Yan Wang 13/12/2021
- 1 Load all datasets
- 2 Data preparation
- 2.1 Calculate the dependent variable = the difference of symptom controllability scores at baseline and 8 weeks (at the end of the intervention)
- 2.2 Reshape Df1 because we want one patient has multiple entries for symptoms, baseline contrallability scores, and controllability scores
- 2.3 Merge with OEQ (text data) by Participant ID and Symptom
- 2.4 save the output
- 3 Extract linguistics features (predictors) for analysis
- 4 Load the full dataset for analysis
- 5 Merge the datasets to obtain the sociodemographic factors
- 6 Descriptive stats of the controllability score changes @ baseline and 8 wks
- 7 Preliminary bivariate correlation between the dependent variable and potential predictors - pearson correlation
- 8 Model buiding - Mixed-effect regression model
- 9 variable loadings on component 3, 7, 13, 22, 24, 26, 30, 33
- 10 compare toy 13 and 11 and select toy 13 based on low AIC value
- 11 Final model
- 12 Explanation of the results
This script is a new replacement
SQR is a data set that contains participants ID, time of controllability assessmen [Administration Number], 3 symptoms [S1, S2, S3] and 3 controllability scores [S1Cont, S2Cont, S3Cont]; OEQ contains ID, symptom names and answers for the three questions: 1. “How does the symptom make you feel and what’s the cause?”, 2. “How does the symptom affect you?” 3.“Have you tried anything? Is it helpful?”; Metadata contains ID and sociodemographics, including age, marrige status, employment, education, race, ethnicity.
2.1 Calculate the dependent variable = the difference of symptom controllability scores at baseline and 8 weeks (at the end of the intervention)
SQR_8<- SQR %>% filter(`Administration Number`== "8 week f/u")
SQR_BL<- SQR %>% filter(`Administration Number`=="Baseline (week 0)")
names(SQR_8)[3:9]<-c("Time", "S1Cont", "S2Cont", "S3Cont" , "S1", "S2", "S3")
names(SQR_BL)[3:9]<-c("Time", "S1Cont", "S2Cont", "S3Cont" , "S1", "S2", "S3")
sum(is.na(SQR_BL$S1Cont))[1] 0
sum(is.na(SQR_8$S1Cont))[1] 41
SQR_8BL<-inner_join(SQR_8, SQR_BL, by=c("Participant ID", "GOGID", "S1", "S2", "S3"))
names(SQR_8BL) [1] "Participant ID" "GOGID" "Time.x" "S1Cont.x"
[5] "S2Cont.x" "S3Cont.x" "S1" "S2"
[9] "S3" "Time.y" "S1Cont.y" "S2Cont.y"
[13] "S3Cont.y"
Df1<-SQR_8BL %>% dplyr::select(1,2, 7:9, 11:13, 4:6)
names(Df1)[1:11]<-c("ID", "GOGID", "S1", "S2", "S3", "BLS1", "BLS2", "BLS3", "8wkS1","8wkS2", "8wkS3" )2.2 Reshape Df1 because we want one patient has multiple entries for symptoms, baseline contrallability scores, and controllability scores
Df1_1<-Df1 %>% melt(id.vars = c("ID", "GOGID"), measure.vars =c("S1", "S2", "S3"), variable.name = "SymptomNo", value.name = "Symptom")
Df1_2<-Df1 %>% melt(id.vars = c("ID", "GOGID"), measure.vars =c( "8wkS1", "8wkS2", "8wkS3"), variable.name = "toy", value.name = "8wkContr")
Df1_3<-Df1 %>% melt(id.vars = c("ID", "GOGID"), measure.vars =c("BLS1", "BLS2", "BLS3"), variable.name = "toy", value.name = "BSContr")
colnames(Df1_1)[1] "ID" "GOGID" "SymptomNo" "Symptom"
colnames(Df1_2)[1] "ID" "GOGID" "toy" "8wkContr"
colnames(Df1_3)[1] "ID" "GOGID" "toy" "BSContr"
levels(Df1_2$toy)[1]<-"S1"
levels(Df1_2$toy)[2]<-"S2"
levels(Df1_2$toy)[3]<-"S3"
names(Df1_2)[3]<-"SymptomNo"
levels(Df1_3$toy)[1]<-"S1"
levels(Df1_3$toy)[2]<-"S2"
levels(Df1_3$toy)[3]<-"S3"
names(Df1_3)[3]<-"SymptomNo"
Df1_4<- inner_join(Df1_1, Df1_2, by= c("ID", "GOGID", "SymptomNo")) %>% inner_join(., Df1_3, by= c("ID", "GOGID", "SymptomNo")) %>% na.omit()names(OEQ)[1:2]<-c("ID", "Symptom")
Df8BL<-left_join(Df1_4,OEQ, by=c("ID", "Symptom")) %>% na.omit()
names(Df8BL)[7:9]<-c("FeelingCause", "Effect", "Strategy")write.csv(Df8BL, "~/Desktop/Df8BL for text analysis.csv")I used the software LIWC 2015 (http://liwc.wpengine.com/) to extract multiple existing and 10 self-designed word categories; LightSide (http://ankara.lti.cs.cmu.edu/side/) to tag and calculate frequency of the word “control” as verb and noun, respectively. Existing word categories in LIWC include 4 summary language variables (analytical thinking, clout [confidence], authenticity, and emotional tone), 3 general descriptor categories (words per sentence, percent of target words captured by the dictionary, and percent of words in the text that are longer than six letters), 21 standard linguistic dimensions (e.g., percentage of words in the text that are pronouns, articles, auxiliary verbs, etc.), 41 word categories tapping psychological constructs (e.g., affect, cognition, biological processes, drives), 6 personal concern categories (e.g., work, home, leisure activities), 5 informal language markers (assents, fillers, swear words, netspeak), and 12 punctuation categories (periods, commas, etc). I saved all the results in the file “Result8BLSDWRITE.xlsx”
Df<-read.csv("~/Desktop/DS4Ling-2021/old repository/private/DS4Lingdataset/Results8BLSDWRITE.csv")
Df$controlNN[is.na(Df$controlNN)]=0
Df$controlVB[is.na(Df$controlVB)]=0
Df[is.na(Df)]character(0)
Df$controlNN<-Df$controlNN/Df$WC*100
Df$controlVB<-Df$controlVB/Df$WC*100
Df<- Df %>% dplyr::select(2, 4:7,11,106:112, 12:105)
names(Df) [1] "ID" "SymptomNo" "Symptom" "X8wkContr" "BSContr"
[6] "WC" "symptom" "effort" "impact" "positive.adj"
[11] "negative.adj" "controlled" "uncontrolled" "controlNN" "controlVB"
[16] "Analytic" "Clout" "Authentic" "Tone" "WPS"
[21] "Sixltr" "Dic" "function." "pronoun" "ppron"
[26] "i" "we" "you" "shehe" "they"
[31] "ipron" "article" "prep" "auxverb" "adverb"
[36] "conj" "negate" "verb" "adj" "compare"
[41] "interrog" "number" "quant" "affect" "posemo"
[46] "negemo" "anx" "anger" "sad" "social"
[51] "family" "friend" "female" "male" "cogproc"
[56] "insight" "cause" "discrep" "tentat" "certain"
[61] "differ" "percept" "see" "hear" "feel"
[66] "bio" "body" "health" "sexual" "ingest"
[71] "drives" "affiliation" "achieve" "power" "reward"
[76] "risk" "focuspast" "focuspresent" "focusfuture" "relativ"
[81] "motion" "space" "time" "work" "leisure"
[86] "home" "money" "relig" "death" "informal"
[91] "swear" "netspeak" "assent" "nonflu" "filler"
[96] "AllPunc" "Period" "Comma" "Colon" "SemiC"
[101] "QMark" "Exclam" "Dash" "Quote" "Apostro"
[106] "Parenth" "OtherP"
Df description: “ID”- participant ID “Employment” - Employed vs
unemployed “Marriage” - Currently married, divorced, Living with
partner/significant other, never married, separated, widowed “race” -
American Indian, bi/Multi-racial Black or African American, White,
other, unknown “ethinicity” - latino, not latino, don not know “Age”
“Formaleducationyears” - years of formal education “SymptomNo”- Symptom
number participant worked on (i.e., S1, S2, S3) “Symptom” - Symptom
participant worked on (e.g., pain, nausea)
“X8wkContro” - symptom controllability score changes at 8 week post
intervention “BSContr” - baseline controllability score “WC” - the total
number of words in participant posts “WPS” - the number of words per
post
The rest of the variable are the percentage of that specific word category or punctuation category of the total number words in the posts. For example, “symptom” - the percentage of symptom word category (e.g., drowsy, lose hair) of the total number of words in the post (range:0-100) “positive.adj” - the percentage of positive.adj symptom word category (e.g., steady, mild, good) of the total number of words in the post (0-100)
names(Metadata) [1] "ID"
[2] "USERID"
[3] "Employment"
[4] "CASENUM"
[5] "#modules"
[6] "WC"
[7] "MWC"
[8] "Marriage"
[9] "race"
[10] "ethinicity (latio)"
[11] "Age"
[12] "Formaleducationyears"
[13] "AnxietySTA"
[14] "socialsupport"
[15] "Optimismscalescore"
[16] "Comorbidityindex"
[17] "baselineweight"
[18] "total_chemo_coursecancerstage"
[19] "CESDdepression"
[20] "QOL_overall"
[21] "treatmentchoiceon_chemotherapy1yes2No3Uknown"
[22] "HCPcomm"
[23] "Selfmagementbarriers"
Metadata<- Metadata[, c(1, 3, 8:12)]
Dftoy<-left_join(Df, Metadata, by="ID") %>% select(1 , 108:113, everything())Df<- Dftoy %>% na.omit()
ggplot(Df)+
geom_histogram(binwidth=0.05, color="blue",aes(x= BSContr, y=..density.., fill=..count..))+
stat_function(fun=dnorm,color="blue",
args=list(mean=mean(Df$BSContr),sd=sd(Df$BSContr)))+xlab("Baseline controllability score ")
#QQplots
qq<-data.frame(c(Df,qqnorm(Df$BSContr)))
ggplot(qq,aes(x=x,y=y,legend.position="none"))+
geom_point()+
geom_smooth(method="lm")+
labs(title="Q-Q Normal Plot",x="Theoretic",y="Observed")+
theme_bw()`geom_smooth()` using formula 'y ~ x'
##boxplots
ggplot(Df)+
geom_boxplot(aes(Df$BSContr))+
theme_bw()+
theme(legend.position="none")+xlab("Baseline controllability score ")
shapiro.test(Df$BSContr) Shapiro-Wilk normality test
data: Df$BSContr
W = 0.98677, p-value = 0.007485
#install.packages("pastecs")
library(pastecs)Attaching package: 'pastecs'
The following objects are masked from 'package:dplyr':
first, last
stat.desc(Df$BSContr) nbr.val nbr.null nbr.na min max range
300.00000000 1.00000000 0.00000000 0.00000000 4.00000000 4.00000000
sum median mean SE.mean CI.mean.0.95 var
674.20000000 2.20000000 2.24733333 0.04201023 0.08267317 0.52945775
std.dev coef.var
0.72763847 0.32377861
quantile(Df$BSContr, 0.75)-quantile(Df$BSContr, 0.25)75%
1
table(Df$BSContr) 0 0.2 0.4 0.6 0.8 1 1.2 1.4 1.6 1.8 2 2.2 2.4 2.5 2.6 2.8 3 3.2 3.4 3.6
1 1 4 4 3 5 8 14 25 22 40 28 34 2 26 27 24 12 8 6
3.8 4
4 2
Df<- Df %>% na.omit()
ggplot(Df)+
geom_histogram(binwidth=0.05, color="blue",aes(x= X8wkContr, y=..density.., fill=..count..))+
stat_function(fun=dnorm,color="blue",
args=list(mean=mean(Df$BSContr),sd=sd(Df$X8wkContr)))+xlab("Controllability score at 8 weeks")
#QQplots
qq<-data.frame(c(Df,qqnorm(Df$X8wkContr)))
ggplot(qq,aes(x=x,y=y,legend.position="none"))+
geom_point()+
geom_smooth(method="lm")+
labs(title="Q-Q Normal Plot",x="Theoretic",y="Observed")+
theme_bw()`geom_smooth()` using formula 'y ~ x'
##boxplots
ggplot(Df)+
geom_boxplot(aes(Df$X8wkContr))+
theme_bw()+
theme(legend.position="none")+xlab("Controllability score at 8 weeks")
shapiro.test(Df$X8wkContr) Shapiro-Wilk normality test
data: Df$X8wkContr
W = 0.97076, p-value = 8.779e-06
stat.desc(Df$X8wkContr) nbr.val nbr.null nbr.na min max range
300.00000000 2.00000000 0.00000000 0.00000000 4.00000000 4.00000000
sum median mean SE.mean CI.mean.0.95 var
741.40000000 2.60000000 2.47133333 0.04095208 0.08059082 0.50312196
std.dev coef.var
0.70931091 0.28701547
quantile(Df$X8wkContr, 0.75)-quantile(Df$X8wkContr, 0.25)75%
1
table(Df$X8wkContr) 0 0.4 0.6 0.8 1 1.2 1.4 1.6 1.8 2 2.2 2.4 2.6 2.8 3 3.2 3.4 3.6 3.8 4
2 2 2 1 3 2 9 19 16 40 26 26 28 22 51 28 6 8 2 7
# Sample descriptive stats - sample size (157 participants) is bigger
than the sample size (112 participants and 314 posts) I use in mixed
effect model. Participants are predominantly married or Living with
partner/significant other(75.16%), white (93%), non-hispanic (96.18%),
unemployed (59.24%). The mean of age is 58.18 (SD = 9.72). The average
of formal years of education is 14.4 (SD=2.72)
names(Metadata)[1] "ID" "Employment" "Marriage"
[4] "race" "ethinicity (latio)" "Age"
[7] "Formaleducationyears"
Metadata<-na.omit(Metadata)
table(Metadata$Marriage) %>% addmargins() Currently married Divorced
106 13
Living with partner/significant other Never married
12 7
Separated Widowed
6 13
Sum
157
table(Metadata$race)%>% addmargins() American Indian Bi/Multi-racial Black or African American
1 2 4
Other unknown White
1 3 146
Sum
157
table(Metadata$`ethinicity (latio)`)%>% addmargins()Do not know MV No Yes Sum
2 1 151 3 157
table(Metadata$Employment)%>% addmargins() Never employed No Yes Sum
1 92 64 157
#since there is only one individual chose never employed, I will code it as No
Metadata$Employment[Metadata$Employment=="Never employed"]<-"No"
#Age----
library(pastecs)
stat.desc(Metadata$Age) nbr.val nbr.null nbr.na min max range
157.0000000 0.0000000 0.0000000 25.0000000 81.0000000 56.0000000
sum median mean SE.mean CI.mean.0.95 var
9135.0000000 58.0000000 58.1847134 0.7757145 1.5322591 94.4720725
std.dev coef.var
9.7196745 0.1670486
quantile(Metadata$Age, 0.75,na.rm = TRUE)-quantile(Metadata$Age, 0.25, na.rm = TRUE)75%
12
ggplot(Metadata)+
geom_histogram(binwidth=0.1, color="blue",aes(x=Age, y=..density.., fill=..count..))+
stat_function(fun=dnorm,color="blue",
args=list(mean=mean(Metadata$Age),sd=sd(Metadata$Age)))
#QQplots
qq<-data.frame(c(Metadata,qqnorm(Metadata$Age)))
ggplot(qq,aes(x=x,y=y,legend.position="none"))+
geom_point()+
geom_smooth(method="lm")+
labs(title="Q-Q Normal Plot",x="Theoretic",y="Observed")+
theme_bw()`geom_smooth()` using formula 'y ~ x'
##boxplots
ggplot(Metadata)+
geom_boxplot(aes(Metadata$Age))+
theme_bw()+
theme(legend.position="none")Warning: Use of `Metadata$Age` is discouraged. Use `Age` instead.
shapiro.test(Metadata$Age) Shapiro-Wilk normality test
data: Metadata$Age
W = 0.98115, p-value = 0.0305
#Formal years of education normality is violated-----
stat.desc(Metadata$Formaleducationyears) nbr.val nbr.null nbr.na min max range
157.0000000 0.0000000 0.0000000 10.0000000 22.0000000 12.0000000
sum median mean SE.mean CI.mean.0.95 var
2261.0000000 14.0000000 14.4012739 0.2172272 0.4290861 7.4084599
std.dev coef.var
2.7218486 0.1890005
quantile(Metadata$Formaleducationyears, 0.75,na.rm = TRUE)-quantile(Metadata$Formaleducationyears, 0.25, na.rm = TRUE)75%
4
ggplot(Metadata)+
geom_histogram(binwidth=0.5, color="blue",aes(x=Formaleducationyears, y=..density.., fill=..count..))+
stat_function(fun=dnorm,color="blue",
args=list(mean=mean(Metadata$Formaleducationyears),sd=sd(Metadata$Formaleducationyears)))
#QQplots
qq<-data.frame(c(Metadata,qqnorm(Metadata$Formaleducationyears)))
ggplot(qq,aes(x=x,y=y,legend.position="none"))+
geom_point()+
geom_smooth(method="lm")+
labs(title="Q-Q Normal Plot",x="Theoretic",y="Observed")+
theme_bw()`geom_smooth()` using formula 'y ~ x'
##boxplots
ggplot(Metadata)+
geom_boxplot(aes(Metadata$Formaleducationyears))+
theme_bw()+
theme(legend.position="none")Warning: Use of `Metadata$Formaleducationyears` is discouraged. Use
`Formaleducationyears` instead.
shapiro.test(Metadata$Formaleducationyears) Shapiro-Wilk normality test
data: Metadata$Formaleducationyears
W = 0.87253, p-value = 2.529e-10
7 Preliminary bivariate correlation between the dependent variable and potential predictors - pearson correlation
names(Dftoy) [1] "ID" "Employment" "Marriage"
[4] "race" "ethinicity (latio)" "Age"
[7] "Formaleducationyears" "SymptomNo" "Symptom"
[10] "X8wkContr" "BSContr" "WC"
[13] "symptom" "effort" "impact"
[16] "positive.adj" "negative.adj" "controlled"
[19] "uncontrolled" "controlNN" "controlVB"
[22] "Analytic" "Clout" "Authentic"
[25] "Tone" "WPS" "Sixltr"
[28] "Dic" "function." "pronoun"
[31] "ppron" "i" "we"
[34] "you" "shehe" "they"
[37] "ipron" "article" "prep"
[40] "auxverb" "adverb" "conj"
[43] "negate" "verb" "adj"
[46] "compare" "interrog" "number"
[49] "quant" "affect" "posemo"
[52] "negemo" "anx" "anger"
[55] "sad" "social" "family"
[58] "friend" "female" "male"
[61] "cogproc" "insight" "cause"
[64] "discrep" "tentat" "certain"
[67] "differ" "percept" "see"
[70] "hear" "feel" "bio"
[73] "body" "health" "sexual"
[76] "ingest" "drives" "affiliation"
[79] "achieve" "power" "reward"
[82] "risk" "focuspast" "focuspresent"
[85] "focusfuture" "relativ" "motion"
[88] "space" "time" "work"
[91] "leisure" "home" "money"
[94] "relig" "death" "informal"
[97] "swear" "netspeak" "assent"
[100] "nonflu" "filler" "AllPunc"
[103] "Period" "Comma" "Colon"
[106] "SemiC" "QMark" "Exclam"
[109] "Dash" "Quote" "Apostro"
[112] "Parenth" "OtherP"
Df<-Dftoy[, -110]
library(Hmisc)Loading required package: lattice
Loading required package: survival
Loading required package: Formula
Attaching package: 'Hmisc'
The following objects are masked from 'package:dplyr':
src, summarize
The following objects are masked from 'package:base':
format.pval, units
library(corrgram)Attaching package: 'corrgram'
The following object is masked from 'package:lattice':
panel.fill
# there is no variation in "Quote"
res <- cor(Df[10:112])
round(res, 2)[,1] X8wkContr BSContr WC symptom effort impact
1.00 0.46 0.16 0.18 0.00 0.04
positive.adj negative.adj controlled uncontrolled controlNN controlVB
-0.04 -0.03 0.16 -0.07 0.16 0.02
Analytic Clout Authentic Tone WPS Sixltr
0.11 0.06 -0.02 -0.05 0.03 0.09
Dic function. pronoun ppron i we
-0.02 -0.05 -0.08 -0.02 -0.02 -0.06
you shehe they ipron article prep
0.02 0.00 0.02 -0.10 0.06 0.10
auxverb adverb conj negate verb adj
-0.03 -0.08 0.04 -0.10 -0.01 0.09
compare interrog number quant affect posemo
0.10 0.00 0.04 -0.02 0.05 0.02
negemo anx anger sad social family
0.04 0.20 -0.08 -0.02 0.03 0.00
friend female male cogproc insight cause
-0.01 -0.04 0.01 -0.06 0.03 -0.04
discrep tentat certain differ percept see
0.08 -0.02 -0.10 -0.06 -0.05 -0.03
hear feel bio body health sexual
0.07 -0.10 -0.01 -0.19 0.03 0.01
ingest drives affiliation achieve power reward
0.13 0.09 0.00 -0.03 0.07 0.09
risk focuspast focuspresent focusfuture relativ motion
0.05 0.02 -0.13 0.06 0.03 -0.03
space time work leisure home money
0.04 0.02 -0.02 -0.01 -0.06 -0.13
relig death informal swear netspeak assent
0.08 0.01 0.12 0.02 0.00 0.02
nonflu filler AllPunc Period Comma Colon
0.16 0.03 0.02 -0.01 -0.02 -0.03
SemiC QMark Exclam Dash Apostro Parenth
0.04 0.08 -0.01 0.06 0.02 -0.01
OtherP
-0.07
#significant and marginal significant
cor.test(Df$X8wkContr, Df$BSContr) Pearson's product-moment correlation
data: Df$X8wkContr and Df$BSContr
t = 9.1835, df = 312, p-value < 2.2e-16
alternative hypothesis: true correlation is not equal to 0
95 percent confidence interval:
0.3694745 0.5441926
sample estimates:
cor
0.4612939
cor.test(Df$X8wkContr, Df$WC) Pearson's product-moment correlation
data: Df$X8wkContr and Df$WC
t = 2.9049, df = 312, p-value = 0.003936
alternative hypothesis: true correlation is not equal to 0
95 percent confidence interval:
0.05253597 0.26814414
sample estimates:
cor
0.1622764
cor.test(Df$X8wkContr, Df$symptom) Pearson's product-moment correlation
data: Df$X8wkContr and Df$symptom
t = 3.2346, df = 312, p-value = 0.001349
alternative hypothesis: true correlation is not equal to 0
95 percent confidence interval:
0.07085505 0.28512584
sample estimates:
cor
0.1801264
cor.test(Df$X8wkContr, Df$controlled) Pearson's product-moment correlation
data: Df$X8wkContr and Df$controlled
t = 2.9362, df = 312, p-value = 0.003569
alternative hypothesis: true correlation is not equal to 0
95 percent confidence interval:
0.05428192 0.26976842
sample estimates:
cor
0.1639807
cor.test(Df$X8wkContr, Df$controlNN) Pearson's product-moment correlation
data: Df$X8wkContr and Df$controlNN
t = 2.9087, df = 312, p-value = 0.00389
alternative hypothesis: true correlation is not equal to 0
95 percent confidence interval:
0.0527485 0.2683419
sample estimates:
cor
0.1624839
cor.test(Df$X8wkContr, Df$anx) Pearson's product-moment correlation
data: Df$X8wkContr and Df$anx
t = 3.5492, df = 312, p-value = 0.0004459
alternative hypothesis: true correlation is not equal to 0
95 percent confidence interval:
0.08823656 0.30111502
sample estimates:
cor
0.1969966
cor.test(Df$X8wkContr, Df$feel) Pearson's product-moment correlation
data: Df$X8wkContr and Df$feel
t = -1.7491, df = 312, p-value = 0.08125
alternative hypothesis: true correlation is not equal to 0
95 percent confidence interval:
-0.20697057 0.01227386
sample estimates:
cor
-0.09854403
cor.test(Df$X8wkContr, Df$body) Pearson's product-moment correlation
data: Df$X8wkContr and Df$body
t = -3.4943, df = 312, p-value = 0.000544
alternative hypothesis: true correlation is not equal to 0
95 percent confidence interval:
-0.29834226 -0.08521303
sample estimates:
cor
-0.1940667
cor.test(Df$X8wkContr, Df$ingest) Pearson's product-moment correlation
data: Df$X8wkContr and Df$ingest
t = 2.3347, df = 312, p-value = 0.02019
alternative hypothesis: true correlation is not equal to 0
95 percent confidence interval:
0.02065436 0.23826702
sample estimates:
cor
0.1310388
cor.test(Df$X8wkContr, Df$focuspresent) Pearson's product-moment correlation
data: Df$X8wkContr and Df$focuspresent
t = -2.3641, df = 312, p-value = 0.01869
alternative hypothesis: true correlation is not equal to 0
95 percent confidence interval:
-0.23982058 -0.02230132
sample estimates:
cor
-0.1326579
cor.test(Df$X8wkContr, Df$money) Pearson's product-moment correlation
data: Df$X8wkContr and Df$money
t = -2.3909, df = 312, p-value = 0.0174
alternative hypothesis: true correlation is not equal to 0
95 percent confidence interval:
-0.24123888 -0.02380591
sample estimates:
cor
-0.1341365
cor.test(Df$X8wkContr, Df$informal) Pearson's product-moment correlation
data: Df$X8wkContr and Df$informal
t = 2.1241, df = 312, p-value = 0.03445
alternative hypothesis: true correlation is not equal to 0
95 percent confidence interval:
0.008825569 0.227076182
sample estimates:
cor
0.119393
cor.test(Df$X8wkContr, Df$nonflu) Pearson's product-moment correlation
data: Df$X8wkContr and Df$nonflu
t = 2.9473, df = 312, p-value = 0.003447
alternative hypothesis: true correlation is not equal to 0
95 percent confidence interval:
0.05490122 0.27034427
sample estimates:
cor
0.1645851
cor.test(Df$X8wkContr, Df$negate) Pearson's product-moment correlation
data: Df$X8wkContr and Df$negate
t = -1.8085, df = 312, p-value = 0.0715
alternative hypothesis: true correlation is not equal to 0
95 percent confidence interval:
-0.21016689 0.00893254
sample estimates:
cor
-0.1018522
cor.test(Df$X8wkContr, Df$discrep) Pearson's product-moment correlation
data: Df$X8wkContr and Df$discrep
t = 1.4805, df = 312, p-value = 0.1398
alternative hypothesis: true correlation is not equal to 0
95 percent confidence interval:
-0.02741416 0.19242837
sample estimates:
cor
0.08352329
cor.test(Df$X8wkContr, Df$certain) Pearson's product-moment correlation
data: Df$X8wkContr and Df$certain
t = -1.823, df = 312, p-value = 0.06925
alternative hypothesis: true correlation is not equal to 0
95 percent confidence interval:
-0.210950612 0.008112522
sample estimates:
cor
-0.1026637
cor.test(Df$X8wkContr, Df$Analytic) Pearson's product-moment correlation
data: Df$X8wkContr and Df$Analytic
t = 1.8692, df = 312, p-value = 0.06253
alternative hypothesis: true correlation is not equal to 0
95 percent confidence interval:
-0.005512423 0.213433755
sample estimates:
cor
0.1052358
cor.test(Df$X8wkContr, Df$Sixltr) Pearson's product-moment correlation
data: Df$X8wkContr and Df$Sixltr
t = 1.5205, df = 312, p-value = 0.1294
alternative hypothesis: true correlation is not equal to 0
95 percent confidence interval:
-0.02515633 0.19460314
sample estimates:
cor
0.08576649
cor.test(Df$X8wkContr, Df$prep) Pearson's product-moment correlation
data: Df$X8wkContr and Df$prep
t = 1.6987, df = 312, p-value = 0.09038
alternative hypothesis: true correlation is not equal to 0
95 percent confidence interval:
-0.01511768 0.20424646
sample estimates:
cor
0.09572651
corrgram(Df %>% select (X8wkContr,BSContr, WC, symptom, controlled, controlNN,anx,feel, body, ingest, focuspresent, money, informal, nonflu, negate, discrep, certain, Analytic, Sixltr, prep), order=FALSE, lower.panel=panel.shade,
upper.panel=panel.pie, text.panel=panel.txt,
main="IVs and DV correlation")
