Step 1: Import

Load libraries, import data, separate into pre, post, and follow-up dataframes, and create race_minority and gender_minority variables.

Little’s MCAR was used earlier to check if data is missing completely at random. Test is not significant for full dataset (X2 (192, N = 258) = 203.92, p = .264) or for dataset with unit nonresponders dropped (X2 (127, N = 204) = 118.52, p = .692). Data is missing complete at random and we’re okay to proceed with analysis.

Multiple imputation done using Amelia package in R. Tutorial: https://cran.r-project.org/web/packages/Amelia/vignettes/using-amelia.html.

library(psych)
library(DT)
library(ggplot2)
library(afex)
library(emmeans)
library(rstatix)
library(dplyr)
library(stringr)
library(kableExtra)
library(Rmisc)
library(nFactors)
library(car)
library(ggpubr)

# import file
import <- read.csv(file="Z:\\purdue\\Values Affirmation Intervention\\Values Affirmation Intervention Final Data Imputed 10-29-22.csv", header = T)

long <- import
names(import)
##  [1] "UID"              "Q2_1"             "Q2_2"             "Q2_3"            
##  [5] "Q2_4"             "Q2_5"             "Q2_6"             "Q2_7"            
##  [9] "Q2_8"             "Q2_9"             "Q2_10"            "Q2_11"           
## [13] "Q3_1"             "Q3_2"             "Q3_3"             "Q3_4"            
## [17] "Q3_5"             "Q3_6"             "Q3_7"             "Q3_8"            
## [21] "Q3_9"             "Q3_10"            "Q3_11"            "Q3_12"           
## [25] "Q4_1"             "Q4_2"             "Q4_3"             "Q4_4"            
## [29] "Q4_5"             "condition"        "ins"              "Q47_1_pmaps"     
## [33] "Q47_2_pmaps"      "Q47_3_pmaps"      "Q47_4_pmaps"      "Q47_5_pmaps"     
## [37] "Q47_6_pmaps"      "Q47_7_pmaps"      "Q47_8_pmaps"      "re_other"        
## [41] "re_asian"         "re_black"         "re_white"         "re_latin"        
## [45] "re_mena"          "re_nhpi"          "re_aian"          "ge_other"        
## [49] "ge_cis"           "ge_m"             "ge_gq"            "ge_nb"           
## [53] "ge_w"             "pre_misscount"    "post_misscount"   "follow_misscount"
## [57] "timepoint"
pre <- subset(import, timepoint == 1, select=c(2:29))
colnames(pre) <- paste(colnames(pre),"_pre",sep="")
post <- subset(import, timepoint == 2, select=c(2:29))
colnames(post) <- paste(colnames(post),"_post",sep="")
follow <- subset(import, timepoint == 3, select=c(2:29))
colnames(follow) <- paste(colnames(follow),"_follow",sep="")
bws <- subset(import, select=c(UID, 30:57), timepoint == 1)
wide <- cbind.data.frame(bws, pre, post, follow)
rm(pre, post, follow, bws, import)

Step 2: Assumption Checking

Examining Skew & Kurtosis

Decided to drop interest subscale of identity measure based on kurtosis and skew (2_9, 2_10, and 2_11).

Pre-Test Table

desc_pre <- data.frame(describe(subset(wide, select=c(grep("_pre",colnames(wide))))))
desc_post <- describe(subset(wide, select=c(grep("_post",colnames(wide)))))
desc_follow <- describe(subset(wide, select=c(grep("_follow",colnames(wide)))))
desc_pmaps <- describe(subset(wide, select=c(grep("_pmaps",colnames(wide)))))

kbl(round(desc_pre, digits = 2)) %>%
  kable_styling() %>%
  row_spec(which(desc_pre$kurtosis > 2), bold = T) %>%
  row_spec(which(desc_pre$kurtosis < -2), bold = T) %>%
  row_spec(which(desc_pre$skew > 2), italic = T) %>%
  row_spec(which(desc_pre$skew < -2), italic = T)
vars n mean sd median trimmed mad min max range skew kurtosis se
Q2_1_pre 1 231 5.42 1.15 5.06 5.49 1.39 2.00 7.00 5.00 -0.47 -0.25 0.08
Q2_2_pre 2 231 5.54 1.24 6.00 5.67 1.48 2.00 7.00 5.00 -0.70 -0.10 0.08
Q2_3_pre 3 231 6.03 1.00 6.00 6.16 1.48 2.00 7.00 5.00 -1.08 1.28 0.07
Q2_4_pre 4 231 5.53 1.11 6.00 5.61 1.48 2.00 7.00 5.00 -0.61 0.09 0.07
Q2_5_pre 5 231 5.95 1.17 6.00 6.11 1.48 2.00 7.67 5.67 -1.02 0.34 0.08
Q2_6_pre 6 231 5.53 1.18 6.00 5.59 1.48 2.00 7.41 5.41 -0.39 -0.74 0.08
Q2_7_pre 7 231 5.53 1.16 6.00 5.62 1.48 2.00 7.00 5.00 -0.75 0.23 0.08
Q2_8_pre 8 231 4.96 1.74 5.00 5.14 1.48 1.00 7.26 6.26 -0.61 -0.49 0.11
Q2_9_pre 9 231 6.62 0.73 7.00 6.76 0.00 1.00 7.00 6.00 -3.26 17.06 0.05
Q2_10_pre 10 231 6.43 0.88 7.00 6.61 0.00 1.00 7.00 6.00 -2.38 8.75 0.06
Q2_11_pre 11 231 6.23 1.04 7.00 6.42 0.00 1.00 7.00 6.00 -1.66 3.43 0.07
Q3_1_pre 12 231 6.34 0.81 7.00 6.47 0.00 3.00 7.00 4.00 -1.19 1.19 0.05
Q3_2_pre 13 231 6.55 0.67 7.00 6.67 0.00 4.00 7.17 3.17 -1.44 1.76 0.04
Q3_3_pre 14 231 6.54 0.71 7.00 6.67 0.00 3.00 7.22 4.22 -1.70 3.38 0.05
Q3_4_pre 15 231 5.61 1.27 6.00 5.74 1.48 2.00 7.61 5.61 -0.79 0.10 0.08
Q3_5_pre 16 231 5.76 1.28 6.00 5.91 1.48 2.00 7.09 5.09 -0.74 -0.44 0.08
Q3_6_pre 17 231 5.82 1.15 6.00 5.97 1.48 2.00 7.16 5.16 -0.93 0.34 0.08
Q3_7_pre 18 231 6.28 1.02 7.00 6.48 0.00 2.00 7.49 5.49 -1.53 2.02 0.07
Q3_8_pre 19 231 5.51 1.21 6.00 5.61 1.48 1.00 7.00 6.00 -0.78 0.71 0.08
Q3_9_pre 20 231 5.87 1.03 6.00 5.99 1.48 3.00 7.00 4.00 -0.74 -0.04 0.07
Q3_10_pre 21 231 5.12 1.32 5.00 5.20 1.48 1.00 7.00 6.00 -0.61 0.37 0.09
Q3_11_pre 22 231 5.55 1.14 6.00 5.63 1.48 1.00 7.00 6.00 -0.67 0.38 0.08
Q3_12_pre 23 231 5.53 1.02 6.00 5.57 1.48 3.00 7.00 4.00 -0.37 -0.40 0.07
Q4_1_pre 24 231 3.65 1.82 4.00 3.60 2.97 1.00 7.00 6.00 0.23 -1.07 0.12
Q4_2_pre 25 231 4.49 1.59 5.00 4.53 1.48 1.00 7.61 6.61 -0.25 -0.72 0.10
Q4_3_pre 26 231 4.66 1.83 5.00 4.76 1.48 1.00 7.60 6.60 -0.38 -1.01 0.12
Q4_4_pre 27 231 4.15 1.80 4.00 4.16 1.48 1.00 7.42 6.42 -0.07 -1.07 0.12
Q4_5_pre 28 231 3.77 1.79 4.00 3.71 1.48 0.56 7.00 6.44 0.20 -0.97 0.12

Pre-Test Plots

ggplot(gather(subset(wide, select=c(grep("_pre",colnames(wide))))), aes(value)) + 
  geom_histogram(bins = 7) + 
  facet_wrap(~key)
## Warning: Removed 812 rows containing non-finite values (stat_bin).

Post-Test Table

kbl(round(desc_post, digits = 2)) %>%
  kable_styling() %>%
  row_spec(which(desc_post$kurtosis > 2), bold = T) %>%
  row_spec(which(desc_post$kurtosis < -2), bold = T) %>%
  row_spec(which(desc_post$skew > 2), italic = T) %>%
  row_spec(which(desc_post$skew < -2), italic = T)
vars n mean sd median trimmed mad min max range skew kurtosis se
Q2_1_post 1 231 5.69 1.09 6.00 5.80 1.48 2.00 7.00 5.00 -0.74 0.15 0.07
Q2_2_post 2 231 5.75 1.07 6.00 5.85 1.48 3.00 7.00 4.00 -0.69 -0.18 0.07
Q2_3_post 3 231 6.14 0.91 6.00 6.25 1.48 3.00 7.00 4.00 -0.89 0.22 0.06
Q2_4_post 4 231 5.74 1.04 6.00 5.85 1.48 2.00 7.00 5.00 -0.86 0.75 0.07
Q2_5_post 5 231 6.01 1.09 6.00 6.17 1.48 2.00 7.00 5.00 -1.10 0.64 0.07
Q2_6_post 6 231 5.77 1.06 6.00 5.87 1.48 2.00 7.22 5.22 -0.71 0.00 0.07
Q2_7_post 7 231 5.67 1.09 6.00 5.78 1.48 1.00 7.00 6.00 -0.84 1.00 0.07
Q2_8_post 8 231 5.32 1.50 5.41 5.49 2.09 1.00 7.00 6.00 -0.79 0.04 0.10
Q2_9_post 9 231 6.56 0.70 7.00 6.69 0.00 3.00 7.17 4.17 -1.74 3.45 0.05
Q2_10_post 10 231 6.33 0.86 7.00 6.46 0.09 2.00 7.28 5.28 -1.39 2.35 0.06
Q2_11_post 11 231 6.21 0.95 6.33 6.36 0.99 2.00 7.07 5.07 -1.21 1.23 0.06
Q3_1_post 12 231 6.29 0.77 6.00 6.39 1.48 4.00 7.00 3.00 -0.85 0.09 0.05
Q3_2_post 13 231 6.44 0.71 7.00 6.57 0.00 4.00 7.00 3.00 -1.13 0.65 0.05
Q3_3_post 14 231 6.45 0.76 7.00 6.59 0.00 3.00 7.03 4.03 -1.35 1.58 0.05
Q3_4_post 15 231 5.62 1.38 6.00 5.83 1.48 1.00 7.10 6.10 -1.26 1.59 0.09
Q3_5_post 16 231 5.80 1.14 6.00 5.93 1.48 2.00 7.00 5.00 -0.82 0.13 0.08
Q3_6_post 17 231 5.79 1.11 6.00 5.92 1.48 2.00 7.02 5.02 -0.82 0.15 0.07
Q3_7_post 18 231 6.21 0.98 6.48 6.37 0.77 1.00 7.00 6.00 -1.46 2.98 0.06
Q3_8_post 19 231 5.59 1.12 6.00 5.69 1.48 2.00 7.00 5.00 -0.78 0.53 0.07
Q3_9_post 20 231 5.82 1.01 6.00 5.93 1.48 2.00 7.00 5.00 -0.90 0.98 0.07
Q3_10_post 21 231 5.27 1.24 5.00 5.38 1.48 1.00 7.00 6.00 -0.87 0.92 0.08
Q3_11_post 22 231 5.67 1.01 6.00 5.76 1.48 1.00 7.00 6.00 -0.90 1.62 0.07
Q3_12_post 23 231 5.64 1.03 6.00 5.74 1.48 1.00 7.00 6.00 -1.01 2.06 0.07
Q4_1_post 24 231 3.70 1.88 3.88 3.65 2.78 1.00 7.00 6.00 0.19 -1.19 0.12
Q4_2_post 25 231 4.68 1.66 5.00 4.77 1.48 1.00 7.00 6.00 -0.39 -0.82 0.11
Q4_3_post 26 231 4.66 1.85 5.00 4.76 1.56 1.00 7.00 6.00 -0.37 -1.04 0.12
Q4_4_post 27 231 4.14 1.91 4.00 4.16 2.97 1.00 7.00 6.00 0.00 -1.19 0.13
Q4_5_post 28 231 3.89 1.92 4.00 3.87 2.97 0.84 7.00 6.16 0.14 -1.19 0.13

Post-Test Plots

ggplot(gather(subset(wide, select=c(grep("_post",colnames(wide))))), aes(value)) + 
  geom_histogram(bins = 7) + 
  facet_wrap(~key)
## Warning: Removed 812 rows containing non-finite values (stat_bin).

Follow-Up Table

kbl(round(desc_follow, digits = 2)) %>%
  kable_styling() %>%
  row_spec(which(desc_follow$kurtosis > 2), bold = T) %>%
  row_spec(which(desc_follow$kurtosis < -2), bold = T) %>%
  row_spec(which(desc_follow$skew > 2), italic = T) %>%
  row_spec(which(desc_follow$skew < -2), italic = T)
vars n mean sd median trimmed mad min max range skew kurtosis se
Q2_1_follow 1 231 5.84 1.03 6.00 5.94 1.48 1 7.00 6.00 -0.98 1.91 0.07
Q2_2_follow 2 231 5.91 1.07 6.00 6.04 1.48 1 7.00 6.00 -1.29 2.63 0.07
Q2_3_follow 3 231 6.19 0.97 6.00 6.34 1.48 1 7.03 6.03 -1.69 4.40 0.06
Q2_4_follow 4 231 5.84 0.99 6.00 5.94 1.48 3 7.00 4.00 -0.75 0.30 0.07
Q2_5_follow 5 231 6.08 1.05 6.00 6.26 1.48 2 7.07 5.07 -1.28 1.49 0.07
Q2_6_follow 6 231 5.93 1.04 6.00 6.06 1.48 1 7.41 6.41 -1.26 2.64 0.07
Q2_7_follow 7 231 5.83 1.07 6.00 5.95 1.48 1 7.39 6.39 -0.91 1.28 0.07
Q2_8_follow 8 231 5.66 1.21 6.00 5.79 1.48 2 7.00 5.00 -0.65 -0.25 0.08
Q2_9_follow 9 231 6.48 0.89 7.00 6.66 0.00 1 7.12 6.12 -2.78 11.56 0.06
Q2_10_follow 10 231 6.37 0.93 7.00 6.55 0.00 2 7.00 5.00 -2.26 6.83 0.06
Q2_11_follow 11 231 6.24 0.98 6.74 6.40 0.39 2 7.02 5.02 -1.64 3.45 0.06
Q3_1_follow 12 231 6.28 0.97 6.85 6.43 0.22 1 7.09 6.09 -2.06 6.77 0.06
Q3_2_follow 13 231 6.30 0.93 6.64 6.43 0.54 1 7.50 6.50 -2.18 8.27 0.06
Q3_3_follow 14 231 6.27 1.00 6.61 6.45 0.58 1 7.00 6.00 -2.20 7.04 0.07
Q3_4_follow 15 231 5.90 1.16 6.00 6.07 1.48 1 7.71 6.71 -1.26 2.26 0.08
Q3_5_follow 16 231 5.85 1.21 6.00 6.03 1.48 1 7.01 6.01 -1.33 2.45 0.08
Q3_6_follow 17 231 5.92 1.21 6.00 6.12 1.48 1 7.00 6.00 -1.38 1.98 0.08
Q3_7_follow 18 231 6.27 1.12 7.00 6.50 0.00 1 7.29 6.29 -2.38 7.16 0.07
Q3_8_follow 19 231 5.71 1.03 6.00 5.80 1.48 2 7.00 5.00 -0.64 0.22 0.07
Q3_9_follow 20 231 5.92 0.93 6.00 6.01 1.48 2 7.00 5.00 -0.81 0.85 0.06
Q3_10_follow 21 231 5.44 1.13 5.63 5.50 0.94 2 7.03 5.03 -0.41 -0.42 0.07
Q3_11_follow 22 231 5.76 1.03 6.00 5.87 1.48 1 7.15 6.15 -0.92 1.69 0.07
Q3_12_follow 23 231 5.73 0.98 6.00 5.82 1.48 2 7.00 5.00 -0.61 0.28 0.06
Q4_1_follow 24 231 3.57 1.87 3.00 3.48 1.76 1 7.00 6.00 0.29 -1.08 0.12
Q4_2_follow 25 231 4.48 1.69 5.00 4.55 1.48 1 7.00 6.00 -0.34 -0.66 0.11
Q4_3_follow 26 231 4.56 1.92 5.00 4.68 2.20 1 7.00 6.00 -0.40 -1.02 0.13
Q4_4_follow 27 231 4.00 1.93 4.00 4.00 2.97 1 7.00 6.00 -0.02 -1.18 0.13
Q4_5_follow 28 231 3.60 1.85 3.33 3.50 1.98 1 7.00 6.00 0.26 -0.96 0.12

Follow-Up Plots

ggplot(gather(subset(wide, select=c(grep("_follow",colnames(wide))))), aes(value)) + 
  geom_histogram(bins = 7) + 
  facet_wrap(~key)
## Warning: Removed 812 rows containing non-finite values (stat_bin).

