Aggregate data
Ruben Arslan
4/9/2021
Load individual datasets
##############################################################################################
library(tidyverse)## ── Attaching packages ─────────────────────────────────────── tidyverse 1.3.0 ──## ✓ ggplot2 3.3.3 ✓ purrr 0.3.4
## ✓ tibble 3.0.1 ✓ dplyr 1.0.0
## ✓ tidyr 1.1.0 ✓ stringr 1.4.0
## ✓ readr 1.3.1 ✓ forcats 0.5.0## ── Conflicts ────────────────────────────────────────── tidyverse_conflicts() ──
## x dplyr::filter() masks stats::filter()
## x dplyr::lag() masks stats::lag()##alles zu einem Datensatz machen
#alle zusammen gefuegten Rohdaten einlesen
daten1 <- rio::import("Rohdaten_und_einzelne_Datensaetze/Daten_PSK_Stimme_zusammen_roh/daten1.xlsx")
daten2 <- rio::import("Rohdaten_und_einzelne_Datensaetze/Daten_PSK_Stimme_zusammen_roh/daten2.xlsx")
daten3 <- rio::import("Rohdaten_und_einzelne_Datensaetze/Daten_PSK_Stimme_zusammen_roh/daten3.xlsx")
daten4 <- rio::import("Rohdaten_und_einzelne_Datensaetze/Daten_PSK_Stimme_zusammen_roh/daten4.xlsx")
daten5 <- rio::import("Rohdaten_und_einzelne_Datensaetze/Daten_PSK_Stimme_zusammen_roh/daten5.xlsx")
daten6 <- rio::import("Rohdaten_und_einzelne_Datensaetze/Daten_PSK_Stimme_zusammen_roh/daten6.xlsx")
daten7 <- rio::import("Rohdaten_und_einzelne_Datensaetze/Daten_PSK_Stimme_zusammen_roh/daten7.xlsx")
daten8 <- rio::import("Rohdaten_und_einzelne_Datensaetze/Daten_PSK_Stimme_zusammen_roh/daten8.xlsx")
daten9 <- rio::import("Rohdaten_und_einzelne_Datensaetze/Daten_PSK_Stimme_zusammen_roh/daten9.xlsx")
daten10 <- rio::import("Rohdaten_und_einzelne_Datensaetze/Daten_PSK_Stimme_zusammen_roh/daten10_new.xlsx")
daten11 <- rio::import("Rohdaten_und_einzelne_Datensaetze/Daten_PSK_Stimme_zusammen_roh/daten11.xlsx")
data_complete_all_raw <- rbind(daten1, daten2, daten3, daten4, daten5, daten6, daten7, daten8, daten9, daten10, daten11)Export raw
rio::export(data_complete_all_raw, "data_complete_untrans.rds")
rio::export(data_complete_all_raw, "datarelease/data_personality_voices_untransformed.rds")
rio::export(data_complete_all_raw, "datarelease/data_personality_voices_untransformed.sav")
rio::export(data_complete_all_raw, "datarelease/data_personality_voices_untransformed.xlsx")
rio::export(data_complete_all_raw, "datarelease/data_personality_voices_untransformed.csv")Compute POMP (percentage of maximum possible)
Add min/max for each scale
############################################################################################
#Persoenlichkeitsdaten alle auf -2 bis +2 bringen (wie von Ruben empfohlen)
#Formel: (x-5)/10*5 fuer 9er Skala. Sonst immer die 9 austauschen durch Länge der Skala und die erste 5 in der Klammer durch den Skalenmittelpunkt
#Daten1: SOIR 1-5
psych::describe(daten1)## Warning in FUN(newX[, i], ...): no non-missing arguments to min; returning Inf
## Warning in FUN(newX[, i], ...): no non-missing arguments to min; returning Inf
## Warning in FUN(newX[, i], ...): no non-missing arguments to min; returning Inf
## Warning in FUN(newX[, i], ...): no non-missing arguments to min; returning Inf
## Warning in FUN(newX[, i], ...): no non-missing arguments to min; returning Inf
## Warning in FUN(newX[, i], ...): no non-missing arguments to min; returning Inf
## Warning in FUN(newX[, i], ...): no non-missing arguments to min; returning Inf## Warning in FUN(newX[, i], ...): no non-missing arguments to max; returning -Inf
## Warning in FUN(newX[, i], ...): no non-missing arguments to max; returning -Inf
## Warning in FUN(newX[, i], ...): no non-missing arguments to max; returning -Inf
## Warning in FUN(newX[, i], ...): no non-missing arguments to max; returning -Inf
## Warning in FUN(newX[, i], ...): no non-missing arguments to max; returning -Inf
## Warning in FUN(newX[, i], ...): no non-missing arguments to max; returning -Inf
## Warning in FUN(newX[, i], ...): no non-missing arguments to max; returning -Inf## vars n mean sd median trimmed mad min max
## voice_id 1 339 170.00 98.01 170.00 170.00 126.02 1.00 339.00
## ID 2 339 1545.37 170.02 1562.00 1559.52 176.43 1006.00 1798.00
## dataset 3 339 1.00 0.00 1.00 1.00 0.00 1.00 1.00
## sex 4 339 -1.00 0.00 -1.00 -1.00 0.00 -1.00 -1.00
## age 5 339 20.68 3.21 20.00 20.03 1.48 18.00 35.00
## f0 6 339 211.00 22.25 209.75 210.08 20.25 145.50 303.37
## f1 7 339 470.63 56.24 460.44 466.45 48.99 320.83 718.77
## f2 8 339 1678.81 188.45 1670.17 1671.33 190.60 1229.63 2479.51
## f3 9 339 2913.68 127.94 2903.64 2911.40 121.26 2574.70 3245.74
## f4 10 339 3993.69 177.61 4011.59 4001.69 172.20 3492.43 4509.54
## pf 11 0 NaN NA NA NaN NA Inf -Inf
## neuro 12 0 NaN NA NA NaN NA Inf -Inf
## extra 13 0 NaN NA NA NaN NA Inf -Inf
## openn 14 0 NaN NA NA NaN NA Inf -Inf
## agree 15 0 NaN NA NA NaN NA Inf -Inf
## consc 16 0 NaN NA NA NaN NA Inf -Inf
## dominance 17 0 NaN NA NA NaN NA Inf -Inf
## behavior 18 339 2.10 0.92 2.00 2.02 0.99 1.00 5.00
## attitude 19 339 3.32 1.13 3.33 3.38 0.99 1.00 5.00
## desire 20 339 2.72 0.98 2.67 2.70 0.99 1.00 5.00
## soir_full 21 339 2.71 0.79 2.78 2.71 0.82 1.00 4.78
## range skew kurtosis se
## voice_id 338.00 0.00 -1.21 5.32
## ID 792.00 -0.70 0.12 9.23
## dataset 0.00 NaN NaN 0.00
## sex 0.00 NaN NaN 0.00
## age 17.00 2.05 4.35 0.17
## f0 157.88 0.49 0.81 1.21
## f1 397.94 0.82 1.24 3.05
## f2 1249.88 0.49 0.61 10.24
## f3 671.04 0.16 -0.17 6.95
## f4 1017.11 -0.34 -0.20 9.65
## pf -Inf NA NA NA
## neuro -Inf NA NA NA
## extra -Inf NA NA NA
## openn -Inf NA NA NA
## agree -Inf NA NA NA
## consc -Inf NA NA NA
## dominance -Inf NA NA NA
## behavior 4.00 0.58 -0.49 0.05
## attitude 4.00 -0.39 -0.69 0.06
## desire 4.00 0.17 -0.90 0.05
## soir_full 3.78 0.01 -0.58 0.04names(daten1)## [1] "voice_id" "ID" "dataset" "sex" "age" "f0"
## [7] "f1" "f2" "f3" "f4" "pf" "neuro"
## [13] "extra" "openn" "agree" "consc" "dominance" "behavior"
## [19] "attitude" "desire" "soir_full"daten1$soi_minp <- 1
daten1$soi_maxp <- 5
##################################################################
#Daten2: SOIR 1-5
#dominance auch 1-5?
#BFI auch 1-5
psych::describe(daten2)## Warning in FUN(newX[, i], ...): no non-missing arguments to min; returning Inf
## Warning in FUN(newX[, i], ...): no non-missing arguments to max; returning -Inf## vars n mean sd median trimmed mad
## voice_id 1 383 531.00 110.71 531.00 531.00 142.33
## ID 2 383 1645586.39 506570.17 2020902.00 1647804.94 356135.35
## dataset 3 383 2.00 0.00 2.00 2.00 0.00
## sex 4 383 -0.01 1.00 -1.00 -0.01 0.00
## age 5 382 32.76 7.36 32.00 32.32 8.90
## f0 6 383 158.61 51.93 158.15 157.16 72.63
## f1 7 383 488.74 82.27 479.87 481.04 80.00
## f2 8 383 1675.95 142.67 1683.39 1673.36 178.04
## f3 9 383 2785.36 243.38 2749.97 2775.83 317.79
## f4 10 383 3847.34 343.85 3783.67 3823.67 411.55
## pf 11 0 NaN NA NA NaN NA
## neuro 12 382 2.62 0.72 2.58 2.59 0.74
## extra 13 382 3.51 0.57 3.50 3.53 0.62
## openn 14 382 3.83 0.51 3.88 3.84 0.56
## agree 15 382 3.81 0.52 3.83 3.83 0.50
## consc 16 382 3.82 0.61 3.83 3.84 0.62
## dominance 17 381 3.20 0.51 3.25 3.22 0.49
## behavior 18 381 2.67 1.13 2.67 2.64 1.48
## attitude 19 382 3.16 1.13 3.33 3.18 1.48
## desire 20 382 3.07 0.98 3.00 3.04 0.99
## soir_full 21 381 2.97 0.83 3.00 2.96 0.83
## min max range skew kurtosis se
## voice_id 340.00 722.00 382.00 0.00 -1.21 5.66
## ID 1020901.00 2261113.00 1240212.00 -0.02 -1.93 25884.53
## dataset 2.00 2.00 0.00 NaN NaN 0.00
## sex -1.00 1.00 2.00 0.02 -2.00 0.05
## age 18.00 54.00 36.00 0.48 -0.39 0.38
## f0 83.51 265.56 182.04 0.13 -1.61 2.65
## f1 341.17 847.28 506.11 1.03 1.67 4.20
## f2 1348.88 2087.72 738.84 0.14 -0.83 7.29
## f3 2387.15 3460.33 1073.18 0.25 -1.22 12.44
## f4 3305.17 4631.96 1326.78 0.37 -1.19 17.57
## pf Inf -Inf -Inf NA NA NA
## neuro 1.08 4.58 3.50 0.37 -0.34 0.04
## extra 1.83 4.75 2.92 -0.29 -0.33 0.03
## openn 1.92 4.92 3.00 -0.43 0.21 0.03
## agree 2.00 4.92 2.92 -0.40 -0.13 0.03
## consc 1.92 5.00 3.08 -0.40 -0.09 0.03
## dominance 1.58 4.58 3.00 -0.27 0.07 0.03
## behavior 1.00 5.00 4.00 0.13 -1.03 0.06
## attitude 1.00 5.00 4.00 -0.11 -1.05 0.06
## desire 1.00 5.00 4.00 0.17 -0.86 0.05
## soir_full 1.00 5.00 4.00 0.06 -0.65 0.04daten2$soi_minp <- 1
daten2$soi_maxp <- 5
daten2$dominance_minp <- 1
daten2$dominance_maxp <- 5
daten2$big5_minp <- 1
daten2$big5_maxp <- 5
##################################################################
#Daten3: SOIR 1-5
#dominance -3 bis +3?