## As we can see posemo and tone are highly correlated, symptom and health category are highly correlated as well. 
Based on theories and bivariate correlation test results , I have some ideas of candidate predictors. Since this analysis is exploratory in nature and we have a lot of potential predictors. I will explore a few approaches of model building
Random effect: use participant ID Covariate:social demographic factors and symptom selected, e.g., marriage status, age, education, employment, race, ethnicity because patients were predominantly non-Hispanic white) Full purpose-oriented predictor selection
##Get convergence code for a single model
##verbose: TRUE to return full convergence code, FALSE to return logical
##checkSingular: TRUE to count singular fit as nonconvergence, FALSE to ignore singular fit
getConvCode <- function (x, verbose=FALSE, checkSingular=TRUE) {
library(lme4)
library(purrr)
##Get convergence messages
convMsg <-
x %>%
attr("optinfo") %>%
pluck("conv", "lme4")
##Get singular-fit status
if (checkSingular) sgFit <- x %>% isSingular()
##If not verbose, get convergence code as logical
if (!verbose) {
convCode <- length(convMsg)==0
if (checkSingular) convCode <- convCode & !sgFit
##If verbose, get convergence code as character
} else {
convCode <- character(0L)
if (length(convMsg) > 0) {
convCode <- convMsg %>%
pluck("messages") %>%
paste(collapse="\n")
}
if (checkSingular) {
if (sgFit) {
convCode <- paste(c(convCode, "Singular fit"), collapse="\n")
}
}
##If nothing has been added to convCode, return the good news.
if (length(convCode)==0) {
convCode <- "Converged"
if (checkSingular) convCode <- paste0(convCode, ", no singular fit")
}
}
convCode
}
##Convenience function for checking if something is an error
is.error <- function(x) "error" %in% class(x)
##Get Fox & Monette's (1992) GVIF, which is the square of the "GVIF" reported
## by car::vif() and thus is comparable to the 'VIF < 10' criterion.
##Whereas car::vif() returns either a vector or a matrix, this function always
## returns a dataframe
vif <- function(mod, decreasing=TRUE) {
library(car)
library(dplyr)
vifReturn <- tryCatch(car::vif(mod),
##Catch and return "fewer than 2 terms" error"
error = function(e) e)
if (is.error(vifReturn)) {
return(NA)
}
##Turn vector VIF into dataframe
if (is.numeric(vifReturn) & !is.matrix(vifReturn)) {
ret <- data.frame(Term = names(vifReturn),
GVIF = vifReturn,
Df = rep(1, length(vifReturn))) %>%
mutate(`GVIF^(1/(2*Df))` = sqrt(vifReturn),
`GVIF^(1/Df)` = vifReturn)
}
##Turn matrix VIF into dataframe
if (is.numeric(vifReturn) & is.matrix(vifReturn)) {
ret <- as.data.frame(vifReturn) %>%
rownames_to_column("Term") %>%
select(Term, everything()) %>%
mutate(`GVIF^(1/Df)` = GVIF ^ (1/Df))
}
ret
}
##Get maximum VIF from model
getMaxVIF <- function(mod, decreasing=TRUE) {
library(dplyr)
library(purrr)
##If just one term, return NA
if (length(labels(terms(mod))) < 2)
return(NA)
vif(mod) %>%
##Get unique max GVIF (in case there are ties)
arrange(desc(`GVIF^(1/Df)`)) %>%
slice(1) %>%
pull(`GVIF^(1/Df)`, name=Term)
}#install.packages("lmerTest")
library(lmerTest)Attaching package: 'lmerTest'
The following object is masked from 'package:lme4':
lmer
The following object is masked from 'package:stats':
step
# since there is very little variations in race and enthnicity, so I wont include them.
toy1<-lmer(X8wkContr ~ BSContr+Marriage+Age+
Formaleducationyears+Employment+(1|ID), Df)
summary(toy1)Linear mixed model fit by REML. t-tests use Satterthwaite's method [
lmerModLmerTest]
Formula: X8wkContr ~ BSContr + Marriage + Age + Formaleducationyears +
Employment + (1 | ID)
Data: Df
REML criterion at convergence: 523.3
Scaled residuals:
Min 1Q Median 3Q Max
-2.8850 -0.4099 0.0026 0.5385 3.8917
Random effects:
Groups Name Variance Std.Dev.
ID (Intercept) 0.2004 0.4476
Residual 0.1904 0.4364
Number of obs: 300, groups: ID, 114
Fixed effects:
Estimate Std. Error df
(Intercept) 1.073978 0.640891 100.611746
BSContr 0.393284 0.046577 277.078093
MarriageDivorced 0.211995 0.188425 100.800822
MarriageLiving with partner/significant other 0.305969 0.198777 95.949027
MarriageNever married -0.037369 0.225237 94.318976
MarriageSeparated 0.020527 0.239254 93.686455
MarriageWidowed 0.051874 0.211430 120.917110
Age -0.004874 0.005453 100.897446
Formaleducationyears 0.041446 0.018180 101.149037
EmploymentNo 0.252269 0.523890 93.747169
EmploymentYes 0.002143 0.524525 93.602658
t value Pr(>|t|)
(Intercept) 1.676 0.0969 .
BSContr 8.444 1.75e-15 ***
MarriageDivorced 1.125 0.2632
MarriageLiving with partner/significant other 1.539 0.1270
MarriageNever married -0.166 0.8686
MarriageSeparated 0.086 0.9318
MarriageWidowed 0.245 0.8066
Age -0.894 0.3736
Formaleducationyears 2.280 0.0247 *
EmploymentNo 0.482 0.6313
EmploymentYes 0.004 0.9967
---
Signif. codes: 0 '***' 0.001 '**' 0.01 '*' 0.05 '.' 0.1 ' ' 1
Correlation of Fixed Effects:
(Intr) BSCntr MrrgDv MLwp/o MrrgNm MrrgSp MrrgWd Age Frmldc
BSContr -0.188
MarrigDvrcd 0.097 -0.106
MrrgLwprt/o 0.052 0.003 0.109
MrrgNvrmrrd -0.114 0.050 0.057 0.030
MarrigSprtd -0.039 0.044 0.066 0.086 0.065
MarriagWdwd 0.009 0.000 0.096 0.125 0.056 0.073
Age -0.453 -0.011 -0.112 -0.203 0.214 0.023 -0.127
Frmldctnyrs -0.333 -0.059 -0.070 0.114 0.018 0.057 0.141 0.011
EmploymentN -0.736 0.072 -0.027 -0.039 -0.034 -0.041 -0.034 -0.069 -0.099
EmploymntYs -0.755 0.063 -0.037 -0.034 -0.046 -0.022 -0.050 -0.016 -0.107
EmplyN
BSContr
MarrigDvrcd
MrrgLwprt/o
MrrgNvrmrrd
MarrigSprtd
MarriagWdwd
Age
Frmldctnyrs
EmploymentN
EmploymntYs 0.979
anova(toy1)Type III Analysis of Variance Table with Satterthwaite's method
Sum Sq Mean Sq NumDF DenDF F value Pr(>F)
BSContr 13.5777 13.5777 1 277.078 71.2960 1.746e-15 ***
Marriage 0.6429 0.1286 5 100.425 0.6752 0.64321
Age 0.1521 0.1521 1 100.897 0.7987 0.37361
Formaleducationyears 0.9898 0.9898 1 101.149 5.1974 0.02472 *
Employment 1.0662 0.5331 2 97.492 2.7992 0.06576 .
---
Signif. codes: 0 '***' 0.001 '**' 0.01 '*' 0.05 '.' 0.1 ' ' 1
toy2<-lmer(X8wkContr ~ BSContr+Marriage+ Age+ Formaleducationyears+Employment+ symptom +(1|ID), Df)
summary(toy2)Linear mixed model fit by REML. t-tests use Satterthwaite's method [
lmerModLmerTest]
Formula: X8wkContr ~ BSContr + Marriage + Age + Formaleducationyears +
Employment + symptom + (1 | ID)
Data: Df
REML criterion at convergence: 523.3
Scaled residuals:
Min 1Q Median 3Q Max
-2.8457 -0.4345 0.0071 0.5143 3.4963
Random effects:
Groups Name Variance Std.Dev.
ID (Intercept) 0.1930 0.4393
Residual 0.1884 0.4340
Number of obs: 300, groups: ID, 114
Fixed effects:
Estimate Std. Error df
(Intercept) 1.076153 0.631216 99.773958
BSContr 0.392466 0.046210 276.888271
MarriageDivorced 0.187484 0.185836 100.663207
MarriageLiving with partner/significant other 0.279872 0.196011 95.606949
MarriageNever married -0.053926 0.221867 93.575523
MarriageSeparated 0.035430 0.235634 92.929336
MarriageWidowed 0.052814 0.208419 120.210697
Age -0.004633 0.005372 100.111659
Formaleducationyears 0.037972 0.017959 101.491969
EmploymentNo 0.204271 0.516167 93.064289
EmploymentYes -0.030878 0.516599 92.806812
symptom 0.046279 0.018551 240.538933
t value Pr(>|t|)
(Intercept) 1.705 0.0913 .
BSContr 8.493 1.25e-15 ***
MarriageDivorced 1.009 0.3155
MarriageLiving with partner/significant other 1.428 0.1566
MarriageNever married -0.243 0.8085
MarriageSeparated 0.150 0.8808
MarriageWidowed 0.253 0.8004
Age -0.862 0.3905
Formaleducationyears 2.114 0.0369 *
EmploymentNo 0.396 0.6932
EmploymentYes -0.060 0.9525
symptom 2.495 0.0133 *
---
Signif. codes: 0 '***' 0.001 '**' 0.01 '*' 0.05 '.' 0.1 ' ' 1
Correlation of Fixed Effects:
(Intr) BSCntr MrrgDv MLwp/o MrrgNm MrrgSp MrrgWd Age Frmldc
BSContr -0.190
MarrigDvrcd 0.097 -0.106
MrrgLwprt/o 0.052 0.004 0.111
MrrgNvrmrrd -0.114 0.051 0.059 0.032
MarrigSprtd -0.039 0.044 0.065 0.085 0.064
MarriagWdwd 0.010 0.000 0.096 0.125 0.056 0.073
Age -0.452 -0.011 -0.113 -0.204 0.214 0.024 -0.127
Frmldctnyrs -0.332 -0.058 -0.065 0.117 0.021 0.055 0.140 0.010
EmploymentN -0.736 0.073 -0.025 -0.037 -0.033 -0.042 -0.034 -0.069 -0.096
EmploymntYs -0.754 0.064 -0.036 -0.032 -0.045 -0.023 -0.050 -0.016 -0.105
symptom 0.002 -0.012 -0.052 -0.053 -0.030 0.025 0.002 0.018 -0.077
EmplyN EmplyY
BSContr
MarrigDvrcd
MrrgLwprt/o
MrrgNvrmrrd
MarrigSprtd
MarriagWdwd
Age
Frmldctnyrs
EmploymentN
EmploymntYs 0.979
symptom -0.038 -0.026
# we will keep age, formal years of education as covariates. Age is never a significant predictor, but it is an important patient characteristic.
toy3<-lmer(X8wkContr ~ BSContr+ Age+ Formaleducationyears+ WC+ symptom +(1|ID), Df)
toy3.1<-lmer(X8wkContr ~ BSContr+ Age+ Formaleducationyears+Employment+ WC+ symptom +(1|ID), Df)
summary(toy3)Linear mixed model fit by REML. t-tests use Satterthwaite's method [
lmerModLmerTest]
Formula: X8wkContr ~ BSContr + Age + Formaleducationyears + WC + symptom +
(1 | ID)
Data: Df
REML criterion at convergence: 532.4
Scaled residuals:
Min 1Q Median 3Q Max
-2.8652 -0.4340 -0.0488 0.5633 3.4644
Random effects:
Groups Name Variance Std.Dev.
ID (Intercept) 0.1857 0.4309
Residual 0.1878 0.4334
Number of obs: 300, groups: ID, 114
Fixed effects:
Estimate Std. Error df t value Pr(>|t|)
(Intercept) 8.744e-01 3.935e-01 1.132e+02 2.222 0.02828 *
BSContr 3.839e-01 4.554e-02 2.835e+02 8.430 1.77e-15 ***
Age 8.524e-04 4.777e-03 1.043e+02 0.178 0.85873
Formaleducationyears 3.165e-02 1.730e-02 1.090e+02 1.829 0.07011 .
WC 4.160e-04 1.733e-04 2.935e+02 2.400 0.01701 *
symptom 5.803e-02 1.862e-02 2.474e+02 3.116 0.00205 **
---
Signif. codes: 0 '***' 0.001 '**' 0.01 '*' 0.05 '.' 0.1 ' ' 1
Correlation of Fixed Effects:
(Intr) BSCntr Age Frmldc WC
BSContr -0.196
Age -0.726 -0.017
Frmldctnyrs -0.644 -0.064 0.049
WC -0.057 -0.074 -0.019 -0.074
symptom -0.031 -0.023 -0.011 -0.092 0.159
anova(toy3, toy3.1)refitting model(s) with ML (instead of REML)
Data: Df
Models:
toy3: X8wkContr ~ BSContr + Age + Formaleducationyears + WC + symptom + (1 | ID)
toy3.1: X8wkContr ~ BSContr + Age + Formaleducationyears + Employment + WC + symptom + (1 | ID)
npar AIC BIC logLik deviance Chisq Df Pr(>Chisq)
toy3 8 502.99 532.62 -243.5 486.99
toy3.1 10 502.01 539.04 -241.0 482.01 4.9893 2 0.08253 .
---
Signif. codes: 0 '***' 0.001 '**' 0.01 '*' 0.05 '.' 0.1 ' ' 1
summary(toy3)Linear mixed model fit by REML. t-tests use Satterthwaite's method [
lmerModLmerTest]
Formula: X8wkContr ~ BSContr + Age + Formaleducationyears + WC + symptom +
(1 | ID)
Data: Df
REML criterion at convergence: 532.4
Scaled residuals:
Min 1Q Median 3Q Max
-2.8652 -0.4340 -0.0488 0.5633 3.4644
Random effects:
Groups Name Variance Std.Dev.
ID (Intercept) 0.1857 0.4309
Residual 0.1878 0.4334
Number of obs: 300, groups: ID, 114
Fixed effects:
Estimate Std. Error df t value Pr(>|t|)
(Intercept) 8.744e-01 3.935e-01 1.132e+02 2.222 0.02828 *
BSContr 3.839e-01 4.554e-02 2.835e+02 8.430 1.77e-15 ***
Age 8.524e-04 4.777e-03 1.043e+02 0.178 0.85873
Formaleducationyears 3.165e-02 1.730e-02 1.090e+02 1.829 0.07011 .
WC 4.160e-04 1.733e-04 2.935e+02 2.400 0.01701 *
symptom 5.803e-02 1.862e-02 2.474e+02 3.116 0.00205 **
---
Signif. codes: 0 '***' 0.001 '**' 0.01 '*' 0.05 '.' 0.1 ' ' 1
Correlation of Fixed Effects:
(Intr) BSCntr Age Frmldc WC
BSContr -0.196
Age -0.726 -0.017
Frmldctnyrs -0.644 -0.064 0.049
WC -0.057 -0.074 -0.019 -0.074
symptom -0.031 -0.023 -0.011 -0.092 0.159
anova(toy2, toy3)refitting model(s) with ML (instead of REML)
Data: Df
Models:
toy3: X8wkContr ~ BSContr + Age + Formaleducationyears + WC + symptom + (1 | ID)
toy2: X8wkContr ~ BSContr + Marriage + Age + Formaleducationyears + Employment + symptom + (1 | ID)
npar AIC BIC logLik deviance Chisq Df Pr(>Chisq)
toy3 8 502.99 532.62 -243.50 486.99
toy2 14 512.27 564.12 -242.13 484.27 2.7296 6 0.8419
#toy 4 is significant better than toy3.1 no collinearity issue
toy4<-lmer(X8wkContr ~ BSContr+ Age+ Formaleducationyears+ WC+ symptom+anx +(1|ID), Df)
summary(toy4)Linear mixed model fit by REML. t-tests use Satterthwaite's method [
lmerModLmerTest]
Formula: X8wkContr ~ BSContr + Age + Formaleducationyears + WC + symptom +
anx + (1 | ID)
Data: Df
REML criterion at convergence: 532.3
Scaled residuals:
Min 1Q Median 3Q Max
-2.8023 -0.4545 -0.0160 0.5311 3.2970
Random effects:
Groups Name Variance Std.Dev.
ID (Intercept) 0.1818 0.4264
Residual 0.1856 0.4308
Number of obs: 300, groups: ID, 114
Fixed effects:
Estimate Std. Error df t value Pr(>|t|)
(Intercept) 8.575e-01 3.900e-01 1.134e+02 2.199 0.02994 *
BSContr 3.693e-01 4.569e-02 2.820e+02 8.084 1.86e-14 ***
Age 9.637e-04 4.733e-03 1.044e+02 0.204 0.83906
Formaleducationyears 3.116e-02 1.715e-02 1.091e+02 1.817 0.07191 .
WC 4.428e-04 1.725e-04 2.927e+02 2.567 0.01075 *
symptom 5.877e-02 1.850e-02 2.471e+02 3.176 0.00168 **
anx 7.205e-02 3.141e-02 2.255e+02 2.294 0.02270 *
---
Signif. codes: 0 '***' 0.001 '**' 0.01 '*' 0.05 '.' 0.1 ' ' 1
Correlation of Fixed Effects:
(Intr) BSCntr Age Frmldc WC symptm
BSContr -0.192
Age -0.726 -0.019
Frmldctnyrs -0.643 -0.062 0.049
WC -0.058 -0.083 -0.018 -0.075
symptom -0.031 -0.025 -0.011 -0.092 0.159
anx -0.019 -0.141 0.010 -0.012 0.067 0.015
anova(toy3.1, toy4)refitting model(s) with ML (instead of REML)
Data: Df
Models:
toy4: X8wkContr ~ BSContr + Age + Formaleducationyears + WC + symptom + anx + (1 | ID)
toy3.1: X8wkContr ~ BSContr + Age + Formaleducationyears + Employment + WC + symptom + (1 | ID)
npar AIC BIC logLik deviance Chisq Df Pr(>Chisq)
toy4 9 499.65 532.98 -240.82 481.65
toy3.1 10 502.01 539.04 -241.00 482.01 0 1 1
vif(toy4)Loading required package: carData
Attaching package: 'car'
The following object is masked _by_ '.GlobalEnv':
vif
The following object is masked from 'package:dplyr':
recode
Term GVIF Df GVIF^(1/(2*Df))
BSContr BSContr 1.031570 1 1.015662
Age Age 1.002975 1 1.001486
Formaleducationyears Formaleducationyears 1.019631 1 1.009768
WC WC 1.040503 1 1.020051
symptom symptom 1.033237 1 1.016483
anx anx 1.023870 1 1.011865
GVIF^(1/Df)
BSContr 1.031570
Age 1.002975
Formaleducationyears 1.019631
WC 1.040503
symptom 1.033237
anx 1.023870
#toy 5 is significant better than toy4
toy5<-lmer(X8wkContr ~ BSContr+ Age+ Formaleducationyears+ WC+ symptom+anx+feel+(1|ID), Df)
summary(toy5)Linear mixed model fit by REML. t-tests use Satterthwaite's method [
lmerModLmerTest]
Formula: X8wkContr ~ BSContr + Age + Formaleducationyears + WC + symptom +
anx + feel + (1 | ID)
Data: Df
REML criterion at convergence: 533.9
Scaled residuals:
Min 1Q Median 3Q Max
-2.94612 -0.44090 -0.00612 0.55773 3.08014
Random effects:
Groups Name Variance Std.Dev.
ID (Intercept) 0.1804 0.4247
Residual 0.1828 0.4276
Number of obs: 300, groups: ID, 114
Fixed effects:
Estimate Std. Error df t value Pr(>|t|)
(Intercept) 9.798e-01 3.919e-01 1.168e+02 2.500 0.013809 *
BSContr 3.675e-01 4.538e-02 2.804e+02 8.097 1.73e-14 ***
Age 5.016e-04 4.714e-03 1.043e+02 0.106 0.915464
Formaleducationyears 2.942e-02 1.708e-02 1.091e+02 1.723 0.087790 .