PMAPS Table

kbl(round(desc_pmaps, digits = 2)) %>%
  kable_styling() %>%
  row_spec(which(desc_pmaps$kurtosis > 2), bold = T) %>%
  row_spec(which(desc_pmaps$kurtosis < -2), bold = T) %>%
  row_spec(which(desc_pmaps$skew > 2), italic = T) %>%
  row_spec(which(desc_pmaps$skew < -2), italic = T)
vars n mean sd median trimmed mad min max range skew kurtosis se
Q47_1_pmaps 1 212 2.84 1.66 3 2.67 1.48 1 7 6 0.66 -0.48 0.11
Q47_2_pmaps 2 212 2.52 1.56 2 2.31 1.48 1 7 6 0.97 0.39 0.11
Q47_3_pmaps 3 212 2.62 1.65 2 2.41 1.48 1 7 6 0.95 0.05 0.11
Q47_4_pmaps 4 212 2.91 1.83 2 2.68 1.48 1 7 6 0.76 -0.47 0.13
Q47_5_pmaps 5 212 4.61 1.48 5 4.67 1.48 1 7 6 -0.41 -0.30 0.10
Q47_6_pmaps 6 212 5.00 1.36 5 5.11 1.48 1 7 6 -0.79 0.73 0.09
Q47_7_pmaps 7 212 5.08 1.38 5 5.21 1.48 1 7 6 -1.03 1.03 0.09
Q47_8_pmaps 8 212 4.67 1.60 5 4.78 1.48 1 7 6 -0.54 -0.23 0.11

PMAPS Plots

ggplot(gather(subset(wide, select=c(grep("_pmaps",colnames(wide))))), aes(value)) + 
  geom_histogram(bins = 7) + 
  facet_wrap(~key)
## Warning: Removed 384 rows containing non-finite values (stat_bin).

long2 <- subset(long, select=-c(Q2_9,Q2_10,Q2_11))
wide2 <- subset(wide, select=-c(grep("Q2_9",colnames(wide)), grep("Q2_10",colnames(wide)), grep("Q2_11",colnames(wide))))

Screen for Outliers

Remove participants who are marked as outliers at three timepoints.

d1 <- na.omit(subset(long2, timepoint == 1, select=(1:26)))

m_dist <- mahalanobis(d1[-1], colMeans(d1[-1]), cov(d1[-1]))
d1$MD <- round(m_dist, 1)
plot(d1$MD)
describe(m_dist)
##    vars   n  mean    sd median trimmed   mad  min   max range skew kurtosis
## X1    1 231 24.89 12.46  22.62   23.54 11.91 5.02 74.17 69.16 1.08     1.27
##      se
## X1 0.82
cut <- qchisq(.99, df=(ncol(d1)-1))
abline(a=cut, b=0, col="red")

d1$outlier <- F
d1$outlier[d1$MD > cut] <- T
table(d1$outlier)
## 
## FALSE  TRUE 
##   213    18
outs <- cbind.data.frame(d1$UID, d1$outlier)

d1 <- na.omit(subset(long2, timepoint == 2, select=(1:26)))

m_dist <- mahalanobis(d1[-1], colMeans(d1[-1]), cov(d1[-1]))
d1$MD <- round(m_dist, 1)
plot(d1$MD)
describe(m_dist)
##    vars   n  mean   sd median trimmed   mad  min   max range skew kurtosis   se
## X1    1 231 24.89 15.5  21.49   22.93 13.03 3.02 98.44 95.42 1.39     2.59 1.02
abline(a=cut, b=0, col="red")

d1$outlier <- F
d1$outlier[d1$MD > cut] <- T
table(d1$outlier)
## 
## FALSE  TRUE 
##   211    20
outs <- cbind.data.frame(outs, d1$outlier)

d1 <- na.omit(subset(long2, timepoint == 3, select=(1:26)))

m_dist <- mahalanobis(d1[-1], colMeans(d1[-1]), cov(d1[-1]))
d1$MD <- round(m_dist, 1)
plot(d1$MD)
describe(m_dist)
##    vars   n  mean    sd median trimmed   mad  min    max  range skew kurtosis
## X1    1 231 24.89 16.43  21.46   22.74 13.57 3.23 115.24 112.02 1.65     4.51
##      se
## X1 1.08
abline(a=cut, b=0, col="red")

d1$outlier <- F
d1$outlier[d1$MD > cut] <- T
table(d1$outlier)
## 
## FALSE  TRUE 
##   209    22
outs <- cbind.data.frame(outs, d1$outlier)

outs$sum <- rowSums(outs[2:4])
table(outs$sum)
## 
##   0   1   2   3 
## 190  25  13   3
outs2 <- subset(outs, sum == 3)

wide3 <- subset(wide2, !(UID %in% outs2$`long2$UID`))
long3 <- subset(long2, !(UID %in% outs2$`long2$UID`))

Step 3: Composites & Check

Create Factor Scores

attach(long3)
long3$bel <- (Q2_1 + Q2_2 + Q2_3 + Q2_4)/4
long3$rec <- (Q2_5 + Q2_6 + Q2_7 + Q2_8)/4
long3$inst <- (Q3_1 + Q3_2 + Q3_3)/3
long3$pof <- (Q3_4 + Q3_5 + Q3_6 + Q3_7)/4
long3$exp <- (Q3_8 + Q3_9 + Q3_10 + Q3_11 + Q3_12)/5
long3$anx <- (Q4_1 + Q4_2 + Q4_3 + Q4_4 + Q4_5)/5
long3$triv <- (Q47_1_pmaps + Q47_2_pmaps + Q47_3_pmaps + Q47_4_pmaps)/4
long3$vig <- (Q47_5_pmaps + Q47_6_pmaps + Q47_7_pmaps + Q47_8_pmaps)/4
detach(long3)

long3$re_minority <- NA
long3$re_minority[long3$re_white == T] <- "re_maj"
long3$re_minority[long3$re_latin == T | long3$re_mena == T | long3$re_nhpi == T | long3$re_black == T | long3$re_other == T | long3$re_aian== T | long3$re_asian == T] <- "re_min"
table(long3$re_minority, useNA = "always")
## 
## re_maj re_min   <NA> 
##    345    333    102
long3$ge_minority <- NA
long3$ge_minority[long3$ge_m == T] <- "ge_maj"
long3$ge_minority[long3$ge_w == T | long3$ge_gq == T | long3$ge_other == T | long3$ge_nb == T | long3$ge_w == T] <- "ge_min"
table(long3$ge_minority, useNA = "always")
## 
## ge_maj ge_min   <NA> 
##    492    183    105
long3$gere <- NA
long3$gere[long3$re_minority == "re_maj" & long3$ge_minority == "ge_maj"] <- "white men"
long3$gere[long3$re_minority == "re_maj" & long3$ge_minority == "ge_min"] <- "white women/nb"
long3$gere[long3$re_minority == "re_min" & long3$ge_minority == "ge_maj"] <- "men of color"
long3$gere[long3$re_minority == "re_min" & long3$ge_minority == "ge_min"] <- "women/nb of color"
table(long3$gere, useNA = "always")
## 
##      men of color         white men    white women/nb women/nb of color 
##               255               237               105                78 
##              <NA> 
##               105
table(long3$gere, long3$condition, useNA = "always")
##                    
##                     con equi int <NA>
##   men of color       69   81 105    0
##   white men          72   84  81    0
##   white women/nb     33   42  30    0
##   women/nb of color  21   30  27    0
##   <NA>               42   36  27    0
# long3$gere <- NA
# long3$gere[long3$re_minority == "re_whi" & long3$ge_minority == "ge_maj"] <- "white men"
# long3$gere[long3$re_minority == "re_whi" & long3$ge_minority == "ge_min"] <- "white women/nb"
# long3$gere[long3$re_minority == "re_asi" & long3$ge_minority == "ge_maj"] <- "asian men"
# long3$gere[long3$re_minority == "re_asi" & long3$ge_minority == "ge_min"] <- "asian women/nb"
# long3$gere[long3$re_minority == "re_min" & long3$ge_minority == "ge_maj"] <- "minority men"
# long3$gere[long3$re_minority == "re_min" & long3$ge_minority == "ge_min"] <- "minority women/nb"
# table(long3$gere, useNA = "always")
# table(long3$gere, long3$condition, useNA = "always")

long3$gere <- as.factor(long3$gere)
long3$ge_minority <- as.factor(long3$ge_minority)
long3$re_minority <- as.factor(long3$re_minority)

Check Cell Sizes

table(long3$condition, long3$timepoint)
##       
##         1  2  3
##   con  79 79 79
##   equi 91 91 91
##   int  90 90 90

Examining Skew & Kurtosis Main

Pre-Test Table

desc_pre <- data.frame(describe(subset(long3, select=c(bel, rec, inst, pof, exp, anx), timepoint == "1")))

kbl(round(desc_pre, digits = 2)) %>%
  kable_styling() %>%
  row_spec(which(desc_pre$kurtosis > 2), bold = T) %>%
  row_spec(which(desc_pre$kurtosis < -2), bold = T) %>%
  row_spec(which(desc_pre$skew > 2), italic = T) %>%
  row_spec(which(desc_pre$skew < -2), italic = T)
vars n mean sd median trimmed mad min max range skew kurtosis se
bel 1 231 5.63 0.99 5.75 5.70 1.11 2.00 7.00 5.00 -0.64 0.23 0.07
rec 2 231 5.49 1.02 5.56 5.56 1.02 2.25 7.15 4.90 -0.60 0.01 0.07
inst 3 231 6.48 0.61 6.67 6.58 0.49 4.00 7.08 3.08 -1.36 1.89 0.04
pof 4 231 5.87 1.00 6.00 5.98 1.11 2.75 7.34 4.59 -0.81 -0.02 0.07
exp 5 231 5.51 1.02 5.60 5.59 0.89 2.40 7.00 4.60 -0.62 0.02 0.07
anx 6 231 4.14 1.37 4.20 4.12 1.48 1.00 7.03 6.03 0.08 -0.77 0.09

Post-Test Table

desc_post <- data.frame(describe(subset(long3, select=c(bel, rec, inst, pof, exp, anx), timepoint == "2")))

kbl(round(desc_post, digits = 2)) %>%
  kable_styling() %>%
  row_spec(which(desc_post$kurtosis > 2), bold = T) %>%
  row_spec(which(desc_post$kurtosis < -2), bold = T) %>%
  row_spec(which(desc_post$skew > 2), italic = T) %>%
  row_spec(which(desc_post$skew < -2), italic = T)
vars n mean sd median trimmed mad min max range skew kurtosis se
bel 1 231 5.83 0.90 6.00 5.91 1.11 2.75 7 4.25 -0.70 0.04 0.06
rec 2 231 5.69 0.89 5.75 5.75 1.11 2.75 7 4.25 -0.65 0.25 0.06
inst 3 231 6.40 0.63 6.67 6.49 0.49 4.33 7 2.67 -0.95 0.23 0.04
pof 4 231 5.86 0.97 6.00 5.97 1.11 2.25 7 4.75 -0.94 0.53 0.06
exp 5 231 5.60 0.94 5.80 5.66 0.89 2.20 7 4.80 -0.76 0.77 0.06
anx 6 231 4.22 1.57 4.20 4.21 2.08 1.00 7 6.00 0.01 -1.03 0.10

Follow-Up Table

desc_follow <- data.frame(describe(subset(long3, select=c(bel, rec, inst, pof, exp, anx, condition), timepoint == "3")))

kbl(round(desc_follow, digits = 2)) %>%
  kable_styling() %>%
  row_spec(which(desc_follow$kurtosis > 2), bold = T) %>%
  row_spec(which(desc_follow$kurtosis < -2), bold = T) %>%
  row_spec(which(desc_follow$skew > 2), italic = T) %>%
  row_spec(which(desc_follow$skew < -2), italic = T)
vars n mean sd median trimmed mad min max range skew kurtosis se
bel 1 231 5.94 0.91 6.00 6.04 0.74 1.50 7.00 5.50 -1.22 2.86 0.06
rec 2 231 5.88 0.88 6.00 5.95 0.74 1.75 7.06 5.31 -1.10 2.67 0.06
inst 3 231 6.28 0.88 6.42 6.42 0.87 1.00 7.11 6.11 -2.54 10.91 0.06
pof 4 231 5.99 1.04 6.25 6.14 1.11 1.25 7.23 5.98 -1.57 3.34 0.07
exp 5 231 5.71 0.88 5.80 5.77 0.89 2.20 7.00 4.80 -0.54 0.40 0.06
anx 6 231 4.04 1.54 4.13 4.03 1.67 1.00 7.00 6.00 0.06 -0.69 0.10
condition* 7 260 2.04 0.81 2.00 2.05 1.48 1.00 3.00 2.00 -0.08 -1.47 0.05

Examining Skew & Kurtosis PMAPs

desc_follow <- data.frame(describe(subset(long3, select=c(triv, vig, condition), timepoint == "3")))

kbl(round(desc_follow, digits = 2)) %>%
  kable_styling() %>%
  row_spec(which(desc_follow$kurtosis > 2), bold = T) %>%
  row_spec(which(desc_follow$kurtosis < -2), bold = T) %>%
  row_spec(which(desc_follow$skew > 2), italic = T) %>%
  row_spec(which(desc_follow$skew < -2), italic = T)
vars n mean sd median trimmed mad min max range skew kurtosis se
triv 1 212 2.72 1.54 2.25 2.55 1.85 1 7 6 0.81 -0.02 0.11
vig 2 212 4.84 1.24 4.88 4.91 1.30 1 7 6 -0.65 0.75 0.09
condition* 3 260 2.04 0.81 2.00 2.05 1.48 1 3 2 -0.08 -1.47 0.05

Examine QQ-plots and Trim (Trivialization)

# Examination of residuals demonstrated some normality challenges. Trim
qqPlot(long3$triv)

## [1] 114 189
res.aovB <- aov(triv ~ gere*condition, data = long3)
qqPlot(res.aovB)

## [1] 259 519
# bu <- long3$triv
# long3$triv <- bu

upper_quantile <- quantile(long3$triv, 0.98, na.rm = TRUE)
lower_quantile <- quantile(long3$triv, 0.02, na.rm = TRUE)
# where 99th percentile becomes NA
long3$triv[long3$triv >= upper_quantile] <- NA
long3$triv[long3$triv <= lower_quantile] <- NA

res.aovB <- aov(triv ~ gere*condition, data = long3)
qqPlot(res.aovB)

## [1]  28 288

Examine QQ-plots and Trim (Vigilance)

qqPlot(long3$vig)

## [1] 189 224
res.aovB <- aov(vig ~ gere*condition, data = long3)
qqPlot(res.aovB)

## [1] 224 484

Step 4: Profile Analysis (Canceled)

Information

Do different groups have parallel profiles? This is commonly known as the test of parallelism and is the primary question addressed by profile analysis. - Two of the major tests of profile analysis (parallelism and flatness) test difference scores between DVs measured on adjacent occasions - Difference scores are called segments in profile analysis.

Parallelism of profiles
  • When using the profile approach to univariate repeated-measures ANOVA, the parallelism test is the test of interaction.
  • For example, do traditional and computer-assisted instructions lead to the same pattern of gains in achievement over the course of testing? Or, do changes in achievement depend on the method of instruction used?
Overall difference among groups
  • In profile analysis jargon, the “levels” hypothesis addresses the same question as the between-subjects main effect in repeated-measures ANOVA.
  • Does one group, on average, score higher on the collected set of measures than another? For example, does one method of instruction lead to greateroverall math achievement than the other method?
Flatness of profiles
  • The third question addressed by profile analysis concerns the similarity of response to all DVs, independent of groups. Do all the DVs elicit the same average response?
  • In the instructional example, the flatness test evaluates whether achievement changes over the period of testing. In this context the flatness test evaluates the same hypothesis as the within-subjects main effect in repeated-measures ANOVA.
Contrasts
  • With more than two groups or more than two measures, differences in parallelism, flatness, and/or level can result from a variety of sources.
Parameter estimates
  • Parameters are estimated whenever statistically significant differences are found between groups or measures the major description of results is typically a plot of profiles in which the means for each of the DVs are plotted for each of the groups
  • If the null hypothesis regarding levels is rejected (signifcant main effect) assessment of group means and group standard deviations, standard errors, or confidence intervals is helpful.
  • If the null hypothesis regarding flatness is rejected (signifcant effect of time) a plot of scores combined over groups is instructive, along with standard deviations or standard errors.
To apply profile analysis
  • All measures must have the same range of possible scores, with the same score value having the same meaning on all the measures.
Limitations of PA
  • DVs must be commensurate
  • There should be more research units in the smallest group than there are DVs.
  • Unequal sample sizes typically provide no special difficulty in profile analysis unless there are fewer cases than DVs in the smallest group and highly unequal n
  • Deviation from normality of sampling distributions is not expected
  • Profile analysis is extremely sensitive to outliers.
library(ggplot2)
library(afex)
library(emmeans)

Step 4: Analyses

Preliminary

Dropping post-test time point for clearer interpretation. Comparing scores at pre-test to examine random assignment.

long3 <- subset(long3, timepoint != 2)

long3$condition <- as.factor(long3$condition)

long3$belc <- scale(long3$bel, center = T, scale = T)
long3$recc <- scale(long3$rec, center = T, scale = T)
long3$instc <- scale(long3$inst, center = T, scale = T)
long3$pofc <- scale(long3$pof, center = T, scale = T)
long3$expc <- scale(long3$exp, center = T, scale = T)
long3$anxc <- scale(long3$anx, center = T, scale = T)

long4 <- subset(long3, timepoint == 1)

fit <- aov_ez(id="UID", dv="bel", data=long4, between=c("condition"))
## Warning: More than one observation per cell, aggregating the data using mean
## (i.e, fun_aggregate = mean)!
## Warning: Missing values for following ID(s):
## DEZFY5154Y, GLBEZ3029S, IQKYL9589B, IYGKJ9212Q, JDRGO5214D, JTIXL1749F, KHTGW5971J, KNGSJ3794D, MOLUH1393I, NGWDC4182S, NOTVI7273I, OIJRE9661O, PPDBO1961W, PSAJP5500H, QFYEE1921U, QXQOZ9556W, SJOBN2309Y, SRAGV3585D, STHXN7574W, UGNMS7589N, UHUGQ9517B, VETKA3898N, VPMAK8781R, WIGYE4790M, WVCXC4155L, WVPCP3440D, YZKBO8058U, YZVMS4925G, ZQCVQ9164X
## Removing those cases from the analysis.
## Contrasts set to contr.sum for the following variables: condition
nice(fit)
## Anova Table (Type 3 tests)
## 
## Response: bel
##      Effect     df  MSE    F  ges p.value
## 1 condition 2, 225 0.96 1.88 .016    .154
## ---
## Signif. codes:  0 '***' 0.001 '**' 0.01 '*' 0.05 '+' 0.1 ' ' 1
fit <- aov_ez(id="UID", dv="rec", data=long4, between=c("condition"))
## Warning: More than one observation per cell, aggregating the data using mean (i.e, fun_aggregate = mean)!

