R/score_emotb.R
In PennStateDEPENdLab/dependlab: Useful functions for the UNC Chapel Hill DEPENd Lab (Director: Michael Hallquist)
Defines functions
score_emotb
Documented in
score_emotb
#' Score Adult EMOTB
#'
#' Computes scores for the NIH Toolbox Emotion Measures, Adult Forms (ages 18-52) (2019).
#'
#' @param df The data frame containing the assessment items.
#' @param scales A character vector specifying the scales to include in the scoring.
#' The default is "all", which includes all scales. Other possible values are:
#' "percstrs", "self", "mp", "sad", "gls", "posaf", "angaf", "angpa",
#' "anghost", "emosup", "fearaf", "fearsoma", "friend", "instrsup",
#' "lone", "perchost", "percrej", and "apathy".
#' @param max_impute The threshold for mean imputation. Items with missing values
#' below this threshold will be mean imputed. The default is 0.2.
#' @param drop_items Logical indicating whether to drop the individual item columns
#' from the resulting data frame. The default is FALSE.
#' @param keep_reverse_codes Logical indicating whether to keep the reverse-coded
#' items in the resulting data frame. The default is FALSE.
#' @param add_alphas Logical indicating whether to compute Cronbach's alpha coefficients
#' for the scales and append them to the resulting data frame. The default is TRUE.
#' @param item_prefix A character vector specifying the item prefixes as they appear in the data.frame.
#' The vector must be the same length as the `scales` parameter and prefixes must be in the
#' same corresponding order. If `scales` is set to "all", the prefixes must appear in
#' the order of possible values for the `scales` parameter provided above
#' (Perceived Stress, Self Efficacy, etc.).
#'
#' @return A data frame with the computed scores and optional alpha coefficients.
#'
#' @details The function scores the Adult EMOTB assessment based on the provided data frame.
#' It supports various scoring options, including selecting specific scales,
#' mean imputation of missing values, dropping individual item columns,
#' keeping reverse-coded items, and computing alpha coefficients.
#'
#' The function adds three columns per scale to the data frame `df`, containing
#' the raw scores (\code{_Raw}), the theta scores (\code{_Theta}), and the
#' t-scores (\code{_T}) for the NIH Toolbox Emotion Measures, respectively.
#'
#' Note: The scoring is based on the 2019 raw score to t-score conversion tables manual.
#' Note: Some measures were omitted from the manual, in which case older versions were used.
#' Note: The function assumes that pasting `TB_`, the subscale, and the item numbers
#' together yields the items from each of the tests.
#'
#' See: \url{http://www.healthmeasures.net/images/nihtoolbox/Technical_Manuals/Emotion/NIH_TB_Emotion_Raw_Score_to_T-Score_Conversion_Tables_Manual_3.19.19.pdf}
#'
#' @examples
#' # Score all scales with default options
#' scores <- score_emotb(data_frame)
#'
#' # Score specific scales and keep reverse-coded items, while specifying item-level prefixes as they appear in the data
#' scores <- score_emotb(data_frame, scales = c("self", "angaf"), keep_reverse_codes = TRUE, item_prefix=c("Self_","AngAf_"))
#'
#' @importFrom dplyr case_when mutate
#'
#' @author Michael Hallquist, Melanie Glatz, Zach Vig
#' @export
score_emotb <- function(df, scales="all", max_impute=0.2, drop_items=FALSE,
keep_reverse_codes=FALSE, add_alphas=TRUE,
item_prefix=c("TB_PercStrs_",
"TB_Self_",