WC 4.218e-04 1.716e-04 2.915e+02 2.458 0.014547 *
symptom 6.662e-02 1.872e-02 2.421e+02 3.560 0.000447 ***
anx 6.784e-02 3.124e-02 2.251e+02 2.172 0.030905 *
feel -3.512e-02 1.584e-02 2.433e+02 -2.217 0.027572 *
---
Signif. codes: 0 '***' 0.001 '**' 0.01 '*' 0.05 '.' 0.1 ' ' 1
Correlation of Fixed Effects:
(Intr) BSCntr Age Frmldc WC symptm anx
BSContr -0.192
Age -0.724 -0.018
Frmldctnyrs -0.643 -0.061 0.050
WC -0.065 -0.082 -0.016 -0.072
symptom -0.004 -0.028 -0.019 -0.099 0.146
anx -0.027 -0.140 0.013 -0.009 0.070 0.004
feel -0.141 0.017 0.044 0.047 0.055 -0.191 0.060
anova(toy5, toy4)refitting model(s) with ML (instead of REML)
Data: Df
Models:
toy4: X8wkContr ~ BSContr + Age + Formaleducationyears + WC + symptom + anx + (1 | ID)
toy5: X8wkContr ~ BSContr + Age + Formaleducationyears + WC + symptom + anx + feel + (1 | ID)
npar AIC BIC logLik deviance Chisq Df Pr(>Chisq)
toy4 9 499.65 532.98 -240.82 481.65
toy5 10 496.66 533.69 -238.33 476.66 4.9946 1 0.02543 *
---
Signif. codes: 0 '***' 0.001 '**' 0.01 '*' 0.05 '.' 0.1 ' ' 1
vif(toy5) Term GVIF Df GVIF^(1/(2*Df))
BSContr BSContr 1.031832 1 1.015791
Age Age 1.004937 1 1.002465
Formaleducationyears Formaleducationyears 1.021783 1 1.010833
WC WC 1.043609 1 1.021572
symptom symptom 1.072277 1 1.035508
anx anx 1.027573 1 1.013693
feel feel 1.052467 1 1.025898
GVIF^(1/Df)
BSContr 1.031832
Age 1.004937
Formaleducationyears 1.021783
WC 1.043609
symptom 1.072277
anx 1.027573
feel 1.052467
#toy 6 is significant better than toy5, no collinearity issue
toy6<-lmer(X8wkContr ~ BSContr+ Age+ Formaleducationyears+ WC+ symptom+anx+feel+ focuspresent+(1|ID), Df)
summary(toy6)Linear mixed model fit by REML. t-tests use Satterthwaite's method [
lmerModLmerTest]
Formula: X8wkContr ~ BSContr + Age + Formaleducationyears + WC + symptom +
anx + feel + focuspresent + (1 | ID)
Data: Df
REML criterion at convergence: 536.8
Scaled residuals:
Min 1Q Median 3Q Max
-2.82512 -0.43991 -0.00717 0.56171 2.97512
Random effects:
Groups Name Variance Std.Dev.
ID (Intercept) 0.1853 0.4305
Residual 0.1780 0.4219
Number of obs: 300, groups: ID, 114
Fixed effects:
Estimate Std. Error df t value Pr(>|t|)
(Intercept) 1.404e+00 4.399e-01 1.584e+02 3.191 0.00171 **
BSContr 3.624e-01 4.506e-02 2.781e+02 8.043 2.54e-14 ***
Age -3.957e-04 4.762e-03 1.052e+02 -0.083 0.93393
Formaleducationyears 2.723e-02 1.721e-02 1.091e+02 1.582 0.11659
WC 3.648e-04 1.723e-04 2.888e+02 2.117 0.03515 *
symptom 6.083e-02 1.868e-02 2.373e+02 3.256 0.00130 **
anx 6.661e-02 3.089e-02 2.222e+02 2.156 0.03213 *
feel -3.490e-02 1.568e-02 2.399e+02 -2.226 0.02696 *
focuspresent -2.042e-02 9.455e-03 2.581e+02 -2.159 0.03174 *
---
Signif. codes: 0 '***' 0.001 '**' 0.01 '*' 0.05 '.' 0.1 ' ' 1
Correlation of Fixed Effects:
(Intr) BSCntr Age Frmldc WC symptm anx feel
BSContr -0.188
Age -0.686 -0.014
Frmldctnyrs -0.604 -0.057 0.056
WC -0.123 -0.074 -0.002 -0.061
symptom -0.061 -0.023 -0.007 -0.088 0.163
anx -0.029 -0.139 0.014 -0.008 0.072 0.007
feel -0.122 0.016 0.043 0.046 0.053 -0.189 0.061
focuspresnt -0.445 0.042 0.087 0.064 0.148 0.131 0.012 -0.004
anova(toy6, toy5)refitting model(s) with ML (instead of REML)
Data: Df
Models:
toy5: X8wkContr ~ BSContr + Age + Formaleducationyears + WC + symptom + anx + feel + (1 | ID)
toy6: X8wkContr ~ BSContr + Age + Formaleducationyears + WC + symptom + anx + feel + focuspresent + (1 | ID)
npar AIC BIC logLik deviance Chisq Df Pr(>Chisq)
toy5 10 496.66 533.69 -238.33 476.66
toy6 11 493.98 534.72 -235.99 471.98 4.6731 1 0.03064 *
---
Signif. codes: 0 '***' 0.001 '**' 0.01 '*' 0.05 '.' 0.1 ' ' 1
vif(toy6) Term GVIF Df GVIF^(1/(2*Df))
BSContr BSContr 1.033272 1 1.016500
Age Age 1.012557 1 1.006259
Formaleducationyears Formaleducationyears 1.025384 1 1.012613
WC WC 1.066923 1 1.032920
symptom symptom 1.090605 1 1.044320
anx anx 1.027801 1 1.013805
feel feel 1.052013 1 1.025677
focuspresent focuspresent 1.053668 1 1.026483
GVIF^(1/Df)
BSContr 1.033272
Age 1.012557
Formaleducationyears 1.025384
WC 1.066923
symptom 1.090605
anx 1.027801
feel 1.052013
focuspresent 1.053668
#toy 7 is significant better than toy6, no collinearity issue
toy7<-lmer(X8wkContr ~ BSContr+ Age+ Formaleducationyears+ WC+ symptom+anx+feel+ focuspresent+money+(1|ID), Df)
summary(toy7)Linear mixed model fit by REML. t-tests use Satterthwaite's method [
lmerModLmerTest]
Formula: X8wkContr ~ BSContr + Age + Formaleducationyears + WC + symptom +
anx + feel + focuspresent + money + (1 | ID)
Data: Df
REML criterion at convergence: 535.8
Scaled residuals:
Min 1Q Median 3Q Max
-2.69844 -0.43143 -0.02449 0.54561 2.95450
Random effects:
Groups Name Variance Std.Dev.
ID (Intercept) 0.1801 0.4244
Residual 0.1775 0.4213
Number of obs: 300, groups: ID, 114
Fixed effects:
Estimate Std. Error df t value Pr(>|t|)
(Intercept) 1.364e+00 4.364e-01 1.575e+02 3.126 0.00211 **
BSContr 3.654e-01 4.489e-02 2.781e+02 8.138 1.35e-14 ***
Age -2.110e-04 4.712e-03 1.044e+02 -0.045 0.96436
Formaleducationyears 2.947e-02 1.707e-02 1.091e+02 1.726 0.08718 .
WC 4.140e-04 1.733e-04 2.889e+02 2.389 0.01754 *
symptom 5.711e-02 1.874e-02 2.351e+02 3.047 0.00257 **
anx 6.184e-02 3.092e-02 2.214e+02 2.000 0.04670 *
feel -3.588e-02 1.564e-02 2.394e+02 -2.294 0.02268 *
focuspresent -1.906e-02 9.446e-03 2.594e+02 -2.018 0.04465 *
money -1.879e-01 9.548e-02 2.520e+02 -1.968 0.05014 .
---
Signif. codes: 0 '***' 0.001 '**' 0.01 '*' 0.05 '.' 0.1 ' ' 1
Correlation of Fixed Effects:
(Intr) BSCntr Age Frmldc WC symptm anx feel fcsprs
BSContr -0.190
Age -0.685 -0.014
Frmldctnyrs -0.604 -0.055 0.057
WC -0.129 -0.070 0.001 -0.051
symptom -0.057 -0.026 -0.009 -0.095 0.145
anx -0.025 -0.141 0.012 -0.014 0.059 0.015
feel -0.121 0.015 0.043 0.044 0.048 -0.185 0.063
focuspresnt -0.449 0.043 0.089 0.069 0.156 0.123 0.006 -0.007
money 0.043 -0.027 -0.020 -0.069 -0.142 0.108 0.082 0.031 -0.069
anova(toy7, toy6)refitting model(s) with ML (instead of REML)
Data: Df
Models:
toy6: X8wkContr ~ BSContr + Age + Formaleducationyears + WC + symptom + anx + feel + focuspresent + (1 | ID)
toy7: X8wkContr ~ BSContr + Age + Formaleducationyears + WC + symptom + anx + feel + focuspresent + money + (1 | ID)
npar AIC BIC logLik deviance Chisq Df Pr(>Chisq)
toy6 11 493.98 534.72 -235.99 471.98
toy7 12 492.00 536.44 -234.00 468.00 3.9841 1 0.04593 *
---
Signif. codes: 0 '***' 0.001 '**' 0.01 '*' 0.05 '.' 0.1 ' ' 1
vif(toy7) Term GVIF Df GVIF^(1/(2*Df))
BSContr BSContr 1.034145 1 1.016929
Age Age 1.013113 1 1.006535
Formaleducationyears Formaleducationyears 1.030624 1 1.015196
WC WC 1.089011 1 1.043557
symptom symptom 1.103666 1 1.050555
anx anx 1.034728 1 1.017216
feel feel 1.053232 1 1.026271
focuspresent focuspresent 1.059155 1 1.029153
money money 1.059961 1 1.029544
GVIF^(1/Df)
BSContr 1.034145
Age 1.013113
Formaleducationyears 1.030624
WC 1.089011
symptom 1.103666
anx 1.034728
feel 1.053232
focuspresent 1.059155
money 1.059961
#toy 8 is marginally significant better than toy7 but with lower AIC, no collinearity issue
toy8<-lmer(X8wkContr ~ BSContr+ Age+ Formaleducationyears+ WC+ symptom+anx+feel+ focuspresent+money+
informal+(1|ID), Df)
summary(toy8)Linear mixed model fit by REML. t-tests use Satterthwaite's method [
lmerModLmerTest]
Formula: X8wkContr ~ BSContr + Age + Formaleducationyears + WC + symptom +
anx + feel + focuspresent + money + informal + (1 | ID)
Data: Df
REML criterion at convergence: 536.1
Scaled residuals:
Min 1Q Median 3Q Max
-2.71199 -0.48661 -0.00619 0.52685 2.93946
Random effects:
Groups Name Variance Std.Dev.
ID (Intercept) 0.1747 0.4179
Residual 0.1774 0.4212
Number of obs: 300, groups: ID, 114
Fixed effects:
Estimate Std. Error df t value Pr(>|t|)
(Intercept) 1.343e+00 4.326e-01 1.569e+02 3.104 0.00226 **
BSContr 3.640e-01 4.476e-02 2.780e+02 8.131 1.42e-14 ***
Age -4.639e-04 4.662e-03 1.036e+02 -0.100 0.92093
Formaleducationyears 2.928e-02 1.689e-02 1.081e+02 1.734 0.08579 .
WC 4.195e-04 1.726e-04 2.882e+02 2.430 0.01572 *
symptom 6.094e-02 1.880e-02 2.378e+02 3.241 0.00136 **
anx 5.542e-02 3.108e-02 2.196e+02 1.783 0.07596 .
feel -3.766e-02 1.564e-02 2.388e+02 -2.408 0.01681 *
focuspresent -1.861e-02 9.424e-03 2.594e+02 -1.975 0.04934 *
money -1.812e-01 9.535e-02 2.516e+02 -1.901 0.05847 .
informal 1.138e-01 6.091e-02 2.280e+02 1.868 0.06302 .
---
Signif. codes: 0 '***' 0.001 '**' 0.01 '*' 0.05 '.' 0.1 ' ' 1
Correlation of Fixed Effects:
(Intr) BSCntr Age Frmldc WC symptm anx feel fcsprs
BSContr -0.190
Age -0.683 -0.013
Frmldctnyrs -0.603 -0.056 0.057
WC -0.130 -0.070 0.000 -0.051
symptom -0.059 -0.027 -0.012 -0.096 0.145
anx -0.023 -0.137 0.015 -0.013 0.056 0.002
feel -0.120 0.018 0.045 0.044 0.047 -0.190 0.069
focuspresnt -0.452 0.042 0.089 0.069 0.157 0.124 0.003 -0.008
money 0.042 -0.027 -0.021 -0.070 -0.140 0.112 0.077 0.028 -0.067
informal -0.023 -0.024 -0.029 -0.003 0.014 0.102 -0.115 -0.059 0.020
money
BSContr
Age
Frmldctnyrs
WC
symptom
anx
feel
focuspresnt
money
informal 0.043
anova(toy8, toy7)refitting model(s) with ML (instead of REML)
Data: Df
Models:
toy7: X8wkContr ~ BSContr + Age + Formaleducationyears + WC + symptom + anx + feel + focuspresent + money + (1 | ID)
toy8: X8wkContr ~ BSContr + Age + Formaleducationyears + WC + symptom + anx + feel + focuspresent + money + informal + (1 | ID)
npar AIC BIC logLik deviance Chisq Df Pr(>Chisq)
toy7 12 492.0 536.44 -234.0 468.0
toy8 13 490.4 538.55 -232.2 464.4 3.5995 1 0.0578 .
---
Signif. codes: 0 '***' 0.001 '**' 0.01 '*' 0.05 '.' 0.1 ' ' 1
vif(toy8) Term GVIF Df GVIF^(1/(2*Df))
BSContr BSContr 1.034838 1 1.017270
Age Age 1.014196 1 1.007073
Formaleducationyears Formaleducationyears 1.031115 1 1.015438
WC WC 1.089312 1 1.043701
symptom symptom 1.115568 1 1.056205
anx anx 1.048493 1 1.023959
feel feel 1.057234 1 1.028219
focuspresent focuspresent 1.060075 1 1.029600
money money 1.061772 1 1.030423
informal informal 1.029564 1 1.014674
GVIF^(1/Df)
BSContr 1.034838
Age 1.014196
Formaleducationyears 1.031115
WC 1.089312
symptom 1.115568
anx 1.048493
feel 1.057234
focuspresent 1.060075
money 1.061772
informal 1.029564
# nonflu (non-fluency) and informal (informality) are signficantlt positive correlated; toy 8.1 is worse than toy 8
cor.test(Df$nonflu, Df$informal) Pearson's product-moment correlation
data: Df$nonflu and Df$informal
t = 15.226, df = 312, p-value < 2.2e-16
alternative hypothesis: true correlation is not equal to 0
95 percent confidence interval:
0.5844543 0.7121240
sample estimates:
cor
0.6529022
toy8.1<-lmer(X8wkContr ~ BSContr+ Age+ Formaleducationyears+ WC+ symptom+anx+feel+ focuspresent+money+
nonflu+(1|ID), Df)
summary(toy8.1)Linear mixed model fit by REML. t-tests use Satterthwaite's method [
lmerModLmerTest]
Formula: X8wkContr ~ BSContr + Age + Formaleducationyears + WC + symptom +
anx + feel + focuspresent + money + nonflu + (1 | ID)
Data: Df
REML criterion at convergence: 537.3
Scaled residuals:
Min 1Q Median 3Q Max
-2.69995 -0.44092 -0.02639 0.56696 2.95963
Random effects:
Groups Name Variance Std.Dev.
ID (Intercept) 0.1764 0.4199
Residual 0.1785 0.4225
Number of obs: 300, groups: ID, 114
Fixed effects:
Estimate Std. Error df t value Pr(>|t|)
(Intercept) 1.403e+00 4.357e-01 1.587e+02 3.221 0.00155 **
BSContr 3.584e-01 4.535e-02 2.783e+02 7.904 6.35e-14 ***
Age -5.369e-04 4.688e-03 1.040e+02 -0.115 0.90905
Formaleducationyears 2.853e-02 1.698e-02 1.084e+02 1.680 0.09579 .
WC 4.233e-04 1.734e-04 2.883e+02 2.441 0.01524 *
symptom 5.370e-02 1.902e-02 2.337e+02 2.823 0.00517 **
anx 6.219e-02 3.098e-02 2.205e+02 2.008 0.04590 *
feel -3.657e-02 1.567e-02 2.391e+02 -2.333 0.02047 *
focuspresent -1.922e-02 9.456e-03 2.586e+02 -2.032 0.04313 *
money -1.903e-01 9.560e-02 2.515e+02 -1.991 0.04756 *
nonflu 1.081e-01 9.176e-02 2.380e+02 1.178 0.23987
---
Signif. codes: 0 '***' 0.001 '**' 0.01 '*' 0.05 '.' 0.1 ' ' 1
Correlation of Fixed Effects:
(Intr) BSCntr Age Frmldc WC symptm anx feel fcsprs
BSContr -0.199
Age -0.686 -0.005
Frmldctnyrs -0.604 -0.049 0.060
WC -0.126 -0.075 -0.002 -0.053
symptom -0.069 -0.002 0.001 -0.088 0.136
anx -0.025 -0.140 0.012 -0.014 0.058 0.013
feel -0.124 0.021 0.045 0.045 0.047 -0.177 0.062
focuspresnt -0.451 0.045 0.091 0.070 0.156 0.125 0.006 -0.006
money 0.042 -0.024 -0.019 -0.069 -0.141 0.109 0.082 0.031 -0.068
nonflu 0.080 -0.140 -0.060 -0.043 0.041 -0.162 0.004 -0.035 -0.021
money
BSContr
Age
Frmldctnyrs
WC
symptom
anx
feel
focuspresnt
money
nonflu -0.014
anova(toy8, toy8.1)refitting model(s) with ML (instead of REML)
Data: Df
Models:
toy8: X8wkContr ~ BSContr + Age + Formaleducationyears + WC + symptom + anx + feel + focuspresent + money + informal + (1 | ID)
toy8.1: X8wkContr ~ BSContr + Age + Formaleducationyears + WC + symptom + anx + feel + focuspresent + money + nonflu + (1 | ID)
npar AIC BIC logLik deviance Chisq Df Pr(>Chisq)
toy8 13 490.40 538.55 -232.20 464.40
toy8.1 13 492.56 540.71 -233.28 466.56 0 0
#toy9 is not significantly better than toy 8, but has a bit lower AIC
toy9<-lmer(X8wkContr ~ BSContr+ Age+ Formaleducationyears+ WC+ symptom+anx+feel+ focuspresent+money+
informal+ body+(1|ID), Df)
summary(toy9)Linear mixed model fit by REML. t-tests use Satterthwaite's method [
lmerModLmerTest]
Formula: X8wkContr ~ BSContr + Age + Formaleducationyears + WC + symptom +
anx + feel + focuspresent + money + informal + body + (1 | ID)
Data: Df
REML criterion at convergence: 540.5
Scaled residuals:
Min 1Q Median 3Q Max
-2.76220 -0.46443 -0.01638 0.50509 2.98145
Random effects:
Groups Name Variance Std.Dev.
ID (Intercept) 0.1685 0.4105
Residual 0.1788 0.4228
Number of obs: 300, groups: ID, 114
Fixed effects:
Estimate Std. Error df t value Pr(>|t|)
(Intercept) 1.429e+00 4.333e-01 1.606e+02 3.298 0.00120 **
BSContr 3.519e-01 4.562e-02 2.782e+02 7.712 2.21e-13 ***
Age -4.167e-04 4.608e-03 1.019e+02 -0.090 0.92812
Formaleducationyears 3.024e-02 1.672e-02 1.068e+02 1.809 0.07322 .
WC 3.876e-04 1.739e-04 2.875e+02 2.229 0.02656 *
symptom 6.204e-02 1.884e-02 2.381e+02 3.293 0.00114 **
anx 5.745e-02 3.117e-02 2.192e+02 1.843 0.06667 .
feel -2.977e-02 1.659e-02 2.390e+02 -1.795 0.07394 .
focuspresent -2.133e-02 9.628e-03 2.569e+02 -2.216 0.02758 *
money -1.850e-01 9.545e-02 2.512e+02 -1.938 0.05372 .
informal 1.203e-01 6.114e-02 2.293e+02 1.968 0.05024 .
body -2.162e-02 1.483e-02 2.290e+02 -1.458 0.14619
---
Signif. codes: 0 '***' 0.001 '**' 0.01 '*' 0.05 '.' 0.1 ' ' 1
Correlation of Fixed Effects:
(Intr) BSCntr Age Frmldc WC symptm anx feel fcsprs
BSContr -0.213
Age -0.674 -0.014
Frmldctnyrs -0.588 -0.064 0.058
WC -0.147 -0.044 0.000 -0.057
symptom -0.056 -0.031 -0.012 -0.096 0.140
anx -0.017 -0.142 0.016 -0.012 0.050 0.002
feel -0.067 -0.047 0.045 0.057 0.001 -0.171 0.077
focuspresnt -0.471 0.078 0.087 0.060 0.180 0.117 -0.005 -0.074
money 0.039 -0.022 -0.021 -0.071 -0.136 0.112 0.076 0.021 -0.061
informal -0.014 -0.035 -0.029 -0.001 0.005 0.102 -0.112 -0.036 0.007
body -0.141 0.196 -0.007 -0.044 0.131 -0.027 -0.039 -0.329 0.202
money infrml
BSContr
Age
Frmldctnyrs
WC
symptom
anx
feel
focuspresnt
money
informal 0.042
body 0.018 -0.062
anova(toy8, toy9)refitting model(s) with ML (instead of REML)
Data: Df
Models:
toy8: X8wkContr ~ BSContr + Age + Formaleducationyears + WC + symptom + anx + feel + focuspresent + money + informal + (1 | ID)
toy9: X8wkContr ~ BSContr + Age + Formaleducationyears + WC + symptom + anx + feel + focuspresent + money + informal + body + (1 | ID)
npar AIC BIC logLik deviance Chisq Df Pr(>Chisq)
toy8 13 490.40 538.55 -232.2 464.40
toy9 14 490.21 542.06 -231.1 462.21 2.1897 1 0.1389
#toy10 is not significantly better than toy 9.
toy10<-lmer(X8wkContr ~ BSContr+ Age+ Formaleducationyears+ WC+ symptom+anx+feel+ focuspresent+money+
informal+ body+ingest+(1|ID), Df)
summary(toy10)Linear mixed model fit by REML. t-tests use Satterthwaite's method [
lmerModLmerTest]
Formula: X8wkContr ~ BSContr + Age + Formaleducationyears + WC + symptom +
anx + feel + focuspresent + money + informal + body + ingest +
(1 | ID)
Data: Df
REML criterion at convergence: 546.9
Scaled residuals:
Min 1Q Median 3Q Max
-2.7410 -0.4665 -0.0188 0.5016 2.9849
Random effects:
Groups Name Variance Std.Dev.
ID (Intercept) 0.1694 0.4115
Residual 0.1790 0.4231
Number of obs: 300, groups: ID, 114
Fixed effects:
Estimate Std. Error df t value Pr(>|t|)
(Intercept) 1.412e+00 4.348e-01 1.599e+02 3.247 0.00142 **
BSContr 3.612e-01 4.802e-02 2.795e+02 7.522 7.4e-13 ***
Age -2.634e-04 4.623e-03 1.020e+02 -0.057 0.95469
Formaleducationyears 3.094e-02 1.678e-02 1.070e+02 1.844 0.06798 .
WC 3.858e-04 1.741e-04 2.863e+02 2.216 0.02748 *
symptom 6.146e-02 1.887e-02 2.364e+02 3.257 0.00129 **
anx 5.603e-02 3.127e-02 2.172e+02 1.792 0.07455 .
feel -2.904e-02 1.664e-02 2.383e+02 -1.745 0.08227 .
focuspresent -2.151e-02 9.639e-03 2.555e+02 -2.231 0.02655 *
money -1.861e-01 9.555e-02 2.501e+02 -1.947 0.05261 .
informal 1.194e-01 6.120e-02 2.280e+02 1.952 0.05220 .
body -2.321e-02 1.506e-02 2.293e+02 -1.541 0.12459
ingest -8.456e-03 1.336e-02 2.681e+02 -0.633 0.52726
---
Signif. codes: 0 '***' 0.001 '**' 0.01 '*' 0.05 '.' 0.1 ' ' 1
Correlation matrix not shown by default, as p = 13 > 12.