n_distinct(daten3$dominance)## [1] 37items <- 8
all(round(daten3$dominance*items) == daten3$dominance*items,na.rm=T)## [1] TRUE# 8 items. closest whole digit to min/max is -3/3
#BFI auch 1-5
psych::describe(daten3)## Warning in FUN(newX[, i], ...): no non-missing arguments to min; returning Inf
## Warning in FUN(newX[, i], ...): no non-missing arguments to max; returning -Inf## vars n mean sd median trimmed mad min max
## voice_id 1 285 865.00 82.42 865.00 865.00 105.26 723.00 1007.00
## ID 2 285 260.65 171.17 245.00 250.56 204.60 11.00 643.00
## dataset 3 285 3.00 0.00 3.00 3.00 0.00 3.00 3.00
## sex 4 285 -0.01 1.00 -1.00 -0.01 0.00 -1.00 1.00
## age 5 284 23.73 2.73 23.00 23.59 2.97 19.00 30.00
## f0 6 285 166.12 50.18 170.92 165.63 70.24 84.93 265.02
## f1 7 285 511.82 47.33 507.58 508.05 42.18 409.34 745.47
## f2 8 285 1588.13 145.70 1566.75 1581.00 167.71 1286.97 2022.02
## f3 9 285 2698.48 219.04 2695.78 2700.03 297.24 2216.26 3095.68
## f4 10 285 3805.49 284.59 3810.24 3807.16 378.50 3057.45 4313.86
## pf 11 0 NaN NA NA NaN NA Inf -Inf
## neuro 12 284 20.02 5.17 20.00 19.95 5.93 9.00 33.00
## extra 13 284 28.57 5.69 29.00 28.72 5.93 12.00 39.00
## openn 14 284 38.36 6.18 39.00 38.73 5.93 16.00 50.00
## agree 15 284 26.80 4.40 27.00 26.86 4.45 14.00 39.00
## consc 16 284 31.08 5.60 31.00 31.01 5.93 15.00 45.00
## dominance 17 283 0.67 0.86 0.62 0.67 0.74 -2.62 2.75
## behavior 18 284 2.57 1.38 2.20 2.37 1.19 1.00 8.20
## attitude 19 284 6.19 2.04 6.60 6.38 1.93 1.00 9.00
## desire 20 284 4.33 1.90 4.40 4.28 2.37 1.00 9.00
## soir_full 21 284 4.36 1.37 4.40 4.39 1.38 1.07 8.40
## range skew kurtosis se
## voice_id 284.00 0.00 -1.21 4.88
## ID 632.00 0.38 -0.92 10.14
## dataset 0.00 NaN NaN 0.00
## sex 2.00 0.02 -2.01 0.06
## age 11.00 0.36 -0.92 0.16
## f0 180.09 0.03 -1.59 2.97
## f1 336.13 1.09 2.45 2.80
## f2 735.05 0.41 -0.57 8.63
## f3 879.41 -0.03 -1.38 12.97
## f4 1256.41 -0.04 -1.39 16.86
## pf -Inf NA NA NA
## neuro 24.00 0.15 -0.54 0.31
## extra 27.00 -0.23 -0.34 0.34
## openn 34.00 -0.70 0.86 0.37
## agree 25.00 -0.14 -0.15 0.26
## consc 30.00 0.03 -0.06 0.33
## dominance 5.38 -0.09 0.25 0.05
## behavior 7.20 1.31 1.73 0.08
## attitude 8.00 -0.68 -0.31 0.12
## desire 8.00 0.20 -0.90 0.11
## soir_full 7.33 -0.08 -0.21 0.08daten3$soi_minp <- 1
daten3$soi_maxp <- 9
daten3$dominance_minp <- -3
daten3$dominance_maxp <- 3
daten3$big5_minp <- 1
daten3$big5_maxp <- 5
# turn sum scores into means
daten3$neuro <- daten3$neuro/7
daten3$extra <- daten3$extra/8
daten3$openn <- daten3$openn/10
daten3$agree <- daten3$agree/8
daten3$consc <- daten3$consc/9
psych::describe(daten3)[,c("min", "max")]## Warning in FUN(newX[, i], ...): no non-missing arguments to min; returning Inf
## Warning in FUN(newX[, i], ...): no non-missing arguments to max; returning -Inf## min max
## voice_id 723.00 1007.00
## ID 11.00 643.00
## dataset 3.00 3.00
## sex -1.00 1.00
## age 19.00 30.00
## f0 84.93 265.02
## f1 409.34 745.47
## f2 1286.97 2022.02
## f3 2216.26 3095.68
## f4 3057.45 4313.86
## pf Inf -Inf
## neuro 1.29 4.71
## extra 1.50 4.88
## openn 1.60 5.00
## agree 1.75 4.88
## consc 1.67 5.00
## dominance -2.62 2.75
## behavior 1.00 8.20
## attitude 1.00 9.00
## desire 1.00 9.00
## soir_full 1.07 8.40
## soi_minp 1.00 1.00
## soi_maxp 9.00 9.00
## dominance_minp -3.00 -3.00
## dominance_maxp 3.00 3.00
## big5_minp 1.00 1.00
## big5_maxp 5.00 5.00##################################################################
#Daten4: SOIR 1-5
#BFI auch 1-5
psych::describe(daten4)[,c("min", "max")]## Warning in psych::describe(daten4): NAs introduced by coercion## Warning in FUN(newX[, i], ...): no non-missing arguments to min; returning Inf
## Warning in FUN(newX[, i], ...): no non-missing arguments to min; returning Inf
## Warning in FUN(newX[, i], ...): no non-missing arguments to min; returning Inf## Warning in FUN(newX[, i], ...): no non-missing arguments to max; returning -Inf
## Warning in FUN(newX[, i], ...): no non-missing arguments to max; returning -Inf
## Warning in FUN(newX[, i], ...): no non-missing arguments to max; returning -Inf## min max
## voice_id 1008.00 1272.00
## ID* Inf -Inf
## dataset 4.00 4.00
## sex -1.00 -1.00
## age 18.00 35.00
## f0 155.91 272.93
## f1 305.37 478.59
## f2 1451.97 2005.61
## f3 2553.46 3074.81
## f4 3682.98 4408.68
## pf Inf -Inf
## neuro 1.25 4.75
## extra 1.50 4.88
## openn 1.70 4.90
## agree 1.78 4.89
## consc 1.67 5.00
## dominance Inf -Inf
## behavior 1.00 5.00
## attitude 1.00 5.00
## desire 1.00 5.00
## soir_full 1.00 4.67daten4$soi_minp <- 1
daten4$soi_maxp <- 5
daten4$big5_minp <- 1
daten4$big5_maxp <- 5
##################################################################
#Daten5: SOIR 1-5
psych::describe(daten5)[,c("min", "max")]## Warning in FUN(newX[, i], ...): no non-missing arguments to min; returning Inf## Warning in FUN(newX[, i], ...): no non-missing arguments to min; returning Inf
## Warning in FUN(newX[, i], ...): no non-missing arguments to min; returning Inf
## Warning in FUN(newX[, i], ...): no non-missing arguments to min; returning Inf
## Warning in FUN(newX[, i], ...): no non-missing arguments to min; returning Inf
## Warning in FUN(newX[, i], ...): no non-missing arguments to min; returning Inf
## Warning in FUN(newX[, i], ...): no non-missing arguments to min; returning Inf## Warning in FUN(newX[, i], ...): no non-missing arguments to max; returning -Inf
## Warning in FUN(newX[, i], ...): no non-missing arguments to max; returning -Inf
## Warning in FUN(newX[, i], ...): no non-missing arguments to max; returning -Inf
## Warning in FUN(newX[, i], ...): no non-missing arguments to max; returning -Inf
## Warning in FUN(newX[, i], ...): no non-missing arguments to max; returning -Inf
## Warning in FUN(newX[, i], ...): no non-missing arguments to max; returning -Inf
## Warning in FUN(newX[, i], ...): no non-missing arguments to max; returning -Inf## min max
## voice_id 1273.00 1459.00
## ID 1017.00 2210.00
## dataset 5.00 5.00
## sex -1.00 1.00
## age 18.00 27.00
## f0 82.84 245.95
## f1 354.38 589.63
## f2 1331.52 1969.18
## f3 2276.41 3080.25
## f4 3115.03 4234.61
## pf Inf -Inf
## neuro Inf -Inf
## extra Inf -Inf
## openn Inf -Inf
## agree Inf -Inf
## consc Inf -Inf
## dominance Inf -Inf
## behavior 1.00 6.00
## attitude 1.00 8.67
## desire 1.00 8.67
## soir_full 1.00 6.89names(daten5)## [1] "voice_id" "ID" "dataset" "sex" "age" "f0"
## [7] "f1" "f2" "f3" "f4" "pf" "neuro"
## [13] "extra" "openn" "agree" "consc" "dominance" "behavior"
## [19] "attitude" "desire" "soir_full"daten5$soi_minp <- 1
daten5$soi_maxp <- 9
##################################################################
#Daten6: SOIR 1-5