"TB_MP_",
"TB_Sad_",
"TB_GLS_",
"TB_PosAf_",
"TB_AngAf_",
"TB_AngPA_",
"TB_AngHost_",
"TB_EmoSup_",
"TB_FearAf_",
"TB_FearSoma_",
"TB_Friend_",
"TB_InstrSup_",
"TB_Lone_",
"TB_PercHost_",
"TB_PercRej_",
"TB_Apathy_")) {
#set-up logicals
if("all" %in% scales) {scales <- c("percstrs", "self", "mp", "sad", "gls", "posaf", "angaf", "angpa",
"anghost", "emosup", "fearaf", "fearsoma", "friend", "instrsup",
"lone", "perchost", "percrej", "apathy")}
if("percstrs" %in% scales) percstrs <- TRUE else percstrs <- FALSE
if("self" %in% scales) self <- TRUE else self <- FALSE
if("mp" %in% scales) mp <- TRUE else mp <- FALSE
if("sad" %in% scales) sad <- TRUE else sad <- FALSE
if("gls" %in% scales) gls <- TRUE else gls <- FALSE
if("posaf" %in% scales) posaf <- TRUE else posaf <- FALSE
if("angaf" %in% scales) angaf <- TRUE else angaf <- FALSE
if("angpa" %in% scales) angpa <- TRUE else angpa <- FALSE
if("anghost" %in% scales) anghost <- TRUE else anghost <- FALSE
if("emosup" %in% scales) emosup <- TRUE else emosup <- FALSE
if("fearaf" %in% scales) fearaf <- TRUE else fearaf <- FALSE
if("fearsoma" %in% scales) fearsoma <- TRUE else fearsoma <- FALSE
if("friend" %in% scales) friend <- TRUE else friend <- FALSE
if("instrsup" %in% scales) instrsup <- TRUE else instrsup <- FALSE
if("lone" %in% scales) lone <- TRUE else lone <- FALSE
if("perchost" %in% scales) perchost <- TRUE else perchost <- FALSE
if("percrej" %in% scales) percrej <- TRUE else percrej <- FALSE
if("apathy" %in% scales) apathy <- TRUE else apathy <- FALSE
#set-up prefixes
stopifnot(length(scales)==length(item_prefix))
if(percstrs) PercStrs_prefix <- paste0(item_prefix[which(scales=="percstrs")])
if(self) Self_prefix <- paste0(item_prefix[which(scales=="self")])
if(mp) MP_prefix <- paste0(item_prefix[which(scales=="mp")])
if(sad) Sad_prefix <- paste0(item_prefix[which(scales=="sad")])
if(gls) GLS_prefix <- paste0(item_prefix[which(scales=="gls")])
if(posaf) PosAf_prefix <- paste0(item_prefix[which(scales=="posaf")])
if(angaf) AngAf_prefix <- paste0(item_prefix[which(scales=="angaf")])
if(angpa) AngPA_prefix <- paste0(item_prefix[which(scales=="angpa")])
if(anghost) AngHost_prefix <- paste0(item_prefix[which(scales=="anghost")])
if(emosup) EmoSup_prefix <- paste0(item_prefix[which(scales=="emosup")])
if(fearaf) FearAf_prefix <- paste0(item_prefix[which(scales=="fearaf")])
if(fearsoma) FearSoma_prefix <- paste0(item_prefix[which(scales=="fearsoma")])
if(friend) Friend_prefix <- paste0(item_prefix[which(scales=="friend")])
if(instrsup) InstrSup_prefix <- paste0(item_prefix[which(scales=="instrsup")])
if(lone) Lone_prefix <- paste0(item_prefix[which(scales=="lone")])
if(perchost) PercHost_prefix <- paste0(item_prefix[which(scales=="perchost")])
if(percrej) PercRej_prefix <- paste0(item_prefix[which(scales=="percrej")])
if(apathy) Apathy_prefix <- paste0(item_prefix[which(scales=="apathy")])
#set-up items
if(percstrs) PercStrs_items <- paste0(PercStrs_prefix,1:10) else PercStrs_items <- NULL
if(self) Self_items <- paste0(Self_prefix,1:10) else Self_items <- NULL
if(mp) MP_items <- paste0(MP_prefix,1:7) else MP_items <- NULL
if(sad) Sad_items <- paste0(Sad_prefix,1:8) else Sad_items <- NULL
if(gls) GLS_items <- paste0(GLS_prefix,1:5) else GLS_items <- NULL
if(posaf) PosAf_items <- paste0(PosAf_prefix,1:15) else PosAf_items <- NULL
if(angaf) AngAf_items <- paste0(AngAf_prefix,1:5) else AngAf_items <- NULL
if(angpa) AngPA_items <- paste0(AngPA_prefix,1:5) else AngPA_items <- NULL