Use print(x, correlation=TRUE) or
vcov(x) if you need it
anova(toy10, toy9)refitting model(s) with ML (instead of REML)
Data: Df
Models:
toy9: X8wkContr ~ BSContr + Age + Formaleducationyears + WC + symptom + anx + feel + focuspresent + money + informal + body + (1 | ID)
toy10: X8wkContr ~ BSContr + Age + Formaleducationyears + WC + symptom + anx + feel + focuspresent + money + informal + body + ingest + (1 | ID)
npar AIC BIC logLik deviance Chisq Df Pr(>Chisq)
toy9 14 490.21 542.06 -231.1 462.21
toy10 15 491.79 547.35 -230.9 461.79 0.417 1 0.5184
#toy11: although age is an important covariate, but it does not explain much variation; after dropping age, AIC is much lower
toy11<-lmer(X8wkContr ~ BSContr+ Formaleducationyears+ WC+ symptom+anx+feel+ focuspresent+money+
informal+(1|ID), Df)
summary(toy11)Linear mixed model fit by REML. t-tests use Satterthwaite's method [
lmerModLmerTest]
Formula: X8wkContr ~ BSContr + Formaleducationyears + WC + symptom + anx +
feel + focuspresent + money + informal + (1 | ID)
Data: Df
REML criterion at convergence: 527.2
Scaled residuals:
Min 1Q Median 3Q Max
-2.72123 -0.48823 -0.00718 0.52831 2.92969
Random effects:
Groups Name Variance Std.Dev.
ID (Intercept) 0.1723 0.4151
Residual 0.1774 0.4212
Number of obs: 300, groups: ID, 114
Fixed effects:
Estimate Std. Error df t value Pr(>|t|)
(Intercept) 1.313e+00 3.149e-01 1.926e+02 4.169 4.62e-05 ***
BSContr 3.642e-01 4.470e-02 2.789e+02 8.147 1.26e-14 ***
Formaleducationyears 2.934e-02 1.678e-02 1.088e+02 1.748 0.08322 .
WC 4.199e-04 1.724e-04 2.892e+02 2.436 0.01545 *
symptom 6.103e-02 1.879e-02 2.384e+02 3.248 0.00133 **
anx 5.555e-02 3.106e-02 2.201e+02 1.788 0.07511 .
feel -3.761e-02 1.561e-02 2.398e+02 -2.409 0.01675 *
focuspresent -1.849e-02 9.377e-03 2.627e+02 -1.972 0.04968 *
money -1.819e-01 9.525e-02 2.524e+02 -1.909 0.05737 .
informal 1.139e-01 6.085e-02 2.288e+02 1.872 0.06244 .
---
Signif. codes: 0 '***' 0.001 '**' 0.01 '*' 0.05 '.' 0.1 ' ' 1
Correlation of Fixed Effects:
(Intr) BSCntr Frmldc WC symptm anx feel fcsprs money
BSContr -0.273
Frmldctnyrs -0.772 -0.055
WC -0.178 -0.071 -0.051
symptom -0.093 -0.027 -0.096 0.145
anx -0.016 -0.137 -0.014 0.056 0.002
feel -0.124 0.018 0.042 0.048 -0.190 0.068
focuspresnt -0.539 0.043 0.065 0.158 0.126 0.002 -0.012
money 0.038 -0.027 -0.069 -0.140 0.112 0.077 0.029 -0.065
informal -0.059 -0.024 -0.002 0.014 0.101 -0.114 -0.058 0.023 0.042
anova(toy9, toy11)refitting model(s) with ML (instead of REML)
Data: Df
Models:
toy11: X8wkContr ~ BSContr + Formaleducationyears + WC + symptom + anx + feel + focuspresent + money + informal + (1 | ID)
toy9: X8wkContr ~ BSContr + Age + Formaleducationyears + WC + symptom + anx + feel + focuspresent + money + informal + body + (1 | ID)
npar AIC BIC logLik deviance Chisq Df Pr(>Chisq)
toy11 12 488.41 532.85 -232.21 464.41
toy9 14 490.21 542.06 -231.10 462.21 2.1999 2 0.3329
anova(toy8, toy11)refitting model(s) with ML (instead of REML)
Data: Df
Models:
toy11: X8wkContr ~ BSContr + Formaleducationyears + WC + symptom + anx + feel + focuspresent + money + informal + (1 | ID)
toy8: X8wkContr ~ BSContr + Age + Formaleducationyears + WC + symptom + anx + feel + focuspresent + money + informal + (1 | ID)
npar AIC BIC logLik deviance Chisq Df Pr(>Chisq)
toy11 12 488.41 532.85 -232.21 464.41
toy8 13 490.40 538.55 -232.20 464.40 0.0103 1 0.9192
This residual plot does not indicate any deviations from a linear form. It also shows relatively constant variance across the fitted range. The slight reduction in apparent variance on the right and left of the graph are likely a result of there being fewer observation in these predicted areas.
DfNA<-Df %>% na.omit()
# Linearity of the predictors are assumed
ggplot(data.frame(x1=DfNA$Formaleducationyears,pearson=residuals(toy11,type="pearson")),
aes(x=x1,y=pearson)) +
geom_point() +
theme_bw()+xlab("Years of formal education")
ggplot(data.frame(x2=DfNA$WC,pearson=residuals(toy11,type="pearson")),
aes(x=x2,y=pearson)) +
geom_point() +
theme_bw()+xlab("Total word count")
ggplot(data.frame(x2=DfNA$BSContr,pearson=residuals(toy11,type="pearson")),
aes(x=x2,y=pearson)) +
geom_point() +
theme_bw()+xlab("Baseline controllability score")
ggplot(data.frame(x2=DfNA$symptom,pearson=residuals(toy11,type="pearson")),
aes(x=x2,y=pearson)) +
geom_point() +
theme_bw()+xlab("Word category Symptom")
ggplot(data.frame(x2=DfNA$anx,pearson=residuals(toy11,type="pearson")),
aes(x=x2,y=pearson)) +
geom_point() +
theme_bw()+xlab("Word category anxiety")
ggplot(data.frame(x2=DfNA$feel,pearson=residuals(toy11,type="pearson")),
aes(x=x2,y=pearson)) +
geom_point() +
theme_bw()+xlab("Word category feel")
ggplot(data.frame(x2=DfNA$focuspresent,pearson=residuals(toy11,type="pearson")),
aes(x=x2,y=pearson)) +
geom_point() +
theme_bw()+xlab("Word category focus present")
ggplot(data.frame(x2=DfNA$money,pearson=residuals(toy11,type="pearson")),
aes(x=x2,y=pearson)) +
geom_point() +
theme_bw()+xlab("Word category money")
ggplot(data.frame(x2=DfNA$informal,pearson=residuals(toy11,type="pearson")),
aes(x=x2,y=pearson)) +
geom_point() +
theme_bw()+xlab("Word category informal")
# Homogenity is assumed
plot(toy11)
# normality of residuals is assumed
qqnorm(resid(toy11))
fixef(toy11) (Intercept) BSContr Formaleducationyears
1.3125376339 0.3641982228 0.0293420633
WC symptom anx
0.0004199339 0.0610343156 0.0555478219
feel focuspresent money
-0.0376148315 -0.0184900944 -0.1818553261
informal
0.1139350304
confint.merMod(toy11)Computing profile confidence intervals ...
2.5 % 97.5 %
.sig01 3.330604e-01 0.4943690796
.sigma 3.747489e-01 0.4605494913
(Intercept) 7.038722e-01 1.9257762630
BSContr 2.772824e-01 0.4513289024
Formaleducationyears -3.181742e-03 0.0621534581
WC 8.633059e-05 0.0007534760
symptom 2.459139e-02 0.0977411199
anx -4.504202e-03 0.1158319369
feel -6.785587e-02 -0.0074302576
focuspresent -3.664360e-02 -0.0002840905
money -3.669866e-01 0.0025398478
informal -3.914554e-03 0.2326161226
names(Df) [1] "ID" "Employment" "Marriage"
[4] "race" "ethinicity (latio)" "Age"
[7] "Formaleducationyears" "SymptomNo" "Symptom"
[10] "X8wkContr" "BSContr" "WC"
[13] "symptom" "effort" "impact"
[16] "positive.adj" "negative.adj" "controlled"
[19] "uncontrolled" "controlNN" "controlVB"
[22] "Analytic" "Clout" "Authentic"
[25] "Tone" "WPS" "Sixltr"
[28] "Dic" "function." "pronoun"
[31] "ppron" "i" "we"
[34] "you" "shehe" "they"
[37] "ipron" "article" "prep"
[40] "auxverb" "adverb" "conj"
[43] "negate" "verb" "adj"
[46] "compare" "interrog" "number"
[49] "quant" "affect" "posemo"
[52] "negemo" "anx" "anger"
[55] "sad" "social" "family"
[58] "friend" "female" "male"
[61] "cogproc" "insight" "cause"
[64] "discrep" "tentat" "certain"
[67] "differ" "percept" "see"
[70] "hear" "feel" "bio"
[73] "body" "health" "sexual"
[76] "ingest" "drives" "affiliation"
[79] "achieve" "power" "reward"
[82] "risk" "focuspast" "focuspresent"
[85] "focusfuture" "relativ" "motion"
[88] "space" "time" "work"
[91] "leisure" "home" "money"
[94] "relig" "death" "informal"
[97] "swear" "netspeak" "assent"
[100] "nonflu" "filler" "AllPunc"
[103] "Period" "Comma" "Colon"
[106] "SemiC" "QMark" "Exclam"
[109] "Dash" "Apostro" "Parenth"
[112] "OtherP"
toyDf<-Df %>% select (13:112)
head(toyDf) symptom effort impact positive.adj negative.adj controlled uncontrolled
1 3.27 0.00 0.00 0 1.31 0 0.00
2 0.66 3.29 0.66 0 1.32 0 0.00
3 1.42 0.71 0.71 0 0.71 0 0.35
4 1.90 0.00 0.63 0 0.00 0 0.00
5 1.30 0.78 0.13 0 0.00 0 0.00
6 1.51 0.50 0.17 0 1.01 0 0.00
controlNN controlVB Analytic Clout Authentic Tone WPS Sixltr Dic
1 0 0.0000000 14.20 3.39 90.19 2.56 17.00 7.84 98.04
2 0 0.0000000 24.83 5.62 55.66 50.49 7.55 13.91 86.09
3 0 0.0000000 79.95 18.85 83.55 2.97 15.72 24.38 85.87
4 0 0.0000000 60.88 14.09 97.60 2.80 14.36 25.95 87.34
5 0 0.0000000 75.90 27.39 95.38 34.87 15.30 18.17 89.02
6 0 0.3305785 79.49 19.51 96.08 23.19 13.75 18.68 87.93
function. pronoun ppron i we you shehe they ipron article prep auxverb
1 63.40 21.57 11.76 11.76 0.00 0 0 0.00 9.80 6.54 11.11 15.03
2 51.66 17.88 7.28 7.28 0.00 0 0 0.00 10.60 4.64 7.95 8.61
3 48.06 11.31 9.54 8.48 0.00 0 0 1.06 1.77 5.65 14.84 8.48
4 51.90 14.56 6.33 6.33 0.00 0 0 0.00 8.23 5.70 13.92 10.76
5 52.16 11.11 8.50 7.84 0.13 0 0 0.52 2.61 6.41 15.82 8.10
6 51.74 13.06 10.25 9.75 0.00 0 0 0.50 2.81 6.94 15.70 6.45
adverb conj negate verb adj compare interrog number quant affect posemo
1 5.88 7.19 3.92 24.84 5.88 5.23 1.31 0.00 5.88 2.61 0.00
2 6.62 5.96 3.97 18.54 6.62 3.97 1.32 1.99 0.66 3.97 2.65
3 2.83 6.71 0.35 17.31 5.65 2.47 0.71 2.83 1.77 4.59 1.06
4 3.80 5.70 1.90 21.52 7.59 3.80 0.00 1.90 0.00 3.80 0.63
5 5.10 8.50 0.52 15.42 4.31 2.61 0.78 4.31 1.44 3.66 2.09
6 3.31 8.93 0.33 14.71 9.09 3.80 1.16 2.64 2.48 5.45 2.48
negemo anx anger sad social family friend female male cogproc insight cause
1 2.61 1.31 0.00 1.31 1.31 0.00 0.00 0.00 0.00 19.61 3.92 2.61
2 1.32 0.00 0.00 0.00 1.99 0.00 0.00 0.00 0.00 17.88 0.66 4.64
3 3.53 0.00 0.00 0.71 2.12 0.00 0.00 0.00 0.00 8.83 1.77 1.77
4 3.16 0.63 0.00 0.63 1.27 0.63 0.00 0.00 0.00 10.76 3.16 1.90
5 1.57 0.26 0.00 0.78 5.23 2.09 0.52 0.39 0.65 9.28 1.44 1.18
6 2.64 0.83 0.17 0.00 3.64 0.17 0.00 0.00 0.17 9.09 2.15 1.65
discrep tentat certain differ percept see hear feel bio body health sexual
1 3.27 7.19 1.31 3.92 4.58 0.00 0.00 4.58 4.58 0.65 1.96 0.65
2 2.65 3.31 2.65 6.62 5.96 1.32 0.00 3.97 7.95 4.64 4.64 0.00
3 1.77 4.24 0.35 2.12 2.47 0.71 0.00 2.12 9.19 3.53 5.30 0.00
4 0.00 1.90 0.63 3.80 1.27 0.00 0.00 1.27 6.96 1.27 2.53 0.63
5 1.05 3.92 0.78 3.01 2.09 0.65 0.26 1.44 6.01 1.31 2.88 0.00
6 1.16 3.80 0.50 2.15 5.12 0.33 0.17 3.80 6.12 2.98 2.48 0.17
ingest drives affiliation achieve power reward risk focuspast focuspresent
1 2.61 3.92 0.65 2.61 0.65 1.31 0.65 2.61 24.18
2 0.66 7.28 1.32 2.65 1.99 1.32 1.32 2.65 15.89
3 0.71 8.48 0.71 2.47 2.47 1.77 1.41 3.53 12.01
4 3.16 10.13 0.00 1.27 5.06 1.90 1.90 5.06 13.29
5 1.96 7.58 2.35 1.70 1.44 1.44 0.78 2.09 11.63
6 0.99 5.29 0.66 1.32 1.49 1.32 1.32 4.46 8.76
focusfuture relativ motion space time work leisure home money relig death
1 0.00 12.42 1.96 4.58 5.88 0.65 0.65 0.00 0.65 0 0
2 0.66 9.93 0.66 4.64 4.64 1.99 1.99 0.00 0.00 0 0
3 0.35 16.25 5.30 4.59 6.71 3.18 0.35 0.71 0.00 0 0
4 0.00 19.62 1.27 10.13 8.23 0.63 0.00 1.27 0.00 0 0
5 1.70 19.87 2.09 6.01 12.29 1.83 2.09 2.09 0.13 0 0
6 1.49 18.68 1.98 7.11 9.42 0.66 1.32 1.32 0.00 0 0
informal swear netspeak assent nonflu filler AllPunc Period Comma Colon SemiC
1 0.00 0 0.00 0.00 0.00 0 16.34 5.88 5.23 0.00 0
2 0.66 0 0.66 0.00 0.00 0 23.84 15.23 3.31 0.00 0
3 0.35 0 0.35 0.00 0.00 0 15.55 6.36 6.36 0.00 0
4 0.00 0 0.00 0.00 0.00 0 12.66 6.96 3.16 0.00 0
5 0.26 0 0.00 0.00 0.26 0 14.12 6.41 4.31 0.78 0
6 0.83 0 0.00 0.66 0.17 0 10.58 6.94 2.48 0.17 0
QMark Exclam Dash Apostro Parenth OtherP
1 0.00 0.00 0.00 5.23 0.00 0.00
2 0.66 0.00 3.31 1.32 0.00 0.00
3 0.00 0.00 1.77 0.35 0.71 0.00
4 0.00 0.00 0.63 1.27 0.00 0.63
5 0.00 0.13 0.78 0.65 0.65 0.39
6 0.00 0.33 0.50 0.00 0.00 0.17
#install.packages("factoextra")
library(factoextra)Welcome! Want to learn more? See two factoextra-related books at https://goo.gl/ve3WBa
pca<-prcomp(toyDf, scale=TRUE, center = TRUE)
fviz_eig(pca)
summary(pca)Importance of components:
PC1 PC2 PC3 PC4 PC5 PC6 PC7
Standard deviation 2.82340 2.42158 2.31142 1.9799 1.92183 1.81331 1.80990
Proportion of Variance 0.07972 0.05864 0.05343 0.0392 0.03693 0.03288 0.03276
Cumulative Proportion 0.07972 0.13836 0.19178 0.2310 0.26791 0.30080 0.33355
PC8 PC9 PC10 PC11 PC12 PC13 PC14
Standard deviation 1.72148 1.64892 1.59650 1.54936 1.50142 1.47058 1.4211
Proportion of Variance 0.02963 0.02719 0.02549 0.02401 0.02254 0.02163 0.0202
Cumulative Proportion 0.36319 0.39038 0.41587 0.43987 0.46241 0.48404 0.5042
PC15 PC16 PC17 PC18 PC19 PC20 PC21
Standard deviation 1.35722 1.3415 1.33253 1.30280 1.3002 1.24132 1.23542
Proportion of Variance 0.01842 0.0180 0.01776 0.01697 0.0169 0.01541 0.01526
Cumulative Proportion 0.52265 0.5406 0.55841 0.57538 0.5923 0.60769 0.62296
PC22 PC23 PC24 PC25 PC26 PC27 PC28
Standard deviation 1.20135 1.17432 1.15958 1.15304 1.13666 1.09916 1.08353
Proportion of Variance 0.01443 0.01379 0.01345 0.01329 0.01292 0.01208 0.01174
Cumulative Proportion 0.63739 0.65118 0.66463 0.67792 0.69084 0.70292 0.71466
PC29 PC30 PC31 PC32 PC33 PC34 PC35
Standard deviation 1.07089 1.04702 1.02803 1.01275 1.00450 0.97810 0.96859
Proportion of Variance 0.01147 0.01096 0.01057 0.01026 0.01009 0.00957 0.00938
Cumulative Proportion 0.72613 0.73709 0.74766 0.75792 0.76801 0.77757 0.78696
PC36 PC37 PC38 PC39 PC40 PC41 PC42
Standard deviation 0.95571 0.93846 0.92882 0.90661 0.89963 0.87636 0.86220
Proportion of Variance 0.00913 0.00881 0.00863 0.00822 0.00809 0.00768 0.00743
Cumulative Proportion 0.79609 0.80490 0.81352 0.82174 0.82984 0.83752 0.84495
PC43 PC44 PC45 PC46 PC47 PC48 PC49
Standard deviation 0.85684 0.84733 0.82561 0.81776 0.79830 0.79556 0.77293
Proportion of Variance 0.00734 0.00718 0.00682 0.00669 0.00637 0.00633 0.00597
Cumulative Proportion 0.85229 0.85947 0.86629 0.87298 0.87935 0.88568 0.89165
PC50 PC51 PC52 PC53 PC54 PC55 PC56
Standard deviation 0.75883 0.73188 0.72574 0.7209 0.71168 0.69416 0.69090
Proportion of Variance 0.00576 0.00536 0.00527 0.0052 0.00506 0.00482 0.00477
Cumulative Proportion 0.89741 0.90277 0.90803 0.9132 0.91830 0.92311 0.92789
PC57 PC58 PC59 PC60 PC61 PC62 PC63
Standard deviation 0.67758 0.66529 0.64865 0.64423 0.62831 0.6167 0.61057
Proportion of Variance 0.00459 0.00443 0.00421 0.00415 0.00395 0.0038 0.00373
Cumulative Proportion 0.93248 0.93690 0.94111 0.94526 0.94921 0.9530 0.95674
PC64 PC65 PC66 PC67 PC68 PC69 PC70
Standard deviation 0.58785 0.57921 0.55418 0.55118 0.52655 0.51087 0.4997
Proportion of Variance 0.00346 0.00335 0.00307 0.00304 0.00277 0.00261 0.0025
Cumulative Proportion 0.96020 0.96355 0.96662 0.96966 0.97243 0.97504 0.9775
PC71 PC72 PC73 PC74 PC75 PC76 PC77
Standard deviation 0.48082 0.4691 0.44613 0.43047 0.40902 0.40595 0.36625
Proportion of Variance 0.00231 0.0022 0.00199 0.00185 0.00167 0.00165 0.00134
Cumulative Proportion 0.97985 0.9820 0.98404 0.98590 0.98757 0.98922 0.99056
PC78 PC79 PC80 PC81 PC82 PC83 PC84
Standard deviation 0.36544 0.35248 0.3323 0.30887 0.29521 0.27104 0.24705
Proportion of Variance 0.00134 0.00124 0.0011 0.00095 0.00087 0.00073 0.00061
Cumulative Proportion 0.99189 0.99314 0.9942 0.99519 0.99607 0.99680 0.99741
PC85 PC86 PC87 PC88 PC89 PC90 PC91
Standard deviation 0.22945 0.1992 0.18269 0.16750 0.14890 0.12951 0.12877
Proportion of Variance 0.00053 0.0004 0.00033 0.00028 0.00022 0.00017 0.00017
Cumulative Proportion 0.99794 0.9983 0.99867 0.99895 0.99917 0.99934 0.99950
PC92 PC93 PC94 PC95 PC96 PC97 PC98
Standard deviation 0.11251 0.10563 0.09814 0.08368 0.07046 0.06353 0.0123
Proportion of Variance 0.00013 0.00011 0.00010 0.00007 0.00005 0.00004 0.0000
Cumulative Proportion 0.99963 0.99974 0.99984 0.99991 0.99996 1.00000 1.0000
PC99 PC100
Standard deviation 0.00108 0.0008434
Proportion of Variance 0.00000 0.0000000
Cumulative Proportion 1.00000 1.0000000
#Eigenvalues
eig.val<-get_eigenvalue(pca)
eig.val eigenvalue variance.percent cumulative.variance.percent
Dim.1 7.971565e+00 7.971565e+00 7.971565
Dim.2 5.864029e+00 5.864029e+00 13.835595
Dim.3 5.342651e+00 5.342651e+00 19.178246
Dim.4 3.919836e+00 3.919836e+00 23.098082
Dim.5 3.693412e+00 3.693412e+00 26.791494
Dim.6 3.288080e+00 3.288080e+00 30.079574
Dim.7 3.275731e+00 3.275731e+00 33.355304
Dim.8 2.963484e+00 2.963484e+00 36.318789
Dim.9 2.718953e+00 2.718953e+00 39.037742
Dim.10 2.548802e+00 2.548802e+00 41.586544
Dim.11 2.400513e+00 2.400513e+00 43.987057
Dim.12 2.254262e+00 2.254262e+00 46.241318
Dim.13 2.162599e+00 2.162599e+00 48.403917
Dim.14 2.019535e+00 2.019535e+00 50.423452
Dim.15 1.842045e+00 1.842045e+00 52.265497
Dim.16 1.799591e+00 1.799591e+00 54.065088
Dim.17 1.775623e+00 1.775623e+00 55.840711
Dim.18 1.697295e+00 1.697295e+00 57.538006
Dim.19 1.690464e+00 1.690464e+00 59.228471
Dim.20 1.540876e+00 1.540876e+00 60.769347
Dim.21 1.526271e+00 1.526271e+00 62.295618
Dim.22 1.443250e+00 1.443250e+00 63.738869
Dim.23 1.379023e+00 1.379023e+00 65.117891
Dim.24 1.344632e+00 1.344632e+00 66.462523
Dim.25 1.329498e+00 1.329498e+00 67.792022
Dim.26 1.292000e+00 1.292000e+00 69.084022
Dim.27 1.208159e+00 1.208159e+00 70.292181
Dim.28 1.174028e+00 1.174028e+00 71.466208
Dim.29 1.146808e+00 1.146808e+00 72.613016
Dim.30 1.096251e+00 1.096251e+00 73.709267
Dim.31 1.056845e+00 1.056845e+00 74.766113
Dim.32 1.025659e+00 1.025659e+00 75.791772
Dim.33 1.009026e+00 1.009026e+00 76.800798
Dim.34 9.566879e-01 9.566879e-01 77.757486
Dim.35 9.381615e-01 9.381615e-01 78.695648
Dim.36 9.133785e-01 9.133785e-01 79.609026
Dim.37 8.807083e-01 8.807083e-01 80.489734
Dim.38 8.626977e-01 8.626977e-01 81.352432
Dim.39 8.219506e-01 8.219506e-01 82.174383
Dim.40 8.093382e-01 8.093382e-01 82.983721
Dim.41 7.680133e-01 7.680133e-01 83.751734
Dim.42 7.433952e-01 7.433952e-01 84.495129
Dim.43 7.341702e-01 7.341702e-01 85.229300
Dim.44 7.179600e-01 7.179600e-01 85.947260
Dim.45 6.816385e-01 6.816385e-01 86.628898
Dim.46 6.687269e-01 6.687269e-01 87.297625