psych::describe(daten6)[,c("min", "max")]## Warning in FUN(newX[, i], ...): no non-missing arguments to min; returning Inf## Warning in FUN(newX[, i], ...): no non-missing arguments to min; returning Inf
## Warning in FUN(newX[, i], ...): no non-missing arguments to min; returning Inf
## Warning in FUN(newX[, i], ...): no non-missing arguments to min; returning Inf
## Warning in FUN(newX[, i], ...): no non-missing arguments to min; returning Inf
## Warning in FUN(newX[, i], ...): no non-missing arguments to min; returning Inf
## Warning in FUN(newX[, i], ...): no non-missing arguments to min; returning Inf## Warning in FUN(newX[, i], ...): no non-missing arguments to max; returning -Inf
## Warning in FUN(newX[, i], ...): no non-missing arguments to max; returning -Inf
## Warning in FUN(newX[, i], ...): no non-missing arguments to max; returning -Inf
## Warning in FUN(newX[, i], ...): no non-missing arguments to max; returning -Inf
## Warning in FUN(newX[, i], ...): no non-missing arguments to max; returning -Inf
## Warning in FUN(newX[, i], ...): no non-missing arguments to max; returning -Inf
## Warning in FUN(newX[, i], ...): no non-missing arguments to max; returning -Inf## min max
## voice_id 1460.00 1643.00
## ID 1.00 186.00
## dataset 6.00 6.00
## sex 1.00 1.00
## age 18.00 56.00
## f0 86.96 177.71
## f1 295.49 440.08
## f2 1278.61 1681.44
## f3 2209.98 2709.71
## f4 3026.11 3996.97
## pf Inf -Inf
## neuro Inf -Inf
## extra Inf -Inf
## openn Inf -Inf
## agree Inf -Inf
## consc Inf -Inf
## dominance Inf -Inf
## behavior 1.00 9.00
## attitude 1.00 9.00
## desire 1.00 9.00
## soir_full 1.11 8.56names(daten6)## [1] "voice_id" "ID" "dataset" "sex" "age" "f0"
## [7] "f1" "f2" "f3" "f4" "pf" "neuro"
## [13] "extra" "openn" "agree" "consc" "dominance" "behavior"
## [19] "attitude" "desire" "soir_full"daten6$soi_minp <- 1
daten6$soi_maxp <- 9
##################################################################
#Daten7: SOIR 1-9
#BFI auch 1-5
#dominance 1-5
daten7$soi_minp <- 1
daten7$soi_maxp <- 9
daten7$dominance_minp <- 1
daten7$dominance_maxp <- 5
daten7$big5_minp <- 1
daten7$big5_maxp <- 5
psych::describe(daten7)## Warning in FUN(newX[, i], ...): no non-missing arguments to min; returning Inf
## Warning in FUN(newX[, i], ...): no non-missing arguments to max; returning -Inf## vars n mean sd median trimmed mad min max
## voice_id 1 164 1725.50 47.49 1725.50 1725.50 60.79 1644.00 1807.00
## ID 2 164 83.74 48.31 84.50 83.80 62.27 1.00 166.00
## dataset 3 164 7.00 0.00 7.00 7.00 0.00 7.00 7.00
## sex 4 164 1.00 0.00 1.00 1.00 0.00 1.00 1.00
## age 5 164 24.30 3.25 24.00 24.08 2.97 18.00 34.00
## f0 6 164 118.73 15.27 117.30 118.03 14.31 87.46 166.08
## f1 7 164 481.04 34.85 479.66 480.90 32.52 390.69 584.62
## f2 8 164 1396.66 82.01 1394.13 1393.93 76.76 1133.20 1652.04
## f3 9 164 2444.48 94.80 2450.25 2444.33 84.44 2208.07 2777.29
## f4 10 164 3429.84 150.81 3441.89 3430.10 113.10 2968.93 3902.35
## pf 11 0 NaN NA NA NaN NA Inf -Inf
## neuro 12 164 2.67 0.71 2.64 2.65 0.74 1.00 4.57
## extra 13 164 3.51 0.69 3.62 3.53 0.74 1.62 5.00
## openn 14 164 3.76 0.57 3.80 3.76 0.59 2.30 5.00
## agree 15 164 3.44 0.59 3.50 3.45 0.56 1.88 4.62
## consc 16 164 3.26 0.67 3.22 3.27 0.82 1.22 4.89
## dominance 17 164 3.48 0.59 3.60 3.51 0.59 1.60 4.80
## behavior 18 164 3.13 1.91 2.33 2.90 1.48 1.00 9.00
## attitude 19 164 6.43 2.07 6.67 6.63 2.47 1.00 9.00
## desire 20 164 5.17 1.98 5.33 5.19 2.47 1.00 9.00
## soir_full 21 164 4.91 1.48 4.78 4.89 1.32 1.11 8.67
## soi_minp 22 164 1.00 0.00 1.00 1.00 0.00 1.00 1.00
## soi_maxp 23 164 9.00 0.00 9.00 9.00 0.00 9.00 9.00
## dominance_minp 24 164 1.00 0.00 1.00 1.00 0.00 1.00 1.00
## dominance_maxp 25 164 5.00 0.00 5.00 5.00 0.00 5.00 5.00
## big5_minp 26 164 1.00 0.00 1.00 1.00 0.00 1.00 1.00
## big5_maxp 27 164 5.00 0.00 5.00 5.00 0.00 5.00 5.00
## range skew kurtosis se
## voice_id 163.00 0.00 -1.22 3.71
## ID 165.00 -0.01 -1.24 3.77
## dataset 0.00 NaN NaN 0.00
## sex 0.00 NaN NaN 0.00
## age 16.00 0.61 0.09 0.25
## f0 78.62 0.50 0.30 1.19
## f1 193.93 0.01 0.21 2.72
## f2 518.84 0.23 0.50 6.40
## f3 569.22 0.16 0.50 7.40
## f4 933.43 -0.05 0.69 11.78
## pf -Inf NA NA NA
## neuro 3.57 0.26 0.04 0.06
## extra 3.38 -0.26 -0.42 0.05
## openn 2.70 -0.03 -0.60 0.04
## agree 2.75 -0.30 -0.43 0.05
## consc 3.67 -0.14 -0.50 0.05
## dominance 3.20 -0.47 0.03 0.05
## behavior 8.00 0.89 -0.13 0.15
## attitude 8.00 -0.67 -0.21 0.16
## desire 8.00 -0.12 -1.01 0.15
## soir_full 7.56 0.11 -0.16 0.12
## soi_minp 0.00 NaN NaN 0.00
## soi_maxp 0.00 NaN NaN 0.00
## dominance_minp 0.00 NaN NaN 0.00
## dominance_maxp 0.00 NaN NaN 0.00
## big5_minp 0.00 NaN NaN 0.00
## big5_maxp 0.00 NaN NaN 0.00################
##################################################################
#Daten8: SOIR 1-5
#BFI auch 1-5
#dominance -2:2
daten8$soi_minp <- 1
daten8$soi_maxp <- 9
daten8$dominance_minp <- -2
daten8$dominance_maxp <- 2
daten8$big5_minp <- 1
daten8$big5_maxp <- 5
psych::describe(daten8)## Warning in FUN(newX[, i], ...): no non-missing arguments to min; returning Inf
## Warning in FUN(newX[, i], ...): no non-missing arguments to max; returning -Inf## vars n mean sd median trimmed mad min max
## voice_id 1 157 1886.00 45.47 1886.00 1886.00 57.82 1808.00 1964.00
## ID 2 157 191.35 50.81 193.00 191.70 63.75 100.00 279.00
## dataset 3 157 8.00 0.00 8.00 8.00 0.00 8.00 8.00
## sex 4 157 -1.00 0.00 -1.00 -1.00 0.00 -1.00 -1.00
## age 5 157 23.06 3.44 23.00 22.86 2.97 18.00 34.00
## f0 6 157 211.83 17.73 212.60 211.51 17.34 158.41 264.89
## f1 7 157 446.89 69.85 438.42 444.98 72.94 285.37 600.67
## f2 8 157 1706.03 79.48 1713.90 1710.34 72.90 1389.02 1885.53
## f3 9 157 2845.21 138.94 2829.02 2839.73 143.02 2594.67 3255.57
## f4 10 157 4009.92 131.96 4007.99 4012.65 136.17 3629.25 4317.39
## pf 11 0 NaN NA NA NaN NA Inf -Inf
## neuro 12 142 2.89 0.66 2.88 2.89 0.74 1.25 4.50
## extra 13 142 3.57 0.70 3.62 3.61 0.74 1.62 4.88
## openn 14 142 3.71 0.64 3.80 3.75 0.74 2.10 4.90
## agree 15 142 3.68 0.61 3.72 3.71 0.58 2.00 4.89
## consc 16 142 3.46 0.66 3.44 3.46 0.66 2.00 4.78
## dominance 17 157 0.52 0.60 0.62 0.52 0.74 -1.12 1.88
## behavior 18 142 2.32 1.08 2.33 2.25 1.48 1.00 5.00
## attitude 19 142 3.26 1.03 3.33 3.29 0.99 1.00 5.00
## desire 20 142 2.81 0.90 2.67 2.80 0.99 1.00 5.00
## soir_full 21 142 2.80 0.76 2.78 2.78 0.82 1.22 4.89
## soi_minp 22 157 1.00 0.00 1.00 1.00 0.00 1.00 1.00
## soi_maxp 23 157 9.00 0.00 9.00 9.00 0.00 9.00 9.00
## dominance_minp 24 157 -2.00 0.00 -2.00 -2.00 0.00 -2.00 -2.00
## dominance_maxp 25 157 2.00 0.00 2.00 2.00 0.00 2.00 2.00
## big5_minp 26 157 1.00 0.00 1.00 1.00 0.00 1.00 1.00
## big5_maxp 27 157 5.00 0.00 5.00 5.00 0.00 5.00 5.00
## range skew kurtosis se
## voice_id 156.00 0.00 -1.22 3.63
## ID 179.00 -0.06 -1.16 4.06
## dataset 0.00 NaN NaN 0.00
## sex 0.00 NaN NaN 0.00
## age 16.00 0.54 -0.37 0.27
## f0 106.47 0.16 0.47 1.41
## f1 315.30 0.24 -0.65 5.58
## f2 496.51 -0.64 0.93 6.34
## f3 660.90 0.37 -0.45 11.09
## f4 688.13 -0.17 -0.32 10.53
## pf -Inf NA NA NA
## neuro 3.25 0.01 -0.40 0.06
## extra 3.25 -0.51 -0.38 0.06
## openn 2.80 -0.43 -0.43 0.05
## agree 2.89 -0.45 -0.14 0.05
## consc 2.78 -0.05 -0.76 0.06
## dominance 3.00 -0.11 -0.50 0.05
## behavior 4.00 0.39 -1.03 0.09
## attitude 4.00 -0.23 -0.75 0.09
## desire 4.00 0.17 -0.52 0.08
## soir_full 3.67 0.24 -0.49 0.06
## soi_minp 0.00 NaN NaN 0.00
## soi_maxp 0.00 NaN NaN 0.00
## dominance_minp 0.00 NaN NaN 0.00
## dominance_maxp 0.00 NaN NaN 0.00
## big5_minp 0.00 NaN NaN 0.00
## big5_maxp 0.00 NaN NaN 0.00###############
##################################################################
