phase1<-read.csv("./data/phase1.csv")Phase 1
Phase 1 Analyses
Tasks to do:
- Demographics
- Composites (Afterlife Beliefs, FTP, SF-36, RCI)
- Add social partner preferences
- Make a correlation matrix of continuous variables
Read in dataset
For the code to run, the working directory should be set to the “Analyses” folder in box: (Box>Time Perception and Goals Study>Phase 1>Analyses)
Make Composites
FTP
#Reverse score items 8, 9, and 10
phase1$FTP_8.r <- 8-phase1$FTP_8
phase1$FTP_9.r <- 8-phase1$FTP_9
phase1$FTP_10.r <- 8-phase1$FTP_10
phase1$FTP_mean <- rowMeans(
as.data.frame(
lapply(phase1[,c("FTP_1", "FTP_2","FTP_3","FTP_4","FTP_5",
"FTP_6", "FTP_7", "FTP_8.r", "FTP_9.r", "FTP_10.r")],
function(x) as.numeric(as.character(x)))
), na.rm = TRUE
)ART
#ART
phase1$ART_mean <- rowMeans(
as.data.frame(
lapply(phase1[,c("ART_1", "ART_2", "ART_3", "ART_4")],
function(x) as.numeric(as.character(x)))
), na.rm = TRUE
)RCI-10
phase1$RCI_mean <- rowMeans(
as.data.frame(
lapply(phase1[,c("RCI.10_1", "RCI.10_2","RCI.10_3","RCI.10_4","RCI.10_5",
"RCI.10_6", "RCI.10_7", "RCI.10_8", "RCI.10_9", "RCI.10_10")],
function(x) as.numeric(as.character(x)))
), na.rm = TRUE
)SF-36
Here’s the scoring guide: SF-36 Scoring Guide
A higher score represents better health
#STEP ONE. Recode items
#Actually, first I will rename them
library(dplyr)
Attaching package: 'dplyr'The following objects are masked from 'package:stats':
filter, lagThe following objects are masked from 'package:base':
intersect, setdiff, setequal, union#the old column names are confusing...
colnames(phase1[c(46:80)]) [1] "SF.36_1" "SF.36_2" "SF_3_1" "SF_3_2" "SF_3_3"
[6] "SF_3_4" "SF_3_5" "SF_3_6" "SF_3_7" "SF_3_8"
[11] "SF_3_9" "SF_3_10" "SF.36_13_1" "SF.36_13_2" "SF.36_13_3"
[16] "SF.36_13_4" "SF.36_17_1" "SF.36_17_2" "SF.36_17_3" "SF.36_20"
[21] "SF.36_21" "SF.36_22" "SF.36_23_1" "SF.36_23_2" "SF.36_23_3"
[26] "SF.36_23_4" "SF.36_23_5" "SF.36_23_6" "SF.36_23_7" "SF.36_23_8"
[31] "SF.36_23_9" "SF.36_32" "SF.36_33_1" "SF.36_33_2" "SF.36_33_3"#So we can rename them SF36_1-36
phase1 <- phase1 %>%
rename_with(
.fn = ~ paste0("SF36_", seq_along(.)),
.cols = 46:80
)
phase1[1,46:80] SF36_1 SF36_2 SF36_3 SF36_4 SF36_5 SF36_6 SF36_7 SF36_8 SF36_9 SF36_10
1 1 3 3 3 3 3 3 3 3 3
SF36_11 SF36_12 SF36_13 SF36_14 SF36_15 SF36_16 SF36_17 SF36_18 SF36_19
1 3 3 2 2 2 2 2 1 1
SF36_20 SF36_21 SF36_22 SF36_23 SF36_24 SF36_25 SF36_26 SF36_27 SF36_28
1 3 2 1 4 NA 5 5 4 4
SF36_29 SF36_30 SF36_31 SF36_32 SF36_33 SF36_34 SF36_35
1 3 4 3 5 5 1 4#NOTE: ITEM 36 IS MISSING
# Step 1 Item numbers 1, 2, 20, 22, 34, 36 get the following scores:
#1 turns to 100, 2 to 75, 3 to 50, 4 to 25, and 5 to 0
#Make a subset with all relevant columns
SF<- phase1[,46:80]
#Recode values according to the scoring guide
SF <- SF |>
mutate(across(c(1, 2, 20, 22, 34), ~ case_when(
. == 1 ~ 100,
. == 2 ~ 75,
. == 3 ~ 50,
. == 4 ~ 25,
. == 5 ~ 0,
TRUE ~ . # Keep original value if it's not 1–5
)))
SF <- SF |>
mutate(across(c(3, 4, 5, 6, 7, 8, 9, 10, 11, 12 ), ~ case_when(