Dim.47 6.372855e-01 6.372855e-01 87.934911
Dim.48 6.329082e-01 6.329082e-01 88.567819
Dim.49 5.974269e-01 5.974269e-01 89.165246
Dim.50 5.758280e-01 5.758280e-01 89.741074
Dim.51 5.356539e-01 5.356539e-01 90.276728
Dim.52 5.267027e-01 5.267027e-01 90.803430
Dim.53 5.196558e-01 5.196558e-01 91.323086
Dim.54 5.064886e-01 5.064886e-01 91.829575
Dim.55 4.818527e-01 4.818527e-01 92.311427
Dim.56 4.773383e-01 4.773383e-01 92.788766
Dim.57 4.591097e-01 4.591097e-01 93.247875
Dim.58 4.426042e-01 4.426042e-01 93.690480
Dim.59 4.207465e-01 4.207465e-01 94.111226
Dim.60 4.150336e-01 4.150336e-01 94.526260
Dim.61 3.947707e-01 3.947707e-01 94.921030
Dim.62 3.803107e-01 3.803107e-01 95.301341
Dim.63 3.727988e-01 3.727988e-01 95.674140
Dim.64 3.455640e-01 3.455640e-01 96.019704
Dim.65 3.354832e-01 3.354832e-01 96.355187
Dim.66 3.071109e-01 3.071109e-01 96.662298
Dim.67 3.038003e-01 3.038003e-01 96.966098
Dim.68 2.772508e-01 2.772508e-01 97.243349
Dim.69 2.609883e-01 2.609883e-01 97.504337
Dim.70 2.496897e-01 2.496897e-01 97.754027
Dim.71 2.311834e-01 2.311834e-01 97.985210
Dim.72 2.200630e-01 2.200630e-01 98.205273
Dim.73 1.990284e-01 1.990284e-01 98.404302
Dim.74 1.853023e-01 1.853023e-01 98.589604
Dim.75 1.672993e-01 1.672993e-01 98.756903
Dim.76 1.647943e-01 1.647943e-01 98.921698
Dim.77 1.341388e-01 1.341388e-01 99.055837
Dim.78 1.335457e-01 1.335457e-01 99.189382
Dim.79 1.242400e-01 1.242400e-01 99.313622
Dim.80 1.104549e-01 1.104549e-01 99.424077
Dim.81 9.540005e-02 9.540005e-02 99.519477
Dim.82 8.714766e-02 8.714766e-02 99.606625
Dim.83 7.346386e-02 7.346386e-02 99.680089
Dim.84 6.103322e-02 6.103322e-02 99.741122
Dim.85 5.264880e-02 5.264880e-02 99.793771
Dim.86 3.967079e-02 3.967079e-02 99.833442
Dim.87 3.337729e-02 3.337729e-02 99.866819
Dim.88 2.805491e-02 2.805491e-02 99.894874
Dim.89 2.216997e-02 2.216997e-02 99.917044
Dim.90 1.677222e-02 1.677222e-02 99.933816
Dim.91 1.658129e-02 1.658129e-02 99.950397
Dim.92 1.265781e-02 1.265781e-02 99.963055
Dim.93 1.115851e-02 1.115851e-02 99.974214
Dim.94 9.631803e-03 9.631803e-03 99.983845
Dim.95 7.001609e-03 7.001609e-03 99.990847
Dim.96 4.964277e-03 4.964277e-03 99.995811
Dim.97 4.035665e-03 4.035665e-03 99.999847
Dim.98 1.512369e-04 1.512369e-04 99.999998
Dim.99 1.167221e-06 1.167221e-06 99.999999
Dim.100 7.114000e-07 7.114000e-07 100.000000
#results for variables
result.var<-get_pca_var(pca)
head(result.var$coord) Dim.1 Dim.2 Dim.3 Dim.4 Dim.5
symptom -0.15006906 0.45357191 -0.005839691 0.32280418 -0.22603977
effort 0.02967201 0.14867102 -0.004461888 -0.08820889 -0.06595448
impact -0.04180244 0.28067453 0.036160929 0.05181740 0.17162450
positive.adj -0.04139276 0.02069494 0.171093920 -0.05291939 -0.30376667
negative.adj 0.04051528 0.30029470 0.082814581 -0.28445983 -0.07720504
controlled -0.04927962 0.16256150 -0.044968047 0.18935354 -0.21945971
Dim.6 Dim.7 Dim.8 Dim.9 Dim.10
symptom 0.15119904 -0.04083519 0.20618634 -0.25187454 0.001469486
effort -0.02797639 -0.07859082 -0.10953240 0.44804315 0.043756257
impact 0.13903763 0.03566462 0.09681443 0.40218407 0.155671179
positive.adj 0.01276775 0.06615737 -0.23331434 -0.01173285 0.046366860
negative.adj 0.30453195 0.03634215 -0.18019598 -0.03696745 -0.243530658
controlled -0.03795510 0.26073263 0.47729391 -0.23950868 0.213246829
Dim.11 Dim.12 Dim.13 Dim.14 Dim.15
symptom -0.05349429 0.090709972 -0.01554618 0.12359388 -0.08658422
effort 0.24688249 -0.016357867 -0.08799463 0.08781852 0.12796417
impact 0.06343173 0.065776990 -0.07576899 -0.03343537 -0.16082075
positive.adj -0.17235924 0.110891358 0.13693408 -0.05257370 0.03804841
negative.adj -0.12739031 0.048948995 -0.03017091 -0.02229087 0.03527749
controlled -0.03017373 0.009259714 -0.14816638 0.24318296 -0.02834701
Dim.16 Dim.17 Dim.18 Dim.19 Dim.20
symptom 0.17051877 -0.16360394 0.008875469 -0.045160106 0.106905663
effort 0.05356207 -0.04147137 0.292565312 -0.002407799 0.142035539
impact -0.06070459 0.09865611 0.197664024 0.074738022 0.009055056
positive.adj 0.06194144 0.06754268 0.042641172 0.238709263 -0.135972862
negative.adj -0.40370265 0.10639380 0.099914567 -0.131875265 0.089210971
controlled -0.01968542 -0.01683723 0.055390272 0.139490894 0.240344483
Dim.21 Dim.22 Dim.23 Dim.24 Dim.25
symptom -0.05551345 0.2724879 -0.09148741 0.08587028 -0.12948663
effort -0.05741432 0.1254985 -0.11122196 0.30710831 -0.04902733
impact 0.06895730 0.1095497 -0.05849515 0.10628632 -0.02163478
positive.adj 0.14867508 -0.2345263 -0.18024909 0.04326070 -0.21677169
negative.adj 0.15735152 -0.1265856 0.15441005 -0.07491472 0.02710858
controlled -0.02701010 -0.2356116 0.05727279 0.11947638 0.04112219
Dim.26 Dim.27 Dim.28 Dim.29 Dim.30
symptom -0.10432237 0.009695423 0.068968879 -0.07563156 0.057340794
effort 0.16782195 0.323064826 -0.129621213 0.13207707 -0.037567825
impact -0.07838213 -0.085590880 0.097762555 -0.03148811 -0.060881063
positive.adj -0.06895013 -0.376445310 -0.051254117 -0.07947377 0.010095820
negative.adj 0.03346851 0.021044082 0.088074208 0.13772620 0.031742664
controlled -0.14589792 0.044244525 0.008637477 0.02123529 -0.001663236
Dim.31 Dim.32 Dim.33 Dim.34 Dim.35
symptom 0.088197095 -0.06477663 -0.046744870 0.035155235 0.087983120
effort -0.094201160 -0.08474527 -0.008866516 -0.125046393 0.009056018
impact -0.037630574 -0.10334394 -0.138939974 0.364669242 0.119756603
positive.adj -0.113882921 -0.04075667 0.208825342 -0.053102260 -0.133922025
negative.adj 0.017501926 -0.15477346 0.109649674 0.003506023 -0.058137141
controlled 0.009078176 -0.06371984 0.089683352 -0.061209532 -0.097114500
Dim.36 Dim.37 Dim.38 Dim.39 Dim.40
symptom -0.10794183 -0.034182427 -0.073950385 0.008698401 0.02543223
effort -0.05251280 0.008833194 0.039547828 -0.123475299 0.02858925
impact -0.15250801 0.013332257 -0.134485773 0.124245077 0.02203870
positive.adj -0.10874268 0.073623406 -0.046451989 0.106645564 0.09844162
negative.adj 0.08684876 -0.068247129 0.048608472 -0.014919302 0.16100749
controlled 0.20514857 -0.103769638 0.009192439 -0.014901536 0.02329415
Dim.41 Dim.42 Dim.43 Dim.44 Dim.45
symptom 0.01091711 -0.004001565 -0.05497024 -0.033582137 0.045153763
effort -0.09005823 0.061664892 -0.08217083 0.015391529 0.139486442
impact 0.22901474 -0.077952835 -0.05963439 0.265294205 -0.110378811
positive.adj -0.10956899 -0.095588589 -0.11200988 0.039106545 0.004484338
negative.adj -0.06697619 -0.014317548 0.16341095 -0.009493154 0.107051491
controlled -0.01238166 0.055446842 -0.01133274 0.040639598 0.049913728
Dim.46 Dim.47 Dim.48 Dim.49 Dim.50
symptom 0.006969656 -0.01902312 -0.02333354 -0.01532109 -0.0003572222
effort 0.019578407 -0.04255886 -0.16059528 -0.08155656 0.1222812193
impact -0.136464134 0.10024800 0.08650248 -0.05739115 0.0486839275
positive.adj -0.057263469 -0.02736933 -0.21770520 0.03240192 -0.0392305352
negative.adj -0.061525047 -0.11800944 0.02615026 -0.09714277 0.0019077989
controlled -0.010978259 0.05891339 -0.03283221 -0.06824986 -0.1231076133
Dim.51 Dim.52 Dim.53 Dim.54 Dim.55
symptom -0.07147207 0.179112090 0.06876905 0.03203289 -0.02259866
effort -0.06109134 0.124515975 -0.05882143 -0.01650253 -0.01625784
impact 0.07741556 -0.023656000 -0.01417747 -0.02915799 0.02452023
positive.adj -0.11755584 0.161644835 0.09194351 0.10459818 0.01873456
negative.adj 0.01896467 0.004232823 0.02988965 -0.05228291 -0.10877022
controlled 0.06737516 -0.101602558 0.04339292 0.03998399 0.13977948
Dim.56 Dim.57 Dim.58 Dim.59 Dim.60
symptom -0.03132057 -0.07648518 -0.0755692231 0.150550284 0.027676867
effort 0.02510641 -0.11834179 0.0189387578 -0.084093940 0.058866872
impact 0.05677482 0.11478407 0.0117110525 -0.023407662 -0.111464143
positive.adj -0.03132602 0.09414849 0.0150382120 -0.049331496 0.016608003
negative.adj -0.10052440 0.11253198 0.1115302601 0.023312237 -0.010689356
controlled 0.03205007 0.05510990 0.0006019581 -0.004004517 0.002037644
Dim.61 Dim.62 Dim.63 Dim.64 Dim.65
symptom -0.04500644 -0.04942980 -0.04247987 -0.11235241 0.091786236
effort -0.04562068 -0.04331692 0.01771433 0.04150901 0.005231124
impact 0.07673066 -0.02978424 -0.04269006 -0.05361393 -0.027808443
positive.adj 0.05870110 -0.04910514 0.12579247 0.03820046 -0.105783461
negative.adj 0.01735582 0.07621992 -0.06764822 0.14721623 -0.022229883
controlled -0.02362182 -0.10798719 0.09577691 0.03719767 -0.033463041
Dim.66 Dim.67 Dim.68 Dim.69 Dim.70
symptom 0.07577481 0.09393880 -0.009637830 -0.189593159 0.024483328
effort -0.05681098 -0.03348702 0.054094909 -0.002878758 0.026128058
impact -0.09329957 0.06839856 0.102791966 -0.064110510 -0.005583458
positive.adj 0.02519025 -0.11646539 0.043059179 -0.103692767 0.029144232
negative.adj 0.01563726 0.13617930 -0.013350075 -0.109931362 -0.058386692
controlled -0.05691295 -0.01634409 0.004853922 0.048538121 -0.022664928
Dim.71 Dim.72 Dim.73 Dim.74 Dim.75
symptom 0.037907343 0.02883982 -0.143624447 -0.000752045 -0.112294975
effort -0.116330327 -0.04358420 -0.055210264 0.105454551 0.038441465
impact -0.038239969 -0.04053784 0.008788743 0.002447861 0.042169219
positive.adj -0.029682982 -0.02988502 0.028065765 -0.018084753 -0.003011684
negative.adj -0.008555313 -0.02917264 -0.091284156 0.034732849 0.016170349
controlled -0.001224768 0.01820773 -0.046611336 0.001696754 0.043718925
Dim.76 Dim.77 Dim.78 Dim.79 Dim.80
symptom 0.03643199 0.0080425762 0.011278280 0.033429437 -0.031686083
effort 0.02035375 -0.0397406571 -0.056852362 -0.053213896 -0.024286531
impact -0.06020405 -0.0292436123 -0.005624226 -0.001558525 -0.007281740
positive.adj -0.01106365 0.0006102466 -0.011017498 0.027148586 0.006374897
negative.adj -0.06053864 0.0825555724 -0.020694523 0.019943194 -0.037304308
controlled -0.07256713 -0.0302828761 0.043141126 -0.010418655 -0.125036936
Dim.81 Dim.82 Dim.83 Dim.84 Dim.85
symptom 0.011028464 0.021752267 0.009562487 -0.015716088 0.008366435
effort 0.002943488 -0.004892684 0.005883115 0.018771986 0.015652536
impact 0.015053866 -0.001011287 0.008296702 0.017823961 0.016315116
positive.adj 0.014561184 0.002233016 -0.002028628 0.015117689 0.002950444
negative.adj -0.004352636 -0.046816171 -0.008206005 0.004865248 0.010046634
controlled -0.015127041 0.055681790 0.067452592 -0.044705195 0.019063879
Dim.86 Dim.87 Dim.88 Dim.89 Dim.90
symptom 0.0005209647 0.004434445 -0.0067699002 -0.0011890725 -0.002992696
effort 0.0072686880 -0.004303070 -0.0012042721 0.0005525816 0.001828320
impact 0.0027578313 -0.001421599 -0.0043551229 -0.0005509361 0.002683848
positive.adj 0.0044726307 -0.005998010 0.0008680002 0.0006431230 -0.002895717
negative.adj 0.0007999350 -0.003290348 -0.0050261898 0.0024535099 -0.002218942
controlled 0.0087463373 0.002634565 -0.0066626884 -0.0049872440 0.001159278
Dim.91 Dim.92 Dim.93 Dim.94
symptom -0.0024239499 6.045381e-04 0.0013835464 0.0026178417
effort -0.0007537617 -6.080341e-04 -0.0013711938 -0.0050996075
impact -0.0030951983 -2.877400e-03 0.0014048249 -0.0035080189
positive.adj -0.0017724914 2.784810e-04 -0.0007219726 -0.0001249445
negative.adj -0.0009255146 5.323516e-05 0.0009180388 -0.0022978837
controlled 0.0001587212 -2.467065e-03 0.0012349327 -0.0003329511
Dim.95 Dim.96 Dim.97 Dim.98
symptom 0.0003310313 -0.0012480428 -0.0002483704 -1.003494e-05
effort 0.0004047550 0.0001799279 0.0001487770 3.953200e-05
impact -0.0001042192 -0.0010481643 0.0004964307 7.265957e-06
positive.adj 0.0005753315 -0.0022196551 -0.0009307020 1.918453e-05
negative.adj 0.0021868340 -0.0008901426 0.0004933362 6.454179e-06
controlled 0.0006857743 -0.0007637883 -0.0001919997 1.683500e-05
Dim.99 Dim.100
symptom -5.585917e-08 -9.889080e-09
effort 5.039901e-08 1.593870e-07
impact -2.421768e-08 -1.668795e-08
positive.adj 3.569308e-08 4.040396e-08
negative.adj -3.663423e-08 2.364260e-08
controlled -1.188663e-07 -6.322823e-08
head(result.var$contrib) # Contributions to the PCs Dim.1 Dim.2 Dim.3 Dim.4 Dim.5
symptom 0.28251317 3.508295543 0.0006382971 2.65833925 1.3833814
effort 0.01104460 0.376926355 0.0003726323 0.19849829 0.1177771
impact 0.02192097 1.343414044 0.0244749790 0.06849886 0.7975002
positive.adj 0.02149340 0.007303523 0.5479138990 0.07144334 2.4983454
negative.adj 0.02059178 1.537797723 0.1283680011 2.06430559 0.1613851
controlled 0.03046429 0.450649886 0.0378487219 0.91470055 1.3040128
Dim.6 Dim.7 Dim.8 Dim.9 Dim.10 Dim.11
symptom 0.69527361 0.05090507 1.4345548 2.33328000 8.472168e-05 0.11920949
effort 0.02380352 0.18855386 0.4048392 7.38308733 7.511803e-02 2.53908104
impact 0.58792564 0.03882997 0.3162842 5.94905607 9.507805e-01 0.16761355
positive.adj 0.00495777 0.13361285 1.8368776 0.00506297 8.434886e-02 1.23755683
negative.adj 2.82048261 0.04031929 1.0956896 0.05026172 2.326865e+00 0.67603434
controlled 0.04381249 2.07530806 7.6872168 2.10979791 1.784140e+00 0.03792748
Dim.12 Dim.13 Dim.14 Dim.15 Dim.16 Dim.17
symptom 0.365010839 0.01117562 0.75638443 0.40698388 1.61573627 1.50742862
effort 0.011869953 0.35804399 0.38187468 0.88894829 0.15941925 0.09686036
impact 0.191930366 0.26546488 0.05535551 1.40405436 0.20477131 0.54814725
positive.adj 0.545495409 0.86705609 0.13686290 0.07859098 0.21320074 0.25692471
negative.adj 0.106287761 0.04209214 0.02460384 0.06756086 9.05626855 0.63750252
controlled 0.003803565 1.01513412 2.92829580 0.04362288 0.02153355 0.01596580
Dim.18 Dim.19 Dim.20 Dim.21 Dim.22 Dim.23
symptom 0.004641146 0.1206434818 0.74170908 0.2019133 5.1446146 0.6069477
effort 5.042991933 0.0003429528 1.30926097 0.2159777 1.0912779 0.8970355
impact 2.301960542 0.3304282432 0.00532126 0.3115508 0.8315352 0.2481237
positive.adj 0.107127478 3.3707962866 1.19987680 1.4482540 3.8110216 2.3559969
negative.adj 0.588166430 1.0287755992 0.51649808 1.6222220 1.1102657 1.7289390
controlled 0.180763025 1.1510274383 3.74887095 0.0477992 3.8463741 0.2378621
Dim.24 Dim.25 Dim.26 Dim.27 Dim.28 Dim.29
symptom 0.5483810 1.26113630 0.8423494 0.007780535 0.405161376 0.49878745
effort 7.0142261 0.18079594 2.1798916 8.638837484 1.431112795 1.52112252
impact 0.8401394 0.03520603 0.4755230 0.606360551 0.814079446 0.08645750
positive.adj 0.1391822 3.53441329 0.3679659 11.729506279 0.223758334 0.55075315
negative.adj 0.4173794 0.05527461 0.0866982 0.036655229 0.660722655 1.65402655
controlled 1.0615995 0.12719344 1.6475385 0.162029851 0.006354707 0.03932112
Dim.30 Dim.31 Dim.32 Dim.33 Dim.34
symptom 0.2999282018 0.736032688 0.4091039 0.216553603 0.12918430
effort 0.1287425223 0.839655315 0.7002092 0.007791184 1.63445156
impact 0.3381071527 0.133989332 1.0412787 1.913162780 13.90042213
positive.adj 0.0092976486 1.227172859 0.1619550 4.321792367 0.29475130
negative.adj 0.0919129351 0.028984128 2.3355540 1.191549762 0.00128487
controlled 0.0002523466 0.007798044 0.3958642 0.797115333 0.39162269
Dim.35 Dim.36 Dim.37 Dim.38 Dim.39
symptom 0.825127594 1.2756419 0.132670288 0.633902151 0.009205198
effort 0.008741721 0.3019114 0.008859383 0.181295324 1.854874121
impact 1.528696715 2.5464464 0.020182514 2.096495952 1.878073772
positive.adj 1.911729360 1.2946408 0.615459819 0.250120898 1.383693360
negative.adj 0.360271364 0.8258031 0.528855063 0.273883126 0.027080162
controlled 1.005288130 4.6077212 1.222667846 0.009794964 0.027015707
Dim.40 Dim.41 Dim.42 Dim.43 Dim.44
symptom 0.07991695 0.01551839 0.002153971 0.41158404 0.15707837
effort 0.10098933 1.05603436 0.511512431 0.91968404 0.03299615
impact 0.06001254 6.82901597 0.817417765 0.48439183 9.80291562
positive.adj 1.19736760 1.56317131 1.229114503 1.70889706 0.21300934
negative.adj 3.20303810 0.58407976 0.027575129 3.63718631 0.01255223
controlled 0.06704461 0.01996132 0.413555574 0.01749335 0.23003745
Dim.45 Dim.46 Dim.47 Dim.48 Dim.49
symptom 0.299111983 0.007263968 0.05678446 0.08602416 0.03929111
effort 2.854367649 0.057319964 0.28421427 4.07497442 1.11335331
impact 1.787381823 2.784763127 1.57694810 1.18226923 0.55132165
positive.adj 0.002950139 0.490350386 0.11754231 7.48853609 0.17573435
negative.adj 1.681246360 0.566050432 2.18524164 0.10804665 1.57956010
controlled 0.365498782 0.018022631 0.54462052 0.17031765 0.77968421
Dim.50 Dim.51 Dim.52 Dim.53 Dim.54
symptom 2.216073e-05 0.95364870 6.09093944 0.91006047 0.20259212
effort 2.596730e+00 0.69674694 2.94363951 0.66581775 0.05376895
impact 4.116029e-01 1.11885102 0.10624711 0.03867959 0.16785937
positive.adj 2.672734e-01 2.57990765 4.96087339 1.62677075 2.16012338
negative.adj 6.320805e-04 0.06714387 0.00340169 0.17191978 0.53969672
controlled 2.631946e+00 0.84745249 1.95994445 0.36234468 0.31564766
Dim.55 Dim.56 Dim.57 Dim.58 Dim.59 Dim.60
symptom 0.10598663 0.2055101 1.2742016 1.290252e+00 5.386946272 0.18456552
effort 0.05485441 0.1320514 3.0504214 8.103777e-02 1.680772346 0.83494650
impact 0.12477705 0.6752822 2.8697681 3.098677e-02 0.130225361 2.99355398
positive.adj 0.07284047 0.2055816 1.9306797 5.109482e-02 0.578399693 0.06645866
negative.adj 2.45530631 2.1169797 2.7582617 2.810411e+00 0.129165752 0.02753086
controlled 4.05482902 0.2151947 0.6615197 8.186853e-05 0.003811359 0.00100040
Dim.61 Dim.62 Dim.63 Dim.64 Dim.65 Dim.66
symptom 0.51310284 0.6424497 0.48405180 3.6528870 2.511217504 1.86962512
effort 0.52720388 0.4933744 0.08417342 0.4986045 0.008156788 1.05091914
impact 1.49139600 0.2332569 0.48885378 0.8318149 0.230506168 2.83441916
positive.adj 0.87286617 0.6340381 4.24458025 0.4222880 3.335529239 0.20661880