if(anghost) AngHost_items <- paste0(AngHost_prefix,1:5) else AngHost_items <- NULL
if(emosup) EmoSup_items <- paste0(EmoSup_prefix,1:8) else EmoSup_items <- NULL
if(fearaf) FearAf_items <- paste0(FearAf_prefix,1:7) else FearAf_items <- NULL
if(fearsoma) FearSoma_items <- paste0(FearSoma_prefix,1:6) else FearSoma_items <- NULL
if(friend) Friend_items <- paste0(Friend_prefix,1:8) else Friend_items <- NULL
if(instrsup) InstrSup_items <- paste0(InstrSup_prefix,1:8) else InstrSup_items <- NULL
if(lone) Lone_items <- paste0(Lone_prefix,1:5) else Lone_items <- NULL
if(perchost) PercHost_items <- paste0(PercHost_prefix,1:8) else PercHost_items <- NULL
if(percrej) PercRej_items <- paste0(PercRej_prefix,1:8) else PercRej_items <- NULL
if(apathy) Apathy_items <- paste0(Apathy_prefix,1:7) else Apathy_items <- NULL
total_items <- c(PercStrs_items, Self_items, MP_items, Sad_items, GLS_items, PosAf_items,
AngAf_items, AngPA_items, AngHost_items, EmoSup_items, FearAf_items, FearSoma_items,
Friend_items, InstrSup_items, Lone_items, PercHost_items, PercRej_items, Apathy_items)
#validate data.frame and items
stopifnot(is.data.frame(df))
stopifnot(all(total_items %in% names(df)))
#reverse coding
if(percstrs) {
for(i in c(4,5,7,8)){ PercStrs_items[i] <- paste0(PercStrs_items[i],"r") }
reverse_items <- paste0("TB_PercStrs_",c(4,5,7,8))
reverse_items_recode <- paste0(reverse_items,"r")
df[,reverse_items_recode] <- lapply(df[,reverse_items], function(x) { 6 - x }) #1-5 scoring by default
}
#mean impute, if requested
if (max_impute > 0) {
if(percstrs) df <- mean_impute_items(df, PercStrs_items, thresh=max_impute)
if(self) df <- mean_impute_items(df, Self_items, thresh=max_impute)
if(mp) df <- mean_impute_items(df, MP_items, thresh=max_impute)
if(sad) df <- mean_impute_items(df, Sad_items, thresh=max_impute)
if(gls) df <- mean_impute_items(df, GLS_items, thresh=max_impute)
if(posaf) df <- mean_impute_items(df, PosAf_items, thresh=max_impute)
if(angaf) df <- mean_impute_items(df, AngAf_items, thresh=max_impute)
if(angpa) df <- mean_impute_items(df, AngPA_items, thresh=max_impute)
if(anghost) df <- mean_impute_items(df, AngHost_items, thresh=max_impute)
if(emosup) df <- mean_impute_items(df, EmoSup_items, thresh=max_impute)
if(fearaf) df <- mean_impute_items(df, FearAf_items, thresh=max_impute)
if(fearsoma) df <- mean_impute_items(df, FearSoma_items, thresh=max_impute)
if(friend) df <- mean_impute_items(df, Friend_items, thresh=max_impute)
if(instrsup) df <- mean_impute_items(df, InstrSup_items, thresh=max_impute)
if(lone) df <- mean_impute_items(df, Lone_items, thresh=max_impute)
if(perchost) df <- mean_impute_items(df, PercHost_items, thresh=max_impute)
if(percrej) df <- mean_impute_items(df, PercRej_items, thresh=max_impute)
if(apathy) df <- mean_impute_items(df, Apathy_items, thresh=max_impute)
}
#compute row sums
if(percstrs) df <- df %>% mutate(PercStrs_Raw = rowSums(across(all_of(PercStrs_items))))
if(self) df <- df %>% mutate(Self_Raw = rowSums(across(all_of(Self_items))))
if(mp) df <- df %>% mutate(MP_Raw = rowSums(across(all_of(MP_items))))
if(sad) df <- df %>% mutate(Sad_Raw = rowSums(across(all_of(Sad_items))))
if(gls) df <- df %>% mutate(GLS_Raw = rowSums(across(all_of(GLS_items))))
if(posaf) df <- df %>% mutate(PosAf_Raw = rowSums(across(all_of(PosAf_items))))
if(angaf) df <- df %>% mutate(AngAf_Raw = rowSums(across(all_of(AngAf_items))))
