear 4.323e-03 4.552e-03 2.903e+03
## likelihood_rating:change_typemovement 1.012e-02 7.616e-03 3.766e+03
## likelihood_rating:change_typereplacement 3.478e-04 1.984e-02 3.689e+03
## likelihood_rating:change_typesize 8.675e-03 1.023e-02 3.078e+03
## t value Pr(>|t|)
## (Intercept) 46.540 1.15e-09 ***
## likelihood_rating -4.002 6.44e-05 ***
## change_typedisappear -1.975 0.04841 *
## change_typemovement -2.719 0.00662 **
## change_typereplacement -0.318 0.75050
## change_typesize -2.381 0.01741 *
## likelihood_rating:change_typedisappear 0.950 0.34235
## likelihood_rating:change_typemovement 1.328 0.18416
## likelihood_rating:change_typereplacement 0.018 0.98601
## likelihood_rating:change_typesize 0.848 0.39632
## ---
## Signif. codes: 0 '***' 0.001 '**' 0.01 '*' 0.05 '.' 0.1 ' ' 1
##
## Correlation of Fixed Effects:
## (Intr) lklhd_ chng_typd chng_typm chng_typr chng_typs
## liklhd_rtng -0.627
## chng_typdsp -0.593 0.722
## chng_typmvm -0.429 0.544 0.486
## chng_typrpl -0.162 0.202 0.180 0.166
## chang_typsz -0.318 0.397 0.356 0.347 0.138
## lklhd_rtng:chng_typd 0.530 -0.837 -0.862 -0.460 -0.170 -0.336
## lklhd_rtng:chng_typm 0.310 -0.499 -0.371 -0.833 -0.110 -0.240
## lklhd_rtng:chng_typr 0.129 -0.204 -0.150 -0.119 -0.882 -0.092
## lklhd_rtng:chng_typs 0.238 -0.376 -0.279 -0.240 -0.083 -0.833
## lklhd_rtng:chng_typd lklhd_rtng:chng_typm
## liklhd_rtng
## chng_typdsp
## chng_typmvm
## chng_typrpl
## chang_typsz
## lklhd_rtng:chng_typd
## lklhd_rtng:chng_typm 0.432
## lklhd_rtng:chng_typr 0.173 0.112
## lklhd_rtng:chng_typs 0.321 0.230
## lklhd_rtng:chng_typr
## liklhd_rtng
## chng_typdsp
## chng_typmvm
## chng_typrpl
## chang_typsz
## lklhd_rtng:chng_typd
## lklhd_rtng:chng_typm
## lklhd_rtng:chng_typr
## lklhd_rtng:chng_typs 0.088corr <- tbl_all %>%
group_by(image) %>%
dplyr::summarize(detection_rt = mean(detection_rt), likelihood_rating = mean(likelihood_rating), change_type = unique(change_type))
corr$log <- log10(corr$detection_rt)
corr %>%
ggscatter(y = "log", x = "likelihood_rating", ylab = "LogChange Detection RT (sec)", xlab = "Likelihood of Detecting Change", title = "N = 173", add = "reg.line", cor.coef = TRUE, cor.coeff.args = list(method = "pearson", label.x = 1, label.y = 1.75), conf.int = TRUE, xlim = c(1, 5), ylim = c(0.75, 1.75))
corr %>%
ggscatter(y = "log", x = "likelihood_rating", color = "change_type", palette = c("#0d2240", "#00a8e1", "#f7a800", "#E31818", "#dfdddc"), ylab = "Log Change Detection RT (sec)", xlab = "Likelihood of Detecting Change", title = "N = 173", add = "reg.line", conf.int = TRUE, xlim = c(1, 5), ylim = c(0.75, 1.75)) + stat_cor(aes(color = change_type), label.x = 1, label.y = c(1.50, 1.5625, 1.625, 1.6875, 1.75), method = "pearson")
Drop image-116 from Wolfe1.