negative.adj 0.07630371 1.5275605 1.22754735 6.2716651 0.147300272 0.07962075
controlled 0.14134543 3.0662385 2.46063467 0.4004083 0.333779771 1.05469527
Dim.67 Dim.68 Dim.69 Dim.70 Dim.71
symptom 2.9047040 0.033503153 13.772864521 0.2400713 0.6215699548
effort 0.3691177 1.055455665 0.003175333 0.2734095 5.8536833886
impact 1.5399468 3.811057943 1.574843536 0.0124855 0.6325260223
positive.adj 4.4648373 0.668742156 4.119797566 0.3401767 0.3811170618
negative.adj 6.1042739 0.064282775 4.630438941 1.3652969 0.0316603087
controlled 0.0879292 0.008497924 0.902702954 0.2057349 0.0006488596
Dim.72 Dim.73 Dim.74 Dim.75 Dim.76
symptom 0.3779531 10.36433923 0.0003052157 7.537485892 0.80542252
effort 0.8631992 1.53152656 6.0013616273 0.883294923 0.25138924
impact 0.7467483 0.03880953 0.0032336485 1.062911183 2.19942588
positive.adj 0.4058449 0.39576614 0.1764998331 0.005421563 0.07427707
negative.adj 0.3867268 4.18673714 0.6510284382 0.156294849 2.22394099
controlled 0.1506485 1.09161123 0.0015536626 1.142470024 3.19549268
Dim.77 Dim.78 Dim.79 Dim.80 Dim.81
symptom 0.0482209574 0.09524797 0.899490962 0.90897528 0.127491577
effort 1.1773769983 2.42028798 2.279233336 0.53400575 0.009081883
impact 0.6375400828 0.02368621 0.001955089 0.04800487 0.237545864
positive.adj 0.0002776235 0.09089416 0.593243669 0.03679267 0.222251529
negative.adj 5.0808718987 0.32068662 0.320131288 1.25989084 0.019858939
controlled 0.6836592556 1.39364763 0.087369932 14.15440413 0.239860862
Dim.82 Dim.83 Dim.84 Dim.85 Dim.86
symptom 0.542941858 0.124470933 0.4046901 0.13295125 0.0006841412
effort 0.027468734 0.047113011 0.5773699 0.46535128 0.1331806841
impact 0.001173528 0.093699486 0.5205257 0.50558228 0.0191718744
positive.adj 0.005721740 0.005601845 0.3744592 0.01653431 0.0504260873
negative.adj 2.514988931 0.091662098 0.0387832 0.19171349 0.0016130158
controlled 3.557710966 6.193319982 3.2745355 0.69029396 0.1928331193
Dim.87 Dim.88 Dim.89 Dim.90 Dim.91
symptom 0.058915224 0.163363742 0.006377516 0.053399197 0.0354347070
effort 0.055476091 0.005169403 0.001377297 0.019930310 0.0034264918
impact 0.006054845 0.067607046 0.001369107 0.042946269 0.0577774685
positive.adj 0.107786232 0.002685535 0.001865619 0.049994445 0.0189474091
negative.adj 0.032436400 0.090046931 0.027152545 0.029356302 0.0051659251
controlled 0.020795377 0.158230479 0.112190513 0.008012811 0.0001519327
Dim.92 Dim.93 Dim.94 Dim.95 Dim.96
symptom 2.887279e-03 0.017154633 0.0711507005 0.0015650935 0.0313763896
effort 2.920768e-03 0.016849679 0.2700013486 0.0023398428 0.0006521403
impact 6.540965e-02 0.017686356 0.1277662873 0.0001551305 0.0221310880
positive.adj 6.126781e-04 0.004671274 0.0001620789 0.0047275760 0.0992464563
negative.adj 2.238919e-05 0.007552940 0.0548211983 0.0683020564 0.0159611121
controlled 4.808422e-02 0.013667231 0.0011509417 0.0067168338 0.0117514121
Dim.97 Dim.98 Dim.99 Dim.100
symptom 0.0015285675 6.658431e-05 2.673227e-07 1.374668e-08
effort 0.0005484743 1.033332e-03 2.176161e-07 3.571019e-06
impact 0.0061066372 3.490824e-05 5.024721e-08 3.914641e-08
positive.adj 0.0214637795 2.433575e-04 1.091478e-07 2.294743e-07
negative.adj 0.0060307446 2.754383e-05 1.149797e-07 7.857360e-08
controlled 0.0009134524 1.873996e-04 1.210499e-06 5.619636e-07
head(result.var$cos2) #Quality of representation Dim.1 Dim.2 Dim.3 Dim.4 Dim.5
symptom 0.0225207225 0.2057274732 3.410199e-05 0.104202541 0.051093977
effort 0.0008804279 0.0221030713 1.990845e-05 0.007780808 0.004349993
impact 0.0017474443 0.0787781912 1.307613e-03 0.002685043 0.029454969
positive.adj 0.0017133607 0.0004282807 2.927313e-02 0.002800462 0.092274191
negative.adj 0.0016414875 0.0901769067 6.858255e-03 0.080917395 0.005960618
controlled 0.0024284809 0.0264262407 2.022125e-03 0.035854762 0.048162566
Dim.6 Dim.7 Dim.8 Dim.9 Dim.10
symptom 0.0228611491 0.001667513 0.042512808 0.0634407821 2.159388e-06
effort 0.0007826787 0.006176517 0.011997346 0.2007426599 1.914610e-03
impact 0.0193314625 0.001271965 0.009373033 0.1617520268 2.423352e-02
positive.adj 0.0001630154 0.004376797 0.054435582 0.0001376598 2.149886e-03
negative.adj 0.0927397106 0.001320752 0.032470590 0.0013665924 5.930718e-02
controlled 0.0014405894 0.067981506 0.227809474 0.0573644094 4.547421e-02
Dim.11 Dim.12 Dim.13 Dim.14 Dim.15
symptom 0.0028616388 0.0082282990 0.0002416838 0.0152754470 0.0074968266
effort 0.0609509625 0.0002675798 0.0077430541 0.0077120922 0.0163748284
impact 0.0040235846 0.0043266124 0.0057409397 0.0011179239 0.0258633147
positive.adj 0.0297077087 0.0122968932 0.0187509423 0.0027639939 0.0014476813
negative.adj 0.0162282901 0.0023960041 0.0009102839 0.0004968831 0.0012445015
controlled 0.0009104539 0.0000857423 0.0219532755 0.0591379538 0.0008035532
Dim.16 Dim.17 Dim.18 Dim.19 Dim.20
symptom 0.0290766515 0.0267662500 7.877395e-05 2.039435e-03 1.142882e-02
effort 0.0028688952 0.0017198749 8.559446e-02 5.797495e-06 2.017409e-02
impact 0.0036850469 0.0097330288 3.907107e-02 5.585772e-03 8.199404e-05
positive.adj 0.0038367422 0.0045620143 1.818270e-03 5.698211e-02 1.848862e-02
negative.adj 0.1629758335 0.0113196416 9.982921e-03 1.739109e-02 7.958597e-03
controlled 0.0003875159 0.0002834924 3.068082e-03 1.945771e-02 5.776547e-02
Dim.21 Dim.22 Dim.23 Dim.24 Dim.25
symptom 0.0030817435 0.07424967 0.008369947 0.007373706 0.0167667862
effort 0.0032964040 0.01574987 0.012370324 0.094315514 0.0024036790
impact 0.0047551091 0.01200113 0.003421682 0.011296782 0.0004680636
positive.adj 0.0221042791 0.05500258 0.032489736 0.001871488 0.0469899660
negative.adj 0.0247595014 0.01602391 0.023842463 0.005612216 0.0007348751
controlled 0.0007295453 0.05551281 0.003280172 0.014274605 0.0016910347
Dim.26 Dim.27 Dim.28 Dim.29 Dim.30
symptom 0.010883157 9.400122e-05 4.756706e-03 0.0057201334 3.287967e-03
effort 0.028164206 1.043709e-01 1.680166e-02 0.0174443518 1.411342e-03
impact 0.006143758 7.325799e-03 9.557517e-03 0.0009915014 3.706504e-03
positive.adj 0.004754120 1.417111e-01 2.626985e-03 0.0063160801 1.019256e-04
negative.adj 0.001120141 4.428534e-04 7.757066e-03 0.0189685056 1.007597e-03
controlled 0.021286202 1.957578e-03 7.460601e-05 0.0004509377 2.766352e-06
Dim.31 Dim.32 Dim.33 Dim.34 Dim.35
symptom 7.778728e-03 0.004196012 0.0021850829 0.0012358905 7.741029e-03
effort 8.873858e-03 0.007181760 0.0000786151 0.0156366003 8.201146e-05
impact 1.416060e-03 0.010679970 0.0193043164 0.1329836564 1.434164e-02
positive.adj 1.296932e-02 0.001661106 0.0436080234 0.0028198500 1.793511e-02
negative.adj 3.063174e-04 0.023954824 0.0120230510 0.0000122922 3.379927e-03
controlled 8.241327e-05 0.004060218 0.0080431037 0.0037466068 9.431226e-03
Dim.36 Dim.37 Dim.38 Dim.39 Dim.40
symptom 0.011651439 1.168438e-03 5.468660e-03 7.566219e-05 0.0006467984
effort 0.002757594 7.802532e-05 1.564031e-03 1.524615e-02 0.0008173452
impact 0.023258693 1.777491e-04 1.808642e-02 1.543684e-02 0.0004857044
positive.adj 0.011824970 5.420406e-03 2.157787e-03 1.137328e-02 0.0096907531
negative.adj 0.007542707 4.657671e-03 2.362784e-03 2.225856e-04 0.0259234104
controlled 0.042085934 1.076814e-02 8.450093e-05 2.220558e-04 0.0005426176
Dim.41 Dim.42 Dim.43 Dim.44 Dim.45
symptom 0.0001191833 1.601252e-05 0.003021727 1.127760e-03 2.038862e-03
effort 0.0081104843 3.802559e-03 0.006752046 2.368992e-04 1.945647e-02
impact 0.0524477506 6.076645e-03 0.003556260 7.038102e-02 1.218348e-02
positive.adj 0.0120053635 9.137178e-03 0.012546213 1.529322e-03 2.010928e-05
negative.adj 0.0044858102 2.049922e-04 0.026703137 9.011997e-05 1.146002e-02
controlled 0.0001533056 3.074352e-03 0.000128431 1.651577e-03 2.491380e-03
Dim.46 Dim.47 Dim.48 Dim.49 Dim.50
symptom 4.857611e-05 0.0003618791 0.0005444539 0.0002347357 1.276077e-07
effort 3.833140e-04 0.0018112564 0.0257908453 0.0066514725 1.495270e-02
impact 1.862246e-02 0.0100496621 0.0074826783 0.0032937440 2.370125e-03
positive.adj 3.279105e-03 0.0007490801 0.0473955553 0.0010498843 1.539035e-03
negative.adj 3.785331e-03 0.0139262288 0.0006838360 0.0094367174 3.639697e-06
controlled 1.205222e-04 0.0034707878 0.0010779543 0.0046580434 1.515548e-02
Dim.51 Dim.52 Dim.53 Dim.54 Dim.55
symptom 0.0051082566 3.208114e-02 0.0047291823 0.0010261060 0.0005106995
effort 0.0037321522 1.550423e-02 0.0034599607 0.0002723336 0.0002643175
impact 0.0059931693 5.596063e-04 0.0002010007 0.0008501886 0.0006012416
positive.adj 0.0138193762 2.612905e-02 0.0084536090 0.0109407791 0.0003509838
negative.adj 0.0003596588 1.791679e-05 0.0008933912 0.0027335025 0.0118309599
controlled 0.0045394124 1.032308e-02 0.0018829452 0.0015987195 0.0195383034
Dim.56 Dim.57 Dim.58 Dim.59 Dim.60
symptom 0.0009809783 0.005849983 5.710707e-03 2.266539e-02 7.660089e-04
effort 0.0006303317 0.014004780 3.586765e-04 7.071791e-03 3.465309e-03
impact 0.0032233805 0.013175383 1.371488e-04 5.479187e-04 1.242426e-02
positive.adj 0.0009813196 0.008863938 2.261478e-04 2.433596e-03 2.758258e-04
negative.adj 0.0101051547 0.012663447 1.243900e-02 5.434604e-04 1.142623e-04
controlled 0.0010272067 0.003037101 3.623535e-07 1.603616e-05 4.151994e-06
Dim.61 Dim.62 Dim.63 Dim.64 Dim.65
symptom 0.0020255794 0.0024433047 0.0018045393 0.012623063 8.424713e-03
effort 0.0020812462 0.0018763555 0.0003137975 0.001722998 2.736466e-05
impact 0.0058875937 0.0008871009 0.0018224411 0.002874453 7.733095e-04
positive.adj 0.0034458195 0.0024113146 0.0158237444 0.001459275 1.119014e-02
negative.adj 0.0003012246 0.0058094758 0.0045762818 0.021672618 4.941677e-04
controlled 0.0005579903 0.0116612324 0.0091732166 0.001383667 1.119775e-03
Dim.66 Dim.67 Dim.68 Dim.69 Dim.70
symptom 0.0057418220 0.0088244984 9.288776e-05 3.594557e-02 5.994334e-04
effort 0.0032274869 0.0011213805 2.926259e-03 8.287247e-06 6.826754e-04
impact 0.0087048094 0.0046783624 1.056619e-02 4.110158e-03 3.117501e-05
positive.adj 0.0006345488 0.0135641876 1.854093e-03 1.075219e-02 8.493863e-04
negative.adj 0.0002445240 0.0185448004 1.782245e-04 1.208490e-02 3.409006e-03
controlled 0.0032390838 0.0002671291 2.356056e-05 2.355949e-03 5.136989e-04
Dim.71 Dim.72 Dim.73 Dim.74 Dim.75
symptom 1.436967e-03 0.0008317351 0.0206279816 5.655717e-07 1.261016e-02
effort 1.353275e-02 0.0018995823 0.0030481733 1.112066e-02 1.477746e-03
impact 1.462295e-03 0.0016433169 0.0000772420 5.992026e-06 1.778243e-03
positive.adj 8.810794e-04 0.0008931145 0.0007876871 3.270583e-04 9.070238e-06
negative.adj 7.319338e-05 0.0008510426 0.0083327972 1.206371e-03 2.614802e-04
controlled 1.500056e-06 0.0003315216 0.0021726167 2.878973e-06 1.911344e-03
Dim.76 Dim.77 Dim.78 Dim.79 Dim.80
symptom 0.0013272902 6.468303e-05 1.271996e-04 1.117527e-03 1.004008e-03
effort 0.0004142751 1.579320e-03 3.232191e-03 2.831719e-03 5.898356e-04
impact 0.0036245279 8.551889e-04 3.163192e-05 2.429001e-06 5.302374e-05
positive.adj 0.0001224044 3.724009e-07 1.213853e-04 7.370457e-04 4.063932e-05
negative.adj 0.0036649274 6.815423e-03 4.282633e-04 3.977310e-04 1.391611e-03
controlled 0.0052659890 9.170526e-04 1.861157e-03 1.085484e-04 1.563424e-02
Dim.81 Dim.82 Dim.83 Dim.84 Dim.85
symptom 1.216270e-04 4.731611e-04 9.144116e-05 2.469954e-04 6.999724e-05
effort 8.664121e-06 2.393836e-05 3.461104e-05 3.523875e-04 2.450019e-04
impact 2.266189e-04 1.022702e-06 6.883526e-05 3.176936e-04 2.661830e-04
positive.adj 2.120281e-04 4.986362e-06 4.115332e-06 2.285445e-04 8.705118e-06
negative.adj 1.894544e-05 2.191754e-03 6.733852e-05 2.367064e-05 1.009349e-04
controlled 2.288274e-04 3.100462e-03 4.549852e-03 1.998554e-03 3.634315e-04
Dim.86 Dim.87 Dim.88 Dim.89 Dim.90
symptom 2.714042e-07 1.966430e-05 4.583155e-05 1.413893e-06 8.956230e-06
effort 5.283383e-05 1.851641e-05 1.450271e-06 3.053464e-07 3.342755e-06
impact 7.605633e-06 2.020943e-06 1.896710e-05 3.035306e-07 7.203042e-06
positive.adj 2.000443e-05 3.597612e-05 7.534243e-07 4.136073e-07 8.385177e-06
negative.adj 6.398960e-07 1.082639e-05 2.526258e-05 6.019711e-06 4.923703e-06
controlled 7.649842e-05 6.940933e-06 4.439142e-05 2.487260e-05 1.343926e-06
Dim.91 Dim.92 Dim.93 Dim.94 Dim.95
symptom 5.875533e-06 3.654663e-07 1.914201e-06 6.853095e-06 1.095817e-07
effort 5.681567e-07 3.697054e-07 1.880172e-06 2.600600e-05 1.638266e-07
impact 9.580252e-06 8.279432e-06 1.973533e-06 1.230620e-05 1.086163e-08
positive.adj 3.141726e-06 7.755165e-08 5.212444e-07 1.561112e-08 3.310064e-07
negative.adj 8.565773e-07 2.833982e-09 8.427952e-07 5.280270e-06 4.782243e-06
controlled 2.519241e-08 6.086411e-06 1.525059e-06 1.108564e-07 4.702864e-07
Dim.96 Dim.97 Dim.98 Dim.99 Dim.100
symptom 1.557611e-06 6.168786e-08 1.007000e-10 3.120246e-15 9.779390e-17
effort 3.237405e-08 2.213459e-08 1.562779e-09 2.540060e-15 2.540423e-14
impact 1.098648e-06 2.464434e-07 5.279412e-11 5.864958e-16 2.784876e-16
positive.adj 4.926869e-06 8.662062e-07 3.680463e-10 1.273996e-15 1.632480e-15
negative.adj 7.923538e-07 2.433806e-07 4.165642e-11 1.342067e-15 5.589726e-16
controlled 5.833726e-07 3.686388e-08 2.834173e-10 1.412919e-14 3.997809e-15
I will use 33 components bcz the eigen value >1, explaining 76.8% of the variance
ind.coord<-pca$x
ind.coord1<- as.data.frame(ind.coord[,1:33])
toyDf1<-cbind(Df, ind.coord1)using iterative backward elimination, I chose toy13 due to lower AIC
names(toyDf1) [1] "ID" "Employment" "Marriage"
[4] "race" "ethinicity (latio)" "Age"
[7] "Formaleducationyears" "SymptomNo" "Symptom"
[10] "X8wkContr" "BSContr" "WC"
[13] "symptom" "effort" "impact"
[16] "positive.adj" "negative.adj" "controlled"
[19] "uncontrolled" "controlNN" "controlVB"
[22] "Analytic" "Clout" "Authentic"
[25] "Tone" "WPS" "Sixltr"
[28] "Dic" "function." "pronoun"
[31] "ppron" "i" "we"
[34] "you" "shehe" "they"
[37] "ipron" "article" "prep"
[40] "auxverb" "adverb" "conj"
[43] "negate" "verb" "adj"
[46] "compare" "interrog" "number"
[49] "quant" "affect" "posemo"
[52] "negemo" "anx" "anger"
[55] "sad" "social" "family"
[58] "friend" "female" "male"
[61] "cogproc" "insight" "cause"
[64] "discrep" "tentat" "certain"
[67] "differ" "percept" "see"
[70] "hear" "feel" "bio"
[73] "body" "health" "sexual"
[76] "ingest" "drives" "affiliation"
[79] "achieve" "power" "reward"
[82] "risk" "focuspast" "focuspresent"
[85] "focusfuture" "relativ" "motion"
[88] "space" "time" "work"
[91] "leisure" "home" "money"
[94] "relig" "death" "informal"
[97] "swear" "netspeak" "assent"
[100] "nonflu" "filler" "AllPunc"
[103] "Period" "Comma" "Colon"
[106] "SemiC" "QMark" "Exclam"
[109] "Dash" "Apostro" "Parenth"
[112] "OtherP" "PC1" "PC2"
[115] "PC3" "PC4" "PC5"
[118] "PC6" "PC7" "PC8"
[121] "PC9" "PC10" "PC11"
[124] "PC12" "PC13" "PC14"
[127] "PC15" "PC16" "PC17"
[130] "PC18" "PC19" "PC20"
[133] "PC21" "PC22" "PC23"
[136] "PC24" "PC25" "PC26"
[139] "PC27" "PC28" "PC29"
[142] "PC30" "PC31" "PC32"
[145] "PC33"
toy12<-lmer(X8wkContr ~ BSContr+Formaleducationyears+ PC1+PC2+PC3+PC4+PC5+PC6+PC7+PC8+PC9+PC10+PC11+PC12+PC13+PC14+PC15+PC16+PC17+PC18+PC19+
PC20+PC21+PC22+PC23+PC24+PC25+PC26+PC27+PC28+PC29+PC30+PC31+PC32+PC33+(1|ID), toyDf1)
summary(toy12)Linear mixed model fit by REML. t-tests use Satterthwaite's method [
lmerModLmerTest]
Formula: X8wkContr ~ BSContr + Formaleducationyears + PC1 + PC2 + PC3 +
PC4 + PC5 + PC6 + PC7 + PC8 + PC9 + PC10 + PC11 + PC12 +
PC13 + PC14 + PC15 + PC16 + PC17 + PC18 + PC19 + PC20 + PC21 +
PC22 + PC23 + PC24 + PC25 + PC26 + PC27 + PC28 + PC29 + PC30 +
PC31 + PC32 + PC33 + (1 | ID)
Data: toyDf1
REML criterion at convergence: 644.2
Scaled residuals:
Min 1Q Median 3Q Max
-2.49086 -0.43316 -0.00818 0.53555 2.53755
Random effects:
Groups Name Variance Std.Dev.
ID (Intercept) 0.2245 0.4738
Residual 0.1567 0.3959
Number of obs: 300, groups: ID, 114
Fixed effects:
Estimate Std. Error df t value Pr(>|t|)
(Intercept) 1.129e+00 3.023e-01 1.302e+02 3.735 0.000280 ***
BSContr 3.556e-01 4.898e-02 2.451e+02 7.259 5.13e-12 ***
Formaleducationyears 3.757e-02 1.930e-02 1.108e+02 1.947 0.054112 .
PC1 7.435e-04 1.162e-02 2.524e+02 0.064 0.949039
PC2 5.565e-03 1.314e-02 2.537e+02 0.423 0.672376
PC3 -2.452e-02 1.335e-02 2.372e+02 -1.837 0.067488 .
PC4 2.143e-02 1.553e-02 2.425e+02 1.379 0.169015
PC5 1.643e-02 1.562e-02 2.285e+02 1.052 0.293879
PC6 6.595e-03 1.715e-02 2.474e+02 0.385 0.700821
PC7 4.105e-02 1.868e-02 2.639e+02 2.198 0.028847 *
PC8 1.112e-02 1.706e-02 2.168e+02 0.651 0.515493
PC9 -7.390e-03 1.810e-02 2.343e+02 -0.408 0.683436
PC10 2.336e-02 1.972e-02 2.384e+02 1.185 0.237331
PC11 -2.990e-02 2.013e-02 2.442e+02 -1.485 0.138733
PC12 -6.750e-03 1.986e-02 2.257e+02 -0.340 0.734273
PC13 -4.149e-02 2.132e-02 2.414e+02 -1.947 0.052729 .
PC14 1.117e-03 2.026e-02 2.153e+02 0.055 0.956079
PC15 -1.023e-02 2.312e-02 2.391e+02 -0.443 0.658454
PC16 -1.811e-02 2.228e-02 2.200e+02 -0.813 0.417345
PC17 -1.715e-02 2.236e-02 2.246e+02 -0.767 0.443774
PC18 2.411e-02 2.275e-02 2.202e+02 1.060 0.290270
PC19 3.523e-02 2.143e-02 1.992e+02 1.644 0.101684
PC20 1.600e-02 2.345e-02 2.153e+02 0.682 0.495667
PC21 -1.085e-02 2.416e-02 2.142e+02 -0.449 0.653743
PC22 8.964e-02 2.327e-02 2.032e+02 3.852 0.000157 ***
PC23 -6.046e-03 2.419e-02 2.086e+02 -0.250 0.802873
PC24 6.612e-02 2.378e-02 1.937e+02 2.780 0.005974 **
PC25 -4.301e-02 2.659e-02 2.274e+02 -1.618 0.107152
PC26 -4.443e-02 2.669e-02 2.260e+02 -1.664 0.097432 .