#Daten9:
#BFI auch 1-7
psych::describe(daten9 %>% select(extra, consc, neuro, agree,openn))## vars n mean sd median trimmed mad min max range skew kurtosis
## extra 1 113 3.50 1.24 3.5 3.48 1.48 0.50 6.50 6.0 0.10 -0.47
## consc 2 114 4.01 0.99 4.0 3.98 0.74 1.00 6.00 5.0 0.02 0.12
## neuro 3 114 4.02 1.39 4.0 4.05 1.48 1.00 6.50 5.5 -0.19 -0.66
## agree 4 114 4.27 1.11 4.5 4.31 1.48 1.00 6.50 5.5 -0.31 -0.17
## openn 5 114 4.82 0.93 5.0 4.85 0.99 2.67 6.67 4.0 -0.22 -0.82
## se
## extra 0.12
## consc 0.09
## neuro 0.13
## agree 0.10
## openn 0.09daten9$big5_minp <- 0
daten9$big5_maxp <- 7
##################################################################
#Daten10:
#BFI auch 1-5
psych::describe(daten10)## Warning in FUN(newX[, i], ...): no non-missing arguments to min; returning Inf## Warning in FUN(newX[, i], ...): no non-missing arguments to min; returning Inf
## Warning in FUN(newX[, i], ...): no non-missing arguments to min; returning Inf
## Warning in FUN(newX[, i], ...): no non-missing arguments to min; returning Inf
## Warning in FUN(newX[, i], ...): no non-missing arguments to min; returning Inf
## Warning in FUN(newX[, i], ...): no non-missing arguments to min; returning Inf
## Warning in FUN(newX[, i], ...): no non-missing arguments to min; returning Inf## Warning in FUN(newX[, i], ...): no non-missing arguments to max; returning -Inf
## Warning in FUN(newX[, i], ...): no non-missing arguments to max; returning -Inf
## Warning in FUN(newX[, i], ...): no non-missing arguments to max; returning -Inf
## Warning in FUN(newX[, i], ...): no non-missing arguments to max; returning -Inf
## Warning in FUN(newX[, i], ...): no non-missing arguments to max; returning -Inf
## Warning in FUN(newX[, i], ...): no non-missing arguments to max; returning -Inf
## Warning in FUN(newX[, i], ...): no non-missing arguments to max; returning -Inf## vars n mean sd median trimmed mad min max range
## voice_id 1 88 2141.50 25.55 2141.50 2141.50 32.62 2098.00 2185.00 87.00
## ID 2 0 NaN NA NA NaN NA Inf -Inf -Inf
## dataset 3 88 10.00 0.00 10.00 10.00 0.00 10.00 10.00 0.00
## sex 4 88 1.00 0.00 1.00 1.00 0.00 1.00 1.00 0.00
## age 5 88 24.14 2.74 24.00 23.99 1.48 19.00 31.00 12.00
## f0 6 88 121.51 17.95 118.72 120.62 17.49 91.28 175.82 84.54
## f1 7 88 408.73 85.85 386.51 395.38 40.87 300.05 947.63 647.58
## f2 8 88 1770.63 123.72 1779.76 1770.59 118.65 1435.43 2064.01 628.58
## f3 9 88 2655.71 141.03 2647.06 2651.44 159.63 2363.84 3064.42 700.57
## f4 10 88 3551.65 158.42 3543.07 3532.19 133.58 3330.14 4266.06 935.92
## pf 11 0 NaN NA NA NaN NA Inf -Inf -Inf
## neuro 12 88 2.59 0.75 2.56 2.56 0.65 1.12 4.50 3.38
## extra 13 88 3.44 0.88 3.50 3.48 0.93 1.25 5.00 3.75
## openn 14 88 3.76 0.60 3.85 3.79 0.52 1.40 4.90 3.50
## agree 15 88 3.56 0.61 3.56 3.57 0.66 1.78 4.89 3.11
## consc 16 88 3.32 0.73 3.44 3.33 0.82 1.33 4.78 3.44
## dominance 17 0 NaN NA NA NaN NA Inf -Inf -Inf
## behavior 18 0 NaN NA NA NaN NA Inf -Inf -Inf
## attitude 19 0 NaN NA NA NaN NA Inf -Inf -Inf
## desire 20 0 NaN NA NA NaN NA Inf -Inf -Inf
## soir_full 21 0 NaN NA NA NaN NA Inf -Inf -Inf
## skew kurtosis se
## voice_id 0.00 -1.24 2.72
## ID NA NA NA
## dataset NaN NaN 0.00
## sex NaN NaN 0.00
## age 0.50 -0.08 0.29
## f0 0.54 -0.10 1.91
## f1 3.36 16.37 9.15
## f2 -0.06 -0.24 13.19
## f3 0.27 -0.40 15.03
## f4 1.78 4.78 16.89
## pf NA NA NA
## neuro 0.30 -0.43 0.08
## extra -0.32 -0.65 0.09
## openn -0.79 1.51 0.06
## agree -0.18 0.00 0.07
## consc -0.26 -0.41 0.08
## dominance NA NA NA
## behavior NA NA NA
## attitude NA NA NA
## desire NA NA NA
## soir_full NA NA NAdaten10$big5_minp <- 1
daten10$big5_maxp <- 5
##################################################################
#Daten11: SOIR 1-9
psych::describe(daten11)## Warning in psych::describe(daten11): NAs introduced by coercion## Warning in FUN(newX[, i], ...): no non-missing arguments to min; returning Inf
## Warning in FUN(newX[, i], ...): no non-missing arguments to min; returning Inf
## Warning in FUN(newX[, i], ...): no non-missing arguments to min; returning Inf
## Warning in FUN(newX[, i], ...): no non-missing arguments to min; returning Inf
## Warning in FUN(newX[, i], ...): no non-missing arguments to min; returning Inf
## Warning in FUN(newX[, i], ...): no non-missing arguments to min; returning Inf
## Warning in FUN(newX[, i], ...): no non-missing arguments to min; returning Inf
## Warning in FUN(newX[, i], ...): no non-missing arguments to min; returning Inf## Warning in FUN(newX[, i], ...): no non-missing arguments to max; returning -Inf
## Warning in FUN(newX[, i], ...): no non-missing arguments to max; returning -Inf
## Warning in FUN(newX[, i], ...): no non-missing arguments to max; returning -Inf
## Warning in FUN(newX[, i], ...): no non-missing arguments to max; returning -Inf
## Warning in FUN(newX[, i], ...): no non-missing arguments to max; returning -Inf
## Warning in FUN(newX[, i], ...): no non-missing arguments to max; returning -Inf
## Warning in FUN(newX[, i], ...): no non-missing arguments to max; returning -Inf
## Warning in FUN(newX[, i], ...): no non-missing arguments to max; returning -Inf## vars n mean sd median trimmed mad min max range
## voice_id 1 56 2213.50 16.31 2213.50 2213.50 20.76 2186.00 2241.00 55.00
## ID* 2 56 NaN NA NA NaN NA Inf -Inf -Inf
## dataset 3 56 11.00 0.00 11.00 11.00 0.00 11.00 11.00 0.00
## sex 4 56 1.00 0.00 1.00 1.00 0.00 1.00 1.00 0.00
## age 5 56 19.96 1.21 20.00 19.93 1.48 18.00 23.00 5.00
## f0 6 56 105.92 11.44 102.55 105.13 8.95 88.94 133.76 44.82
## f1 7 56 410.40 33.04 418.16 412.39 28.13 338.15 483.39 145.24
## f2 8 56 1491.34 56.39 1493.32 1491.80 68.56 1381.04 1603.63 222.59
## f3 9 56 2485.94 78.16 2484.18 2482.57 71.33 2311.55 2680.83 369.28
## f4 10 56 3433.50 109.01 3433.66 3431.88 112.39 3192.24 3700.46 508.22
## pf 11 0 NaN NA NA NaN NA Inf -Inf -Inf
## neuro 12 0 NaN NA NA NaN NA Inf -Inf -Inf
## extra 