## Warning: Missing values for following ID(s):
## DEZFY5154Y, GLBEZ3029S, IQKYL9589B, IYGKJ9212Q, JDRGO5214D, JTIXL1749F, KHTGW5971J, KNGSJ3794D, MOLUH1393I, NGWDC4182S, NOTVI7273I, OIJRE9661O, PPDBO1961W, PSAJP5500H, QFYEE1921U, QXQOZ9556W, SJOBN2309Y, SRAGV3585D, STHXN7574W, UGNMS7589N, UHUGQ9517B, VETKA3898N, VPMAK8781R, WIGYE4790M, WVCXC4155L, WVPCP3440D, YZKBO8058U, YZVMS4925G, ZQCVQ9164X
## Removing those cases from the analysis.
## Contrasts set to contr.sum for the following variables: condition
nice(fit)
## Anova Table (Type 3 tests)
## 
## Response: rec
##      Effect     df  MSE      F  ges p.value
## 1 condition 2, 225 1.04 2.61 + .023    .076
## ---
## Signif. codes:  0 '***' 0.001 '**' 0.01 '*' 0.05 '+' 0.1 ' ' 1
fit <- aov_ez(id="UID", dv="inst", data=long4, between=c("condition"))
## Warning: More than one observation per cell, aggregating the data using mean (i.e, fun_aggregate = mean)!

## Warning: Missing values for following ID(s):
## DEZFY5154Y, GLBEZ3029S, IQKYL9589B, IYGKJ9212Q, JDRGO5214D, JTIXL1749F, KHTGW5971J, KNGSJ3794D, MOLUH1393I, NGWDC4182S, NOTVI7273I, OIJRE9661O, PPDBO1961W, PSAJP5500H, QFYEE1921U, QXQOZ9556W, SJOBN2309Y, SRAGV3585D, STHXN7574W, UGNMS7589N, UHUGQ9517B, VETKA3898N, VPMAK8781R, WIGYE4790M, WVCXC4155L, WVPCP3440D, YZKBO8058U, YZVMS4925G, ZQCVQ9164X
## Removing those cases from the analysis.
## Contrasts set to contr.sum for the following variables: condition
nice(fit)
## Anova Table (Type 3 tests)
## 
## Response: inst
##      Effect     df  MSE    F  ges p.value
## 1 condition 2, 225 0.38 1.58 .014    .208
## ---
## Signif. codes:  0 '***' 0.001 '**' 0.01 '*' 0.05 '+' 0.1 ' ' 1
fit <- aov_ez(id="UID", dv="pof", data=long4, between=c("condition"))
## Warning: More than one observation per cell, aggregating the data using mean (i.e, fun_aggregate = mean)!

## Warning: Missing values for following ID(s):
## DEZFY5154Y, GLBEZ3029S, IQKYL9589B, IYGKJ9212Q, JDRGO5214D, JTIXL1749F, KHTGW5971J, KNGSJ3794D, MOLUH1393I, NGWDC4182S, NOTVI7273I, OIJRE9661O, PPDBO1961W, PSAJP5500H, QFYEE1921U, QXQOZ9556W, SJOBN2309Y, SRAGV3585D, STHXN7574W, UGNMS7589N, UHUGQ9517B, VETKA3898N, VPMAK8781R, WIGYE4790M, WVCXC4155L, WVPCP3440D, YZKBO8058U, YZVMS4925G, ZQCVQ9164X
## Removing those cases from the analysis.
## Contrasts set to contr.sum for the following variables: condition
nice(fit)
## Anova Table (Type 3 tests)
## 
## Response: pof
##      Effect     df  MSE      F  ges p.value
## 1 condition 2, 225 0.98 2.53 + .022    .082
## ---
## Signif. codes:  0 '***' 0.001 '**' 0.01 '*' 0.05 '+' 0.1 ' ' 1
fit <- aov_ez(id="UID", dv="exp", data=long4, between=c("condition"))
## Warning: More than one observation per cell, aggregating the data using mean (i.e, fun_aggregate = mean)!

## Warning: Missing values for following ID(s):
## DEZFY5154Y, GLBEZ3029S, IQKYL9589B, IYGKJ9212Q, JDRGO5214D, JTIXL1749F, KHTGW5971J, KNGSJ3794D, MOLUH1393I, NGWDC4182S, NOTVI7273I, OIJRE9661O, PPDBO1961W, PSAJP5500H, QFYEE1921U, QXQOZ9556W, SJOBN2309Y, SRAGV3585D, STHXN7574W, UGNMS7589N, UHUGQ9517B, VETKA3898N, VPMAK8781R, WIGYE4790M, WVCXC4155L, WVPCP3440D, YZKBO8058U, YZVMS4925G, ZQCVQ9164X
## Removing those cases from the analysis.
## Contrasts set to contr.sum for the following variables: condition
nice(fit)
## Anova Table (Type 3 tests)
## 
## Response: exp
##      Effect     df  MSE    F  ges p.value
## 1 condition 2, 225 1.03 0.76 .007    .468
## ---
## Signif. codes:  0 '***' 0.001 '**' 0.01 '*' 0.05 '+' 0.1 ' ' 1
fit <- aov_ez(id="UID", dv="anx", data=long4, between=c("condition"))
## Warning: More than one observation per cell, aggregating the data using mean (i.e, fun_aggregate = mean)!

## Warning: Missing values for following ID(s):
## DEZFY5154Y, GLBEZ3029S, IQKYL9589B, IYGKJ9212Q, JDRGO5214D, JTIXL1749F, KHTGW5971J, KNGSJ3794D, MOLUH1393I, NGWDC4182S, NOTVI7273I, OIJRE9661O, PPDBO1961W, PSAJP5500H, QFYEE1921U, QXQOZ9556W, SJOBN2309Y, SRAGV3585D, STHXN7574W, UGNMS7589N, UHUGQ9517B, VETKA3898N, VPMAK8781R, WIGYE4790M, WVCXC4155L, WVPCP3440D, YZKBO8058U, YZVMS4925G, ZQCVQ9164X
## Removing those cases from the analysis.
## Contrasts set to contr.sum for the following variables: condition
nice(fit)
## Anova Table (Type 3 tests)
## 
## Response: anx
##      Effect     df  MSE      F  ges p.value
## 1 condition 2, 225 1.81 4.12 * .035    .018
## ---
## Signif. codes:  0 '***' 0.001 '**' 0.01 '*' 0.05 '+' 0.1 ' ' 1

RM-ANOVA - Belonging

Results - All

fit <- aov_ez(id="UID", dv="bel", data=long3, between=c("condition"), within="timepoint", covariate = c("recc"), factorize = F)
## Warning: Numerical variables NOT centered on 0 (i.e., likely bogus results): NA
## Warning: More than one observation per cell, aggregating the data using mean
## (i.e, fun_aggregate = mean)!
## Warning: Missing values for following ID(s):
## ADMNF5431W, AFITA6644H, AIKAP6463M, AODNI3144W, AQMMS9810J, AVINR5897Y, AYQSB9061F, BIGPC8577I, BJCYW9313A, BKKKT3701U, BLJCY6370P, BMMVL7994H, BQYKI9156R, BVDSH3315M, BWISR1973A, BXWMH7130Y, CDHNQ4412P, CFOQZ9365X, CHADU6389C, CHKFU7601Z, CHPID9225X, CIEHL6827C, CMDRZ4165X, CMHID2661Y, COSQM7547U, CPHXV3544C, CQVAE8343T, DDWDZ2339J, DEGUG8250Q, DEZFY5154Y, DUOOW9047V, DWEFP0640O, DYJAC5887Z, EDPLI4718X, EGNPY6902N, EKHKA1320A, ELPMT5534R, ENHLY4123X, ENLUW4800F, EQRSC1726I, ETLLC3301O, EWGTH0978U, EYJKI2890J, FBPYG4513B, FDKWO5036F, FDRFT1976E, FOCRY5697B, FSKKR7810Y, FWVEK3877Q, FXXAG6538G, FZWVI1954A, GAHHN6818C, GEVFJ6602J, GFITG4909Q, GLBEZ3029S, GOSED0340T, GQBDI2909G, GUMSQ0837O, GUOQX4544Z, GVAQY8946U, HBIFA5295Q, HDDDR5624S, HHENH9196H, HHTYJ9067S, HMITM1029R, HOQSF9391C, ICNZW6469J, IDVCE3660F, IHEIE1031K, IHLAB2822T, IIBWJ1162K, IPNYM5318I, IPVOL5900Z, IQKYL9589B, IVVAN8695S, IVWLS8788C, IWQBG3288R, IXTMS7827C, IYGKJ9212Q, JDRGO5214D, JQUSH7763H, JTIXL1749F, KGZSZ9095U, KHTGW5971J, KNGSJ3794D, KODOI2800E, KOGCS2814C, KQZOR6889Y, KTTQO0724Q, KVDIE4239W, KYLRY5373V, LAZZJ4798S, LDKDH9352Y, LDTLX9551Z, LHJVD2959Y, LJGWF4027N, LJTCD1624Z, LOPWH7294L, MAJVB6162F, MDPFH5436F, MHKVA5255G, MIBGQ4458A, MOLUH1393I, MRECG6247Z, NECDB3791F, NGWDC4182S, NLEXE0538V, NNQVD9904G, NOAGL7626V, NOTVI7273I, NPEEK2345M, NVKGR6192V, NZDFV1837O, OAUDW7602Q, OEHWZ9390E, OIJRE9661O, ONXIQ1991Q, OQEDN0137J, ORMKY3774B, PAMYY6490X, PCYUT6681A, PDRGN5764W, PESUP1240W, PHIAU9193C, PPDBO1961W, PSAJP5500H, PTGGV6124L, PYGEI1292D, QFEKE7247Z, QFTUS4663W, QFYEE1921U, QNAVN3908M, QONZX2504S, QSDTY7001Q, QUOZW7823S, QXQOZ9556W, RIBRN0153W, RRBBT6871J, RTNDB7987U, RXGVF0879U, SACNZ0120A, SADTX3579I, SBFTP7166E, SBUTD1285I, SDTRI9296D, SJOBN2309Y, SKZTW6437N, SLYQX2976O, SQDCF2656X, SRAGV3585D, STHXN7574W, SVRGK7887X, SXXFS1531Q, TDBIA5306X, TEMIM8896R, TFZAE0194I, TJBJJ8859W, TMHJY2195F, TNOMW4491R, TWCNI6874W, TXDDF1309A, TYMVS5920H, UCYZA4959S, UDHFD1638I, UDYOK1503U, UGHOD6018P, UGNMS7589N, UHUGQ9517B, UIQOX7318R, ULLCV9227K, UVBLS1638O, VELTF4260O, VEMXD5267D, VETKA3898N, VGJRD7424G, VHUIR1579V, VIQDL0582N, VNSSV4633E, VPMAK8781R, VUDWY1712Y, VYPFA1221S, WDOOB5760W, WHDVV0726Z, WIGYE4790M, WKMZJ5561X, WPBCK1207N, WUVZT3628Z, WVCXC4155L, WVPCP3440D, WYSWO1471E, XBYQZ8055T, XGNJP1929D, XKCRD7820V, XVCKT5197T, XWLZJ8035G, XXJXG2725P, XZQWA3103I, YAIBB9238C, YHKSR3208E, YIGNG1445M, YIWQF4610I, YOGUC5356J, YTUFO0314P, YVRMZ9228Y, YWQXI5864U, YXZRO8024H, YZKBO8058U, YZTVA8918Z, YZVMS4925G, ZCFIN4382O, ZCNDI2519Z, ZIKXQ5287N, ZQCVQ9164X, ZUOTR6815P, ZWPKK7562T, ZWTSA1605F, ZYFYX4806X, ZZCEK2517F
## Removing those cases from the analysis.
## Contrasts set to contr.sum for the following variables: condition
nice(fit)
## Anova Table (Type 3 tests)
## 
## Response: bel
##                Effect    df  MSE         F  ges p.value
## 1           condition 2, 35 0.48      0.97 .043    .388
## 2                recc 1, 35 0.48 51.36 *** .543   <.001
## 3           timepoint 1, 35 0.11    7.27 * .038    .011
## 4 condition:timepoint 2, 35 0.11      1.70 .018    .198
## 5      recc:timepoint 1, 35 0.11    5.91 * .031    .020
## ---
## Signif. codes:  0 '***' 0.001 '**' 0.01 '*' 0.05 '+' 0.1 ' ' 1
afex_plot(fit, x = "timepoint", error = "within")

Results - Race and Gender

fit <- aov_ez(id="UID", dv="bel", data=long3, between=c("condition","gere"), within="timepoint")
## Warning: More than one observation per cell, aggregating the data using mean
## (i.e, fun_aggregate = mean)!
## Warning: Missing values for following ID(s):
## DEZFY5154Y, EKHKA1320A, GLBEZ3029S, IQKYL9589B, IYGKJ9212Q, JDRGO5214D, JTIXL1749F, KHTGW5971J, KNGSJ3794D, MOLUH1393I, NGWDC4182S, NOTVI7273I, OIHCW0935B, OIJRE9661O, PPDBO1961W, PSAJP5500H, QFYEE1921U, QXQOZ9556W, SBUTD1285I, SJOBN2309Y, SKZTW6437N, SRAGV3585D, STHXN7574W, UGNMS7589N, UHUGQ9517B, VBCFZ4637X, VETKA3898N, VPMAK8781R, WIGYE4790M, WVCXC4155L, WVPCP3440D, XKCRD7820V, YZKBO8058U, YZVMS4925G, ZQCVQ9164X
## Removing those cases from the analysis.
## Contrasts set to contr.sum for the following variables: condition, gere
nice(fit)
## Anova Table (Type 3 tests)
## 
## Response: bel
##                     Effect     df  MSE         F   ges p.value
## 1                condition 2, 210 1.23    2.62 +  .019    .075
## 2                     gere 3, 210 1.23 11.47 ***  .112   <.001
## 3           condition:gere 6, 210 1.23      0.67  .015    .674
## 4                timepoint 1, 210 0.36 35.00 ***  .036   <.001
## 5      condition:timepoint 2, 210 0.36      0.38 <.001    .687
## 6           gere:timepoint 3, 210 0.36      2.06  .007    .106
## 7 condition:gere:timepoint 6, 210 0.36      0.43  .003    .860
## ---
## Signif. codes:  0 '***' 0.001 '**' 0.01 '*' 0.05 '+' 0.1 ' ' 1

RM-ANOVA - Recognition

Results - All

fit <- aov_ez(id="UID", dv="rec", data=long3, between=c("condition"), within="timepoint", covariate = c("belc"), factorize = F)
## Warning: Numerical variables NOT centered on 0 (i.e., likely bogus results): NA
## Warning: More than one observation per cell, aggregating the data using mean
## (i.e, fun_aggregate = mean)!
## Warning: Missing values for following ID(s):
## AFITA6644H, AODNI3144W, AVINR5897Y, AYQSB9061F, BIGPC8577I, BJCYW9313A, BLJCY6370P, BMMVL7994H, BOMVH4712H, BQYKI9156R, BVDSH3315M, BWHCJ5688Q, BWISR1973A, BXWMH7130Y, CDHNQ4412P, CFOQZ9365X, CGXNN1861X, CHADU6389C, CHKFU7601Z, CHPID9225X, CIEHL6827C, CJDVL5022X, CMDRZ4165X, CMHID2661Y, COSQM7547U, CPHXV3544C, CQVAE8343T, DDWDZ2339J, DEGUG8250Q, DEZFY5154Y, DUOOW9047V, DWEFP0640O, ECLDJ0558Y, EDPLI4718X, EGNPY6902N, EKHKA1320A, ELPMT5534R, ENGNR4476F, ENLUW4800F, EPVXZ9711C, ETLLC3301O, EWGTH0978U, EYJKI2890J, FBPYG4513B, FDKWO5036F, FDRFT1976E, FOCRY5697B, FSKKR7810Y, FWVEK3877Q, FXXAG6538G, GAHHN6818C, GFITG4909Q, GLBEZ3029S, GOSED0340T, GQBDI2909G, GUMSQ0837O, GUOQX4544Z, GVAQY8946U, HBIFA5295Q, HDDDR5624S, HHENH9196H, HHTYJ9067S, HMITM1029R, HOQSF9391C, HVTWX3105O, ICNZW6469J, IDVCE3660F, IHEIE1031K, IHLAB2822T, IIBWJ1162K, IQKYL9589B, IVVAN8695S, IWQBG3288R, IXTMS7827C, IYGKJ9212Q, JAQOX6224H, JDRGO5214D, JQUSH7763H, JTIXL1749F, KGZSZ9095U, KHTGW5971J, KNGSJ3794D, KODOI2800E, KOGCS2814C, KQZOR6889Y, KTTQO0724Q, KVDIE4239W, KYLRY5373V, LAZZJ4798S, LDKDH9352Y, LDTLX9551Z, LHJVD2959Y, LJGWF4027N, LOPWH7294L, LQYVS8333P, MAJVB6162F, MHKVA5255G, MIBGQ4458A, MOLUH1393I, MSMVZ2231V, NBUDI3965W, NECDB3791F, NGWDC4182S, NLEXE0538V, NNQVD9904G, NOTVI7273I, NPEEK2345M, NVKGR6192V, NXSHA4361K, NZDFV1837O, OAUDW7602Q, OEHWZ9390E, OIJRE9661O, ONXIQ1991Q, OQEDN0137J, ORMKY3774B, PCYUT6681A, PDRGN5764W, PESUP1240W, PIFPO0872F, PIMZK0270T, PLKLW0979X, PPDBO1961W, PQSIQ7727D, PSAJP5500H, PYGEI1292D, QFTUS4663W, QFYEE1921U, QNAVN3908M, QONZX2504S, QSDTY7001Q, QUOZW7823S, QWQPC1344Y, QXQOZ9556W, RIBRN0153W, RRBBT6871J, RTNDB7987U, RXGVF0879U, SACNZ0120A, SADTX3579I, SBFTP7166E, SDTRI9296D, SGUWP5939A, SJOBN2309Y, SKZTW6437N, SLYQX2976O, SQDCF2656X, SRAGV3585D, STHXN7574W, SVRGK7887X, SXXFS1531Q, SZSPH6169O, TDBIA5306X, TEMIM8896R, TFZAE0194I, TJBJJ8859W, TJDHC0458U, TNOMW4491R, TWCNI6874W, UCYZA4959S, UDHFD1638I, UDYOK1503U, UEWZT9177C, UGHOD6018P, UGNMS7589N, UHUGQ9517B, ULLCV9227K, UVBLS1638O, VELTF4260O, VEMXD5267D, VETKA3898N, VIQDL0582N, VNSSV4633E, VPMAK8781R, VUDWY1712Y, VYPFA1221S, WDOOB5760W, WHDVV0726Z, WIGYE4790M, WKMZJ5561X, WPBCK1207N, WUVZT3628Z, WVCXC4155L, WVPCP3440D, WYSWO1471E, XBYQZ8055T, XGNJP1929D, XKCRD7820V, XVCKT5197T, XWLZJ8035G, XXJXG2725P, YAIBB9238C, YHKSR3208E, YIGNG1445M, YVRMZ9228Y, YXZRO8024H, YZKBO8058U, YZTVA8918Z, YZVMS4925G, ZCFIN4382O, ZIKXQ5287N, ZQCVQ9164X, ZUOTR6815P, ZWPKK7562T, ZWTSA1605F, ZYFYX4806X, ZZCEK2517F
## Removing those cases from the analysis.
## Contrasts set to contr.sum for the following variables: condition
nice(fit)
## Anova Table (Type 3 tests)
## 
## Response: rec
##                Effect    df  MSE         F  ges p.value
## 1           condition 2, 46 0.57      1.08 .036    .347
## 2                belc 1, 46 0.57 38.13 *** .400   <.001
## 3           timepoint 1, 46 0.14    3.84 + .016    .056
## 4 condition:timepoint 2, 46 0.14    2.74 + .023    .075
## 5      belc:timepoint 1, 46 0.14      1.16 .005    .288
## ---
## Signif. codes:  0 '***' 0.001 '**' 0.01 '*' 0.05 '+' 0.1 ' ' 1
afex_plot(fit, x = "timepoint", error = "within")

afex_plot(fit, x = "timepoint", trace = "condition")
## Warning: Panel(s) show a mixed within-between-design.
## Error bars do not allow comparisons across all means.
## Suppress error bars with: error = "none"