PC27 8.786e-03 2.631e-02 2.008e+02 0.334 0.738763
PC28 1.529e-02 2.742e-02 2.209e+02 0.557 0.577816
PC29 -5.706e-03 2.628e-02 2.050e+02 -0.217 0.828328
PC30 8.765e-02 2.752e-02 2.188e+02 3.185 0.001658 **
PC31 -9.237e-03 2.914e-02 2.285e+02 -0.317 0.751527
PC32 2.269e-03 2.885e-02 2.154e+02 0.079 0.937402
PC33 -6.001e-02 2.709e-02 1.957e+02 -2.215 0.027900 *
---
Signif. codes: 0 '***' 0.001 '**' 0.01 '*' 0.05 '.' 0.1 ' ' 1
Correlation matrix not shown by default, as p = 36 > 12.
Use print(x, correlation=TRUE) or
vcov(x) if you need it
toy13<-lmer(X8wkContr ~ BSContr+Formaleducationyears+ PC3+PC7+PC13+PC22+PC24+PC26+PC30+PC33+(1|ID), toyDf1)
summary(toy13)Linear mixed model fit by REML. t-tests use Satterthwaite's method [
lmerModLmerTest]
Formula: X8wkContr ~ BSContr + Formaleducationyears + PC3 + PC7 + PC13 +
PC22 + PC24 + PC26 + PC30 + PC33 + (1 | ID)
Data: toyDf1
REML criterion at convergence: 514.2
Scaled residuals:
Min 1Q Median 3Q Max
-2.55135 -0.48626 -0.00489 0.50587 2.69724
Random effects:
Groups Name Variance Std.Dev.
ID (Intercept) 0.2148 0.4634
Residual 0.1546 0.3931
Number of obs: 300, groups: ID, 114
Fixed effects:
Estimate Std. Error df t value Pr(>|t|)
(Intercept) 1.09827 0.27862 125.43539 3.942 0.000134 ***
BSContr 0.35347 0.04466 274.05592 7.914 6.2e-14 ***
Formaleducationyears 0.03984 0.01784 109.01921 2.234 0.027540 *
PC3 -0.02095 0.01255 240.95340 -1.670 0.096285 .
PC7 0.04615 0.01681 258.02543 2.745 0.006481 **
PC13 -0.04653 0.02053 258.79106 -2.266 0.024283 *
PC22 0.08141 0.02232 218.71358 3.647 0.000332 ***
PC24 0.07200 0.02311 217.49177 3.115 0.002086 **
PC26 -0.04140 0.02575 247.57830 -1.608 0.109115
PC30 0.07900 0.02662 236.15700 2.968 0.003302 **
PC33 -0.06117 0.02626 217.72126 -2.329 0.020761 *
---
Signif. codes: 0 '***' 0.001 '**' 0.01 '*' 0.05 '.' 0.1 ' ' 1
Correlation of Fixed Effects:
(Intr) BSCntr Frmldc PC3 PC7 PC13 PC22 PC24 PC26
BSContr -0.307
Frmldctnyrs -0.915 -0.059
PC3 -0.020 -0.081 0.052
PC7 -0.113 0.071 0.094 0.050
PC13 -0.075 0.233 -0.013 -0.032 -0.018
PC22 0.040 0.001 -0.042 -0.104 -0.036 -0.019
PC24 0.003 -0.108 0.040 0.024 0.028 -0.020 0.000
PC26 -0.082 0.127 0.035 -0.007 0.020 0.069 -0.015 -0.100
PC30 0.064 -0.078 -0.036 -0.043 0.001 -0.029 -0.003 0.016 -0.073
PC33 0.013 -0.038 0.003 0.020 0.005 -0.037 0.001 -0.008 -0.064
PC30
BSContr
Frmldctnyrs
PC3
PC7
PC13
PC22
PC24
PC26
PC30
PC33 0.005
anova(toy12, toy13)refitting model(s) with ML (instead of REML)
Data: toyDf1
Models:
toy13: X8wkContr ~ BSContr + Formaleducationyears + PC3 + PC7 + PC13 + PC22 + PC24 + PC26 + PC30 + PC33 + (1 | ID)
toy12: X8wkContr ~ BSContr + Formaleducationyears + PC1 + PC2 + PC3 + PC4 + PC5 + PC6 + PC7 + PC8 + PC9 + PC10 + PC11 + PC12 + PC13 + PC14 + PC15 + PC16 + PC17 + PC18 + PC19 + PC20 + PC21 + PC22 + PC23 + PC24 + PC25 + PC26 + PC27 + PC28 + PC29 + PC30 + PC31 + PC32 + PC33 + (1 | ID)
npar AIC BIC logLik deviance Chisq Df Pr(>Chisq)
toy13 13 478.07 526.22 -226.04 452.07
toy12 38 507.45 648.20 -215.73 431.45 20.621 25 0.7135
toy13.1<-lmer(X8wkContr ~ BSContr+Formaleducationyears+ PC3+PC7+PC13+PC22+PC24+PC30+PC33+(1|ID), toyDf1)
summary(toy13.1)Linear mixed model fit by REML. t-tests use Satterthwaite's method [
lmerModLmerTest]
Formula: X8wkContr ~ BSContr + Formaleducationyears + PC3 + PC7 + PC13 +
PC22 + PC24 + PC30 + PC33 + (1 | ID)
Data: toyDf1
REML criterion at convergence: 511.3
Scaled residuals:
Min 1Q Median 3Q Max
-2.69153 -0.53188 0.00384 0.51652 2.70775
Random effects:
Groups Name Variance Std.Dev.
ID (Intercept) 0.2188 0.4678
Residual 0.1545 0.3931
Number of obs: 300, groups: ID, 114
Fixed effects:
Estimate Std. Error df t value Pr(>|t|)
(Intercept) 1.06209 0.27955 124.53658 3.799 0.000226 ***
BSContr 0.36221 0.04438 272.96748 8.161 1.23e-14 ***
Formaleducationyears 0.04088 0.01795 108.83799 2.277 0.024762 *
PC3 -0.02116 0.01256 241.03080 -1.684 0.093437 .
PC7 0.04661 0.01683 258.11721 2.769 0.006033 **
PC13 -0.04425 0.02052 257.26292 -2.157 0.031941 *
PC22 0.08091 0.02234 218.92367 3.622 0.000363 ***
PC24 0.06831 0.02301 215.13230 2.968 0.003335 **
PC30 0.07623 0.02657 234.69473 2.869 0.004499 **
PC33 -0.06399 0.02623 216.43870 -2.440 0.015500 *
---
Signif. codes: 0 '***' 0.001 '**' 0.01 '*' 0.05 '.' 0.1 ' ' 1
Correlation of Fixed Effects:
(Intr) BSCntr Frmldc PC3 PC7 PC13 PC22 PC24 PC30
BSContr -0.299
Frmldctnyrs -0.917 -0.063
PC3 -0.020 -0.081 0.053
PC7 -0.112 0.069 0.093 0.051
PC13 -0.069 0.228 -0.016 -0.032 -0.020
PC22 0.039 0.003 -0.041 -0.105 -0.036 -0.018
PC24 -0.005 -0.096 0.044 0.024 0.031 -0.013 -0.001
PC30 0.058 -0.070 -0.033 -0.044 0.002 -0.024 -0.005 0.009
PC33 0.007 -0.031 0.006 0.020 0.006 -0.033 0.000 -0.014 0.001
anova(toy13, toy13.1)refitting model(s) with ML (instead of REML)
Data: toyDf1
Models:
toy13.1: X8wkContr ~ BSContr + Formaleducationyears + PC3 + PC7 + PC13 + PC22 + PC24 + PC30 + PC33 + (1 | ID)
toy13: X8wkContr ~ BSContr + Formaleducationyears + PC3 + PC7 + PC13 + PC22 + PC24 + PC26 + PC30 + PC33 + (1 | ID)
npar AIC BIC logLik deviance Chisq Df Pr(>Chisq)
toy13.1 12 478.73 523.18 -227.37 454.73
toy13 13 478.07 526.22 -226.04 452.07 2.6592 1 0.103
toy13.2<-lmer(X8wkContr ~ BSContr+Formaleducationyears+PC7+PC13+PC22+PC24+PC30+PC33+(1|ID), toyDf1)
summary(toy13.2)Linear mixed model fit by REML. t-tests use Satterthwaite's method [
lmerModLmerTest]
Formula: X8wkContr ~ BSContr + Formaleducationyears + PC7 + PC13 + PC22 +
PC24 + PC30 + PC33 + (1 | ID)
Data: toyDf1
REML criterion at convergence: 507.2
Scaled residuals:
Min 1Q Median 3Q Max
-2.73529 -0.49929 0.01541 0.50294 2.72165
Random effects:
Groups Name Variance Std.Dev.
ID (Intercept) 0.2158 0.4646
Residual 0.1568 0.3960
Number of obs: 300, groups: ID, 114
Fixed effects:
Estimate Std. Error df t value Pr(>|t|)
(Intercept) 1.05161 0.27869 125.12647 3.773 0.000247 ***
BSContr 0.35681 0.04443 274.17412 8.030 2.89e-14 ***
Formaleducationyears 0.04242 0.01787 109.05356 2.374 0.019321 *
PC7 0.04811 0.01690 259.67103 2.848 0.004756 **
PC13 -0.04533 0.02061 259.43236 -2.200 0.028698 *
PC22 0.07696 0.02235 217.86794 3.444 0.000688 ***
PC24 0.06920 0.02315 217.13030 2.989 0.003120 **
PC30 0.07376 0.02670 236.67164 2.763 0.006178 **
PC33 -0.06299 0.02638 218.22441 -2.388 0.017810 *
---
Signif. codes: 0 '***' 0.001 '**' 0.01 '*' 0.05 '.' 0.1 ' ' 1
Correlation of Fixed Effects:
(Intr) BSCntr Frmldc PC7 PC13 PC22 PC24 PC30
BSContr -0.304
Frmldctnyrs -0.916 -0.060
PC7 -0.112 0.074 0.091
PC13 -0.070 0.225 -0.014 -0.018
PC22 0.037 -0.005 -0.036 -0.030 -0.022
PC24 -0.005 -0.095 0.043 0.029 -0.013 0.001
PC30 0.058 -0.074 -0.031 0.004 -0.025 -0.009 0.010
PC33 0.008 -0.028 0.005 0.005 -0.031 0.002 -0.015 0.001
anova(toy13, toy13.2)refitting model(s) with ML (instead of REML)
Data: toyDf1
Models:
toy13.2: X8wkContr ~ BSContr + Formaleducationyears + PC7 + PC13 + PC22 + PC24 + PC30 + PC33 + (1 | ID)
toy13: X8wkContr ~ BSContr + Formaleducationyears + PC3 + PC7 + PC13 + PC22 + PC24 + PC26 + PC30 + PC33 + (1 | ID)
npar AIC BIC logLik deviance Chisq Df Pr(>Chisq)
toy13.2 11 479.64 520.39 -228.82 457.64
toy13 13 478.07 526.22 -226.04 452.07 5.5709 2 0.0617 .
---
Signif. codes: 0 '***' 0.001 '**' 0.01 '*' 0.05 '.' 0.1 ' ' 1
vif(toy13) Term GVIF Df GVIF^(1/(2*Df))
BSContr BSContr 1.105753 1 1.051548
Formaleducationyears Formaleducationyears 1.020740 1 1.010317
PC3 PC3 1.025309 1 1.012576
PC7 PC7 1.021201 1 1.010545
PC13 PC13 1.062346 1 1.030702
PC22 PC22 1.014000 1 1.006976
PC24 PC24 1.022894 1 1.011382
PC26 PC26 1.036182 1 1.017930
PC30 PC30 1.014454 1 1.007201
PC33 PC33 1.006390 1 1.003190
GVIF^(1/Df)
BSContr 1.105753
Formaleducationyears 1.020740
PC3 1.025309
PC7 1.021201
PC13 1.062346
PC22 1.014000
PC24 1.022894
PC26 1.036182
PC30 1.014454
PC33 1.006390
# component 3 is immerse writing of symptom ----
p3<-sort(pca$rotation[,3], decreasing = TRUE) %>% as.data.frame()
p3<-cbind(wordCategory = rownames(p3), p3)
rownames(p3) <- 1:nrow(p3)
names(p3)[2]<-"loadings"
View(p3)
ggplot(p3 %>% filter(loadings>0.1))+geom_col(aes(x=wordCategory, y=loadings))+ labs(y="Percent Contribution to PC", x="Word Category") + ggtitle("PC3:Immersive Writing")
# component 7 is confidence and motivation -----
p7<-sort(pca$rotation[,7], decreasing = TRUE) %>% as.data.frame()
p7<-cbind(wordCategory = rownames(p7), p7)
rownames(p7) <- 1:nrow(p7)
names(p7)[2]<-"loadings"
ggplot(p7 %>% filter(loadings>0.13))+geom_col(aes(x=wordCategory, y=loadings))+labs(y="Percent Contribution to PC", x="Word Category") + ggtitle("PC7:Confidence and motivation")
# component 13 is punctuation. People use more periods usually wrote short sentences and emotionally distant from the topic ------
p13<-sort(pca$rotation[,13], decreasing = TRUE) %>% as.data.frame()
p13<-cbind(wordCategory = rownames(p13), p13)
rownames(p13) <- 1:nrow(p13)
names(p13)[2]<-"loadings"
ggplot(p13 %>% filter(loadings>0.13))+geom_col(aes(x=wordCategory, y=loadings))+labs(y="Percent Contribution to PC", x="Word Category") + ggtitle("PC13:Emotionally distant")
# component 22 is venting (confusion and frustration) about symptom ------
p22<-sort(pca$rotation[,22], decreasing = TRUE) %>% as.data.frame()
p22<-cbind(wordCategory = rownames(p22), p22)
rownames(p22) <- 1:nrow(p22)
names(p22)[2]<-"loadings"
ggplot(p22 %>% filter(loadings>0.13))+geom_col(aes(x=wordCategory, y=loadings))+labs(y="Percent Contribution to PC", x="Word Category") + ggtitle("PC22:Venting(confusion and frustration)")
# component 24 is about efforts they make to fight against cancer, treatment, symptoms. Participants used greater past tense in discussing a disclosed event and greater present tense in discussing an undis- closed event. Verb tense differences could indicate increased psychological distance and a higher degree of resolution for disclosed events compared with undisclosed events ----
p24<-sort(pca$rotation[,24], decreasing = TRUE) %>% as.data.frame()
p24<-cbind(wordCategory = rownames(p24), p24)
rownames(p24) <- 1:nrow(p24)
names(p24)[2]<-"loadings"
ggplot(p24 %>% filter(loadings>0.13))+geom_col(aes(x=wordCategory, y=loadings))+labs(y="Percent Contribution to PC", x="Word Category") + ggtitle("PC24:Effort to fight against cancer")
# component 26 is about pressure from work----
p26<-sort(pca$rotation[,26], decreasing = TRUE) %>% as.data.frame()
p26<-cbind(wordCategory = rownames(p26), p26)
rownames(p26) <- 1:nrow(p26)
names(p26)[2]<-"loadings"
ggplot(p26 %>% filter(loadings>0.13))+geom_col(aes(x=wordCategory, y=loadings))+labs(y="Percent Contribution to PC", x="Word Category") + ggtitle("PC26:Pressure and effort to work")
# component 30 is about perceptive process -----
p30<-sort(pca$rotation[,30], decreasing = TRUE) %>% as.data.frame()
p30<-cbind(wordCategory = rownames(p30), p30)
rownames(p30) <- 1:nrow(p30)
names(p30)[2]<-"loadings"
ggplot(p30 %>% filter(loadings>0.13))+geom_col(aes(x=wordCategory, y=loadings))+labs(y="Percent Contribution to PC", x="Word Category") + ggtitle("PC30:detailed.symptom.descriptionk")
# component 33 is about financial stress -----
p33<-sort(pca$rotation[,33], decreasing = TRUE) %>% as.data.frame()
p33<-cbind(wordCategory = rownames(p33), p33)
rownames(p33) <- 1:nrow(p33)
names(p33)[2]<-"loadings"
ggplot(p33 %>% filter(loadings>0.13))+geom_col(aes(x=wordCategory, y=loadings))+labs(y="Percent Contribution to PC", x="Word Category") + ggtitle("PC33:financial.stress")
summary(toy13)Linear mixed model fit by REML. t-tests use Satterthwaite's method [
lmerModLmerTest]
Formula: X8wkContr ~ BSContr + Formaleducationyears + PC3 + PC7 + PC13 +
PC22 + PC24 + PC26 + PC30 + PC33 + (1 | ID)
Data: toyDf1
REML criterion at convergence: 514.2
Scaled residuals:
Min 1Q Median 3Q Max
-2.55135 -0.48626 -0.00489 0.50587 2.69724
Random effects:
Groups Name Variance Std.Dev.
ID (Intercept) 0.2148 0.4634
Residual 0.1546 0.3931
Number of obs: 300, groups: ID, 114
Fixed effects:
Estimate Std. Error df t value Pr(>|t|)
(Intercept) 1.09827 0.27862 125.43539 3.942 0.000134 ***
BSContr 0.35347 0.04466 274.05592 7.914 6.2e-14 ***
Formaleducationyears 0.03984 0.01784 109.01921 2.234 0.027540 *
PC3 -0.02095 0.01255 240.95340 -1.670 0.096285 .
PC7 0.04615 0.01681 258.02543 2.745 0.006481 **
PC13 -0.04653 0.02053 258.79106 -2.266 0.024283 *
PC22 0.08141 0.02232 218.71358 3.647 0.000332 ***
PC24 0.07200 0.02311 217.49177 3.115 0.002086 **
PC26 -0.04140 0.02575 247.57830 -1.608 0.109115
PC30 0.07900 0.02662 236.15700 2.968 0.003302 **
PC33 -0.06117 0.02626 217.72126 -2.329 0.020761 *
---
Signif. codes: 0 '***' 0.001 '**' 0.01 '*' 0.05 '.' 0.1 ' ' 1
Correlation of Fixed Effects:
(Intr) BSCntr Frmldc PC3 PC7 PC13 PC22 PC24 PC26
BSContr -0.307
Frmldctnyrs -0.915 -0.059
PC3 -0.020 -0.081 0.052
PC7 -0.113 0.071 0.094 0.050
PC13 -0.075 0.233 -0.013 -0.032 -0.018
PC22 0.040 0.001 -0.042 -0.104 -0.036 -0.019
PC24 0.003 -0.108 0.040 0.024 0.028 -0.020 0.000
PC26 -0.082 0.127 0.035 -0.007 0.020 0.069 -0.015 -0.100
PC30 0.064 -0.078 -0.036 -0.043 0.001 -0.029 -0.003 0.016 -0.073
PC33 0.013 -0.038 0.003 0.020 0.005 -0.037 0.001 -0.008 -0.064
PC30
BSContr
Frmldctnyrs
PC3
PC7
PC13
PC22
PC24
PC26
PC30
PC33 0.005
AIC(toy13, toy11) df AIC
toy13 13 540.2030
toy11 12 551.1619
BIC(toy13, toy11) df BIC
toy13 13 588.3522
toy11 12 595.6073
summary(toy13)Linear mixed model fit by REML. t-tests use Satterthwaite's method [
lmerModLmerTest]
Formula: X8wkContr ~ BSContr + Formaleducationyears + PC3 + PC7 + PC13 +
PC22 + PC24 + PC26 + PC30 + PC33 + (1 | ID)
Data: toyDf1
REML criterion at convergence: 514.2
Scaled residuals:
Min 1Q Median 3Q Max
-2.55135 -0.48626 -0.00489 0.50587 2.69724
Random effects:
Groups Name Variance Std.Dev.
ID (Intercept) 0.2148 0.4634
Residual 0.1546 0.3931
Number of obs: 300, groups: ID, 114
Fixed effects:
Estimate Std. Error df t value Pr(>|t|)
(Intercept) 1.09827 0.27862 125.43539 3.942 0.000134 ***
BSContr 0.35347 0.04466 274.05592 7.914 6.2e-14 ***
Formaleducationyears 0.03984 0.01784 109.01921 2.234 0.027540 *
PC3 -0.02095 0.01255 240.95340 -1.670 0.096285 .
PC7 0.04615 0.01681 258.02543 2.745 0.006481 **
PC13 -0.04653 0.02053 258.79106 -2.266 0.024283 *
PC22 0.08141 0.02232 218.71358 3.647 0.000332 ***
PC24 0.07200 0.02311 217.49177 3.115 0.002086 **
PC26 -0.04140 0.02575 247.57830 -1.608 0.109115
PC30 0.07900 0.02662 236.15700 2.968 0.003302 **
PC33 -0.06117 0.02626 217.72126 -2.329 0.020761 *
---
Signif. codes: 0 '***' 0.001 '**' 0.01 '*' 0.05 '.' 0.1 ' ' 1
Correlation of Fixed Effects:
(Intr) BSCntr Frmldc PC3 PC7 PC13 PC22 PC24 PC26
BSContr -0.307
Frmldctnyrs -0.915 -0.059
PC3 -0.020 -0.081 0.052
PC7 -0.113 0.071 0.094 0.050
PC13 -0.075 0.233 -0.013 -0.032 -0.018
PC22 0.040 0.001 -0.042 -0.104 -0.036 -0.019
PC24 0.003 -0.108 0.040 0.024 0.028 -0.020 0.000
PC26 -0.082 0.127 0.035 -0.007 0.020 0.069 -0.015 -0.100
PC30 0.064 -0.078 -0.036 -0.043 0.001 -0.029 -0.003 0.016 -0.073
PC33 0.013 -0.038 0.003 0.020 0.005 -0.037 0.001 -0.008 -0.064
PC30
BSContr
Frmldctnyrs
PC3
PC7
PC13
PC22
PC24
PC26
PC30
PC33 0.005
names(toyDf1)[c(115, 119, 125, 134, 136, 138,142, 145)]<-c("immerse.writing.experience", "confidence.and.motivation", "emotionanlly.distant", "venting", "effort.to.fight.cancer", "pressure.and.effort.to.work", "detailed.symptom.description", "financial.stress")
toy13<-lmer(X8wkContr ~ BSContr+Formaleducationyears+`immerse.writing.experience`+ `confidence.and.motivation`+`emotionanlly.distant`+ `venting` + `effort.to.fight.cancer` +`pressure.and.effort.to.work` +`detailed.symptom.description`+ `financial.stress`+ (1|ID), toyDf1)
summary(toy13)Linear mixed model fit by REML. t-tests use Satterthwaite's method [
lmerModLmerTest]
Formula:
X8wkContr ~ BSContr + Formaleducationyears + immerse.writing.experience +
confidence.and.motivation + emotionanlly.distant + venting +
effort.to.fight.cancer + pressure.and.effort.to.work + detailed.symptom.description +
financial.stress + (1 | ID)
Data: toyDf1
REML criterion at convergence: 514.2
Scaled residuals:
Min 1Q Median 3Q Max
-2.55135 -0.48626 -0.00489 0.50587 2.69724
Random effects:
Groups Name Variance Std.Dev.
ID (Intercept) 0.2148 0.4634
Residual 0.1546 0.3931
Number of obs: 300, groups: ID, 114
Fixed effects:
Estimate Std. Error df t value Pr(>|t|)
(Intercept) 1.09827 0.27862 125.43539 3.942 0.000134
BSContr 0.35347 0.04466 274.05592 7.914 6.2e-14
Formaleducationyears 0.03984 0.01784 109.01921 2.234 0.027540
immerse.writing.experience -0.02095 0.01255 240.95340 -1.670 0.096285
confidence.and.motivation 0.04615 0.01681 258.02543 2.745 0.006481
emotionanlly.distant -0.04653 0.02053 258.79106 -2.266 0.024283
venting 0.08141 0.02232 218.71358 3.647 0.000332
effort.to.fight.cancer 0.07200 0.02311 217.49177 3.115 0.002086
pressure.and.effort.to.work -0.04140 0.02575 247.57830 -1.608 0.109115
detailed.symptom.description 0.07900 0.02662 236.15700 2.968 0.003302
financial.stress -0.06117 0.02626 217.72126 -2.329 0.020761
(Intercept) ***
BSContr ***
Formaleducationyears *
immerse.writing.experience .
confidence.and.motivation **
emotionanlly.distant *
venting ***
effort.to.fight.cancer **
pressure.and.effort.to.work
detailed.symptom.description **
financial.stress *
---
Signif. codes: 0 '***' 0.001 '**' 0.01 '*' 0.05 '.' 0.1 ' ' 1
Correlation of Fixed Effects:
(Intr) BSCntr Frmldc immr.. cnfd.. emtnn. ventng eff... pr....