. == 1 ~ 0,
. == 2 ~ 50,
. == 3 ~ 100,
TRUE ~ . # Keep original value if it's not 1–5
)))
SF <- SF |>
mutate(across(c(13, 14, 15, 16, 17, 18, 19 ), ~ case_when(
. == 1 ~ 0,
. == 2 ~ 100,
TRUE ~ . # Keep original value if it's not 1–5
)))
SF <- SF |>
mutate(across(c(21, 23, 26, 27, 30), ~ case_when(
. == 1 ~ 100,
. == 2 ~ 80,
. == 3 ~ 60,
. == 4 ~ 40,
. == 5 ~ 20,
. == 6 ~ 0,
TRUE ~ . # Keep original value if it's not 1–5
)))
SF <- SF |>
mutate(across(c(24, 25, 28, 29, 31), ~ case_when(
. == 1 ~ 0,
. == 2 ~ 20,
. == 3 ~ 40,
. == 4 ~ 60,
. == 5 ~ 80,
. == 6 ~ 100,
TRUE ~ . # Keep original value if it's not 1–5
)))
SF <- SF |>
mutate(across(c(32, 33, 35), ~ case_when(
. == 1 ~ 0,
. == 2 ~ 25,
. == 3 ~ 50,
. == 4 ~ 75,
. == 5 ~ 100,
TRUE ~ . # Keep original value if it's not 1–5
)))
#STEP 2. Average items to form scales
#Physical functioning
SF$phys_fun <- rowMeans(SF[, 3:12], na.rm = TRUE)
phase1$phys_fun<-SF$phys_fun
#Role Limitations due to physical health
SF$rl_ph <- rowMeans(SF[, 13:16], na.rm = TRUE)
phase1$rl_ph<-SF$rl_ph
#Role limitations due to emotional problems
SF$rl_ep <- rowMeans(SF[, 17:19], na.rm = TRUE)
phase1$rl_ep<-SF$rl_ep
#Energy/fatigue
SF$ener <- rowMeans(SF[, c(23, 27, 29, 31)], na.rm = TRUE)
phase1$ener<-SF$ener
#Emotional well-being
SF$emo_wel <- rowMeans(SF[, c(24, 25, 26, 28, 30)], na.rm = TRUE)
phase1$emo_wel<-SF$emo_wel
#Social functioning
SF$soc_fun <- rowMeans(SF[, c(20,32)], na.rm = TRUE)
phase1$soc_fun<-SF$soc_fun
#Pain
SF$pain <- rowMeans(SF[, c(21,22)], na.rm = TRUE)
phase1$pain<-SF$pain
#General health
SF$gen_health <- rowMeans(SF[, c(1, 33, 34, 35, 36)], na.rm = TRUE)
phase1$gen_health<-SF$gen_healthCorrelation Matrix
cor.matrix.vars <- data.frame(
FTP_mean = phase1$FTP_mean ,
RCI_mean = phase1$RCI_mean ,
ART_mean = phase1$ART_mean,
gen_health = phase1$gen_health ,
pain = phase1$pain,
soc_fun = phase1$soc_fun,
emo_wel = phase1$emo_wel,
rl_ep = phase1$rl_ep,
rl_ph = phase1$rl_ph,
phys_fun = phase1$phys_fun
)
cor.matrix.vars <- data.frame(lapply(cor.matrix.vars, function(x) as.numeric(as.character(x))))
cor_matrix <- cor(cor.matrix.vars, use = "pairwise.complete.obs")
print(cor_matrix) FTP_mean RCI_mean ART_mean gen_health pain
FTP_mean 1.00000000 0.33276140 -0.06141829 0.4723321 0.4780242
RCI_mean 0.33276140 1.00000000 -0.02095957 0.3298692 0.2078803
ART_mean -0.06141829 -0.02095957 1.00000000 -0.1814282 -0.1732114
gen_health 0.47233212 0.32986925 -0.18142816 1.0000000 0.6941485
pain 0.47802415 0.20788026 -0.17321140 0.6941485 1.0000000
soc_fun 0.39375802 0.48844551 -0.07437738 0.7039997 0.6661411
emo_wel 0.52066429 0.43462896 0.09552651 0.5803051 0.5380105
rl_ep 0.45768326 0.39362040 0.02125515 0.3669501 0.5913309
rl_ph 0.49352737 0.35000346 -0.43683404 0.7895801 0.8118421
phys_fun 0.34040006 0.28579402 -0.33794013 0.8462054 0.7185091
soc_fun emo_wel rl_ep rl_ph phys_fun
FTP_mean 0.39375802 0.52066429 0.45768326 0.4935274 0.3404001
RCI_mean 0.48844551 0.43462896 0.39362040 0.3500035 0.2857940