tbl_all_no_outlier <- tbl_all %>%
filter(image != "image-116")
fit_log_no_outlier <- lmer(log10(detection_rt) ~ likelihood_rating * change_type + (1 | workerId) + (1 | image), data=tbl_all_no_outlier)
summary(fit_log_no_outlier)## Linear mixed model fit by REML. t-tests use Satterthwaite's method [
## lmerModLmerTest]
## Formula: log10(detection_rt) ~ likelihood_rating * change_type + (1 |
## workerId) + (1 | image)
## Data: tbl_all_no_outlier
##
## REML criterion at convergence: -4219.3
##
## Scaled residuals:
## Min 1Q Median 3Q Max
## -2.7709 -0.5449 -0.1241 0.3339 5.2009
##
## Random effects:
## Groups Name Variance Std.Dev.
## image (Intercept) 0.001509 0.03885
## workerId (Intercept) 0.012464 0.11164
## Residual 0.016226 0.12738
## Number of obs: 3921, groups: image, 479; workerId, 173
##
## Fixed effects:
## Estimate Std. Error df
## (Intercept) 1.020e+00 1.730e-02 1.331e+03
## likelihood_rating -1.757e-02 3.984e-03 3.000e+03
## change_typedisappear -3.706e-02 1.739e-02 1.708e+03
## change_typemovement -7.676e-02 2.535e-02 2.174e+03
## change_typereplacement -3.022e-02 7.035e-02 2.799e+03
## change_typesize -9.029e-02 3.443e-02 1.646e+03
## likelihood_rating:change_typedisappear 5.620e-03 4.530e-03 2.894e+03
## likelihood_rating:change_typemovement 1.146e-02 7.590e-03 3.759e+03
## likelihood_rating:change_typereplacement 2.017e-03 1.977e-02 3.682e+03
## likelihood_rating:change_typesize 1.025e-02 1.018e-02 3.062e+03
## t value Pr(>|t|)
## (Intercept) 58.994 < 2e-16 ***
## likelihood_rating -4.409 1.08e-05 ***
## change_typedisappear -2.131 0.03320 *
## change_typemovement -3.027 0.00250 **
## change_typereplacement -0.430 0.66750
## change_typesize -2.623 0.00881 **
## likelihood_rating:change_typedisappear 1.241 0.21482
## likelihood_rating:change_typemovement 1.509 0.13128
## likelihood_rating:change_typereplacement 0.102 0.91872
## likelihood_rating:change_typesize 1.007 0.31419
## ---
## Signif. codes: 0 '***' 0.001 '**' 0.01 '*' 0.05 '.' 0.1 ' ' 1
##
## Correlation of Fixed Effects:
## (Intr) lklhd_ chng_typd chng_typm chng_typr chng_typs
## liklhd_rtng -0.769
## chng_typdsp -0.751 0.729
## chng_typmvm -0.513 0.544 0.489
## chng_typrpl -0.191 0.200 0.181 0.162
## chang_typsz -0.379 0.397 0.359 0.342 0.134
## lklhd_rtng:chng_typd 0.656 -0.838 -0.868 -0.460 -0.170 -0.336
## lklhd_rtng:chng_typm 0.382 -0.499 -0.374 -0.837 -0.109 -0.239
## lklhd_rtng:chng_typr 0.158 -0.204 -0.151 -0.119 -0.884 -0.091
## lklhd_rtng:chng_typs 0.292 -0.376 -0.282 -0.240 -0.083 -0.837
## lklhd_rtng:chng_typd lklhd_rtng:chng_typm
## liklhd_rtng
## chng_typdsp
## chng_typmvm
## chng_typrpl
## chang_typsz
## lklhd_rtng:chng_typd
## lklhd_rtng:chng_typm 0.432
## lklhd_rtng:chng_typr 0.173 0.112
## lklhd_rtng:chng_typs 0.321 0.230
## lklhd_rtng:chng_typr
## liklhd_rtng
## chng_typdsp
## chng_typmvm
## chng_typrpl
## chang_typsz
## lklhd_rtng:chng_typd
## lklhd_rtng:chng_typm
## lklhd_rtng:chng_typr
## lklhd_rtng:chng_typs 0.087corr$log <- log10(corr$detection_rt)
corr %>%
filter(image!="image-116") %>%
ggscatter(y = "log", x = "likelihood_rating", ylab = "Raw Change Detection RT (sec)", xlab = "Likelihood of Detecting Change", title = "N = 173", add = "reg.line", cor.coef = TRUE, cor.coeff.args = list(method = "pearson", label.x = 1, label.y = 1.4), conf.int = TRUE, xlim = c(1, 5), ylim = c(.75, 1.5))
corr %>%
filter(image!="image-116") %>%
ggscatter(y = "log", x = "likelihood_rating", color = "change_type", palette = c("#0d2240", "#00a8e1", "#f7a800", "#E31818", "#dfdddc"), ylab = "Raw Change Detection RT (sec)", xlab = "Likelihood of Detecting Change", title = "N = 173", add = "reg.line", conf.int = TRUE, xlim = c(1, 5), ylim = c(.75, 1.5)) + stat_cor(aes(color = change_type), label.x = c(1, 1, 2.5, 2.5, 4), label.y = c(1.4, 1.35, 1.4, 1.35, 1.4), method = "pearson")
6 Predicting self vs. other.
Note: Some subjects have only 1 “good” image and some images have only 1 meta-awareness rating. Correlations cannot be computed for these instances (both NA). Therefore, images with only 1 meta-awareness rating and subjects with only 1 good image are omitted.