if(angpa) df <- df %>% mutate(AngPA_Raw = rowSums(across(all_of(AngPA_items))))
if(anghost) df <- df %>% mutate(AngHost_Raw = rowSums(across(all_of(AngHost_items))))
if(emosup) df <- df %>% mutate(EmoSup_Raw = rowSums(across(all_of(EmoSup_items))))
if(fearaf) df <- df %>% mutate(FearAf_Raw = rowSums(across(all_of(FearAf_items))))
if(fearsoma) df <- df %>% mutate(FearSoma_Raw = rowSums(across(all_of(FearSoma_items))))
if(friend) df <- df %>% mutate(Friend_Raw = rowSums(across(all_of(Friend_items))))
if(instrsup) df <- df %>% mutate(InstrSup_Raw = rowSums(across(all_of(InstrSup_items))))
if(lone) df <- df %>% mutate(Lone_Raw = rowSums(across(all_of(Lone_items))))
if(perchost) df <- df %>% mutate(PercHost_Raw = rowSums(across(all_of(PercHost_items))))
if(percrej) df <- df %>% mutate(PercRej_Raw = rowSums(across(all_of(PercRej_items))))
if(apathy) df <- df %>% mutate(Apathy_Raw = rowSums(across(all_of(Apathy_items))))
#compute Theta scores
if(percstrs) df <- df %>% mutate(PercStrs_Theta = case_when(PercStrs_Raw%%1==0~percstrs_adult(PercStrs_Raw), PercStrs_Raw%%1!=0~percstrs_adult_approx(PercStrs_Raw)))
if(self) df <- df %>% mutate(Self_Theta = case_when(Self_Raw%%1==0~self_adult(Self_Raw),Self_Raw%%1!=0~self_adult_approx(Self_Raw)))
if(mp) df <- df %>% mutate(MP_Theta = case_when(MP_Raw%%1==0~mp_adult(MP_Raw),MP_Raw%%1!=0~mp_adult_approx(MP_Raw)))
if(sad) df <- df %>% mutate(Sad_Theta = case_when(Sad_Raw%%1==0~sad_adult(Sad_Raw),Sad_Raw%%1!=0~sad_adult_approx(Sad_Raw)))
if(gls) df <- df %>% mutate(GLS_Theta = case_when(GLS_Raw%%1==0~gls_adult(GLS_Raw),GLS_Raw%%1!=0~gls_adult_approx(GLS_Raw)))
if(posaf) df <- df %>% mutate(PosAf_Theta = case_when(PosAf_Raw%%1==0~posaf_adult(PosAf_Raw),PosAf_Raw%%1!=0~posaf_adult_approx(PosAf_Raw)))
if(angaf) df <- df %>% mutate(AngAf_Theta = case_when(AngAf_Raw%%1==0~angaf_adult(AngAf_Raw),AngAf_Raw%%1!=0~angaf_adult_approx(AngAf_Raw)))
if(angpa) df <- df %>% mutate(AngPA_Theta = case_when(AngPA_Raw%%1==0~angpa_adult(AngPA_Raw),AngPA_Raw%%1!=0~angpa_adult_approx(AngPA_Raw)))
if(anghost) df <- df %>% mutate(AngHost_Theta = case_when(AngHost_Raw%%1==0~anghost_adult(AngHost_Raw),AngHost_Raw%%1!=0~anghost_adult_approx(AngHost_Raw)))
if(emosup) df <- df %>% mutate(EmoSup_Theta = case_when(EmoSup_Raw%%1==0~emosup_adult(EmoSup_Raw),EmoSup_Raw%%1!=0~emosup_adult_approx(EmoSup_Raw)))
if(fearaf) df <- df %>% mutate(FearAf_Theta = case_when(FearAf_Raw%%1==0~fearaf_adult(FearAf_Raw),FearAf_Raw%%1!=0~fearaf_adult_approx(FearAf_Raw)))
if(fearsoma) df <- df %>% mutate(FearSoma_Theta = case_when(FearSoma_Raw%%1==0~fearsoma_adult(FearSoma_Raw),FearSoma_Raw%%1!=0~fearsoma_adult_approx(FearSoma_Raw)))
if(friend) df <- df %>% mutate(Friend_Theta = case_when(Friend_Raw%%1==0~friend_adult(Friend_Raw),Friend_Raw%%1!=0~friend_adult_approx(Friend_Raw)))
if(instrsup) df <- df %>% mutate(InstrSup_Theta = case_when(InstrSup_Raw%%1==0~instrsup_adult(InstrSup_Raw),InstrSup_Raw%%1!=0~instrsup_adult_approx(InstrSup_Raw)))
if(lone) df <- df %>% mutate(Lone_Theta = case_when(Lone_Raw%%1==0~lone_adult(Lone_Raw),Lone_Raw%%1!=0~lone_adult_approx(Lone_Raw)))
if(perchost) df <- df %>% mutate(PercHost_Theta = case_when(PercHost_Raw%%1==0~perchost_adult(PercHost_Raw),PercHost_Raw%%1!=0~perchost_adult_approx(PercHost_Raw)))
if(percrej) df <- df %>% mutate(PercRej_Theta = case_when(PercRej_Raw%%1==0~percrej_adult(PercRej_Raw),PercRej_Raw%%1!=0~percrej_adult_approx(PercRej_Raw)))