Results - Race & Gender

fit <- aov_ez(id="UID", dv="rec", data=long3, between=c("condition","gere"), within="timepoint", covariate = c("belc"), factorize = F)
## Warning: Numerical variables NOT centered on 0 (i.e., likely bogus results): NA
## Warning: More than one observation per cell, aggregating the data using mean
## (i.e, fun_aggregate = mean)!
## Warning: Missing values for following ID(s):
## AFITA6644H, AODNI3144W, AVINR5897Y, AYQSB9061F, BIGPC8577I, BJCYW9313A, BLJCY6370P, BMMVL7994H, BOMVH4712H, BQYKI9156R, BVDSH3315M, BWHCJ5688Q, BWISR1973A, BXWMH7130Y, CDHNQ4412P, CFOQZ9365X, CGXNN1861X, CHADU6389C, CHKFU7601Z, CHPID9225X, CIEHL6827C, CJDVL5022X, CMDRZ4165X, CMHID2661Y, COSQM7547U, CPHXV3544C, CQVAE8343T, DDWDZ2339J, DEGUG8250Q, DEZFY5154Y, DUOOW9047V, DWEFP0640O, ECLDJ0558Y, EDPLI4718X, EGNPY6902N, EKHKA1320A, ELPMT5534R, ENGNR4476F, ENLUW4800F, EPVXZ9711C, ETLLC3301O, EWGTH0978U, EYJKI2890J, FBPYG4513B, FDKWO5036F, FDRFT1976E, FOCRY5697B, FSKKR7810Y, FWVEK3877Q, FXXAG6538G, GAHHN6818C, GFITG4909Q, GLBEZ3029S, GOSED0340T, GQBDI2909G, GUMSQ0837O, GUOQX4544Z, GVAQY8946U, HBIFA5295Q, HDDDR5624S, HHENH9196H, HHTYJ9067S, HMITM1029R, HOQSF9391C, HVTWX3105O, ICNZW6469J, IDVCE3660F, IHEIE1031K, IHLAB2822T, IIBWJ1162K, IQKYL9589B, IVVAN8695S, IWQBG3288R, IXTMS7827C, IYGKJ9212Q, JAQOX6224H, JDRGO5214D, JQUSH7763H, JTIXL1749F, KGZSZ9095U, KHTGW5971J, KNGSJ3794D, KODOI2800E, KOGCS2814C, KQZOR6889Y, KTTQO0724Q, KVDIE4239W, KYLRY5373V, LAZZJ4798S, LDKDH9352Y, LDTLX9551Z, LHJVD2959Y, LJGWF4027N, LOPWH7294L, LQYVS8333P, MAJVB6162F, MHKVA5255G, MIBGQ4458A, MOLUH1393I, MSMVZ2231V, NBUDI3965W, NECDB3791F, NGWDC4182S, NLEXE0538V, NNQVD9904G, NOTVI7273I, NPEEK2345M, NVKGR6192V, NXSHA4361K, NZDFV1837O, OAUDW7602Q, OEHWZ9390E, OIHCW0935B, OIJRE9661O, ONXIQ1991Q, OQEDN0137J, ORMKY3774B, PCYUT6681A, PDRGN5764W, PESUP1240W, PIFPO0872F, PIMZK0270T, PLKLW0979X, PPDBO1961W, PQSIQ7727D, PSAJP5500H, PYGEI1292D, QFTUS4663W, QFYEE1921U, QNAVN3908M, QONZX2504S, QSDTY7001Q, QUOZW7823S, QWQPC1344Y, QXQOZ9556W, RIBRN0153W, RRBBT6871J, RTNDB7987U, RXGVF0879U, SACNZ0120A, SADTX3579I, SBFTP7166E, SBUTD1285I, SDTRI9296D, SGUWP5939A, SJOBN2309Y, SKZTW6437N, SLYQX2976O, SQDCF2656X, SRAGV3585D, STHXN7574W, SVRGK7887X, SXXFS1531Q, SZSPH6169O, TDBIA5306X, TEMIM8896R, TFZAE0194I, TJBJJ8859W, TJDHC0458U, TNOMW4491R, TWCNI6874W, UCYZA4959S, UDHFD1638I, UDYOK1503U, UEWZT9177C, UGHOD6018P, UGNMS7589N, UHUGQ9517B, ULLCV9227K, UVBLS1638O, VBCFZ4637X, VELTF4260O, VEMXD5267D, VETKA3898N, VIQDL0582N, VNSSV4633E, VPMAK8781R, VUDWY1712Y, VYPFA1221S, WDOOB5760W, WHDVV0726Z, WIGYE4790M, WKMZJ5561X, WPBCK1207N, WUVZT3628Z, WVCXC4155L, WVPCP3440D, WYSWO1471E, XBYQZ8055T, XGNJP1929D, XKCRD7820V, XVCKT5197T, XWLZJ8035G, XXJXG2725P, YAIBB9238C, YHKSR3208E, YIGNG1445M, YVRMZ9228Y, YXZRO8024H, YZKBO8058U, YZTVA8918Z, YZVMS4925G, ZCFIN4382O, ZIKXQ5287N, ZQCVQ9164X, ZUOTR6815P, ZWPKK7562T, ZWTSA1605F, ZYFYX4806X, ZZCEK2517F
## Removing those cases from the analysis.
## Contrasts set to contr.sum for the following variables: condition, gere
nice(fit)
## Anova Table (Type 3 tests)
## 
## Response: rec
##                     Effect    df  MSE         F  ges p.value
## 1                condition 2, 34 0.42      1.97 .080    .155
## 2                     gere 3, 34 0.42    4.23 * .219    .012
## 3                     belc 1, 34 0.42 15.17 *** .251   <.001
## 4           condition:gere 6, 34 0.42      1.27 .144    .297
## 5                timepoint 1, 34 0.14    4.48 * .032    .042
## 6      condition:timepoint 2, 34 0.14    2.91 + .041    .068
## 7           gere:timepoint 3, 34 0.14      0.54 .012    .657
## 8           belc:timepoint 1, 34 0.14      0.65 .005    .426
## 9 condition:gere:timepoint 6, 34 0.14      1.38 .057    .253
## ---
## Signif. codes:  0 '***' 0.001 '**' 0.01 '*' 0.05 '+' 0.1 ' ' 1
afex_plot(fit, x = "condition")
## NOTE: Results may be misleading due to involvement in interactions

afex_plot(fit, x = "gere")
## NOTE: Results may be misleading due to involvement in interactions

afex_plot(fit, x = "timepoint", error = "within")

RM-ANOVA - Instrumentality

Results - All

fit <- aov_ez(id="UID", dv="inst", data=long3, between=c("condition"), within="timepoint", covariate = c("expc","pofc"), factorize = F)
## Warning: Numerical variables NOT centered on 0 (i.e., likely bogus results): NA,
## NA
## Warning: More than one observation per cell, aggregating the data using mean
## (i.e, fun_aggregate = mean)!
## Warning: Missing values for following ID(s):
## ADMNF5431W, AFITA6644H, AIKAP6463M, AODNI3144W, AQMMS9810J, AVINR5897Y, AYQSB9061F, BIGPC8577I, BJCYW9313A, BKKKT3701U, BLJCY6370P, BMMVL7994H, BOMVH4712H, BQYKI9156R, BVDSH3315M, BWHCJ5688Q, BWISR1973A, BXWMH7130Y, CDHNQ4412P, CFOQZ9365X, CGXNN1861X, CHADU6389C, CHKFU7601Z, CHPID9225X, CIEHL6827C, CJDVL5022X, CMDRZ4165X, CMHID2661Y, COSQM7547U, CPHXV3544C, CQVAE8343T, CZHNG1043X, DDWDZ2339J, DEGUG8250Q, DEZFY5154Y, DKMEC3318T, DUOOW9047V, DWEFP0640O, ECLDJ0558Y, EDPLI4718X, EGNPY6902N, EKHKA1320A, ELPMT5534R, ENGNR4476F, ENLUW4800F, EQRSC1726I, ETLLC3301O, EWGTH0978U, EYJKI2890J, FBPYG4513B, FDKWO5036F, FDRFT1976E, FOCRY5697B, FSKKR7810Y, FWVEK3877Q, FXXAG6538G, FZWVI1954A, GAHHN6818C, GEVFJ6602J, GFITG4909Q, GLBEZ3029S, GOSED0340T, GQBDI2909G, GUMSQ0837O, GUOQX4544Z, HBIFA5295Q, HDDDR5624S, HHENH9196H, HHTYJ9067S, HMITM1029R, HOQSF9391C, HVTWX3105O, IACUK2046G, ICNZW6469J, IDVCE3660F, IHEIE1031K, IHLAB2822T, IIBWJ1162K, IPNYM5318I, IPVOL5900Z, IQKYL9589B, ITEDM8731I, IVVAN8695S, IVWLS8788C, IWQBG3288R, IXTMS7827C, IYGKJ9212Q, JAQOX6224H, JDRGO5214D, JQUSH7763H, JRWTN5883X, JTIXL1749F, KGZSZ9095U, KHTGW5971J, KNGSJ3794D, KODOI2800E, KOGCS2814C, KQZOR6889Y, KTTQO0724Q, KVDIE4239W, KYLRY5373V, LAZZJ4798S, LDKDH9352Y, LDTLX9551Z, LHJVD2959Y, LJGWF4027N, LJTCD1624Z, LOPWH7294L, LQYVS8333P, MAJVB6162F, MDPFH5436F, MHKVA5255G, MIBGQ4458A, MOLUH1393I, MRECG6247Z, MSMVZ2231V, NBUDI3965W, NECDB3791F, NGWDC4182S, NLEXE0538V, NNQVD9904G, NOTVI7273I, NPEEK2345M, NVKGR6192V, NXSHA4361K, NZDFV1837O, OAUDW7602Q, ODNEC0022H, OEHWZ9390E, OIHCW0935B, OIJRE9661O, ONXIQ1991Q, OQEDN0137J, ORMKY3774B, OSNKC4976B, PAMYY6490X, PCYUT6681A, PDRGN5764W, PESUP1240W, PHIAU9193C, PIFPO0872F, PIMZK0270T, PLKLW0979X, PPDBO1961W, PQSIQ7727D, PSAJP5500H, PTGGV6124L, PYGEI1292D, QFEKE7247Z, QFTUS4663W, QFYEE1921U, QHDSO1457K, QNAVN3908M, QONZX2504S, QSDTY7001Q, QUOZW7823S, QWQPC1344Y, QXQOZ9556W, RIBRN0153W, RRBBT6871J, RTNDB7987U, RXGVF0879U, SACNZ0120A, SADTX3579I, SBFTP7166E, SBUTD1285I, SDTRI9296D, SGUWP5939A, SJOBN2309Y, SKZTW6437N, SLYQX2976O, SQDCF2656X, SRAGV3585D, STHXN7574W, STTSA5710O, SVRGK7887X, SXXFS1531Q, SZSPH6169O, TDBIA5306X, TEMIM8896R, TFZAE0194I, TJBJJ8859W, TJDHC0458U, TLPKY1497B, TMHJY2195F, TNOMW4491R, TWCNI6874W, TXDDF1309A, TYMVS5920H, UCYZA4959S, UDHFD1638I, UDYOK1503U, UEWZT9177C, UGHOD6018P, UGNMS7589N, UHUGQ9517B, ULLCV9227K, UVBLS1638O, VBCFZ4637X, VELTF4260O, VEMXD5267D, VETKA3898N, VGJRD7424G, VHUIR1579V, VIFDZ9388B, VIQDL0582N, VJOCD9817C, VNSSV4633E, VPMAK8781R, VYPFA1221S, WDOOB5760W, WHDVV0726Z, WIGYE4790M, WKMZJ5561X, WPBCK1207N, WUVZT3628Z, WVCXC4155L, WVPCP3440D, WYSWO1471E, XBYQZ8055T, XGNJP1929D, XKCRD7820V, XVCKT5197T, XWLZJ8035G, XXJXG2725P, XZQWA3103I, YAIBB9238C, YHKSR3208E, YIGNG1445M, YIWQF4610I, YOGUC5356J, YTUFO0314P, YVRMZ9228Y, YWQXI5864U, YXZRO8024H, YZKBO8058U, YZTVA8918Z, YZVMS4925G, ZBGBY1736H, ZCFIN4382O, ZCNDI2519Z, ZIKXQ5287N, ZJCBX9350S, ZQCVQ9164X, ZUOTR6815P, ZWPKK7562T, ZWTSA1605F, ZYFYX4806X
## Removing those cases from the analysis.
## Contrasts set to contr.sum for the following variables: condition
nice(fit)
## Anova Table (Type 3 tests)
## 
## Response: inst
##                Effect   df  MSE    F  ges p.value
## 1           condition 2, 4 0.54 1.56 .397    .316
## 2                expc 1, 4 0.54 0.03 .006    .870
## 3                pofc 1, 4 0.54 3.16 .400    .150
## 4           timepoint 1, 4 0.10 3.07 .107    .154
## 5 condition:timepoint 2, 4 0.10 2.98 .188    .162
## 6      expc:timepoint 1, 4 0.10 0.17 .006    .704
## 7      pofc:timepoint 1, 4 0.10 3.58 .122    .131
## ---
## Signif. codes:  0 '***' 0.001 '**' 0.01 '*' 0.05 '+' 0.1 ' ' 1

Results - Race & Gender

No covariates - empty cells?

fit2 <- aov_ez(id="UID", dv="inst", data=long3, between=c("condition","re_minority"), within="timepoint")
## Warning: More than one observation per cell, aggregating the data using mean
## (i.e, fun_aggregate = mean)!
## Warning: Missing values for following ID(s):
## DEZFY5154Y, GLBEZ3029S, IQKYL9589B, IYGKJ9212Q, JDRGO5214D, JTIXL1749F, KHTGW5971J, KNGSJ3794D, MOLUH1393I, NGWDC4182S, NOTVI7273I, OIHCW0935B, OIJRE9661O, PPDBO1961W, PSAJP5500H, QFYEE1921U, QXQOZ9556W, SBUTD1285I, SJOBN2309Y, SKZTW6437N, SRAGV3585D, STHXN7574W, UGNMS7589N, UHUGQ9517B, VBCFZ4637X, VETKA3898N, VPMAK8781R, WIGYE4790M, WVCXC4155L, WVPCP3440D, XKCRD7820V, YZKBO8058U, YZVMS4925G, ZQCVQ9164X
## Removing those cases from the analysis.
## Contrasts set to contr.sum for the following variables: condition, re_minority
nice(fit2)
## Anova Table (Type 3 tests)
## 
## Response: inst
##                            Effect     df  MSE       F  ges p.value
## 1                       condition 2, 217 0.78    0.77 .005    .463
## 2                     re_minority 1, 217 0.78    2.44 .008    .119
## 3           condition:re_minority 2, 217 0.78    0.81 .005    .447
## 4                       timepoint 1, 217 0.38 9.23 ** .014    .003
## 5             condition:timepoint 2, 217 0.38    2.20 .007    .114
## 6           re_minority:timepoint 1, 217 0.38    0.92 .001    .339
## 7 condition:re_minority:timepoint 2, 217 0.38  3.65 * .011    .028
## ---
## Signif. codes:  0 '***' 0.001 '**' 0.01 '*' 0.05 '+' 0.1 ' ' 1