6.1 Correlation between subject’s change detection RT and meta-awareness likelihood rating.
func1 <- function(tbl_all_no_outlier)
{
return(data.frame(corr_indiv = cor(tbl_all_no_outlier$detection_rt, tbl_all_no_outlier$likelihood_rating)))
}
tbl_corr_indiv <- ddply(tbl_all_no_outlier, .(workerId), func1)
tbl_corr_indiv$fisherZ_indiv <- .5*((log(1+tbl_corr_indiv$corr_indiv)) - (log(1-tbl_corr_indiv$corr_indiv)))
tbl_corr_indiv_no_NAs <- na.omit(tbl_corr_indiv)6.2 Correlation between subject’s meta-awareness likelihood rating and group average change detection RT (without subject).
workerIds <- unique(tbl_all_no_outlier$workerId)
datalist = list()
for (i in 1:length(workerIds)){
df1 <- tbl_all_no_outlier[tbl_all_no_outlier$workerId==workerIds[i],]
image <- df1$image
image <- data.frame(image)
rts <- tbl_all_no_outlier %>%
filter(workerId != workerIds[i])
rts <- rts[(rts$image %in% image$image),]
rts <- rts %>%
group_by(image) %>%
dplyr::summarize(mean_detection_rt = mean(detection_rt))
datalist[[i]] <- left_join(df1, rts, by = "image")
}
df2 = do.call(rbind, datalist)
df2_no_NAs <- na.omit(df2)
func2 <- function(df2_no_NAs)
{
return(data.frame(corr_group = cor(df2_no_NAs$mean_detection_rt, df2_no_NAs$likelihood_rating)))
}
tbl_corr_group <- ddply(df2_no_NAs, .(workerId), func2)
tbl_corr_group$fisherZ_group <- .5*((log(1+tbl_corr_group$corr_group)) - (log(1-tbl_corr_group$corr_group)))
tbl_corr_group_no_NAs <- na.omit(tbl_corr_group)6.3 Compare means.
tbl_predict <- left_join(tbl_corr_indiv_no_NAs, tbl_corr_group_no_NAs, by = "workerId")6.3.1 Correlations
tbl_predict_corrs <- tbl_predict[, -c(3,5)]
tbl_predict_corrs <- tbl_predict_corrs %>%
gather(key = "predicting", value = "corr", corr_indiv, corr_group)
tbl_predict_corrs %>%
group_by(predicting) %>%
get_summary_stats(corr, type = "mean_se")## # A tibble: 2 x 5
## predicting variable n mean se
## <chr> <chr> <dbl> <dbl> <dbl>
## 1 corr_group corr 168 -0.171 0.017
## 2 corr_indiv corr 168 -0.119 0.019tbl_predict_corrs %>%
with(t.test(corr~predicting,paired=TRUE))##
## Paired t-test
##
## data: corr by predicting
## t = -2.3753, df = 167, p-value = 0.01867
## alternative hypothesis: true difference in means is not equal to 0
## 95 percent confidence interval:
## -0.095586188 -0.008813586
## sample estimates:
## mean of the differences
## -0.052199896.3.2 FisherZ-transformed correlations
tbl_predict_fisherZ <- tbl_predict[, -c(2,4)]
tbl_predict_fisherZ <- tbl_predict_fisherZ %>%
gather(key = "predicting", value = "fisherZ", fisherZ_indiv, fisherZ_group)
tbl_predict_fisherZ %>%
group_by(predicting) %>%
get_summary_stats(fisherZ, type = "mean_se")## # A tibble: 2 x 5
## predicting variable n mean se
## <chr> <chr> <dbl> <dbl> <dbl>
## 1 fisherZ_group fisherZ 168 -0.286 0.11
## 2 fisherZ_indiv fisherZ 168 -0.233 0.112tbl_predict_fisherZ %>%
with(t.test(fisherZ~predicting,paired=TRUE))##
## Paired t-test
##
## data: fisherZ by predicting
## t = -2.2257, df = 167, p-value = 0.02737
## alternative hypothesis: true difference in means is not equal to 0
## 95 percent confidence interval:
## -0.100194270 -0.005998137
## sample estimates:
## mean of the differences
## -0.0530962