13 0 NaN NA NA NaN NA Inf -Inf -Inf
## openn 14 0 NaN NA NA NaN NA Inf -Inf -Inf
## agree 15 0 NaN NA NA NaN NA Inf -Inf -Inf
## consc 16 0 NaN NA NA NaN NA Inf -Inf -Inf
## dominance 17 0 NaN NA NA NaN NA Inf -Inf -Inf
## behavior 18 56 3.99 1.62 3.83 3.92 1.73 1.33 7.67 6.33
## attitude 19 56 7.37 1.45 7.67 7.57 1.24 2.67 9.00 6.33
## desire 20 56 6.49 1.38 6.67 6.57 1.48 3.00 9.00 6.00
## soir_full 21 56 5.95 1.11 5.89 6.00 1.15 3.00 7.78 4.78
## skew kurtosis se
## voice_id 0.00 -1.26 2.18
## ID* NA NA NA
## dataset NaN NaN 0.00
## sex NaN NaN 0.00
## age 0.19 -0.68 0.16
## f0 0.65 -0.30 1.53
## f1 -0.53 -0.27 4.41
## f2 -0.08 -0.97 7.53
## f3 0.34 -0.08 10.44
## f4 0.13 -0.03 14.57
## pf NA NA NA
## neuro NA NA NA
## extra NA NA NA
## openn NA NA NA
## agree NA NA NA
## consc NA NA NA
## dominance NA NA NA
## behavior 0.31 -0.82 0.22
## attitude -1.23 1.27 0.19
## desire -0.55 -0.05 0.18
## soir_full -0.33 -0.18 0.15names(daten11)## [1] "voice_id" "ID" "dataset" "sex" "age" "f0"
## [7] "f1" "f2" "f3" "f4" "pf" "neuro"
## [13] "extra" "openn" "agree" "consc" "dominance" "behavior"
## [19] "attitude" "desire" "soir_full"daten11$soi_minp <- 1
daten11$soi_maxp <- 9Test that all values are between min/max
##############################################################################################
#erneut alles zusammen fuegen um zu z-transformieren
data_complete_all <- list(daten1, daten2, daten3, daten4, daten5, daten6, daten7, daten8, daten9, daten10, daten11) %>% map(~ mutate(., ID = as.character(ID))) %>% bind_rows() %>% as_tibble()
outside_range <- function(vec, min, max) {
vec < min | vec > max
}
none <- function(df) {
nrow(df) == 0
}
data_complete_all %>% filter(
outside_range(extra, big5_minp, big5_maxp)
) %>% none() %>%
testthat::expect_true()## Error in get(genname, envir = envir) : object 'testthat_print' not founddata_complete_all %>% filter(
outside_range(agree, big5_minp, big5_maxp)
) %>% none() %>%
testthat::expect_true()
data_complete_all %>% filter(
outside_range(neuro, big5_minp, big5_maxp)
) %>% none() %>%
testthat::expect_true()
data_complete_all %>% filter(
outside_range(openn, big5_minp, big5_maxp)
) %>% none() %>%
testthat::expect_true()
data_complete_all %>% filter(
outside_range(consc, big5_minp, big5_maxp)
) %>% none() %>%
testthat::expect_true()
data_complete_all %>% filter(
outside_range(behavior, soi_minp, soi_maxp)
) %>% none() %>%
testthat::expect_true()
data_complete_all %>% filter(
outside_range(attitude, soi_minp, soi_maxp)
) %>% none() %>%
testthat::expect_true()
data_complete_all %>% filter(
outside_range(desire, soi_minp, soi_maxp)
) %>% none() %>%
testthat::expect_true()
data_complete_all %>% filter(
outside_range(soir_full, soi_minp, soi_maxp)
) %>% none() %>%
testthat::expect_true()
data_complete_all %>% filter(
outside_range(dominance, dominance_minp, dominance_maxp)
) %>% none() %>%
testthat::expect_true()Use min/max to compute POMP
pomp <- function(raw, min, max) {
(raw - min)/(max-min)
}
data_complete_all <- data_complete_all %>%
mutate_at(vars(extra, agree, neuro, openn, consc), ~ pomp(., big5_minp, big5_maxp)) %>%
mutate_at(vars(behavior, attitude, desire, soir_full), ~ pomp(., soi_minp, soi_maxp)) %>%
mutate_at(vars(dominance), ~ pomp(., dominance_minp, dominance_maxp))
data_complete_all %>% select(extra, agree, neuro, openn, consc, behavior, attitude, desire, soir_full, dominance) %>% gather(variable, value) %>%
ggplot(aes(value)) + geom_histogram() + facet_wrap(~ variable)## `stat_bin()` using `bins = 30`. Pick better value with `binwidth`.## Warning: Removed 6229 rows containing non-finite values (stat_bin).
Calculate pf
#Pf bilden
data_complete_all$pf <- (scale(data_complete_all$f1)[,1] +
scale(data_complete_all$f2)[,1] +
scale(data_complete_all$f3)[,1] +
scale(data_complete_all$f4)[,1])/4Add labels
library(labelled)
var_label(data_complete_all$dominance) <- "Dominance"
var_label(data_complete_all$neuro) <- "Neuroticism"
var_label(data_complete_all$agree) <- "Agreeableness"
var_label(data_complete_all$extra) <- "Extraversion"
var_label(data_complete_all$openn) <- "Openness"
var_label(data_complete_all$consc) <- "Conscientiousness"
var_label(data_complete_all$soir_full) <- "Unrestricted sociosexuality"
var_label(data_complete_all$f0) <- "Voice pitch"
var_label(data_complete_all$pf) <- "Formants"
data_complete_all$sex_c <- data_complete_all$sex
data_complete_all$sex <- factor(if_else(data_complete_all$sex == 1, "male", "female"))
contrasts(data_complete_all$sex) <- contr.helmert(2)
var_label(data_complete_all$age) <- "Age"Save pre-standardisation
data_complete_all_unstd <- data_complete_all
rio::export(data_complete_all_unstd, "data_complete_2021_unstd_pomp.rds")Compute z-scores
data_complete_all <- data_complete_all %>% mutate_at(vars(extra, agree, neuro, openn, consc, behavior, attitude, desire, soir_full, dominance, f0, f1, f2, f3, f4, pf), ~ scale(.)[,1])
data_complete_all %>% select(extra, agree, neuro, openn, consc, behavior, attitude, desire, soir_full, dominance, f0, f1, f2, f3, f4, pf) %>% gather(variable, value) %>%
ggplot(aes(value)) + geom_histogram() + facet_wrap(~ variable)## `stat_bin()` using `bins = 30`. Pick better value with `binwidth`.## Warning: Removed 6256 rows containing non-finite values (stat_bin).
Save data
rio::export(data_complete_all, "data_complete_2021_zscored.rds")
rio::export(data_complete_all, "datarelease/data_personality_voices_zscored.rds")
rio::export(data_complete_all, "datarelease/data_personality_voices_zscored.sav")
rio::export(data_complete_all, "datarelease/data_personality_voices_zscored.xlsx")
rio::export(data_complete_all, "datarelease/data_personality_voices_zscored.csv")Standardize within dataset, re-add sex differences
vcs <- data_complete_all_unstd
vcs <- vcs %>% mutate_at(vars(extra, agree, neuro, openn, consc, behavior, attitude, desire, soir_full, dominance), ~ scale(.)[,1])
# vcs10c <- rio::import("daten10_formants.rds")
# vcs10 <- vcs %>% filter(dataset == 10)
# vcs10 <- vcs10 %>% select(-(f0:f4)) %>% left_join(vcs10c)
# vcs <- vcs %>% filter(dataset != 10) %>%
# bind_rows(vcs10)- we first standardise within datasets
svcs <- vcs %>% group_by(dataset) %>%
mutate(f0 = scale(f0)[,1],
f1 = scale(f1)[,1],
f2 = scale(f2)[,1],
f3 = scale(f3)[,1],
f4 = scale(f4)[,1])- we use the HQ datasets with both genders to estimate mean diff of m/f and restricted variance
svcs %>% filter(dataset %in% c(2,3, 5, 9)) %>% ungroup() %>%
summarise_at(vars(f0:f4), ~broom::tidy(t.test(. ~ sex))$estimate)## # A tibble: 1 x 5
## f0 f1 f2 f3 f4
## <dbl> <dbl> <dbl> <dbl> <dbl>
## 1 1.86 1.21 1.55 1.76 1.77svcs %>% filter(dataset %in% c(2,3, 5, 9)) %>%
group_by(dataset) %>%
summarise_at(vars(f0:f4), ~broom::tidy(t.test(. ~ sex))$estimate)## # A tibble: 4 x 6