ART_mean -0.07437738 0.09552651 0.02125515 -0.4368340 -0.3379401
gen_health 0.70399966 0.58030506 0.36695006 0.7895801 0.8462054
pain 0.66614115 0.53801052 0.59133088 0.8118421 0.7185091
soc_fun 1.00000000 0.80682689 0.68989760 0.7243209 0.7041067
emo_wel 0.80682689 1.00000000 0.64264518 0.5444328 0.4893878
rl_ep 0.68989760 0.64264518 1.00000000 0.5818254 0.3792092
rl_ph 0.72432091 0.54443282 0.58182536 1.0000000 0.8284023
phys_fun 0.70410665 0.48938780 0.37920917 0.8284023 1.0000000library(ggcorrplot)Warning: package 'ggcorrplot' was built under R version 4.4.3Loading required package: ggplot2ggcorrplot::ggcorrplot(cor_matrix, lab = TRUE,
lab_size=3, type = "upper",
colors = c("blue", "white", "red"),
title = "Correlation Matrix", tl.cex = 11)
cor_pmat <- function(mat) {
n <- ncol(mat)
p.mat <- matrix(NA, n, n)
colnames(p.mat) <- rownames(p.mat) <- colnames(mat)
for (i in 1:n) {
for (j in 1:n) {
# Only test if i != j (optional), or else cor.test will throw warning about cor(x,x)
if (i == j) {
p.mat[i, j] <- 0
} else {
# Use complete.obs to handle missing values for each pair
test <- cor.test(mat[,i], mat[,j], use = "pairwise.complete.obs")
p.mat[i, j] <- test$p.value
}
}
}
return(p.mat)
}
p_matrix <- cor_pmat(cor.matrix.vars)
p_matrix FTP_mean RCI_mean ART_mean gen_health pain
FTP_mean 0.000000000 0.10409200 0.77055820 1.712065e-02 1.565369e-02
RCI_mean 0.104091999 0.00000000 0.92078638 1.073211e-01 3.186984e-01
ART_mean 0.770558199 0.92078638 0.00000000 3.854304e-01 4.076682e-01
gen_health 0.017120645 0.10732113 0.38543038 0.000000e+00 1.186044e-04
pain 0.015653692 0.31869841 0.40766823 1.186044e-04 0.000000e+00
soc_fun 0.051475207 0.01323517 0.72383531 8.595598e-05 2.778116e-04
emo_wel 0.007620475 0.02992252 0.64966996 2.357542e-03 5.535786e-03
rl_ep 0.021418958 0.05156409 0.91967280 7.117504e-02 1.851378e-03
rl_ph 0.012172728 0.08631662 0.02900759 2.698555e-06 8.388492e-07
phys_fun 0.095909949 0.16608370 0.09849060 9.868921e-08 5.223265e-05
soc_fun emo_wel rl_ep rl_ph phys_fun
FTP_mean 5.147521e-02 7.620475e-03 0.0214189583 1.217273e-02 9.590995e-02
RCI_mean 1.323517e-02 2.992252e-02 0.0515640888 8.631662e-02 1.660837e-01
ART_mean 7.238353e-01 6.496700e-01 0.9196728048 2.900759e-02 9.849060e-02
gen_health 8.595598e-05 2.357542e-03 0.0711750422 2.698555e-06 9.868921e-08
pain 2.778116e-04 5.535786e-03 0.0018513778 8.388492e-07 5.223265e-05
soc_fun 0.000000e+00 1.105625e-06 0.0001357634 4.242025e-05 8.564995e-05
emo_wel 1.105625e-06 0.000000e+00 0.0005317524 4.897020e-03 1.303260e-02
rl_ep 1.357634e-04 5.317524e-04 0.0000000000 2.281416e-03 6.155413e-02
rl_ph 4.242025e-05 4.897020e-03 0.0022814156 0.000000e+00 3.170937e-07
phys_fun 8.564995e-05 1.303260e-02 0.0615541266 3.170937e-07 0.000000e+00ggcorrplot(cor_matrix,
p.mat = p_matrix,
sig.level = 0.05,
insig = "pch",
pch = 15,
pch.cex=14.23,
pch.col = "white",
lab = TRUE,
lab_size = 3,
type = "upper",
colors = c("blue", "white", "red"),
title = "Correlation Matrix",
tl.cex = 11)