if(apathy) df <- df %>% mutate(Apathy_Theta = case_when(Apathy_Raw%%1==0~apathy_adult(Apathy_Raw),Apathy_Raw%%1!=0~apathy_adult_approx(Apathy_Raw)))
#compute T scores
theta_to_t <- function(x) { 50 + 10*x }
df <- df %>% mutate(across(ends_with("_Theta"), .fns = list(T=~theta_to_t(.)), .names = "{gsub('_Theta','_T',{col})}"))
#compute alphas
if (add_alphas) {
if(percstrs) attr(df[["PercStrs_Raw"]],"alpha") <- psych::alpha(df[,PercStrs_items],max=100,warnings = F)$total
if(self) attr(df[["Self_Raw"]],"alpha") <- psych::alpha(df[,Self_items],max=100,warnings = F)$total
if(mp) attr(df[["MP_Raw"]],"alpha") <- psych::alpha(df[,MP_items],max=100,warnings = F)$total
if(sad) attr(df[["Sad_Raw"]],"alpha") <- psych::alpha(df[,Sad_items],max=100,warnings = F)$total
if(gls) attr(df[["GLS_Raw"]],"alpha") <- psych::alpha(df[,GLS_items],max=100,warnings = F)$total
if(posaf) attr(df[["PosAf_Raw"]],"alpha") <- psych::alpha(df[,PosAf_items],max=100,warnings = F)$total
if(angaf) attr(df[["AngAf_Raw"]],"alpha") <- psych::alpha(df[,AngAf_items],max=100,warnings = F)$total
if(angpa) attr(df[["AngPA_Raw"]],"alpha") <- psych::alpha(df[,AngPA_items],max=100,warnings = F)$total
if(anghost) attr(df[["AngHost_Raw"]],"alpha") <- psych::alpha(df[,AngHost_items],max=100,warnings = F)$total
if(emosup) attr(df[["EmoSup_Raw"]],"alpha") <- psych::alpha(df[,EmoSup_items],max=100,warnings = F)$total
if(fearaf) attr(df[["FearAf_Raw"]],"alpha") <- psych::alpha(df[,FearAf_items],max=100,warnings = F)$total
if(fearsoma) attr(df[["FearSoma_Raw"]],"alpha") <- psych::alpha(df[,FearSoma_items],max=100,warnings = F)$total
if(friend) attr(df[["Friend_Raw"]],"alpha") <- psych::alpha(df[,Friend_items],max=100,warnings = F)$total
if(instrsup) attr(df[["InstrSup_Raw"]],"alpha") <- psych::alpha(df[,InstrSup_items],max=100,warnings = F)$total
if(lone) attr(df[["Lone_Raw"]],"alpha") <- psych::alpha(df[,Lone_items],max=100,warnings = F)$total
if(perchost) attr(df[["PercHost_Raw"]],"alpha") <- psych::alpha(df[,PercHost_items],max=100,warnings = F)$total
if(percrej) attr(df[["PercRej_Raw"]],"alpha") <- psych::alpha(df[,PercRej_items],max=100,warnings = F)$total
if(apathy) attr(df[["Apathy_Raw"]],"alpha") <- psych::alpha(df[,Apathy_items],max=100,warnings = F)$total
}
#drop reverse codes and item-level data
if (!keep_reverse_codes & percstrs) { df <- df %>% select(-all_of(reverse_items_recode)) }
if (drop_items) { df <- df %>% select(-starts_with("TB_")) }
return(df)
}
#######################
## ORIGINAL FUNCTION ##
#######################
# score_emotb <- function(df, drop_items=FALSE) {
# df <- df %>% mutate(Apathy_Theta = if_else(Age >= 18, apathy_adult(Apathy_Raw), NA_real_),
# PercStrs_Theta = if_else(Age >= 18, percstrs_adult(PercStrs_Raw), percstrs_teen(PercStrs_Raw)),
# Self_Theta = if_else(Age >= 18, self_adult(Self_Raw), if_else(Age <= 12, self_child(Self_Raw), self_teen(Self_Raw))),
# MP_Theta = if_else(Age >= 18, mp_adult(MP_Raw), NA_real_),
# Sad_Theta = if_else(Age >= 18, sad_adult(Sad_Raw), sad_teen(Sad_Raw)),
# GLS_Theta = if_else(Age >= 18, gls_adult(GLS_Raw), if_else(Age <= 12, gls_child(GLS_Raw), gls_teen(GLS_Raw))),
# PosAf_Theta = if_else(Age >= 18, posaf_adult(PosAf_Raw), if_else(Age <= 12, posaf_child(PosAf_Raw), posaf_teen(PosAf_Raw))),
# AngAf_Theta = if_else(Age >= 18, angaf_adult(AngAf_Raw), angaf_teen(AngAf_Raw)),
# AngPA_Theta = if_else(Age >= 18, angpa_adult(AngPA_Raw), NA_real_),
# AngHost_Theta = if_else(Age >= 18, anghost_adult(AngHost_Raw), NA_real_),
# EmoSup_Theta = if_else(Age >= 18, emosup_adult(EmoSup_Raw), emosup_teen(EmoSup_Raw)),
# FearAf_Theta = if_else(Age >= 18, fearaf_adult(FearAf_Raw), fearaf_teen(FearAf_Raw)),
# FearSoma_Theta = if_else(Age >= 18, fearsoma_adult(FearSoma_Raw), NA_real_),
# Friend_Theta = if_else(Age >= 18, friend_adult(Friend_Raw), friend_teen(Friend_Raw)),
# InstrSup_Theta = if_else(Age >= 18, instrsup_adult(InstrSup_Raw), NA_real_),
# Lone_Theta = if_else(Age >= 18, lone_adult(Lone_Raw), lone_teen(Lone_Raw)),
# PercHost_Theta = if_else(Age >= 18, perchost_adult(PercHost_Raw), perchost_teen(PercHost_Raw)),
# PercRej_Theta = if_else(Age >= 18, percrej_adult(PercRej_Raw), percrej_teen(PercRej_Raw)))
# theta_to_t <- function(x) { 50 + 10*x }
# df <- df %>% mutate_at(vars(ends_with("_Theta")), list(T=~theta_to_t(.)))
#
# if (drop_items) { df <- df %>% select(-starts_with("TB_")) }
# return(df)
# }
PennStateDEPENdLab/dependlab documentation built on April 10, 2024, 5:15 p.m.
R Package Documentation
Browse R Packages
We want your feedback!
Note that we can't provide technical support on individual packages. You should contact the package authors for that.
Defines functions score_emotb
Documented in score_emotb
#' Score Adult EMOTB
#'
#' Computes scores for the NIH Toolbox Emotion Measures, Adult Forms (ages 18-52) (2019).
#'
#' @param df The data frame containing the assessment items.
#' @param scales A character vector specifying the scales to include in the scoring.
#' The default is "all", which includes all scales. Other possible values are:
#' "percstrs", "self", "mp", "sad", "gls", "posaf", "angaf", "angpa",
#' "anghost", "emosup", "fearaf", "fearsoma", "friend", "instrsup",
#' "lone", "perchost", "percrej", and "apathy".
#' @param max_impute The threshold for mean imputation. Items with missing values
#' below this threshold will be mean imputed. The default is 0.2.
#' @param drop_items Logical indicating whether to drop the individual item columns
#' from the resulting data frame. The default is FALSE.
#' @param keep_reverse_codes Logical indicating whether to keep the reverse-coded
#' items in the resulting data frame. The default is FALSE.
#' @param add_alphas Logical indicating whether to compute Cronbach's alpha coefficients
#' for the scales and append them to the resulting data frame. The default is TRUE.
#' @param item_prefix A character vector specifying the item prefixes as they appear in the data.frame.
#' The vector must be the same length as the `scales` parameter and prefixes must be in the
#' same corresponding order. If `scales` is set to "all", the prefixes must appear in
#' the order of possible values for the `scales` parameter provided above
#' (Perceived Stress, Self Efficacy, etc.).
#'
#' @return A data frame with the computed scores and optional alpha coefficients.
#'
#' @details The function scores the Adult EMOTB assessment based on the provided data frame.
#' It supports various scoring options, including selecting specific scales,
#' mean imputation of missing values, dropping individual item columns,
#' keeping reverse-coded items, and computing alpha coefficients.
#'
#' The function adds three columns per scale to the data frame `df`, containing
#' the raw scores (\code{_Raw}), the theta scores (\code{_Theta}), and the
#' t-scores (\code{_T}) for the NIH Toolbox Emotion Measures, respectively.
#'
#' Note: The scoring is based on the 2019 raw score to t-score conversion tables manual.