RM-ANOVA - Perceptions of Future

Results - All

fit <- aov_ez(id="UID", dv="pof", data=long3, between=c("condition"), within="timepoint", covariate = c("expc","instc"), factorize = F)
## Warning: Numerical variables NOT centered on 0 (i.e., likely bogus results): NA,
## NA
## Warning: More than one observation per cell, aggregating the data using mean
## (i.e, fun_aggregate = mean)!
## Warning: Missing values for following ID(s):
## ADMNF5431W, AFITA6644H, AIKAP6463M, AODNI3144W, AQMMS9810J, AVINR5897Y, AYQSB9061F, BIGPC8577I, BJCYW9313A, BKKKT3701U, BLJCY6370P, BMMVL7994H, BOMVH4712H, BQYKI9156R, BVDSH3315M, BWHCJ5688Q, BWISR1973A, BXWMH7130Y, CDHNQ4412P, CFOQZ9365X, CGXNN1861X, CHADU6389C, CHKFU7601Z, CHPID9225X, CIEHL6827C, CJDVL5022X, CMDRZ4165X, CMHID2661Y, COSQM7547U, CPHXV3544C, CQVAE8343T, CZHNG1043X, DDWDZ2339J, DEGUG8250Q, DEZFY5154Y, DKMEC3318T, DUOOW9047V, DWEFP0640O, ECLDJ0558Y, EGNPY6902N, EKHKA1320A, ELPMT5534R, ENGNR4476F, ENHLY4123X, ENLUW4800F, EQRSC1726I, ETLLC3301O, EWGTH0978U, EYJKI2890J, FBPYG4513B, FDKWO5036F, FOCRY5697B, FSKKR7810Y, FWVEK3877Q, FXXAG6538G, FZWVI1954A, GAHHN6818C, GEVFJ6602J, GFITG4909Q, GLBEZ3029S, GOSED0340T, GQBDI2909G, GUMSQ0837O, GUOQX4544Z, HBIFA5295Q, HDDDR5624S, HHENH9196H, HHTYJ9067S, HMITM1029R, HOQSF9391C, HVTWX3105O, IACUK2046G, ICNZW6469J, IDVCE3660F, IHEIE1031K, IHLAB2822T, IIBWJ1162K, IPNYM5318I, IPVOL5900Z, IQKYL9589B, ITEDM8731I, IVVAN8695S, IVWLS8788C, IWQBG3288R, IXTMS7827C, IYGKJ9212Q, JAQOX6224H, JDRGO5214D, JQUSH7763H, JRWTN5883X, JTIXL1749F, KGZSZ9095U, KHTGW5971J, KNGSJ3794D, KODOI2800E, KOGCS2814C, KQZOR6889Y, KTTQO0724Q, KVDIE4239W, KYLRY5373V, LDKDH9352Y, LDTLX9551Z, LHJVD2959Y, LJGWF4027N, LJTCD1624Z, LOPWH7294L, LQYVS8333P, MAJVB6162F, MDPFH5436F, MHKVA5255G, MIBGQ4458A, MOLUH1393I, MRECG6247Z, MSMVZ2231V, NBUDI3965W, NECDB3791F, NGWDC4182S, NLEXE0538V, NNQVD9904G, NOTVI7273I, NPEEK2345M, NVKGR6192V, NXSHA4361K, OAUDW7602Q, OIHCW0935B, OIJRE9661O, ONXIQ1991Q, OQEDN0137J, ORMKY3774B, OSNKC4976B, PAMYY6490X, PCYUT6681A, PDRGN5764W, PESUP1240W, PHIAU9193C, PIFPO0872F, PIMZK0270T, PLKLW0979X, PPDBO1961W, PQSIQ7727D, PSAJP5500H, PTGGV6124L, PYGEI1292D, QFEKE7247Z, QFTUS4663W, QFYEE1921U, QHDSO1457K, QNAVN3908M, QONZX2504S, QUOZW7823S, QWQPC1344Y, QXQOZ9556W, RIBRN0153W, RRBBT6871J, RTNDB7987U, RXGVF0879U, SACNZ0120A, SADTX3579I, SBFTP7166E, SBUTD1285I, SDTRI9296D, SGUWP5939A, SJOBN2309Y, SKZTW6437N, SLYQX2976O, SQDCF2656X, SRAGV3585D, STHXN7574W, STTSA5710O, SVRGK7887X, SXXFS1531Q, SZSPH6169O, TDBIA5306X, TEMIM8896R, TJBJJ8859W, TJDHC0458U, TLPKY1497B, TMHJY2195F, TNOMW4491R, TWCNI6874W, TYMVS5920H, UCYZA4959S, UDHFD1638I, UDYOK1503U, UEWZT9177C, UGHOD6018P, UGNMS7589N, UHUGQ9517B, ULLCV9227K, UVBLS1638O, VBCFZ4637X, VELTF4260O, VEMXD5267D, VETKA3898N, VGJRD7424G, VHUIR1579V, VIFDZ9388B, VIQDL0582N, VJOCD9817C, VNSSV4633E, VPMAK8781R, VYPFA1221S, WDOOB5760W, WHDVV0726Z, WIGYE4790M, WKMZJ5561X, WPBCK1207N, WUVZT3628Z, WVCXC4155L, WVPCP3440D, WYSWO1471E, XBYQZ8055T, XGNJP1929D, XKCRD7820V, XVCKT5197T, XWLZJ8035G, XXJXG2725P, XZQWA3103I, YHKSR3208E, YIWQF4610I, YOGUC5356J, YTUFO0314P, YVRMZ9228Y, YWQXI5864U, YZKBO8058U, YZTVA8918Z, YZVMS4925G, ZBGBY1736H, ZCFIN4382O, ZCNDI2519Z, ZIKXQ5287N, ZJCBX9350S, ZQCVQ9164X, ZUOTR6815P, ZWPKK7562T, ZWTSA1605F, ZYFYX4806X, ZZCEK2517F
## Removing those cases from the analysis.
## Contrasts set to contr.sum for the following variables: condition
nice(fit)
## Anova Table (Type 3 tests)
## 
## Response: pof
##                Effect    df  MSE         F  ges p.value
## 1           condition 2, 14 0.28    6.30 * .396    .011
## 2                expc 1, 14 0.28      1.93 .091    .187
## 3               instc 1, 14 0.28 27.44 *** .588   <.001
## 4           timepoint 1, 14 0.10      0.09 .002    .775
## 5 condition:timepoint 2, 14 0.10      1.08 .040    .367
## 6      expc:timepoint 1, 14 0.10      0.13 .003    .720
## 7     instc:timepoint 1, 14 0.10      0.05 .001    .818
## ---
## Signif. codes:  0 '***' 0.001 '**' 0.01 '*' 0.05 '+' 0.1 ' ' 1

Results - Race & Gender

fit <- aov_ez(id="UID", dv="pof", data=long3, between=c("condition","ge_minority"), within="timepoint", covariate = c("expc","instc"), factorize = F)
## Warning: Numerical variables NOT centered on 0 (i.e., likely bogus results): NA,
## NA
## Warning: More than one observation per cell, aggregating the data using mean
## (i.e, fun_aggregate = mean)!
## Warning: Missing values for following ID(s):
## ADMNF5431W, AFITA6644H, AIKAP6463M, AODNI3144W, AQMMS9810J, AVINR5897Y, AYQSB9061F, BIGPC8577I, BJCYW9313A, BKKKT3701U, BLJCY6370P, BMMVL7994H, BOMVH4712H, BQYKI9156R, BVDSH3315M, BWHCJ5688Q, BWISR1973A, BXWMH7130Y, CDHNQ4412P, CFOQZ9365X, CGXNN1861X, CHADU6389C, CHKFU7601Z, CHPID9225X, CIEHL6827C, CJDVL5022X, CMDRZ4165X, CMHID2661Y, COSQM7547U, CPHXV3544C, CQVAE8343T, CZHNG1043X, DDWDZ2339J, DEGUG8250Q, DEZFY5154Y, DKMEC3318T, DUOOW9047V, DWEFP0640O, ECLDJ0558Y, EGNPY6902N, EKHKA1320A, ELPMT5534R, ENGNR4476F, ENHLY4123X, ENLUW4800F, EQRSC1726I, ETLLC3301O, EWGTH0978U, EYJKI2890J, FBPYG4513B, FDKWO5036F, FOCRY5697B, FSKKR7810Y, FWVEK3877Q, FXXAG6538G, FZWVI1954A, GAHHN6818C, GEVFJ6602J, GFITG4909Q, GLBEZ3029S, GOSED0340T, GQBDI2909G, GUMSQ0837O, GUOQX4544Z, HBIFA5295Q, HDDDR5624S, HHENH9196H, HHTYJ9067S, HMITM1029R, HOQSF9391C, HVTWX3105O, IACUK2046G, ICNZW6469J, IDVCE3660F, IHEIE1031K, IHLAB2822T, IIBWJ1162K, IPNYM5318I, IPVOL5900Z, IQKYL9589B, ITEDM8731I, IVVAN8695S, IVWLS8788C, IWQBG3288R, IXTMS7827C, IYGKJ9212Q, JAQOX6224H, JDRGO5214D, JQUSH7763H, JRWTN5883X, JTIXL1749F, KGZSZ9095U, KHTGW5971J, KNGSJ3794D, KODOI2800E, KOGCS2814C, KQZOR6889Y, KTTQO0724Q, KVDIE4239W, KYLRY5373V, LDKDH9352Y, LDTLX9551Z, LHJVD2959Y, LJGWF4027N, LJTCD1624Z, LOPWH7294L, LQYVS8333P, MAJVB6162F, MDPFH5436F, MHKVA5255G, MIBGQ4458A, MOLUH1393I, MRECG6247Z, MSMVZ2231V, NBUDI3965W, NECDB3791F, NGWDC4182S, NLEXE0538V, NNQVD9904G, NOTVI7273I, NPEEK2345M, NVKGR6192V, NXSHA4361K, OAUDW7602Q, OIHCW0935B, OIJRE9661O, ONXIQ1991Q, OQEDN0137J, ORMKY3774B, OSNKC4976B, PAMYY6490X, PCYUT6681A, PDRGN5764W, PESUP1240W, PHIAU9193C, PIFPO0872F, PIMZK0270T, PLKLW0979X, PPDBO1961W, PQSIQ7727D, PSAJP5500H, PTGGV6124L, PYGEI1292D, QFEKE7247Z, QFTUS4663W, QFYEE1921U, QHDSO1457K, QNAVN3908M, QONZX2504S, QUOZW7823S, QWQPC1344Y, QXQOZ9556W, RIBRN0153W, RRBBT6871J, RTNDB7987U, RXGVF0879U, SACNZ0120A, SADTX3579I, SBFTP7166E, SBUTD1285I, SDTRI9296D, SGUWP5939A, SJOBN2309Y, SKZTW6437N, SLYQX2976O, SQDCF2656X, SRAGV3585D, STHXN7574W, STTSA5710O, SVRGK7887X, SXXFS1531Q, SZSPH6169O, TDBIA5306X, TEMIM8896R, TJBJJ8859W, TJDHC0458U, TLPKY1497B, TMHJY2195F, TNOMW4491R, TWCNI6874W, TYMVS5920H, UCYZA4959S, UDHFD1638I, UDYOK1503U, UEWZT9177C, UGHOD6018P, UGNMS7589N, UHUGQ9517B, ULLCV9227K, UVBLS1638O, VBCFZ4637X, VELTF4260O, VEMXD5267D, VETKA3898N, VGJRD7424G, VHUIR1579V, VIFDZ9388B, VIQDL0582N, VJOCD9817C, VNSSV4633E, VPMAK8781R, VYPFA1221S, WDOOB5760W, WHDVV0726Z, WIGYE4790M, WKMZJ5561X, WPBCK1207N, WUVZT3628Z, WVCXC4155L, WVPCP3440D, WYSWO1471E, XBYQZ8055T, XGNJP1929D, XKCRD7820V, XVCKT5197T, XWLZJ8035G, XXJXG2725P, XZQWA3103I, YHKSR3208E, YIWQF4610I, YOGUC5356J, YTUFO0314P, YVRMZ9228Y, YWQXI5864U, YZKBO8058U, YZTVA8918Z, YZVMS4925G, ZBGBY1736H, ZCFIN4382O, ZCNDI2519Z, ZIKXQ5287N, ZJCBX9350S, ZQCVQ9164X, ZUOTR6815P, ZWPKK7562T, ZWTSA1605F, ZYFYX4806X, ZZCEK2517F
## Removing those cases from the analysis.
## Contrasts set to contr.sum for the following variables: condition, ge_minority
nice(fit)
## Anova Table (Type 3 tests)
## 
## Response: pof
##                             Effect    df  MSE         F   ges p.value
## 1                        condition 2, 11 0.16   9.89 **  .538    .003
## 2                      ge_minority 1, 11 0.16  10.48 **  .381    .008
## 3                             expc 1, 11 0.16      0.01 <.001    .925
## 4                            instc 1, 11 0.16 53.19 ***  .758   <.001
## 5            condition:ge_minority 2, 11 0.16      2.39  .220    .137
## 6                        timepoint 1, 11 0.09      0.10  .003    .761
## 7              condition:timepoint 2, 11 0.09    3.11 +  .166    .085
## 8            ge_minority:timepoint 1, 11 0.09      0.05  .002    .823
## 9                   expc:timepoint 1, 11 0.09      0.00 <.001    .978
## 10                 instc:timepoint 1, 11 0.09      0.29  .009    .601
## 11 condition:ge_minority:timepoint 2, 11 0.09      2.71  .148    .110
## ---
## Signif. codes:  0 '***' 0.001 '**' 0.01 '*' 0.05 '+' 0.1 ' ' 1
fit <- aov_ez(id="UID", dv="pof", data=long3, between=c("condition","re_minority"), within="timepoint", covariate = c("expc","instc"), factorize = F)
## Warning: Numerical variables NOT centered on 0 (i.e., likely bogus results): NA,
## NA
## Warning: More than one observation per cell, aggregating the data using mean
## (i.e, fun_aggregate = mean)!
## Warning: Missing values for following ID(s):
## ADMNF5431W, AFITA6644H, AIKAP6463M, AODNI3144W, AQMMS9810J, AVINR5897Y, AYQSB9061F, BIGPC8577I, BJCYW9313A, BKKKT3701U, BLJCY6370P, BMMVL7994H, BOMVH4712H, BQYKI9156R, BVDSH3315M, BWHCJ5688Q, BWISR1973A, BXWMH7130Y, CDHNQ4412P, CFOQZ9365X, CGXNN1861X, CHADU6389C, CHKFU7601Z, CHPID9225X, CIEHL6827C, CJDVL5022X, CMDRZ4165X, CMHID2661Y, COSQM7547U, CPHXV3544C, CQVAE8343T, CZHNG1043X, DDWDZ2339J, DEGUG8250Q, DEZFY5154Y, DKMEC3318T, DUOOW9047V, DWEFP0640O, ECLDJ0558Y, EGNPY6902N, EKHKA1320A, ELPMT5534R, ENGNR4476F, ENHLY4123X, ENLUW4800F, EQRSC1726I, ETLLC3301O, EWGTH0978U, EYJKI2890J, FBPYG4513B, FDKWO5036F, FOCRY5697B, FSKKR7810Y, FWVEK3877Q, FXXAG6538G, FZWVI1954A, GAHHN6818C, GEVFJ6602J, GFITG4909Q, GLBEZ3029S, GOSED0340T, GQBDI2909G, GUMSQ0837O, GUOQX4544Z, HBIFA5295Q, HDDDR5624S, HHENH9196H, HHTYJ9067S, HMITM1029R, HOQSF9391C, HVTWX3105O, IACUK2046G, ICNZW6469J, IDVCE3660F, IHEIE1031K, IHLAB2822T, IIBWJ1162K, IPNYM5318I, IPVOL5900Z, IQKYL9589B, ITEDM8731I, IVVAN8695S, IVWLS8788C, IWQBG3288R, IXTMS7827C, IYGKJ9212Q, JAQOX6224H, JDRGO5214D, JQUSH7763H, JRWTN5883X, JTIXL1749F, KGZSZ9095U, KHTGW5971J, KNGSJ3794D, KODOI2800E, KOGCS2814C, KQZOR6889Y, KTTQO0724Q, KVDIE4239W, KYLRY5373V, LDKDH9352Y, LDTLX9551Z, LHJVD2959Y, LJGWF4027N, LJTCD1624Z, LOPWH7294L, LQYVS8333P, MAJVB6162F, MDPFH5436F, MHKVA5255G, MIBGQ4458A, MOLUH1393I, MRECG6247Z, MSMVZ2231V, NBUDI3965W, NECDB3791F, NGWDC4182S, NLEXE0538V, NNQVD9904G, NOTVI7273I, NPEEK2345M, NVKGR6192V, NXSHA4361K, OAUDW7602Q, OIHCW0935B, OIJRE9661O, ONXIQ1991Q, OQEDN0137J, ORMKY3774B, OSNKC4976B, PAMYY6490X, PCYUT6681A, PDRGN5764W, PESUP1240W, PHIAU9193C, PIFPO0872F, PIMZK0270T, PLKLW0979X, PPDBO1961W, PQSIQ7727D, PSAJP5500H, PTGGV6124L, PYGEI1292D, QFEKE7247Z, QFTUS4663W, QFYEE1921U, QHDSO1457K, QNAVN3908M, QONZX2504S, QUOZW7823S, QWQPC1344Y, QXQOZ9556W, RIBRN0153W, RRBBT6871J, RTNDB7987U, RXGVF0879U, SACNZ0120A, SADTX3579I, SBFTP7166E, SBUTD1285I, SDTRI9296D, SGUWP5939A, SJOBN2309Y, SKZTW6437N, SLYQX2976O, SQDCF2656X, SRAGV3585D, STHXN7574W, STTSA5710O, SVRGK7887X, SXXFS1531Q, SZSPH6169O, TDBIA5306X, TEMIM8896R, TJBJJ8859W, TJDHC0458U, TLPKY1497B, TMHJY2195F, TNOMW4491R, TWCNI6874W, TYMVS5920H, UCYZA4959S, UDHFD1638I, UDYOK1503U, UEWZT9177C, UGHOD6018P, UGNMS7589N, UHUGQ9517B, ULLCV9227K, UVBLS1638O, VBCFZ4637X, VELTF4260O, VEMXD5267D, VETKA3898N, VGJRD7424G, VHUIR1579V, VIFDZ9388B, VIQDL0582N, VJOCD9817C, VNSSV4633E, VPMAK8781R, VYPFA1221S, WDOOB5760W, WHDVV0726Z, WIGYE4790M, WKMZJ5561X, WPBCK1207N, WUVZT3628Z, WVCXC4155L, WVPCP3440D, WYSWO1471E, XBYQZ8055T, XGNJP1929D, XKCRD7820V, XVCKT5197T, XWLZJ8035G, XXJXG2725P, XZQWA3103I, YHKSR3208E, YIWQF4610I, YOGUC5356J, YTUFO0314P, YVRMZ9228Y, YWQXI5864U, YZKBO8058U, YZTVA8918Z, YZVMS4925G, ZBGBY1736H, ZCFIN4382O, ZCNDI2519Z, ZIKXQ5287N, ZJCBX9350S, ZQCVQ9164X, ZUOTR6815P, ZWPKK7562T, ZWTSA1605F, ZYFYX4806X, ZZCEK2517F
## Removing those cases from the analysis.
## Contrasts set to contr.sum for the following variables: condition, re_minority
nice(fit)
## Anova Table (Type 3 tests)
## 
## Response: pof
##                             Effect    df  MSE         F  ges p.value
## 1                        condition 2, 11 0.23   9.63 ** .554    .004
## 2                      re_minority 1, 11 0.23    3.31 + .176    .096
## 3                             expc 1, 11 0.23      1.18 .071    .301
## 4                            instc 1, 11 0.23 29.87 *** .658   <.001
## 5            condition:re_minority 2, 11 0.23      2.46 .241    .131
## 6                        timepoint 1, 11 0.09      0.99 .025    .342
## 7              condition:timepoint 2, 11 0.09      2.12 .100    .167
## 8            re_minority:timepoint 1, 11 0.09      0.17 .005    .685
## 9                   expc:timepoint 1, 11 0.09      0.98 .025    .344
## 10                 instc:timepoint 1, 11 0.09      0.21 .005    .659
## 11 condition:re_minority:timepoint 2, 11 0.09      2.34 .110    .143
## ---
## Signif. codes:  0 '***' 0.001 '**' 0.01 '*' 0.05 '+' 0.1 ' ' 1