## dataset f0 f1 f2 f3 f4
## <dbl> <dbl> <dbl> <dbl> <dbl> <dbl>
## 1 2 1.88 1.26 1.64 1.81 1.80
## 2 3 1.88 1.10 1.36 1.80 1.84
## 3 5 1.98 1.33 1.84 1.88 1.86
## 4 9 2.03 1.39 1.65 1.73 1.78svcs %>% filter(dataset %in% c(2,3, 5, 9)) %>%
group_by(dataset, sex) %>%
summarise_at(vars(f0:f4), list(mean = ~mean(., na.rm = T), sd = ~sd(., na.rm = T)))## # A tibble: 8 x 12
## # Groups: dataset [4]
## dataset sex f0_mean f1_mean f2_mean f3_mean f4_mean f0_sd f1_sd f2_sd f3_sd
## <dbl> <fct> <dbl> <dbl> <dbl> <dbl> <dbl> <dbl> <dbl> <dbl> <dbl>
## 1 2 fema… 0.934 0.627 0.814 0.899 0.895 0.398 0.943 0.598 0.514
## 2 2 male -0.949 -0.637 -0.826 -0.913 -0.909 0.251 0.556 0.544 0.301
## 3 3 fema… 0.931 0.542 0.671 0.892 0.909 0.359 1.03 0.846 0.432
## 4 3 male -0.951 -0.554 -0.686 -0.911 -0.928 0.309 0.575 0.600 0.428
## 5 5 fema… 0.651 0.435 0.605 0.615 0.611 0.369 0.843 0.527 0.519
## 6 5 male -1.33 -0.892 -1.24 -1.26 -1.25 0.332 0.640 0.433 0.352
## 7 9 fema… 0.590 0.428 0.508 0.534 0.548 0.404 0.805 0.658 0.569
## 8 9 male -1.44 -0.963 -1.14 -1.20 -1.23 0.324 0.676 0.622 0.658
## # … with 1 more variable: f4_sd <dbl># not that much variation in sex diff across datasets, so we pool
sex_diffs <- svcs %>% filter(dataset %in% c(2,3, 5, 9)) %>%
group_by(sex) %>%
summarise_at(vars(f0:f4), list(mean = ~mean(., na.rm = T), sd = ~sd(., na.rm = T)))
sex_diffs## # A tibble: 2 x 11
## sex f0_mean f1_mean f2_mean f3_mean f4_mean f0_sd f1_sd f2_sd f3_sd f4_sd
## <fct> <dbl> <dbl> <dbl> <dbl> <dbl> <dbl> <dbl> <dbl> <dbl> <dbl>
## 1 female 0.813 0.529 0.679 0.773 0.778 0.411 0.926 0.675 0.527 0.531
## 2 male -1.05 -0.676 -0.868 -0.988 -0.994 0.340 0.601 0.588 0.418 0.384- we fudge the data in the single gender datasets
((x + gender_mean) * sd_gender)
svcs %>% filter(!dataset %in% c(2,3, 5, 9)) %>%
group_by(dataset, sex) %>%
summarise_at(vars(f0:f4), list(mean = ~mean(., na.rm = T), sd = ~sd(., na.rm = T)))## # A tibble: 7 x 12
## # Groups: dataset [7]
## dataset sex f0_mean f1_mean f2_mean f3_mean f4_mean f0_sd f1_sd
## <dbl> <fct> <dbl> <dbl> <dbl> <dbl> <dbl> <dbl> <dbl>
## 1 1 fema… -2.10e-16 -2.35e-16 5.83e-16 1.43e-15 -2.80e-16 1 1.00
## 2 4 fema… -7.82e-16 -3.22e-16 -1.08e-15 1.68e-15 -1.42e-15 1 1
## 3 6 male 3.05e-16 8.55e-16 7.23e-16 1.36e-15 -9.86e-16 1.00 1
## 4 7 male -2.80e-16 2.43e-16 -3.06e-16 -6.45e-16 -8.05e-16 1.00 1.00
## 5 8 fema… -5.26e-16 -3.84e-16 -7.00e-16 7.74e-16 1.79e-16 1.00 1
## 6 10 male 2.51e-16 -2.87e-16 4.40e-16 3.95e-16 -1.41e-15 1 1
## 7 11 male 4.69e-16 -2.23e-16 -7.39e-17 -3.06e-16 2.02e-15 1.00 1
## # … with 3 more variables: f2_sd <dbl>, f3_sd <dbl>, f4_sd <dbl>svcs <- svcs %>%
left_join(sex_diffs, by = "sex") %>%
mutate(f0 = if_else(!dataset %in% c(2,3, 5, 9),
f0 * f0_sd + f0_mean, f0),
f1 = if_else(!dataset %in% c(2,3, 5, 9),
f1 * f1_sd + f1_mean, f1),
f2 = if_else(!dataset %in% c(2,3, 5, 9),
f2 * f2_sd + f2_mean, f2),
f3 = if_else(!dataset %in% c(2,3, 5, 9),
f3 * f3_sd + f3_mean, f3),
f4 = if_else(!dataset %in% c(2,3, 5, 9),
f4 * f4_sd + f4_mean, f4))
svcs %>% filter(!dataset %in% c(2,3, 5, 9)) %>%
group_by(dataset, sex) %>%
summarise_at(vars(f0:f4), list(mean = ~mean(., na.rm = T), sd = ~sd(., na.rm = T)))## # A tibble: 7 x 12
## # Groups: dataset [7]
## dataset sex f0_mean f1_mean f2_mean f3_mean f4_mean f0_sd f1_sd f2_sd f3_sd
## <dbl> <fct> <dbl> <dbl> <dbl> <dbl> <dbl> <dbl> <dbl> <dbl> <dbl>
## 1 1 fema… 0.813 0.529 0.679 0.773 0.778 0.411 0.926 0.675 0.527
## 2 4 fema… 0.813 0.529 0.679 0.773 0.778 0.411 0.926 0.675 0.527
## 3 6 male -1.05 -0.676 -0.868 -0.988 -0.994 0.340 0.601 0.588 0.418
## 4 7 male -1.05 -0.676 -0.868 -0.988 -0.994 0.340 0.601 0.588 0.418
## 5 8 fema… 0.813 0.529 0.679 0.773 0.778 0.411 0.926 0.675 0.527
## 6 10 male -1.05 -0.676 -0.868 -0.988 -0.994 0.340 0.601 0.588 0.418
## 7 11 male -1.05 -0.676 -0.868 -0.988 -0.994 0.340 0.601 0.588 0.418
## # … with 1 more variable: f4_sd <dbl>svcs %>% filter(!dataset %in% c(2,3, 5, 9)) %>% ungroup() %>%
summarise_at(vars(f0:f4), ~broom::tidy(t.test(. ~ sex))$estimate)## # A tibble: 1 x 5
## f0 f1 f2 f3 f4
## <dbl> <dbl> <dbl> <dbl> <dbl>
## 1 1.86 1.21 1.55 1.76 1.77svcs %>% filter(dataset %in% c(2,3, 5, 9)) %>% ungroup() %>%
summarise_at(vars(f0:f4), ~broom::tidy(t.test(. ~ sex))$estimate)## # A tibble: 1 x 5
## f0 f1 f2 f3 f4
## <dbl> <dbl> <dbl> <dbl> <dbl>
## 1 1.86 1.21 1.55 1.76 1.77- then we average across f1 to f4 and standardise pf again
svcs %>% ungroup() %>%
summarise_at(vars(f0:f4), list(mean = ~mean(., na.rm = T), sd = ~sd(., na.rm = T)))## # A tibble: 1 x 10
## f0_mean f1_mean f2_mean f3_mean f4_mean f0_sd f1_sd f2_sd f3_sd f4_sd
## <dbl> <dbl> <dbl> <dbl> <dbl> <dbl> <dbl> <dbl> <dbl> <dbl>
## 1 0.0466 0.0313 0.0402 0.0457 0.0460 0.992 1.00 0.995 0.993 0.994svcs <- svcs %>% ungroup() %>% mutate(pf = scale(f1 + f2 + f3 + f4)[,1])
svcs %>% filter(!dataset %in% c(2,3, 5, 9)) %>%
group_by(dataset, sex) %>%
summarise_at(vars(pf, f0:f4), list(mean = ~mean(., na.rm = T), sd = ~sd(., na.rm = T)))## # A tibble: 7 x 14
## # Groups: dataset [7]
## dataset sex pf_mean f0_mean f1_mean f2_mean f3_mean f4_mean pf_sd f0_sd
## <dbl> <fct> <dbl> <dbl> <dbl> <dbl> <dbl> <dbl> <dbl> <dbl>
## 1 1 fema… 0.746 0.813 0.529 0.679 0.773 0.778 0.462 0.411
## 2 4 fema… 0.746 0.813 0.529 0.679 0.773 0.778 0.494 0.411
## 3 6 male -1.06 -1.05 -0.676 -0.868 -0.988 -0.994 0.318 0.340
## 4 7 male -1.06 -1.05 -0.676 -0.868 -0.988 -0.994 0.322 0.340
## 5 8 fema… 0.746 0.813 0.529 0.679 0.773 0.778 0.569 0.411
## 6 10 male -1.06 -1.05 -0.676 -0.868 -0.988 -0.994 0.488 0.340
## 7 11 male -1.06 -1.05 -0.676 -0.868 -0.988 -0.994 0.303 0.340
## # … with 4 more variables: f1_sd <dbl>, f2_sd <dbl>, f3_sd <dbl>, f4_sd <dbl>- let’s check intercorrelations in each dataset after this transformation
vcs %>% select(dataset, f0:f4, sex_c) %>%
group_by(dataset) %>%
mutate_at(vars(f0:f4), ~resid(lm(. ~ sex_c, na.action = na.exclude))) %>%
summarise(corrr::correlate(across(f0:f4)) %>% corrr::shave() %>% corrr::stretch()) %>%
unite(vars, x, y) %>%
drop_na() %>%
ggplot(aes(dataset, r)) +
geom_text(aes(label = dataset)) +
coord_flip() +
facet_wrap(~ vars)##
## Correlation method: 'pearson'
## Missing treated using: 'pairwise.complete.obs'
##
##
## Correlation method: 'pearson'
## Missing treated using: 'pairwise.complete.obs'
##
##
## Correlation method: 'pearson'
## Missing treated using: 'pairwise.complete.obs'
##
##
## Correlation method: 'pearson'
## Missing treated using: 'pairwise.complete.obs'
##
##
## Correlation method: 'pearson'
## Missing treated using: 'pairwise.complete.obs'
##
##
## Correlation method: 'pearson'
## Missing treated using: 'pairwise.complete.obs'
##
##
## Correlation method: 'pearson'
## Missing treated using: 'pairwise.complete.obs'
##
##
## Correlation method: 'pearson'
## Missing treated using: 'pairwise.complete.obs'
##
##
## Correlation method: 'pearson'
## Missing treated using: 'pairwise.complete.obs'
##
##
## Correlation method: 'pearson'
## Missing treated using: 'pairwise.complete.obs'
##
##
## Correlation method: 'pearson'
## Missing treated using: 'pairwise.complete.obs'## `summarise()` regrouping output by 'dataset' (override with `.groups` argument)
svcs %>% select(dataset, f0:f4, sex_c) %>%