#' Note: Some measures were omitted from the manual, in which case older versions were used.
#' Note: The function assumes that pasting `TB_`, the subscale, and the item numbers
#' together yields the items from each of the tests.
#'
#' See: \url{http://www.healthmeasures.net/images/nihtoolbox/Technical_Manuals/Emotion/NIH_TB_Emotion_Raw_Score_to_T-Score_Conversion_Tables_Manual_3.19.19.pdf}
#'
#' @examples
#' # Score all scales with default options
#' scores <- score_emotb(data_frame)
#'
#' # Score specific scales and keep reverse-coded items, while specifying item-level prefixes as they appear in the data
#' scores <- score_emotb(data_frame, scales = c("self", "angaf"), keep_reverse_codes = TRUE, item_prefix=c("Self_","AngAf_"))
#'
#' @importFrom dplyr case_when mutate
#'
#' @author Michael Hallquist, Melanie Glatz, Zach Vig
#' @export
score_emotb <- function(df, scales="all", max_impute=0.2, drop_items=FALSE,
keep_reverse_codes=FALSE, add_alphas=TRUE,
item_prefix=c("TB_PercStrs_",
"TB_Self_",
"TB_MP_",
"TB_Sad_",
"TB_GLS_",
"TB_PosAf_",
"TB_AngAf_",
"TB_AngPA_",
"TB_AngHost_",
"TB_EmoSup_",
"TB_FearAf_",
"TB_FearSoma_",
"TB_Friend_",
"TB_InstrSup_",
"TB_Lone_",
"TB_PercHost_",
"TB_PercRej_",
"TB_Apathy_")) {
#set-up logicals
if("all" %in% scales) {scales <- c("percstrs", "self", "mp", "sad", "gls", "posaf", "angaf", "angpa",
"anghost", "emosup", "fearaf", "fearsoma", "friend", "instrsup",
"lone", "perchost", "percrej", "apathy")}
if("percstrs" %in% scales) percstrs <- TRUE else percstrs <- FALSE
if("self" %in% scales) self <- TRUE else self <- FALSE
if("mp" %in% scales) mp <- TRUE else mp <- FALSE
if("sad" %in% scales) sad <- TRUE else sad <- FALSE
if("gls" %in% scales) gls <- TRUE else gls <- FALSE
if("posaf" %in% scales) posaf <- TRUE else posaf <- FALSE
if("angaf" %in% scales) angaf <- TRUE else angaf <- FALSE
if("angpa" %in% scales) angpa <- TRUE else angpa <- FALSE
if("anghost" %in% scales) anghost <- TRUE else anghost <- FALSE
if("emosup" %in% scales) emosup <- TRUE else emosup <- FALSE
if("fearaf" %in% scales) fearaf <- TRUE else fearaf <- FALSE
if("fearsoma" %in% scales) fearsoma <- TRUE else fearsoma <- FALSE
if("friend" %in% scales) friend <- TRUE else friend <- FALSE
if("instrsup" %in% scales) instrsup <- TRUE else instrsup <- FALSE
if("lone" %in% scales) lone <- TRUE else lone <- FALSE
if("perchost" %in% scales) perchost <- TRUE else perchost <- FALSE
if("percrej" %in% scales) percrej <- TRUE else percrej <- FALSE
if("apathy" %in% scales) apathy <- TRUE else apathy <- FALSE
#set-up prefixes
stopifnot(length(scales)==length(item_prefix))
if(percstrs) PercStrs_prefix <- paste0(item_prefix[which(scales=="percstrs")])
if(self) Self_prefix <- paste0(item_prefix[which(scales=="self")])
if(mp) MP_prefix <- paste0(item_prefix[which(scales=="mp")])
if(sad) Sad_prefix <- paste0(item_prefix[which(scales=="sad")])
if(gls) GLS_prefix <- paste0(item_prefix[which(scales=="gls")])
if(posaf) PosAf_prefix <- paste0(item_prefix[which(scales=="posaf")])
if(angaf) AngAf_prefix <- paste0(item_prefix[which(scales=="angaf")])
if(angpa) AngPA_prefix <- paste0(item_prefix[which(scales=="angpa")])
if(anghost) AngHost_prefix <- paste0(item_prefix[which(scales=="anghost")])
if(emosup) EmoSup_prefix <- paste0(item_prefix[which(scales=="emosup")])
if(fearaf) FearAf_prefix <- paste0(item_prefix[which(scales=="fearaf")])