RM-ANOVA - Expectancy

Results - All

fit <- aov_ez(id="UID", dv="exp", data=long3, between=c("condition"), within="timepoint")
## Warning: More than one observation per cell, aggregating the data using mean
## (i.e, fun_aggregate = mean)!
## Warning: Missing values for following ID(s):
## DEZFY5154Y, GLBEZ3029S, IQKYL9589B, IYGKJ9212Q, JDRGO5214D, JTIXL1749F, KHTGW5971J, KNGSJ3794D, MOLUH1393I, NGWDC4182S, NOTVI7273I, OIJRE9661O, PPDBO1961W, PSAJP5500H, QFYEE1921U, QXQOZ9556W, SJOBN2309Y, SRAGV3585D, STHXN7574W, UGNMS7589N, UHUGQ9517B, VETKA3898N, VPMAK8781R, WIGYE4790M, WVCXC4155L, WVPCP3440D, YZKBO8058U, YZVMS4925G, ZQCVQ9164X
## Removing those cases from the analysis.
## Contrasts set to contr.sum for the following variables: condition
nice(fit)
## Anova Table (Type 3 tests)
## 
## Response: exp
##                Effect     df  MSE         F   ges p.value
## 1           condition 2, 225 1.42      0.64  .004    .529
## 2           timepoint 1, 225 0.38 12.66 ***  .012   <.001
## 3 condition:timepoint 2, 225 0.38      0.25 <.001    .778
## ---
## Signif. codes:  0 '***' 0.001 '**' 0.01 '*' 0.05 '+' 0.1 ' ' 1
afex_plot(fit, x = "timepoint", error = "within")

Results - Race/Ethnicity

fit <- aov_ez(id="UID", dv="exp", data=long3, between=c("condition","re_minority"), within="timepoint")
## Warning: More than one observation per cell, aggregating the data using mean
## (i.e, fun_aggregate = mean)!
## Warning: Missing values for following ID(s):
## DEZFY5154Y, GLBEZ3029S, IQKYL9589B, IYGKJ9212Q, JDRGO5214D, JTIXL1749F, KHTGW5971J, KNGSJ3794D, MOLUH1393I, NGWDC4182S, NOTVI7273I, OIHCW0935B, OIJRE9661O, PPDBO1961W, PSAJP5500H, QFYEE1921U, QXQOZ9556W, SBUTD1285I, SJOBN2309Y, SKZTW6437N, SRAGV3585D, STHXN7574W, UGNMS7589N, UHUGQ9517B, VBCFZ4637X, VETKA3898N, VPMAK8781R, WIGYE4790M, WVCXC4155L, WVPCP3440D, XKCRD7820V, YZKBO8058U, YZVMS4925G, ZQCVQ9164X
## Removing those cases from the analysis.
## Contrasts set to contr.sum for the following variables: condition, re_minority
nice(fit)
## Anova Table (Type 3 tests)
## 
## Response: exp
##                            Effect     df  MSE        F   ges p.value
## 1                       condition 2, 217 1.39     0.77  .006    .465
## 2                     re_minority 1, 217 1.39   4.25 *  .015    .040
## 3           condition:re_minority 2, 217 1.39     0.25  .002    .781
## 4                       timepoint 1, 217 0.39 11.01 **  .011    .001
## 5             condition:timepoint 2, 217 0.39     0.16 <.001    .856
## 6           re_minority:timepoint 1, 217 0.39     0.45 <.001    .502
## 7 condition:re_minority:timepoint 2, 217 0.39     0.42 <.001    .656
## ---
## Signif. codes:  0 '***' 0.001 '**' 0.01 '*' 0.05 '+' 0.1 ' ' 1
afex_plot(fit, x = "timepoint", trace = "re_minority", panel = "condition")
## Warning: Panel(s) show a mixed within-between-design.
## Error bars do not allow comparisons across all means.
## Suppress error bars with: error = "none"

Results - Gender

fit <- aov_ez(id="UID", dv="exp", data=long3, between=c("condition","ge_minority"), within="timepoint")
## Warning: More than one observation per cell, aggregating the data using mean
## (i.e, fun_aggregate = mean)!
## Warning: Missing values for following ID(s):
## DEZFY5154Y, EKHKA1320A, GLBEZ3029S, IQKYL9589B, IYGKJ9212Q, JDRGO5214D, JTIXL1749F, KHTGW5971J, KNGSJ3794D, MOLUH1393I, NGWDC4182S, NOTVI7273I, OIHCW0935B, OIJRE9661O, PPDBO1961W, PSAJP5500H, QFYEE1921U, QXQOZ9556W, SBUTD1285I, SJOBN2309Y, SKZTW6437N, SRAGV3585D, STHXN7574W, UGNMS7589N, UHUGQ9517B, VBCFZ4637X, VETKA3898N, VPMAK8781R, WIGYE4790M, WVCXC4155L, WVPCP3440D, XKCRD7820V, YZKBO8058U, YZVMS4925G, ZQCVQ9164X
## Removing those cases from the analysis.
## Contrasts set to contr.sum for the following variables: condition, ge_minority
nice(fit)
## Anova Table (Type 3 tests)
## 
## Response: exp
##                            Effect     df  MSE         F   ges p.value
## 1                       condition 2, 216 1.34      0.91  .007    .404
## 2                     ge_minority 1, 216 1.34 11.58 ***  .040   <.001
## 3           condition:ge_minority 2, 216 1.34      0.17  .001    .846
## 4                       timepoint 1, 216 0.39   9.31 **  .010    .003
## 5             condition:timepoint 2, 216 0.39      0.13 <.001    .874
## 6           ge_minority:timepoint 1, 216 0.39      0.06 <.001    .799
## 7 condition:ge_minority:timepoint 2, 216 0.39      0.35 <.001    .708
## ---
## Signif. codes:  0 '***' 0.001 '**' 0.01 '*' 0.05 '+' 0.1 ' ' 1
afex_plot(fit, x = "ge_minority")
## NOTE: Results may be misleading due to involvement in interactions

RM-ANOVA - Test Anxiety

Results - All

fit <- aov_ez(id="UID", dv="anx", data=long3, between=c("condition"), within="timepoint")
## Warning: More than one observation per cell, aggregating the data using mean
## (i.e, fun_aggregate = mean)!
## Warning: Missing values for following ID(s):
## DEZFY5154Y, GLBEZ3029S, IQKYL9589B, IYGKJ9212Q, JDRGO5214D, JTIXL1749F, KHTGW5971J, KNGSJ3794D, MOLUH1393I, NGWDC4182S, NOTVI7273I, OIJRE9661O, PPDBO1961W, PSAJP5500H, QFYEE1921U, QXQOZ9556W, SJOBN2309Y, SRAGV3585D, STHXN7574W, UGNMS7589N, UHUGQ9517B, VETKA3898N, VPMAK8781R, WIGYE4790M, WVCXC4155L, WVPCP3440D, YZKBO8058U, YZVMS4925G, ZQCVQ9164X
## Removing those cases from the analysis.
## Contrasts set to contr.sum for the following variables: condition
nice(fit)
## Anova Table (Type 3 tests)
## 
## Response: anx
##                Effect     df  MSE      F  ges p.value
## 1           condition 2, 225 3.45 4.26 * .031    .015
## 2           timepoint 1, 225 0.69 3.20 + .002    .075
## 3 condition:timepoint 2, 225 0.69   1.78 .003    .172
## ---
## Signif. codes:  0 '***' 0.001 '**' 0.01 '*' 0.05 '+' 0.1 ' ' 1
afex_plot(fit, x = "timepoint", error = "within")

afex_plot(fit, x = "condition")

Results - Race/Ethnicity

fit <- aov_ez(id="UID", dv="anx", data=long3, between=c("condition","re_minority"), within="timepoint")
## Warning: More than one observation per cell, aggregating the data using mean
## (i.e, fun_aggregate = mean)!
## Warning: Missing values for following ID(s):
## DEZFY5154Y, GLBEZ3029S, IQKYL9589B, IYGKJ9212Q, JDRGO5214D, JTIXL1749F, KHTGW5971J, KNGSJ3794D, MOLUH1393I, NGWDC4182S, NOTVI7273I, OIHCW0935B, OIJRE9661O, PPDBO1961W, PSAJP5500H, QFYEE1921U, QXQOZ9556W, SBUTD1285I, SJOBN2309Y, SKZTW6437N, SRAGV3585D, STHXN7574W, UGNMS7589N, UHUGQ9517B, VBCFZ4637X, VETKA3898N, VPMAK8781R, WIGYE4790M, WVCXC4155L, WVPCP3440D, XKCRD7820V, YZKBO8058U, YZVMS4925G, ZQCVQ9164X
## Removing those cases from the analysis.
## Contrasts set to contr.sum for the following variables: condition, re_minority
nice(fit)
## Anova Table (Type 3 tests)
## 
## Response: anx
##                            Effect     df  MSE      F   ges p.value
## 1                       condition 2, 217 3.40 4.48 *  .033    .012
## 2                     re_minority 1, 217 3.40 4.44 *  .017    .036
## 3           condition:re_minority 2, 217 3.40   0.09 <.001    .916
## 4                       timepoint 1, 217 0.70 3.37 +  .003    .068
## 5             condition:timepoint 2, 217 0.70   1.51  .002    .224
## 6           re_minority:timepoint 1, 217 0.70   0.27 <.001    .601
## 7 condition:re_minority:timepoint 2, 217 0.70   0.11 <.001    .896
## ---
## Signif. codes:  0 '***' 0.001 '**' 0.01 '*' 0.05 '+' 0.1 ' ' 1

Results - Gender

fit <- aov_ez(id="UID", dv="anx", data=long3, between=c("condition","ge_minority"), within="timepoint")
## Warning: More than one observation per cell, aggregating the data using mean
## (i.e, fun_aggregate = mean)!
## Warning: Missing values for following ID(s):
## DEZFY5154Y, EKHKA1320A, GLBEZ3029S, IQKYL9589B, IYGKJ9212Q, JDRGO5214D, JTIXL1749F, KHTGW5971J, KNGSJ3794D, MOLUH1393I, NGWDC4182S, NOTVI7273I, OIHCW0935B, OIJRE9661O, PPDBO1961W, PSAJP5500H, QFYEE1921U, QXQOZ9556W, SBUTD1285I, SJOBN2309Y, SKZTW6437N, SRAGV3585D, STHXN7574W, UGNMS7589N, UHUGQ9517B, VBCFZ4637X, VETKA3898N, VPMAK8781R, WIGYE4790M, WVCXC4155L, WVPCP3440D, XKCRD7820V, YZKBO8058U, YZVMS4925G, ZQCVQ9164X
## Removing those cases from the analysis.
## Contrasts set to contr.sum for the following variables: condition, ge_minority
nice(fit)
## Anova Table (Type 3 tests)
## 
## Response: anx
##                            Effect     df  MSE      F   ges p.value
## 1                       condition 2, 216 3.44 4.49 *  .034    .012
## 2                     ge_minority 1, 216 3.44   2.37  .009    .125
## 3           condition:ge_minority 2, 216 3.44   0.10 <.001    .903
## 4                       timepoint 1, 216 0.68   1.15 <.001    .285
## 5             condition:timepoint 2, 216 0.68   0.09 <.001    .917
## 6           ge_minority:timepoint 1, 216 0.68   2.00  .002    .158
## 7 condition:ge_minority:timepoint 2, 216 0.68 2.99 +  .005    .052
## ---
## Signif. codes:  0 '***' 0.001 '**' 0.01 '*' 0.05 '+' 0.1 ' ' 1
afex_plot(fit, x = "timepoint", trace="ge_minority", panel="condition")
## Warning: Panel(s) show a mixed within-between-design.
## Error bars do not allow comparisons across all means.
## Suppress error bars with: error = "none"

pairs(emmeans(fit, specs="timepoint", by=c("ge_minority","condition")))
## ge_minority = ge_maj, condition = con:
##  contrast estimate    SE  df t.ratio p.value
##  X1 - X3    0.5654 0.171 216   3.316  0.0011
## 
## ge_minority = ge_min, condition = con:
##  contrast estimate    SE  df t.ratio p.value
##  X1 - X3   -0.2826 0.276 216  -1.026  0.3062
## 
## ge_minority = ge_maj, condition = equi:
##  contrast estimate    SE  df t.ratio p.value
##  X1 - X3    0.1616 0.158 216   1.025  0.3064
## 
## ge_minority = ge_min, condition = equi:
##  contrast estimate    SE  df t.ratio p.value
##  X1 - X3    0.0280 0.239 216   0.117  0.9067
## 
## ge_minority = ge_maj, condition = int:
##  contrast estimate    SE  df t.ratio p.value
##  X1 - X3   -0.0665 0.152 216  -0.437  0.6624
## 
## ge_minority = ge_min, condition = int:
##  contrast estimate    SE  df t.ratio p.value
##  X1 - X3    0.1635 0.268 216   0.610  0.5427
pairs(emmeans(fit, specs="ge_minority", by=c("timepoint","condition")))
## timepoint = X1, condition = con:
##  contrast        estimate    SE  df t.ratio p.value
##  ge_maj - ge_min    0.141 0.375 216   0.377  0.7068
## 
## timepoint = X3, condition = con:
##  contrast        estimate    SE  df t.ratio p.value
##  ge_maj - ge_min   -0.707 0.420 216  -1.683  0.0939
## 
## timepoint = X1, condition = equi:
##  contrast        estimate    SE  df t.ratio p.value
##  ge_maj - ge_min   -0.145 0.331 216  -0.437  0.6624
## 
## timepoint = X3, condition = equi:
##  contrast        estimate    SE  df t.ratio p.value
##  ge_maj - ge_min   -0.278 0.371 216  -0.750  0.4539
## 
## timepoint = X1, condition = int:
##  contrast        estimate    SE  df t.ratio p.value
##  ge_maj - ge_min   -0.538 0.357 216  -1.510  0.1326
## 
## timepoint = X3, condition = int:
##  contrast        estimate    SE  df t.ratio p.value
##  ge_maj - ge_min   -0.308 0.400 216  -0.771  0.4417
pairs(emmeans(fit, specs="condition", by=c("timepoint","ge_minority")))
## timepoint = X1, ge_minority = ge_maj:
##  contrast   estimate    SE  df t.ratio p.value
##  con - equi   0.6107 0.269 216   2.274  0.0616
##  con - int    0.1367 0.264 216   0.517  0.8630
##  equi - int  -0.4740 0.253 216  -1.871  0.1496
## 
## timepoint = X3, ge_minority = ge_maj:
##  contrast   estimate    SE  df t.ratio p.value
##  con - equi   0.2069 0.301 216   0.687  0.7712
##  con - int   -0.4953 0.296 216  -1.671  0.2186
##  equi - int  -0.7022 0.284 216  -2.472  0.0377
## 
## timepoint = X1, ge_minority = ge_min:
##  contrast   estimate    SE  df t.ratio p.value
##  con - equi   0.3250 0.421 216   0.771  0.7211
##  con - int   -0.5427 0.445 216  -1.221  0.4423
##  equi - int  -0.8677 0.415 216  -2.090  0.0942
## 
## timepoint = X3, ge_minority = ge_min:
##  contrast   estimate    SE  df t.ratio p.value
##  con - equi   0.6356 0.473 216   1.345  0.3719
##  con - int   -0.0965 0.499 216  -0.194  0.9795
##  equi - int  -0.7321 0.465 216  -1.573  0.2595
## 
## P value adjustment: tukey method for comparing a family of 3 estimates