group_by(dataset) %>%
mutate_at(vars(f0:f4), ~resid(lm(. ~ sex_c, na.action = na.exclude))) %>%
summarise(corrr::correlate(across(f0:f4)) %>% corrr::shave() %>% corrr::stretch()) %>%
unite(vars, x, y) %>%
drop_na() %>%
ggplot(aes(dataset, r)) +
geom_text(aes(label = dataset)) +
coord_flip() +
facet_wrap(~ vars)##
## Correlation method: 'pearson'
## Missing treated using: 'pairwise.complete.obs'##
## Correlation method: 'pearson'
## Missing treated using: 'pairwise.complete.obs'
##
##
## Correlation method: 'pearson'
## Missing treated using: 'pairwise.complete.obs'
##
##
## Correlation method: 'pearson'
## Missing treated using: 'pairwise.complete.obs'
##
##
## Correlation method: 'pearson'
## Missing treated using: 'pairwise.complete.obs'
##
##
## Correlation method: 'pearson'
## Missing treated using: 'pairwise.complete.obs'
##
##
## Correlation method: 'pearson'
## Missing treated using: 'pairwise.complete.obs'
##
##
## Correlation method: 'pearson'
## Missing treated using: 'pairwise.complete.obs'
##
##
## Correlation method: 'pearson'
## Missing treated using: 'pairwise.complete.obs'
##
##
## Correlation method: 'pearson'
## Missing treated using: 'pairwise.complete.obs'
##
##
## Correlation method: 'pearson'
## Missing treated using: 'pairwise.complete.obs'## `summarise()` regrouping output by 'dataset' (override with `.groups` argument)
bind_rows(stdised = svcs %>% select(dataset, f0:f4, sex_c) %>%
ungroup() %>%
mutate_at(vars(f0:f4), ~resid(lm(. ~ sex_c, na.action = na.exclude))) %>%
summarise(corrr::correlate(across(f0:f4)) %>% corrr::shave() %>% corrr::stretch()) %>%
unite(vars, x, y) %>%
drop_na(),
unstdised = vcs %>% select(dataset, f0:f4, sex_c) %>%
ungroup() %>%
mutate_at(vars(f0:f4), ~resid(lm(. ~ sex_c, na.action = na.exclude))) %>%
summarise(corrr::correlate(across(f0:f4)) %>% corrr::shave() %>% corrr::stretch()) %>%
unite(vars, x, y) %>%
drop_na(), .id = "process") %>%
ggplot(aes(process, r)) +
geom_text(aes(label = sprintf("%.2f", r))) +
coord_flip() +
facet_wrap(~ vars)##
## Correlation method: 'pearson'
## Missing treated using: 'pairwise.complete.obs'##
## Correlation method: 'pearson'
## Missing treated using: 'pairwise.complete.obs'
- let’s check how this looks compared to the simple approach
svcs <- svcs %>% ungroup()
vcs_2 <- svcs %>% select(voice_id, f0:f4, pf) %>%
inner_join(
data_complete_all %>% select(voice_id, f0:f4, pf), by = "voice_id", suffix = c("_within", "_std"))
rcamisc::mtmm(vcs_2 %>% ungroup() %>% select(-voice_id))
Distributions by dataset
svcs$sex <- factor(if_else(svcs$sex_c == 1, "male", "female"))
contrasts(svcs$sex) <- contr.helmert(2)
ggplot(svcs, aes(f0, fill = factor(dataset))) +
geom_histogram(position = "identity", alpha = 0.4)+
facet_wrap(~ sex, scales = "free_x") +
scale_fill_viridis_d()## `stat_bin()` using `bins = 30`. Pick better value with `binwidth`.## Warning: Removed 7 rows containing non-finite values (stat_bin).
ggplot(svcs, aes(f1, fill = factor(dataset))) +
geom_histogram(position = "identity", alpha = 0.4)+
facet_wrap(~ sex, scales = "free_x") +
scale_fill_viridis_d()## `stat_bin()` using `bins = 30`. Pick better value with `binwidth`.## Warning: Removed 4 rows containing non-finite values (stat_bin).
ggplot(svcs, aes(f2, fill = factor(dataset))) +
geom_histogram(position = "identity", alpha = 0.4)+
facet_wrap(~ sex, scales = "free_x") +
scale_fill_viridis_d()## `stat_bin()` using `bins = 30`. Pick better value with `binwidth`.## Warning: Removed 4 rows containing non-finite values (stat_bin).
ggplot(svcs, aes(f3, fill = factor(dataset))) +
geom_histogram(position = "identity", alpha = 0.4)+
facet_wrap(~ sex, scales = "free_x") +
scale_fill_viridis_d()## `stat_bin()` using `bins = 30`. Pick better value with `binwidth`.## Warning: Removed 4 rows containing non-finite values (stat_bin).
ggplot(svcs, aes(f4, fill = factor(dataset))) +
geom_histogram(position = "identity", alpha = 0.4)+
facet_wrap(~ sex, scales = "free_x") +
scale_fill_viridis_d()## `stat_bin()` using `bins = 30`. Pick better value with `binwidth`.## Warning: Removed 4 rows containing non-finite values (stat_bin).
ggplot(svcs, aes(pf, fill = factor(dataset))) +
geom_histogram(position = "identity", alpha = 0.4)+
facet_wrap(~ sex, scales = "free_x") +
scale_fill_viridis_d()## `stat_bin()` using `bins = 30`. Pick better value with `binwidth`.## Warning: Removed 4 rows containing non-finite values (stat_bin).
Save data
rio::export(svcs, "data_complete_2021_within_zscored.rds")Outlier removal
svcs <- data_complete_all
svcs <- svcs %>% # group_by(sex) %>%
mutate(pf_threshold_hi = median(pf, na.rm = T) + 2.5 * mad(pf, na.rm = T),
pf_threshold_lo = median(pf, na.rm = T) - 2.5 * mad(pf, na.rm = T),
f0_threshold_hi = median(f0, na.rm = T) + 2.5 * mad(f0, na.rm = T),
f0_threshold_lo = median(f0, na.rm = T) - 2.5 * mad(f0, na.rm = T))
pf_outliers <- svcs %>%
filter(pf < pf_threshold_lo | pf > pf_threshold_hi)
nrow(pf_outliers)## [1] 3xtabs(~ dataset + sex, pf_outliers)## sex
## dataset female male
## 2 2 0
## 10 0 1f0_outliers <- svcs %>%
filter(f0 < f0_threshold_lo | f0 > f0_threshold_hi)
nrow(f0_outliers)## [1] 0xtabs(~ dataset + sex, f0_outliers)## < table of extent 0 x 2 >vcs <- svcs %>% filter(
pf > pf_threshold_lo,
pf < pf_threshold_hi,
f0 > f0_threshold_lo,
f0 < f0_threshold_hi)
psych::describeBy(vcs %>% ungroup() %>% select(sex_c, f0:f4, pf), vcs$dataset)##