BSContr -0.307
Frmldctnyrs -0.915 -0.059
immrs.wrtn. -0.020 -0.081 0.052
cnfdnc.nd.m -0.113 0.071 0.094 0.050
emtnnlly.ds -0.075 0.233 -0.013 -0.032 -0.018
venting 0.040 0.001 -0.042 -0.104 -0.036 -0.019
effrt.t.fg. 0.003 -0.108 0.040 0.024 0.028 -0.020 0.000
prssr.nd... -0.082 0.127 0.035 -0.007 0.020 0.069 -0.015 -0.100
dtld.sympt. 0.064 -0.078 -0.036 -0.043 0.001 -0.029 -0.003 0.016 -0.073
fnncl.strss 0.013 -0.038 0.003 0.020 0.005 -0.037 0.001 -0.008 -0.064
dtld..
BSContr
Frmldctnyrs
immrs.wrtn.
cnfdnc.nd.m
emtnnlly.ds
venting
effrt.t.fg.
prssr.nd...
dtld.sympt.
fnncl.strss 0.005
vif(toy13) Term GVIF Df
BSContr BSContr 1.105753 1
Formaleducationyears Formaleducationyears 1.020740 1
immerse.writing.experience immerse.writing.experience 1.025309 1
confidence.and.motivation confidence.and.motivation 1.021201 1
emotionanlly.distant emotionanlly.distant 1.062346 1
venting venting 1.014000 1
effort.to.fight.cancer effort.to.fight.cancer 1.022894 1
pressure.and.effort.to.work pressure.and.effort.to.work 1.036182 1
detailed.symptom.description detailed.symptom.description 1.014454 1
financial.stress financial.stress 1.006390 1
GVIF^(1/(2*Df)) GVIF^(1/Df)
BSContr 1.051548 1.105753
Formaleducationyears 1.010317 1.020740
immerse.writing.experience 1.012576 1.025309
confidence.and.motivation 1.010545 1.021201
emotionanlly.distant 1.030702 1.062346
venting 1.006976 1.014000
effort.to.fight.cancer 1.011382 1.022894
pressure.and.effort.to.work 1.017930 1.036182
detailed.symptom.description 1.007201 1.014454
financial.stress 1.003190 1.006390
vif(toy11) Term GVIF Df GVIF^(1/(2*Df))
BSContr BSContr 1.034703 1 1.017204
Formaleducationyears Formaleducationyears 1.027945 1 1.013876
WC WC 1.089354 1 1.043721
symptom symptom 1.115491 1 1.056168
anx anx 1.048245 1 1.023838
feel feel 1.055237 1 1.027248
focuspresent focuspresent 1.051768 1 1.025557
money money 1.061247 1 1.030169
informal informal 1.028593 1 1.014196
GVIF^(1/Df)
BSContr 1.034703
Formaleducationyears 1.027945
WC 1.089354
symptom 1.115491
anx 1.048245
feel 1.055237
focuspresent 1.051768
money 1.061247
informal 1.028593
library(sjPlot)
sjPlot::plot_model(toy11, show.values=TRUE, show.p=TRUE,
title="Effect of linguistic features on controllability")
sjPlot::tab_model(toy11)| X8wkContr | |||
|---|---|---|---|
| Predictors | Estimates | CI | p |
| (Intercept) | 1.31 | 0.69 – 1.93 | <0.001 |
| BSContr | 0.36 | 0.28 – 0.45 | <0.001 |
| Formaleducationyears | 0.03 | -0.00 – 0.06 | 0.081 |
| WC | 0.00 | 0.00 – 0.00 | 0.015 |
| symptom | 0.06 | 0.02 – 0.10 | 0.001 |
| anx | 0.06 | -0.01 – 0.12 | 0.075 |
| feel | -0.04 | -0.07 – -0.01 | 0.017 |
| focuspresent | -0.02 | -0.04 – -0.00 | 0.050 |
| money | -0.18 | -0.37 – 0.01 | 0.057 |
| informal | 0.11 | -0.01 – 0.23 | 0.062 |
| Random Effects | |||
| σ2 | 0.18 | ||
| τ00 ID | 0.17 | ||
| ICC | 0.49 | ||
| N ID | 114 | ||
| Observations | 300 | ||
| Marginal R2 / Conditional R2 | 0.280 / 0.635 | ||
sjPlot::plot_model(toy13, show.values=TRUE, show.p=TRUE,
title="Effect of linguistic features on controllability")
sjPlot::tab_model(toy13)| X8wkContr | |||
|---|---|---|---|
| Predictors | Estimates | CI | p |
| (Intercept) | 1.10 | 0.55 – 1.65 | <0.001 |
| BSContr | 0.35 | 0.27 – 0.44 | <0.001 |
| Formaleducationyears | 0.04 | 0.00 – 0.07 | 0.026 |
|
immerse writing experience |
-0.02 | -0.05 – 0.00 | 0.096 |
| confidence and motivation | 0.05 | 0.01 – 0.08 | 0.006 |
| emotionanlly distant | -0.05 | -0.09 – -0.01 | 0.024 |
| venting | 0.08 | 0.04 – 0.13 | <0.001 |
| effort to fight cancer | 0.07 | 0.03 – 0.12 | 0.002 |
|
pressure and effort to work |
-0.04 | -0.09 – 0.01 | 0.109 |
|
detailed symptom description |
0.08 | 0.03 – 0.13 | 0.003 |
| financial stress | -0.06 | -0.11 – -0.01 | 0.021 |
| Random Effects | |||
| σ2 | 0.15 | ||
| τ00 ID | 0.21 | ||
| ICC | 0.58 | ||
| N ID | 114 | ||
| Observations | 300 | ||
| Marginal R2 / Conditional R2 | 0.278 / 0.698 | ||
#homogeniety
plot(toy13)
#normality assumed
qqnorm(resid(toy13))
toyDf2 <- na.omit(toyDf1)
# Linearity of the predictors are assumed
ggplot(data.frame(x1=toyDf2$immerse.writing.experience,pearson=residuals(toy13,type="pearson")),
aes(x=x1,y=pearson)) +
geom_point() +
theme_bw()+xlab("Symptom experience description")
ggplot(data.frame(x2=toyDf2$confidence.and.motivation,pearson=residuals(toy13,type="pearson")),
aes(x=x2,y=pearson)) +
geom_point() +
theme_bw()+xlab("Confidence and motivation")
ggplot(data.frame(x2=toyDf2$emotionanlly.distant,pearson=residuals(toy13,type="pearson")),
aes(x=x2,y=pearson)) +
geom_point() +
theme_bw()+xlab("Confidence and motivation")
ggplot(data.frame(x2=toyDf2$venting,pearson=residuals(toy13,type="pearson")),
aes(x=x2,y=pearson)) +
geom_point() +
theme_bw()+xlab("Venting")
ggplot(data.frame(x2=toyDf2$emotionanlly.distant,pearson=residuals(toy13,type="pearson")),
aes(x=x2,y=pearson)) +
geom_point() +
theme_bw()+xlab("Emotionally distance")
ggplot(data.frame(x2=toyDf2$effort.to.fight.cancer,pearson=residuals(toy13,type="pearson")),
aes(x=x2,y=pearson)) +
geom_point() +
theme_bw()+xlab("Effort to fight against cancer")
ggplot(data.frame(x2=toyDf2$pressure.and.effort.to.work,pearson=residuals(toy13,type="pearson")),
aes(x=x2,y=pearson)) +
geom_point() +
theme_bw()+xlab("Pressure and effort to work")
ggplot(data.frame(x2=toyDf2$detailed.symptom.description,pearson=residuals(toy13,type="pearson")),
aes(x=x2,y=pearson)) +
geom_point() +
theme_bw()+xlab("Detailed symptom description")
ggplot(data.frame(x2=toyDf2$financial.stress,pearson=residuals(toy13,type="pearson")),
aes(x=x2,y=pearson)) +
geom_point() +
theme_bw()+xlab("Financial stress")
fixef(toy13) (Intercept) BSContr
1.09827360 0.35346600
Formaleducationyears immerse.writing.experience
0.03984447 -0.02095401
confidence.and.motivation emotionanlly.distant
0.04614526 -0.04653064
venting effort.to.fight.cancer
0.08140745 0.07200259
pressure.and.effort.to.work detailed.symptom.description
-0.04140271 0.07900483
financial.stress
-0.06117362
confint.merMod(toy13)Computing profile confidence intervals ...
2.5 % 97.5 %
.sig01 0.382674015 0.544214806
.sigma 0.348648143 0.428320838
(Intercept) 0.558099631 1.641588490
BSContr 0.266721672 0.439996605
Formaleducationyears 0.005190181 0.074647444
immerse.writing.experience -0.045186556 0.003289813
confidence.and.motivation 0.013679839 0.078601007
emotionanlly.distant -0.086143707 -0.006919917
venting 0.038333678 0.124464712
effort.to.fight.cancer 0.027401705 0.116588026
pressure.and.effort.to.work -0.091193602 0.008376619
detailed.symptom.description 0.026933030 0.130934191
financial.stress -0.111851210 -0.010433448
# Coeffiencient with 95% CI band
effects_p3 <- effects::effect(term= "immerse.writing.experience", mod= toy13)
summary(effects_p3) immerse.writing.experience effect
immerse.writing.experience
-22 -15 -8.6 -2.1 4.4
2.934704 2.788026 2.653921 2.517720 2.381519
Lower 95 Percent Confidence Limits
immerse.writing.experience
-22 -15 -8.6 -2.1 4.4
2.382642 2.404942 2.420247 2.407393 2.235587
Upper 95 Percent Confidence Limits
immerse.writing.experience
-22 -15 -8.6 -2.1 4.4
3.486767 3.171111 2.887594 2.628046 2.527451
x_p3<-as.data.frame(effects_p3)
p3_plot <- ggplot() +
geom_point(data=toyDf1, aes(x=immerse.writing.experience, y=X8wkContr))+
geom_point(data=x_p3, aes(x=immerse.writing.experience, y=fit ), color="blue") +
geom_line(data=x_p3, aes(x= immerse.writing.experience, y=fit), color="blue") +
geom_ribbon(data=x_p3, aes(x=immerse.writing.experience, ymin=lower, ymax=upper), alpha= 0.3, fill="blue") + labs(x="Immerse writing of traumatic experience", y="controllability")
p3_plot
effects_p7 <- effects::effect(term= "confidence.and.motivation", mod= toy13)
summary(effects_p7) confidence.and.motivation effect
confidence.and.motivation
-6 -2 1 5 9
2.197737 2.382318 2.520754 2.705335 2.889916
Lower 95 Percent Confidence Limits
confidence.and.motivation
-6 -2 1 5 9
1.976826 2.264721 2.417829 2.513143 2.576355
Upper 95 Percent Confidence Limits
confidence.and.motivation
-6 -2 1 5 9
2.418648 2.499915 2.623678 2.897527 3.203477
x_p7<-as.data.frame(effects_p7)
p7_plot <- ggplot() +
geom_point(data=toyDf1, aes(x=confidence.and.motivation, y=X8wkContr))+
geom_point(data=x_p7, aes(x=confidence.and.motivation, y=fit ), color="blue") +
geom_line(data=x_p7, aes(x= confidence.and.motivation, y=fit), color="blue") +
geom_ribbon(data=x_p7, aes(x=confidence.and.motivation, ymin=lower, ymax=upper), alpha= 0.3, fill="blue") + labs(x="Confidence and motivation", y="controllability")
p7_plot
effects_p13 <- effects::effect(term= "emotionanlly.distant", mod= toy13)
summary(effects_p13) emotionanlly.distant effect
emotionanlly.distant
-6 -3 -0.4 2 5
2.752509 2.612917 2.491938 2.380264 2.240672
Lower 95 Percent Confidence Limits
emotionanlly.distant
-6 -3 -0.4 2 5
2.489553 2.456035 2.392941 2.254842 2.017947
Upper 95 Percent Confidence Limits
emotionanlly.distant
-6 -3 -0.4 2 5
3.015465 2.769800 2.590935 2.505686 2.463398
x_p13<-as.data.frame(effects_p13)
p13_plot <- ggplot() +
geom_point(data=toyDf1, aes(x=emotionanlly.distant, y=X8wkContr))+
geom_point(data=x_p13, aes(x=emotionanlly.distant, y=fit ), color="blue") +
geom_line(data=x_p13, aes(x= emotionanlly.distant, y=fit), color="blue") +
geom_ribbon(data=x_p13, aes(x=emotionanlly.distant, ymin=lower, ymax=upper), alpha= 0.3, fill="blue") + labs(x="Emotionanlly distant", y="controllability")
p13_plot
effects_p22 <- effects::effect(term= "venting", mod= toy13)
summary(effects_p22) venting effect
venting
-5 -2 0.5 3 6
2.066383 2.310606 2.514124 2.717643 2.961865
Lower 95 Percent Confidence Limits
venting
-5 -2 0.5 3 6
1.826752 2.179925 2.414143 2.553197 2.680180
Upper 95 Percent Confidence Limits
venting
-5 -2 0.5 3 6
2.306015 2.441287 2.614106 2.882089 3.243550
x_p22<-as.data.frame(effects_p22)
p22_plot <- ggplot() +
geom_point(data=toyDf1, aes(x=venting, y=X8wkContr))+
geom_point(data=x_p22, aes(x=venting, y=fit ), color="blue") +
geom_line(data=x_p22, aes(x= venting, y=fit), color="blue") +
geom_ribbon(data=x_p22, aes(x=venting, ymin=lower, ymax=upper), alpha= 0.3, fill="blue") + labs(x="Venting", y="controllability")
p22_plot
effects_p24 <- effects::effect(term= "effort.to.fight.cancer", mod= toy13)
summary(effects_p24) effort.to.fight.cancer effect
effort.to.fight.cancer
-8 -5 -2 0.3 3
1.898148 2.114156 2.330164 2.495770 2.690177
Lower 95 Percent Confidence Limits
effort.to.fight.cancer
-8 -5 -2 0.3 3
1.522192 1.867476 2.197452 2.397324 2.521852
Upper 95 Percent Confidence Limits
effort.to.fight.cancer
-8 -5 -2 0.3 3
2.274104 2.360836 2.462875 2.594215 2.858501
x_p24<-as.data.frame(effects_p24)
p24_plot <- ggplot() +
geom_point(data=toyDf1, aes(x=effort.to.fight.cancer, y=X8wkContr))+
geom_point(data=x_p24, aes(x=effort.to.fight.cancer, y=fit ), color="blue") +
geom_line(data=x_p24, aes(x= effort.to.fight.cancer, y=fit), color="blue") +
geom_ribbon(data=x_p24, aes(x=effort.to.fight.cancer, ymin=lower, ymax=upper), alpha= 0.3, fill="blue") + labs(x="Effort to fight cancer", y="controllability")
p24_plot
effects_p26 <- effects::effect(term= "pressure.and.effort.to.work", mod= toy13)
summary(effects_p26) pressure.and.effort.to.work effect
pressure.and.effort.to.work
-4 -2 -0.1 2 4
2.639786 2.556980 2.478315 2.391369 2.308564
Lower 95 Percent Confidence Limits
pressure.and.effort.to.work
-4 -2 -0.1 2 4
2.413377 2.415207 2.380707 2.252030 2.085197
Upper 95 Percent Confidence Limits
pressure.and.effort.to.work
-4 -2 -0.1 2 4
2.866195 2.698754 2.575923 2.530709 2.531930
x_p26<-as.data.frame(effects_p26)
p26_plot <- ggplot() +
geom_point(data=toyDf1, aes(x=pressure.and.effort.to.work, y=X8wkContr))+
geom_point(data=x_p26, aes(x=pressure.and.effort.to.work, y=fit ), color="blue") +
geom_line(data=x_p26, aes(x= pressure.and.effort.to.work, y=fit), color="blue") +
geom_ribbon(data=x_p26, aes(x=pressure.and.effort.to.work, ymin=lower, ymax=upper), alpha= 0.3, fill="blue") + labs(x="Pressure and effort to work", y="controllability")
p26_plot
effects_p30 <- effects::effect(term= "detailed.symptom.description", mod= toy13)
summary(effects_p30) detailed.symptom.description effect
detailed.symptom.description
-3 -1 0.3 2 3
2.236929 2.394938 2.497645 2.631953 2.710958
Lower 95 Percent Confidence Limits
detailed.symptom.description
-3 -1 0.3 2 3
2.052995 2.284890 2.398827 2.488103 2.525135
Upper 95 Percent Confidence Limits
detailed.symptom.description
-3 -1 0.3 2 3
2.420862 2.504987 2.596462 2.775802 2.896781
x_p30<-as.data.frame(effects_p30)
p30_plot <- ggplot() +
geom_point(data=toyDf1, aes(x=detailed.symptom.description, y=X8wkContr))+
geom_point(data=x_p30, aes(x=detailed.symptom.description, y=fit ), color="blue") +
geom_line(data=x_p30, aes(x= detailed.symptom.description, y=fit), color="blue") +
geom_ribbon(data=x_p30, aes(x=detailed.symptom.description, ymin=lower, ymax=upper), alpha= 0.3, fill="blue") + labs(x="detailed.symptom.description", y="controllability")
p30_plot
effects_p33 <- effects::effect(term= "financial.stress", mod= toy13)
summary(effects_p33) financial.stress effect
financial.stress
-4 -2 -0.4 1 3
2.717619 2.595271 2.497394 2.411751 2.289403
Lower 95 Percent Confidence Limits
financial.stress
-4 -2 -0.4 1 3
2.490018 2.454010 2.398059 2.301012 2.105408
Upper 95 Percent Confidence Limits
financial.stress
-4 -2 -0.4 1 3
2.945220 2.736533 2.596728 2.522489 2.473399
x_p33<-as.data.frame(effects_p33)
p33_plot <- ggplot() +
geom_point(data=toyDf1, aes(x=financial.stress, y=X8wkContr))+
geom_point(data=x_p33, aes(x=financial.stress, y=fit ), color="blue") +
geom_line(data=x_p33, aes(x= financial.stress, y=fit), color="blue") +
geom_ribbon(data=x_p33, aes(x=financial.stress, ymin=lower, ymax=upper), alpha= 0.3, fill="blue") + labs(x="Financial stress", y="controllability")
p33_plot
After controlled for participant ID, baseline controllability, and participant education, “symptom experience description” is a marginal negative predictor for controllability at 8 weeks. In the immersive writing of a traumatic event, the more authentic and relevant participants write their symptom experience (cancer), the more likely they relive the event and thus might feel worse and loss of control over the related traumatic event (dealing with multiple ans severe symptoms) they are experiencing now.
After controlled for participant ID, baseline controllability, and participant education, “confidence and motivation” is a significant positive predictor for controllability at 8 weeks. When cancer survivors are confident and motivated fight against cancer, they are more likely to manage the symptoms better.
After controlled for participant ID, baseline controllability, and participant education, “emotionanlly distant” is a significant negative predictor for controllability at 8 weeks. In expressive writing, some participants wrote rich narratives of their experience and emotionally invested while others were simply informative and emotionally distant from the topic. When participants were emotionally distant from the event they are dealing with, they are less likely to deal with the issues.
After controlled for participant ID, baseline controllability, and participant education, “venting” is a significant positive predictor for controllability at 8 weeks. Studies on expressive writing have shown that venting help patients release negative emotions and maintain healthy mental state to manage their symptoms better
After controlled for participant ID, baseline controllability, and participant education, “effort to fight cancer” is a significant positive predictor for controllability at 8 weeks. When participants try hard to fight cancer and survive, they are more motivated to manage their symptoms to have a life with better quality.
After controlled for participant ID, baseline controllability, and participant education, “pressure and effort to work” is a marginally negative predictor for controllability at 8 weeks. When patients talked about their work and job, most of them describe how cancer symptoms affect their employment or work routine and how their job make their symptom worse. On the other hand, some types of works aggravates patients’ symptoms, such as pain, fatigue, abdominal bloating. Here is an example “I have to push myself all day to keep up with my job duties. Shortly after I eat lunch, I usually have that ‘sugar’ drop and want to crawl in a hole and nap.” “I have to lay in bed for a few days before going back to work.”
After controlled for participant ID, baseline controllability, and participant education, “detailed symptom description” is a significant positive predictor for controllability at 8 weeks. Reflecting and describing how symptom affects daily life helps patients think about their symptoms in a different way to create condition for behavior change. Patients might be more aware of the limitations of current belief and behavior.The more participants reflect on their symptoms and how they manage it, the more likely they are willing to try new strategies to better manage their symptoms.
After controlled for participant ID, baseline controllability, and participant education, “financial stress” is a significant negative predictor for controllability at 8 weeks. Many participants talked about how financial struggling affects their emotional and social participationIt is not surprising that general loss of control influence their perceived sense of control over symptoms, such as depression, anxiety.
sessionInfo()R version 4.1.1 (2021-08-10)
Platform: x86_64-apple-darwin17.0 (64-bit)
Running under: macOS Big Sur 10.16
Matrix products: default
BLAS: /Library/Frameworks/R.framework/Versions/4.1/Resources/lib/libRblas.0.dylib
LAPACK: /Library/Frameworks/R.framework/Versions/4.1/Resources/lib/libRlapack.dylib
locale:
[1] en_US.UTF-8/en_US.UTF-8/en_US.UTF-8/C/en_US.UTF-8/en_US.UTF-8
attached base packages:
[1] stats graphics grDevices utils datasets methods base
other attached packages:
[1] sjPlot_2.8.10 factoextra_1.0.7 car_3.0-11 carData_3.0-4
[5] lmerTest_3.1-3 corrgram_1.14 Hmisc_4.6-0 Formula_1.2-4
[9] survival_3.2-11 lattice_0.20-44 pastecs_1.3.21 lme4_1.1-27.1
[13] Matrix_1.3-4 reshape2_1.4.4 ggplot2_3.3.5 dplyr_1.0.7
[17] readxl_1.3.1
loaded via a namespace (and not attached):
[1] minqa_1.2.4 colorspace_2.0-2 ggsignif_0.6.3
[4] ellipsis_0.3.2 rio_0.5.27 sjlabelled_1.1.8
[7] estimability_1.3 htmlTable_2.3.0 parameters_0.15.0
[10] base64enc_0.1-3 rstudioapi_0.13 ggpubr_0.4.0
[13] farver_2.1.0 ggrepel_0.9.1 fansi_0.5.0
[16] mvtnorm_1.1-3 splines_4.1.1 knitr_1.33
[19] effects_4.2-0 sjmisc_2.8.7 nloptr_1.2.2.2
[22] ggeffects_1.1.1 broom_0.7.9 cluster_2.1.2
[25] png_0.1-7 effectsize_0.5 compiler_4.1.1
[28] sjstats_0.18.1 emmeans_1.7.0 backports_1.2.1
[31] assertthat_0.2.1 fastmap_1.1.0 survey_4.1-1
[34] htmltools_0.5.2 tools_4.1.1 coda_0.19-4
[37] gtable_0.3.0 glue_1.4.2 Rcpp_1.0.7
[40] cellranger_1.1.0 vctrs_0.3.8 nlme_3.1-152
[43] insight_0.14.5 xfun_0.25 stringr_1.4.0
[46] openxlsx_4.2.4 lifecycle_1.0.0 rstatix_0.7.0
[49] MASS_7.3-54 scales_1.1.1 hms_1.1.0
[52] RColorBrewer_1.1-2 yaml_2.2.1 curl_4.3.2
[55] gridExtra_2.3 rpart_4.1-15 latticeExtra_0.6-29
[58] stringi_1.7.4 highr_0.9 bayestestR_0.11.0
[61] checkmate_2.0.0 boot_1.3-28 zip_2.2.0
[64] rlang_0.4.11 pkgconfig_2.0.3 evaluate_0.14
[67] purrr_0.3.4 htmlwidgets_1.5.4 labeling_0.4.2
[70] tidyselect_1.1.1 plyr_1.8.6 magrittr_2.0.1
[73] R6_2.5.1 generics_0.1.0 DBI_1.1.1
[76] pillar_1.6.2 haven_2.4.3 foreign_0.8-81
[79] withr_2.4.2 mgcv_1.8-36 datawizard_0.2.1
[82] abind_1.4-5 nnet_7.3-16 tibble_3.1.4
[85] performance_0.8.0 modelr_0.1.8 crayon_1.4.1
[88] utf8_1.2.2 rmarkdown_2.10 jpeg_0.1-9
[91] grid_4.1.1 data.table_1.14.0 forcats_0.5.1
[94] digest_0.6.27 xtable_1.8-4 tidyr_1.1.3
[97] numDeriv_2016.8-1.1 munsell_0.5.0 mitools_2.4