EFA - PMAPs

EFA

names(long3)
##  [1] "UID"              "Q2_1"             "Q2_2"             "Q2_3"            
##  [5] "Q2_4"             "Q2_5"             "Q2_6"             "Q2_7"            
##  [9] "Q2_8"             "Q3_1"             "Q3_2"             "Q3_3"            
## [13] "Q3_4"             "Q3_5"             "Q3_6"             "Q3_7"            
## [17] "Q3_8"             "Q3_9"             "Q3_10"            "Q3_11"           
## [21] "Q3_12"            "Q4_1"             "Q4_2"             "Q4_3"            
## [25] "Q4_4"             "Q4_5"             "condition"        "ins"             
## [29] "Q47_1_pmaps"      "Q47_2_pmaps"      "Q47_3_pmaps"      "Q47_4_pmaps"     
## [33] "Q47_5_pmaps"      "Q47_6_pmaps"      "Q47_7_pmaps"      "Q47_8_pmaps"     
## [37] "re_other"         "re_asian"         "re_black"         "re_white"        
## [41] "re_latin"         "re_mena"          "re_nhpi"          "re_aian"         
## [45] "ge_other"         "ge_cis"           "ge_m"             "ge_gq"           
## [49] "ge_nb"            "ge_w"             "pre_misscount"    "post_misscount"  
## [53] "follow_misscount" "timepoint"        "bel"              "rec"             
## [57] "inst"             "pof"              "exp"              "anx"             
## [61] "triv"             "vig"              "re_minority"      "ge_minority"     
## [65] "gere"             "belc"             "recc"             "instc"           
## [69] "pofc"             "expc"             "anxc"
d <- na.omit(subset(long3, timepoint == 1, select=c(grep("_pmaps",colnames(long3)))))

ev <- eigen(cor(d)) # get eigenvalues
ap <- parallel(subject = nrow(d), var = ncol(d),
               rep = 100,cent = .05)
nS <- nScree(x = ev$values, aparallel = ap$eigen$qevpea)
plotnScree(nS)

fit <- factanal(d, 2, rotation="promax") 
print(fit, digits = 3, cutoff = 0.3, sort = TRUE) 
## 
## Call:
## factanal(x = d, factors = 2, rotation = "promax")
## 
## Uniquenesses:
## Q47_1_pmaps Q47_2_pmaps Q47_3_pmaps Q47_4_pmaps Q47_5_pmaps Q47_6_pmaps 
##       0.220       0.156       0.120       0.317       0.518       0.107 
## Q47_7_pmaps Q47_8_pmaps 
##       0.190       0.540 
## 
## Loadings:
##             Factor1 Factor2
## Q47_1_pmaps  0.905         
## Q47_2_pmaps  0.938         
## Q47_3_pmaps  0.945         
## Q47_4_pmaps  0.807         
## Q47_5_pmaps          0.659 
## Q47_6_pmaps          0.973 
## Q47_7_pmaps          0.920 
## Q47_8_pmaps          0.634 
## 
##                Factor1 Factor2
## SS loadings      3.267   2.638
## Proportion Var   0.408   0.330
## Cumulative Var   0.408   0.738
## 
## Factor Correlations:
##         Factor1 Factor2
## Factor1   1.000  -0.408
## Factor2  -0.408   1.000
## 
## Test of the hypothesis that 2 factors are sufficient.
## The chi square statistic is 18.14 on 13 degrees of freedom.
## The p-value is 0.152

Cronbach’s Alpha

psych::alpha(subset(d, select=c(1:4)))
## 
## Reliability analysis   
## Call: psych::alpha(x = subset(d, select = c(1:4)))
## 
##   raw_alpha std.alpha G6(smc) average_r S/N    ase mean  sd median_r
##       0.94      0.94    0.92      0.79  15 0.0072  2.7 1.5     0.79
## 
##     95% confidence boundaries 
##          lower alpha upper
## Feldt     0.92  0.94  0.95
## Duhachek  0.92  0.94  0.95
## 
##  Reliability if an item is dropped:
##             raw_alpha std.alpha G6(smc) average_r  S/N alpha se   var.r med.r
## Q47_1_pmaps      0.92      0.92    0.89      0.79 11.6   0.0100 0.00349  0.77
## Q47_2_pmaps      0.91      0.91    0.88      0.78 10.4   0.0108 0.00246  0.77
## Q47_3_pmaps      0.90      0.91    0.87      0.76  9.7   0.0115 0.00155  0.75
## Q47_4_pmaps      0.94      0.94    0.91      0.83 15.0   0.0075 0.00076  0.83
## 
##  Item statistics 
##               n raw.r std.r r.cor r.drop mean  sd
## Q47_1_pmaps 212  0.92  0.92  0.88   0.85  2.8 1.7
## Q47_2_pmaps 212  0.93  0.93  0.91   0.87  2.5 1.6
## Q47_3_pmaps 212  0.94  0.94  0.93   0.89  2.6 1.6
## Q47_4_pmaps 212  0.89  0.89  0.82   0.80  2.9 1.8
## 
## Non missing response frequency for each item
##                1    2    3    4    5    6    7 miss
## Q47_1_pmaps 0.27 0.22 0.18 0.15 0.10 0.05 0.03    0
## Q47_2_pmaps 0.35 0.22 0.19 0.13 0.07 0.01 0.03    0
## Q47_3_pmaps 0.32 0.26 0.17 0.08 0.11 0.02 0.04    0
## Q47_4_pmaps 0.29 0.21 0.17 0.12 0.08 0.06 0.06    0
psych::alpha(subset(d, select=c(5:8)))
## 
## Reliability analysis   
## Call: psych::alpha(x = subset(d, select = c(5:8)))
## 
##   raw_alpha std.alpha G6(smc) average_r S/N   ase mean  sd median_r
##       0.87      0.88    0.86      0.64 7.1 0.015  4.8 1.2     0.61
## 
##     95% confidence boundaries 
##          lower alpha upper
## Feldt     0.84  0.87   0.9
## Duhachek  0.84  0.87   0.9
## 
##  Reliability if an item is dropped:
##             raw_alpha std.alpha G6(smc) average_r S/N alpha se  var.r med.r
## Q47_5_pmaps      0.86      0.87    0.84      0.69 6.8    0.017 0.0187  0.61
## Q47_6_pmaps      0.80      0.80    0.73      0.57 4.0    0.024 0.0037  0.60
## Q47_7_pmaps      0.81      0.81    0.75      0.59 4.3    0.023 0.0063  0.61
## Q47_8_pmaps      0.87      0.88    0.85      0.70 7.0    0.016 0.0180  0.65
## 
##  Item statistics 
##               n raw.r std.r r.cor r.drop mean  sd
## Q47_5_pmaps 212  0.81  0.81  0.69   0.65  4.6 1.5
## Q47_6_pmaps 212  0.91  0.91  0.91   0.83  5.0 1.4
## Q47_7_pmaps 212  0.89  0.90  0.88   0.80  5.1 1.4
## Q47_8_pmaps 212  0.81  0.80  0.68   0.64  4.7 1.6
## 
## Non missing response frequency for each item
##                1    2    3    4    5    6    7 miss
## Q47_5_pmaps 0.03 0.08 0.09 0.24 0.29 0.18 0.10    0
## Q47_6_pmaps 0.03 0.03 0.06 0.17 0.35 0.24 0.12    0
## Q47_7_pmaps 0.03 0.02 0.07 0.11 0.33 0.32 0.11    0
## Q47_8_pmaps 0.06 0.06 0.07 0.25 0.24 0.21 0.12    0

ANOVA - PMAPs Trivialization

  • Racial minorities are too worried about being discriminated against.
  • Racial minorities are too sensitive about stereotypes.
  • Minorities today are overly worried about being victims of racism.
  • People are overly concerned about racial issues.

Results - All

fit <- aov_ez(id="UID", dv="triv", data=long3, between=c("condition"), covariate = c("belc","anxc"), factorize = F)
## Warning: Numerical variables NOT centered on 0 (i.e., likely bogus results): NA,
## NA
## Warning: More than one observation per cell, aggregating the data using mean
## (i.e, fun_aggregate = mean)!
## Warning: Missing values for following ID(s):
## AVINR5897Y, BIGPC8577I, BJCYW9313A, BLJCY6370P, BMMVL7994H, CHADU6389C, CPHXV3544C, CZHNG1043X, DDWDZ2339J, DEZFY5154Y, DWEFP0640O, EKHKA1320A, ELPMT5534R, ENHLY4123X, ENLUW4800F, EPVXZ9711C, EQRSC1726I, ETLLC3301O, FBPYG4513B, FDKWO5036F, FDRFT1976E, FWVEK3877Q, FXXAG6538G, GFITG4909Q, GLBEZ3029S, GOSED0340T, GQBDI2909G, HBIFA5295Q, HDDDR5624S, HHTYJ9067S, HMITM1029R, HOQSF9391C, ICNZW6469J, IDVCE3660F, IQKYL9589B, IYGKJ9212Q, JDRGO5214D, JTIXL1749F, KHTGW5971J, KNGSJ3794D, KOGCS2814C, KTTQO0724Q, KVDIE4239W, KYLRY5373V, LHJVD2959Y, LJTCD1624Z, LOPWH7294L, MHKVA5255G, MIBGQ4458A, MOLUH1393I, MRECG6247Z, NGWDC4182S, NLEXE0538V, NOTVI7273I, NPEEK2345M, NVKGR6192V, OIJRE9661O, OSNKC4976B, PESUP1240W, PHIAU9193C, PIMZK0270T, PPDBO1961W, PSAJP5500H, QFYEE1921U, QHDSO1457K, QONZX2504S, QXQOZ9556W, SACNZ0120A, SBUTD1285I, SJOBN2309Y, SLYQX2976O, SRAGV3585D, STHXN7574W, SVRGK7887X, TFZAE0194I, TXDDF1309A, UGNMS7589N, UHUGQ9517B, UIQOX7318R, VETKA3898N, VNSSV4633E, VPMAK8781R, VYPFA1221S, WHDVV0726Z, WIGYE4790M, WUVZT3628Z, WVCXC4155L, WVPCP3440D, XBYQZ8055T, XXJXG2725P, YIWQF4610I, YWQXI5864U, YXZRO8024H, YZKBO8058U, YZTVA8918Z, YZVMS4925G, ZIKXQ5287N, ZQCVQ9164X, ZYFYX4806X
## Removing those cases from the analysis.
## Contrasts set to contr.sum for the following variables: condition
nice(fit)
## Anova Table (Type 3 tests)
## 
## Response: triv
##      Effect     df  MSE    F   ges p.value
## 1 condition 2, 309 1.40 1.11  .007    .329
## 2      belc 1, 309 1.40 0.18 <.001    .672
## 3      anxc 1, 309 1.40 0.00 <.001    .974
## ---
## Signif. codes:  0 '***' 0.001 '**' 0.01 '*' 0.05 '+' 0.1 ' ' 1
afex_plot(fit, x = "condition")

Results - Gender and Race - Control Only

Control group only: - Significant main effect of gender/race (p = .059). Seems to be driven by difference in trivialization between majority women/nb and minority men (p = .05)

long4 <- subset(long3, condition == "con")
fit <- aov_ez(id="UID", dv="triv", data=long4, between=c("gere"), covariate = c("belc","anxc"), factorize = F)
## Warning: Numerical variables NOT centered on 0 (i.e., likely bogus results): NA,
## NA
## Warning: More than one observation per cell, aggregating the data using mean
## (i.e, fun_aggregate = mean)!
## Warning: Missing values for following ID(s):
## EKHKA1320A, EPVXZ9711C, ETLLC3301O, FDKWO5036F, FXXAG6538G, GFITG4909Q, GQBDI2909G, JDRGO5214D, JTIXL1749F, KTTQO0724Q, NGWDC4182S, NLEXE0538V, OIHCW0935B, OIJRE9661O, PESUP1240W, PPDBO1961W, PSAJP5500H, QONZX2504S, STHXN7574W, SVRGK7887X, TFZAE0194I, VBCFZ4637X, VETKA3898N, VPMAK8781R, WUVZT3628Z, XBYQZ8055T, XKCRD7820V, YWQXI5864U, YXZRO8024H, ZQCVQ9164X, ZYFYX4806X
## Removing those cases from the analysis.
## Contrasts set to contr.sum for the following variables: gere
nice(fit)
## Anova Table (Type 3 tests)
## 
## Response: triv
##   Effect    df  MSE      F  ges p.value
## 1   gere 3, 89 1.25   1.92 .061    .132
## 2   belc 1, 89 1.25   0.53 .006    .467
## 3   anxc 1, 89 1.25 5.76 * .061    .018
## ---
## Signif. codes:  0 '***' 0.001 '**' 0.01 '*' 0.05 '+' 0.1 ' ' 1
afex_plot(fit, x = "gere")

pairs(emmeans(fit, specs = "gere"))
##  contrast                               estimate    SE df t.ratio p.value
##  men of color - white men                  0.256 0.276 89   0.928  0.7902
##  men of color - (white women/nb)           0.562 0.369 89   1.524  0.4279
##  men of color - (women/nb of color)       -0.563 0.440 89  -1.278  0.5793
##  white men - (white women/nb)              0.306 0.393 89   0.778  0.8642
##  white men - (women/nb of color)          -0.819 0.468 89  -1.749  0.3051
##  (white women/nb) - (women/nb of color)   -1.125 0.504 89  -2.234  0.1219
## 
## P value adjustment: tukey method for comparing a family of 4 estimates
summ <- summarySE(long4, measurevar="triv", groupvars=c("gere"), na.rm = T)
summ
##                gere  N     triv        sd        se        ci
## 1      men of color 42 3.273810 1.3203491 0.2037343 0.4114497
## 2         white men 32 2.812500 1.0453430 0.1847923 0.3768863
## 3    white women/nb 14 2.678571 1.0535132 0.2815633 0.6082804
## 4 women/nb of color  8 4.000000 0.8451543 0.2988072 0.7065666
## 5              <NA>  6 3.750000 1.0246951 0.4183300 1.0753515

Results - Gender x Race Interaction - All Conditions

  • Equivalent group is where largest difference emerged, with minority men in this condition reporting less trivialization than their control/intervention counterparts.
  • Intervention group has possible effect for minority women/nb, although effect is not significant, perhaps due to small size of group (n = 28).

All conditions: - Significant main effect of condition (.016), gender/race (< .001) - Significant interaction (p = .006)

Post-hoc tests, condition: - Equivalent group significantly lower than control (.016) and intervention (.057)

Post-hoc tests, gender/race: - Majority men significantly higher than majority women/nb (.014) - Majority men significantly higher than minority women/nb (.043) - Minority men significantly higher than majority women/nb (< .001) - Minority men significantly higher than minority women/nb (.001)

Post-hoc tests, interaction: - Minority men in intervention condition significantly lower than minority men in control and equivalent conditions (p = .003 & p = .001) - In control group: Majority men significantly lower than minority men (.099) and majority women/nb significantly lower than minority men (.081) - In equivalent group: Majority men significantly higher than majority women/nb (p = .006) - In intervention group: Minority men significantly higher than majority men (p = .008), majority women/nb (p < .001), and minority women/nb (p < .001); majority men significantly higher than minority women/nb (p - .013) - Majority men did not differ by group - Majority women/nb lower in equivalent when compared to control (p = .071) - Minority men significantly lower in equivalent when compared to control (p = .003) and intervention (p < .001) - Minority women/nb did not differ by group

table(long3$gere, long3$condition)
##                    
##                     con equi int
##   men of color       46   54  70
##   white men          48   56  54
##   white women/nb     22   28  20
##   women/nb of color  14   20  18
fit <- aov_ez(id="UID", dv="triv", data=long3, between=c("condition","gere"), covariate = c("belc","anxc"), factorize = F)
## Warning: Numerical variables NOT centered on 0 (i.e., likely bogus results): NA,
## NA
## Warning: More than one observation per cell, aggregating the data using mean
## (i.e, fun_aggregate = mean)!
## Warning: Missing values for following ID(s):
## AVINR5897Y, BIGPC8577I, BJCYW9313A, BLJCY6370P, BMMVL7994H, CHADU6389C, CPHXV3544C, CZHNG1043X, DDWDZ2339J, DEZFY5154Y, DWEFP0640O, EKHKA1320A, ELPMT5534R, ENHLY4123X, ENLUW4800F, EPVXZ9711C, EQRSC1726I, ETLLC3301O, FBPYG4513B, FDKWO5036F, FDRFT1976E, FWVEK3877Q, FXXAG6538G, GFITG4909Q, GLBEZ3029S, GOSED0340T, GQBDI2909G, HBIFA5295Q, HDDDR5624S, HHTYJ9067S, HMITM1029R, HOQSF9391C, ICNZW6469J, IDVCE3660F, IQKYL9589B, IYGKJ9212Q, JDRGO5214D, JTIXL1749F, KHTGW5971J, KNGSJ3794D, KOGCS2814C, KTTQO0724Q, KVDIE4239W, KYLRY5373V, LHJVD2959Y, LJTCD1624Z, LOPWH7294L, MHKVA5255G, MIBGQ4458A, MOLUH1393I, MRECG6247Z, NGWDC4182S, NLEXE0538V, NOTVI7273I, NPEEK2345M, NVKGR6192V, OIHCW0935B, OIJRE9661O, OSNKC4976B, PESUP1240W, PHIAU9193C, PIMZK0270T, PPDBO1961W, PSAJP5500H, QFYEE1921U, QHDSO1457K, QONZX2504S, QXQOZ9556W, SACNZ0120A, SBUTD1285I, SJOBN2309Y, SKZTW6437N, SLYQX2976O, SRAGV3585D, STHXN7574W, SVRGK7887X, TFZAE0194I, TXDDF1309A, UGNMS7589N, UHUGQ9517B, UIQOX7318R, VBCFZ4637X, VETKA3898N, VNSSV4633E, VPMAK8781R, VYPFA1221S, WHDVV0726Z, WIGYE4790M, WUVZT3628Z, WVCXC4155L, WVPCP3440D, XBYQZ8055T, XKCRD7820V, XXJXG2725P, YIWQF4610I, YWQXI5864U, YXZRO8024H, YZKBO8058U, YZTVA8918Z, YZVMS4925G, ZIKXQ5287N, ZQCVQ9164X, ZYFYX4806X
## Removing those cases from the analysis.
## Contrasts set to contr.sum for the following variables: condition, gere
nice(fit)
## Anova Table (Type 3 tests)
## 
## Response: triv
##           Effect     df  MSE        F   ges p.value
## 1      condition 2, 292 1.27     2.25  .015    .107
## 2           gere 3, 292 1.27     2.07  .021    .104
## 3           belc 1, 292 1.27     0.01 <.001    .938
## 4           anxc 1, 292 1.27     0.00 <.001    .947
## 5 condition:gere 6, 292 1.27 5.60 ***  .103   <.001
## ---
## Signif. codes:  0 '***' 0.001 '**' 0.01 '*' 0.05 '+' 0.1 ' ' 1
afex_plot(fit, x = "condition")
## NOTE: Results may be misleading due to involvement in interactions