## Descriptive statistics by group
## group: 1
## vars n mean sd median trimmed mad min max range skew kurtosis
## sex_c 1 339 -1.00 0.00 -1.00 -1.00 0.00 -1.00 -1.00 0.00 NaN NaN
## f0 2 339 0.77 0.44 0.74 0.75 0.40 -0.53 2.60 3.13 0.49 0.81
## f1 3 339 0.19 0.77 0.05 0.14 0.67 -1.86 3.60 5.46 0.82 1.24
## f2 4 339 0.31 1.15 0.25 0.26 1.17 -2.44 5.20 7.64 0.49 0.61
## f3 5 339 0.82 0.57 0.78 0.81 0.54 -0.68 2.29 2.97 0.16 -0.17
## f4 6 339 0.61 0.58 0.67 0.64 0.56 -1.01 2.29 3.30 -0.34 -0.20
## pf 7 339 0.60 0.60 0.56 0.58 0.58 -1.02 2.84 3.86 0.39 0.19
## se
## sex_c 0.00
## f0 0.02
## f1 0.04
## f2 0.06
## f3 0.03
## f4 0.03
## pf 0.03
## ------------------------------------------------------------
## group: 2
## vars n mean sd median trimmed mad min max range skew kurtosis
## sex_c 1 381 0.00 1.00 -1.00 0.00 0.00 -1.00 1.00 2.00 0.01 -2.01
## f0 2 381 -0.28 1.03 -0.36 -0.31 1.43 -1.76 1.68 3.44 0.12 -1.63
## f1 3 381 0.42 1.08 0.32 0.33 1.10 -1.58 4.46 6.05 0.79 0.62
## f2 4 381 0.28 0.86 0.33 0.27 1.09 -1.71 2.47 4.18 0.10 -0.92
## f3 5 381 0.24 1.06 -0.05 0.20 1.31 -1.51 2.78 4.29 0.22 -1.30
## f4 6 381 0.13 1.11 -0.16 0.05 1.31 -1.62 2.69 4.31 0.36 -1.21
## pf 7 381 0.33 1.16 0.17 0.27 1.50 -1.42 3.05 4.47 0.28 -1.33
## se
## sex_c 0.05
## f0 0.05
## f1 0.06
## f2 0.04
## f3 0.05
## f4 0.06
## pf 0.06
## ------------------------------------------------------------
## group: 3
## vars n mean sd median trimmed mad min max range skew kurtosis
## sex_c 1 285 -0.01 1.00 -1.00 -0.01 0.00 -1.00 1.00 2.00 0.02 -2.01
## f0 2 285 -0.12 0.99 -0.03 -0.13 1.39 -1.73 1.84 3.57 0.03 -1.59
## f1 3 285 0.76 0.65 0.70 0.71 0.58 -0.65 3.96 4.61 1.09 2.45
## f2 4 285 -0.25 0.89 -0.38 -0.29 1.03 -2.09 2.41 4.50 0.41 -0.57
## f3 5 285 -0.13 0.97 -0.14 -0.13 1.32 -2.27 1.63 3.89 -0.03 -1.38
## f4 6 285 0.00 0.92 0.02 0.01 1.23 -2.43 1.65 4.08 -0.04 -1.39
## pf 7 285 0.12 0.93 -0.04 0.10 1.21 -1.76 2.38 4.14 0.15 -1.27
## se
## sex_c 0.06
## f0 0.06
## f1 0.04
## f2 0.05
## f3 0.06
## f4 0.05
## pf 0.06
## ------------------------------------------------------------
## group: 4
## vars n mean sd median trimmed mad min max range skew kurtosis
## sex_c 1 265 -1.00 0.00 -1.00 -1.00 0.00 -1.00 -1.00 0.00 NaN NaN
## f0 2 265 0.79 0.32 0.77 0.78 0.29 -0.33 1.99 2.32 0.18 0.88
## f1 3 265 -0.92 0.44 -0.91 -0.93 0.42 -2.07 0.30 2.38 0.05 0.07
## f2 4 265 0.36 0.55 0.30 0.35 0.59 -1.08 2.31 3.39 0.15 -0.09
## f3 5 265 0.40 0.47 0.43 0.41 0.52 -0.77 1.53 2.31 -0.11 -0.44
## f4 6 265 0.74 0.42 0.76 0.75 0.40 -0.40 1.96 2.36 -0.09 -0.03
## pf 7 265 0.18 0.39 0.15 0.18 0.40 -0.84 1.27 2.11 0.13 -0.41
## se
## sex_c 0.00
## f0 0.02
## f1 0.03
## f2 0.03
## f3 0.03
## f4 0.03
## pf 0.02
## ------------------------------------------------------------
## group: 5
## vars n mean sd median trimmed mad min max range skew kurtosis
## sex_c 1 186 -0.34 0.94 -1.00 -0.43 0.00 -1.00 1.00 2.00 0.73 -1.48
## f0 2 186 0.02 0.94 0.41 0.07 0.74 -1.77 1.46 3.23 -0.54 -1.23
## f1 3 186 0.20 0.62 0.17 0.18 0.61 -1.40 1.83 3.23 0.30 -0.15
## f2 4 186 0.11 0.88 0.43 0.14 0.83 -1.82 2.08 3.90 -0.39 -1.00
## f3 5 186 -0.01 0.91 0.22 0.01 1.00 -2.00 1.56 3.56 -0.34 -1.16
## f4 6 186 -0.23 0.91 0.01 -0.20 0.94 -2.24 1.40 3.64 -0.40 -1.06
## pf 7 186 0.02 0.93 0.33 0.05 0.88 -1.85 1.71 3.56 -0.40 -1.28
## se
## sex_c 0.07
## f0 0.07
## f1 0.05
## f2 0.06
## f3 0.07
## f4 0.07
## pf 0.07
## ------------------------------------------------------------
## group: 6
## vars n mean sd median trimmed mad min max range skew kurtosis
## sex_c 1 184 1.00 0.00 1.00 1.00 0.00 1.00 1.00 0.00 NaN NaN
## f0 2 184 -1.02 0.31 -1.08 -1.04 0.28 -1.69 0.11 1.80 0.67 0.54
## f1 3 184 -1.29 0.38 -1.29 -1.28 0.39 -2.21 -0.23 1.98 -0.16 -0.18
## f2 4 184 -0.95 0.45 -0.96 -0.96 0.44 -2.14 0.32 2.46 0.12 -0.16
## f3 5 184 -1.16 0.44 -1.16 -1.14 0.45 -2.30 -0.08 2.21 -0.19 -0.24
## f4 6 184 -0.95 0.47 -0.93 -0.95 0.44 -2.53 0.62 3.15 0.11 0.65
## pf 7 184 -1.36 0.31 -1.36 -1.35 0.32 -2.10 -0.42 1.67 0.02 -0.12
## se
## sex_c 0.00
## f0 0.02
## f1 0.03
## f2 0.03
## f3 0.03
## f4 0.03
## pf 0.02
## ------------------------------------------------------------
## group: 7
## vars n mean sd median trimmed mad min max range skew kurtosis
## sex_c 1 164 1.00 0.00 1.00 1.00 0.00 1.00 1.00 0.00 NaN NaN
## f0 2 164 -1.06 0.30 -1.09 -1.08 0.28 -1.68 -0.12 1.56 0.50 0.30
## f1 3 164 0.34 0.48 0.32 0.33 0.45 -0.90 1.76 2.66 0.01 0.21
## f2 4 164 -1.42 0.50 -1.43 -1.43 0.47 -3.03 0.14 3.17 0.23 0.50
## f3 5 164 -1.26 0.42 -1.23 -1.26 0.37 -2.30 0.22 2.52 0.16 0.50
## f4 6 164 -1.22 0.49 -1.18 -1.22 0.37 -2.71 0.32 3.03 -0.05 0.69
## pf 7 164 -1.11 0.34 -1.13 -1.11 0.37 -2.01 -0.17 1.84 0.08 -0.08
## se
## sex_c 0.00
## f0 0.02
## f1 0.04
## f2 0.04
## f3 0.03
## f4 0.04
## pf 0.03
## ------------------------------------------------------------
## group: 8
## vars n mean sd median trimmed mad min max range skew kurtosis
## sex_c 1 157 -1.00 0.00 -1.00 -1.00 0.00 -1.00 -1.00 0.00 NaN NaN
## f0 2 157 0.78 0.35 0.80 0.78 0.34 -0.28 1.83 2.11 0.16 0.47
## f1 3 157 -0.13 0.96 -0.25 -0.16 1.00 -2.35 1.98 4.33 0.24 -0.65
## f2 4 157 0.47 0.49 0.52 0.50 0.45 -1.46 1.57 3.04 -0.64 0.93
## f3 5 157 0.52 0.62 0.45 0.49 0.63 -0.59 2.33 2.93 0.37 -0.45
## f4 6 157 0.67 0.43 0.66 0.67 0.44 -0.57 1.66 2.23 -0.17 -0.32
## pf 7 157 0.48 0.58 0.48 0.48 0.59 -1.02 1.79 2.82 -0.06 -0.41
## se
## sex_c 0.00
## f0 0.03
## f1 0.08
## f2 0.04
## f3 0.05
## f4 0.03
## pf 0.05
## ------------------------------------------------------------
## group: 9
## vars n mean sd median trimmed mad min max range skew kurtosis
## sex_c 1 127 -0.42 0.91 -1.00 -0.51 0.00 -1.00 1.00 2.00 0.91 -1.19
## f0 2 127 0.54 1.02 0.91 0.60 0.65 -1.66 2.33 3.99 -0.63 -0.95
## f1 3 127 0.42 0.75 0.46 0.41 0.78 -1.06 2.76 3.82 0.17 -0.23
## f2 4 127 0.35 0.91 0.50 0.38 0.93 -2.12 2.35 4.47 -0.31 -0.44
## f3 5 127 -0.10 0.96 0.07 -0.06 0.94 -2.28 1.79 4.07 -0.46 -0.72
## f4 6 127 -0.03 1.09 0.24 0.03 1.12 -2.78 1.73 4.51 -0.51 -0.72
## pf 7 127 0.20 1.01 0.50 0.25 0.87 -1.76 2.25 4.01 -0.53 -0.95
## se
## sex_c 0.08
## f0 0.09
## f1 0.07
## f2 0.08
## f3 0.09
## f4 0.10
## pf 0.09
## ------------------------------------------------------------
## group: 10
## vars n mean sd median trimmed mad min max range skew kurtosis
## sex_c 1 87 1.00 0.00 1.00 1.00 0.00 1.00 1.00 0.00 NaN NaN
## f0 2 87 -1.00 0.36 -1.06 -1.02 0.34 -1.61 0.07 1.68 0.53 -0.09
## f1 3 87 -0.74 0.87 -0.97 -0.86 0.53 -2.15 2.55 4.70 1.66 3.41
## f2 4 87 0.85 0.74 0.92 0.86 0.72 -1.18 2.45 3.63 -0.15 -0.31
## f3 5 87 -0.34 0.60 -0.37 -0.35 0.69 -1.61 1.04 2.66 0.09 -0.81
## f4 6 87 -0.85 0.45 -0.85 -0.89 0.43 -1.54 0.86 2.40 1.27 2.51
## pf 7 87 -0.34 0.69 -0.43 -0.39 0.62 -1.56 2.14 3.70 0.95 1.59
## se
## sex_c 0.00
## f0 0.04
## f1 0.09
## f2 0.08
## f3 0.06
## f4 0.05
## pf 0.07
## ------------------------------------------------------------
## group: 11
## vars n mean sd median trimmed mad min max range skew kurtosis
## sex_c 1 56 1.00 0.00 1.00 1.00 0.00 1.00 1.00 0.00 NaN NaN
## f0 2 56 -1.32 0.23 -1.38 -1.33 0.18 -1.65 -0.76 0.89 0.65 -0.30
## f1 3 56 -0.63 0.45 -0.53 -0.61 0.39 -1.62 0.37 1.99 -0.53 -0.27
## f2 4 56 -0.84 0.34 -0.83 -0.84 0.42 -1.51 -0.15 1.36 -0.08 -0.97
## f3 5 56 -1.07 0.35 -1.08 -1.09 0.32 -1.85 -0.21 1.63 0.34 -0.08
## f4 6 56 -1.21 0.35 -1.21 -1.21 0.36 -1.99 -0.34 1.65 0.13 -0.03
## pf 7 56 -1.17 0.24 -1.16 -1.17 0.18 -1.71 -0.58 1.12 0.17 0.21
## se
## sex_c 0.00
## f0 0.03
## f1 0.06
## f2 0.05
## f3 0.05
## f4 0.05
## pf 0.03Save data
rio::export(vcs, "data_complete_2021_zscored_no_outliers.rds")