pairs(emmeans(fit, specs = "condition"))
## NOTE: Results may be misleading due to involvement in interactions
##  contrast   estimate    SE  df t.ratio p.value
##  con - equi   0.3849 0.194 292   1.984  0.1180
##  con - int    0.3345 0.203 292   1.651  0.2261
##  equi - int  -0.0504 0.198 292  -0.255  0.9648
## 
## Results are averaged over the levels of: gere 
## P value adjustment: tukey method for comparing a family of 3 estimates
afex_plot(fit, x = "gere")
## NOTE: Results may be misleading due to involvement in interactions

pairs(emmeans(fit, specs = "gere"))
## NOTE: Results may be misleading due to involvement in interactions
##  contrast                               estimate    SE  df t.ratio p.value
##  men of color - white men                -0.0565 0.151 292  -0.375  0.9820
##  men of color - (white women/nb)          0.4839 0.216 292   2.237  0.1158
##  men of color - (women/nb of color)       0.0377 0.243 292   0.155  0.9987
##  white men - (white women/nb)             0.5404 0.224 292   2.407  0.0779
##  white men - (women/nb of color)          0.0943 0.256 292   0.369  0.9829
##  (white women/nb) - (women/nb of color)  -0.4461 0.285 292  -1.564  0.4009
## 
## Results are averaged over the levels of: condition 
## P value adjustment: tukey method for comparing a family of 4 estimates
afex_plot(fit, x = "gere", trace = "condition")

pairs(emmeans(fit, specs = "condition", by=c("gere")))
## gere = men of color:
##  contrast   estimate    SE  df t.ratio p.value
##  con - equi   0.7262 0.251 292   2.896  0.0113
##  con - int   -0.0385 0.247 292  -0.156  0.9867
##  equi - int  -0.7647 0.249 292  -3.072  0.0066
## 
## gere = white men:
##  contrast   estimate    SE  df t.ratio p.value
##  con - equi  -0.6756 0.265 292  -2.552  0.0301
##  con - int   -0.2992 0.269 292  -1.111  0.5080
##  equi - int   0.3765 0.254 292   1.483  0.3007
## 
## gere = white women/nb:
##  contrast   estimate    SE  df t.ratio p.value
##  con - equi   0.5376 0.426 292   1.262  0.4177
##  con - int   -0.2723 0.470 292  -0.579  0.8314
##  equi - int  -0.8099 0.470 292  -1.722  0.1986
## 
## gere = women/nb of color:
##  contrast   estimate    SE  df t.ratio p.value
##  con - equi   0.9513 0.534 292   1.781  0.1779
##  con - int    1.9478 0.535 292   3.639  0.0009
##  equi - int   0.9965 0.506 292   1.971  0.1213
## 
## P value adjustment: tukey method for comparing a family of 3 estimates
afex_plot(fit, x = "condition", trace = "gere")

pairs(emmeans(fit, specs = "gere", by=c("condition")))
## condition = con:
##  contrast                               estimate    SE  df t.ratio p.value
##  men of color - white men                  0.498 0.270 292   1.845  0.2547
##  men of color - (white women/nb)           0.625 0.353 292   1.769  0.2904
##  men of color - (women/nb of color)       -0.699 0.437 292  -1.599  0.3806
##  white men - (white women/nb)              0.127 0.369 292   0.344  0.9860
##  white men - (women/nb of color)          -1.197 0.453 292  -2.640  0.0431
##  (white women/nb) - (women/nb of color)   -1.324 0.501 292  -2.644  0.0427
## 
## condition = equi:
##  contrast                               estimate    SE  df t.ratio p.value
##  men of color - white men                 -0.904 0.247 292  -3.656  0.0017
##  men of color - (white women/nb)           0.436 0.354 292   1.230  0.6080
##  men of color - (women/nb of color)       -0.474 0.403 292  -1.177  0.6421
##  white men - (white women/nb)              1.340 0.352 292   3.812  0.0010
##  white men - (women/nb of color)           0.430 0.403 292   1.067  0.7100
##  (white women/nb) - (women/nb of color)   -0.910 0.468 292  -1.946  0.2115
## 
## condition = int:
##  contrast                               estimate    SE  df t.ratio p.value
##  men of color - white men                  0.237 0.254 292   0.933  0.7872
##  men of color - (white women/nb)           0.391 0.396 292   0.986  0.7573
##  men of color - (women/nb of color)        1.287 0.399 292   3.222  0.0077
##  white men - (white women/nb)              0.154 0.406 292   0.379  0.9814
##  white men - (women/nb of color)           1.050 0.414 292   2.535  0.0567
##  (white women/nb) - (women/nb of color)    0.896 0.505 292   1.775  0.2876
## 
## P value adjustment: tukey method for comparing a family of 4 estimates
summ <- summarySE(long3, measurevar="triv", groupvars=c("condition","gere"), na.rm = T)
## Warning in qt(conf.interval/2 + 0.5, datac$N - 1): NaNs produced
summ
##    condition              gere  N     triv        sd        se        ci
## 1        con      men of color 42 3.273810 1.3203491 0.2037343 0.4114497
## 2        con         white men 32 2.812500 1.0453430 0.1847923 0.3768863
## 3        con    white women/nb 14 2.678571 1.0535132 0.2815633 0.6082804
## 4        con women/nb of color  8 4.000000 0.8451543 0.2988072 0.7065666
## 5        con              <NA>  6 3.750000 1.0246951 0.4183300 1.0753515
## 6       equi      men of color 42 2.583333 1.0457036 0.1613556 0.3258642
## 7       equi         white men 42 3.488095 1.1541975 0.1780965 0.3596733
## 8       equi    white women/nb 14 2.142857 0.6333237 0.1692629 0.3656702
## 9       equi women/nb of color 10 3.050000 1.3270686 0.4196559 0.9493277
## 10      equi              <NA>  2 2.500000 0.0000000 0.0000000 0.0000000
## 11       int      men of color 44 3.397727 1.1950944 0.1801673 0.3633419
## 12       int         white men 40 3.112500 1.1968843 0.1892440 0.3827822
## 13       int    white women/nb 10 2.950000 1.1352924 0.3590110 0.8121393
## 14       int women/nb of color 10 2.050000 0.6101002 0.1929306 0.4364394
## 15       int              <NA>  0      NaN        NA        NA       NaN

ANOVA - PMAPs Vigilance

  • I think about why racial minorities are treated stereotypically.
  • I think about whether people act in a prejudiced or discriminatory manner.
  • I consider whether people’s actions are prejudiced or discriminatory.
  • I am on the lookout for instances of prejudice or discrimination.

Results - All

fit <- aov_ez(id="UID", dv="vig", data=long3, between=c("condition"), covariate = c("belc","anxc"), factorize = F)
## Warning: Numerical variables NOT centered on 0 (i.e., likely bogus results): NA,
## NA
## Warning: More than one observation per cell, aggregating the data using mean
## (i.e, fun_aggregate = mean)!
## Warning: Missing values for following ID(s):
## DEZFY5154Y, ELPMT5534R, ETLLC3301O, FBPYG4513B, FDKWO5036F, GFITG4909Q, GLBEZ3029S, HMITM1029R, HOQSF9391C, ICNZW6469J, IDVCE3660F, IQKYL9589B, IYGKJ9212Q, JDRGO5214D, JTIXL1749F, KHTGW5971J, KNGSJ3794D, KVDIE4239W, LHJVD2959Y, MIBGQ4458A, MOLUH1393I, NGWDC4182S, NLEXE0538V, NOTVI7273I, NPEEK2345M, OIJRE9661O, PESUP1240W, PPDBO1961W, PSAJP5500H, QFYEE1921U, QXQOZ9556W, SJOBN2309Y, SRAGV3585D, STHXN7574W, SVRGK7887X, UGNMS7589N, UHUGQ9517B, VETKA3898N, VPMAK8781R, WIGYE4790M, WVCXC4155L, WVPCP3440D, XXJXG2725P, YZKBO8058U, YZTVA8918Z, YZVMS4925G, ZIKXQ5287N, ZQCVQ9164X
## Removing those cases from the analysis.
## Contrasts set to contr.sum for the following variables: condition
nice(fit)
## Anova Table (Type 3 tests)
## 
## Response: vig
##      Effect     df  MSE    F  ges p.value
## 1 condition 2, 411 1.52 1.61 .008    .201
## 2      belc 1, 411 1.52 2.57 .006    .110
## 3      anxc 1, 411 1.52 2.04 .005    .154
## ---
## Signif. codes:  0 '***' 0.001 '**' 0.01 '*' 0.05 '+' 0.1 ' ' 1

Results - Gender and Race (Control Only)

  • Significant differences in vigilance in control group (p = .005)
  • Majority men significantly higher than minority men (p = .051)
  • Minority men significantly lower than minority women/nb (p = .004)
fit <- aov_ez(id="UID", dv="vig", data=long4, between=c("gere"), covariate = c("belc","anxc"), factorize = F)
## Warning: Numerical variables NOT centered on 0 (i.e., likely bogus results): NA,
## NA
## Warning: More than one observation per cell, aggregating the data using mean
## (i.e, fun_aggregate = mean)!
## Warning: Missing values for following ID(s):
## EKHKA1320A, ETLLC3301O, FDKWO5036F, GFITG4909Q, JDRGO5214D, JTIXL1749F, NGWDC4182S, NLEXE0538V, OIHCW0935B, OIJRE9661O, PESUP1240W, PPDBO1961W, PSAJP5500H, STHXN7574W, SVRGK7887X, VBCFZ4637X, VETKA3898N, VPMAK8781R, XKCRD7820V, ZQCVQ9164X
## Removing those cases from the analysis.
## Contrasts set to contr.sum for the following variables: gere
nice(fit)
## Anova Table (Type 3 tests)
## 
## Response: vig
##   Effect     df  MSE      F   ges p.value
## 1   gere 3, 111 1.46 3.16 *  .079    .027
## 2   belc 1, 111 1.46   0.08 <.001    .781
## 3   anxc 1, 111 1.46   0.00 <.001    .952
## ---
## Signif. codes:  0 '***' 0.001 '**' 0.01 '*' 0.05 '+' 0.1 ' ' 1
afex_plot(fit, x = "gere")

pairs(emmeans(fit, specs = "gere"))
##  contrast                               estimate    SE  df t.ratio p.value
##  men of color - white men                -0.6980 0.264 111  -2.643  0.0457
##  men of color - (white women/nb)         -0.7586 0.350 111  -2.170  0.1380
##  men of color - (women/nb of color)      -0.6868 0.415 111  -1.655  0.3527
##  white men - (white women/nb)            -0.0607 0.365 111  -0.166  0.9984
##  white men - (women/nb of color)          0.0112 0.435 111   0.026  1.0000
##  (white women/nb) - (women/nb of color)   0.0718 0.455 111   0.158  0.9986
## 
## P value adjustment: tukey method for comparing a family of 4 estimates

Results - Race/Ethnicity

  • Equivalent task decreased vigilance for minority women/nb when compared to control (.078) while increasing vigilance for minority men (< .001). Intervention task increased vigilance for majority men when compared to equivalent (.020). Intervention and equivalent tasks increased vigilance for minority men when compared to control (.020 & < .001).
  • No differences in vigilance in intervention group (removing previous differences between majority/minority men and minority men and women/nb)

Significant main effect of gender/race (p = .090): - Majority men higher than minority women/nb (p = .099)

Significant interaction (p < .001): - In control group: majority men significantly higher than minority men (.064); minority men significantly lower than minority women (.004) - In equivalent group: majority men significantly lower than majority women/nb (.003) and minority men (.007) - In intervention group: no differences - Majority men: equivalent group significantly lower than intervention group (.020) - Majority women/nb: no significant differences - Minority men: control significantly lower than equivalent and intervention (< .001 & .020) - Minority women/nb: control significantly higher than equivalent (.078)

fit <- aov_ez(id="UID", dv="vig", data=long3, between=c("condition","gere"), covariate = c("belc","anxc"), factorize = F)
## Warning: Numerical variables NOT centered on 0 (i.e., likely bogus results): NA,
## NA
## Warning: More than one observation per cell, aggregating the data using mean
## (i.e, fun_aggregate = mean)!
## Warning: Missing values for following ID(s):
## DEZFY5154Y, EKHKA1320A, ELPMT5534R, ETLLC3301O, FBPYG4513B, FDKWO5036F, GFITG4909Q, GLBEZ3029S, HMITM1029R, HOQSF9391C, ICNZW6469J, IDVCE3660F, IQKYL9589B, IYGKJ9212Q, JDRGO5214D, JTIXL1749F, KHTGW5971J, KNGSJ3794D, KVDIE4239W, LHJVD2959Y, MIBGQ4458A, MOLUH1393I, NGWDC4182S, NLEXE0538V, NOTVI7273I, NPEEK2345M, OIHCW0935B, OIJRE9661O, PESUP1240W, PPDBO1961W, PSAJP5500H, QFYEE1921U, QXQOZ9556W, SBUTD1285I, SJOBN2309Y, SKZTW6437N, SRAGV3585D, STHXN7574W, SVRGK7887X, UGNMS7589N, UHUGQ9517B, VBCFZ4637X, VETKA3898N, VPMAK8781R, WIGYE4790M, WVCXC4155L, WVPCP3440D, XKCRD7820V, XXJXG2725P, YZKBO8058U, YZTVA8918Z, YZVMS4925G, ZIKXQ5287N, ZQCVQ9164X
## Removing those cases from the analysis.
## Contrasts set to contr.sum for the following variables: condition, gere
nice(fit)
## Anova Table (Type 3 tests)
## 
## Response: vig
##           Effect     df  MSE      F  ges p.value
## 1      condition 2, 390 1.51   0.26 .001    .772
## 2           gere 3, 390 1.51 2.29 + .017    .078
## 3           belc 1, 390 1.51 5.56 * .014    .019
## 4           anxc 1, 390 1.51   1.92 .005    .166
## 5 condition:gere 6, 390 1.51 2.46 * .036    .024
## ---
## Signif. codes:  0 '***' 0.001 '**' 0.01 '*' 0.05 '+' 0.1 ' ' 1
afex_plot(fit, x = "gere")
## NOTE: Results may be misleading due to involvement in interactions

pairs(emmeans(fit, specs = "gere"))
## NOTE: Results may be misleading due to involvement in interactions
##  contrast                               estimate    SE  df t.ratio p.value
##  men of color - white men                 -0.211 0.146 390  -1.446  0.4713
##  men of color - (white women/nb)          -0.384 0.190 390  -2.023  0.1813
##  men of color - (women/nb of color)       -0.447 0.214 390  -2.091  0.1577
##  white men - (white women/nb)             -0.173 0.196 390  -0.885  0.8126
##  white men - (women/nb of color)          -0.236 0.222 390  -1.061  0.7135
##  (white women/nb) - (women/nb of color)   -0.063 0.244 390  -0.258  0.9939
## 
## Results are averaged over the levels of: condition 
## P value adjustment: tukey method for comparing a family of 4 estimates
afex_plot(fit, x = "condition", trace = "gere")

pairs(emmeans(fit, specs = "gere", by = "condition"))
## condition = con:
##  contrast                               estimate    SE  df t.ratio p.value
##  men of color - white men                -0.6760 0.265 390  -2.553  0.0536
##  men of color - (white women/nb)         -0.8436 0.345 390  -2.443  0.0708
##  men of color - (women/nb of color)      -0.7831 0.405 390  -1.934  0.2157
##  white men - (white women/nb)            -0.1676 0.352 390  -0.476  0.9644
##  white men - (women/nb of color)         -0.1071 0.413 390  -0.259  0.9939
##  (white women/nb) - (women/nb of color)   0.0605 0.459 390   0.132  0.9992
## 
## condition = equi:
##  contrast                               estimate    SE  df t.ratio p.value
##  men of color - white men                -0.1610 0.242 390  -0.666  0.9099
##  men of color - (white women/nb)         -0.8032 0.304 390  -2.640  0.0427
##  men of color - (women/nb of color)      -0.2708 0.343 390  -0.789  0.8593
##  white men - (white women/nb)            -0.6422 0.304 390  -2.112  0.1510
##  white men - (women/nb of color)         -0.1098 0.342 390  -0.321  0.9886
##  (white women/nb) - (women/nb of color)   0.5324 0.385 390   1.382  0.5112
## 
## condition = int:
##  contrast                               estimate    SE  df t.ratio p.value
##  men of color - white men                 0.2031 0.244 390   0.831  0.8396
##  men of color - (white women/nb)          0.4938 0.332 390   1.488  0.4458
##  men of color - (women/nb of color)      -0.2880 0.334 390  -0.863  0.8238
##  white men - (white women/nb)             0.2907 0.342 390   0.849  0.8310
##  white men - (women/nb of color)         -0.4911 0.347 390  -1.414  0.4914
##  (white women/nb) - (women/nb of color)  -0.7818 0.410 390  -1.905  0.2275
## 
## P value adjustment: tukey method for comparing a family of 4 estimates
afex_plot(fit, x = "gere", trace = "condition")

pairs(emmeans(fit, specs = "condition", by = "gere"))
## gere = men of color:
##  contrast   estimate    SE  df t.ratio p.value
##  con - equi  -0.3657 0.251 390  -1.458  0.3125
##  con - int   -0.8016 0.245 390  -3.266  0.0034
##  equi - int  -0.4359 0.237 390  -1.836  0.1593
## 
## gere = white men:
##  contrast   estimate    SE  df t.ratio p.value
##  con - equi   0.1493 0.256 390   0.583  0.8291
##  con - int    0.0775 0.260 390   0.298  0.9522
##  equi - int  -0.0718 0.249 390  -0.289  0.9551
## 
## gere = white women/nb:
##  contrast   estimate    SE  df t.ratio p.value
##  con - equi  -0.3252 0.384 390  -0.846  0.6744
##  con - int    0.5358 0.413 390   1.297  0.3978
##  equi - int   0.8611 0.385 390   2.234  0.0668
## 
## gere = women/nb of color:
##  contrast   estimate    SE  df t.ratio p.value
##  con - equi   0.1467 0.459 390   0.319  0.9453
##  con - int   -0.3065 0.460 390  -0.667  0.7831
##  equi - int  -0.4531 0.413 390  -1.097  0.5163
## 
## P value adjustment: tukey method for comparing a family of 3 estimates