if(fearsoma) FearSoma_prefix <- paste0(item_prefix[which(scales=="fearsoma")])
if(friend) Friend_prefix <- paste0(item_prefix[which(scales=="friend")])
if(instrsup) InstrSup_prefix <- paste0(item_prefix[which(scales=="instrsup")])
if(lone) Lone_prefix <- paste0(item_prefix[which(scales=="lone")])
if(perchost) PercHost_prefix <- paste0(item_prefix[which(scales=="perchost")])
if(percrej) PercRej_prefix <- paste0(item_prefix[which(scales=="percrej")])
if(apathy) Apathy_prefix <- paste0(item_prefix[which(scales=="apathy")])
#set-up items
if(percstrs) PercStrs_items <- paste0(PercStrs_prefix,1:10) else PercStrs_items <- NULL
if(self) Self_items <- paste0(Self_prefix,1:10) else Self_items <- NULL
if(mp) MP_items <- paste0(MP_prefix,1:7) else MP_items <- NULL
if(sad) Sad_items <- paste0(Sad_prefix,1:8) else Sad_items <- NULL
if(gls) GLS_items <- paste0(GLS_prefix,1:5) else GLS_items <- NULL
if(posaf) PosAf_items <- paste0(PosAf_prefix,1:15) else PosAf_items <- NULL
if(angaf) AngAf_items <- paste0(AngAf_prefix,1:5) else AngAf_items <- NULL
if(angpa) AngPA_items <- paste0(AngPA_prefix,1:5) else AngPA_items <- NULL
if(anghost) AngHost_items <- paste0(AngHost_prefix,1:5) else AngHost_items <- NULL
if(emosup) EmoSup_items <- paste0(EmoSup_prefix,1:8) else EmoSup_items <- NULL
if(fearaf) FearAf_items <- paste0(FearAf_prefix,1:7) else FearAf_items <- NULL
if(fearsoma) FearSoma_items <- paste0(FearSoma_prefix,1:6) else FearSoma_items <- NULL
if(friend) Friend_items <- paste0(Friend_prefix,1:8) else Friend_items <- NULL
if(instrsup) InstrSup_items <- paste0(InstrSup_prefix,1:8) else InstrSup_items <- NULL
if(lone) Lone_items <- paste0(Lone_prefix,1:5) else Lone_items <- NULL
if(perchost) PercHost_items <- paste0(PercHost_prefix,1:8) else PercHost_items <- NULL
if(percrej) PercRej_items <- paste0(PercRej_prefix,1:8) else PercRej_items <- NULL
if(apathy) Apathy_items <- paste0(Apathy_prefix,1:7) else Apathy_items <- NULL
total_items <- c(PercStrs_items, Self_items, MP_items, Sad_items, GLS_items, PosAf_items,
AngAf_items, AngPA_items, AngHost_items, EmoSup_items, FearAf_items, FearSoma_items,
Friend_items, InstrSup_items, Lone_items, PercHost_items, PercRej_items, Apathy_items)
#validate data.frame and items
stopifnot(is.data.frame(df))
stopifnot(all(total_items %in% names(df)))
#reverse coding
if(percstrs) {
for(i in c(4,5,7,8)){ PercStrs_items[i] <- paste0(PercStrs_items[i],"r") }
reverse_items <- paste0("TB_PercStrs_",c(4,5,7,8))
reverse_items_recode <- paste0(reverse_items,"r")
df[,reverse_items_recode] <- lapply(df[,reverse_items], function(x) { 6 - x }) #1-5 scoring by default
}
#mean impute, if requested
if (max_impute > 0) {
if(percstrs) df <- mean_impute_items(df, PercStrs_items, thresh=max_impute)
if(self) df <- mean_impute_items(df, Self_items, thresh=max_impute)
if(mp) df <- mean_impute_items(df, MP_items, thresh=max_impute)
if(sad) df <- mean_impute_items(df, Sad_items, thresh=max_impute)
if(gls) df <- mean_impute_items(df, GLS_items, thresh=max_impute)
if(posaf) df <- mean_impute_items(df, PosAf_items, thresh=max_impute)
if(angaf) df <- mean_impute_items(df, AngAf_items, thresh=max_impute)
if(angpa) df <- mean_impute_items(df, AngPA_items, thresh=max_impute)
if(anghost) df <- mean_impute_items(df, AngHost_items, thresh=max_impute)
if(emosup) df <- mean_impute_items(df, EmoSup_items, thresh=max_impute)
if(fearaf) df <- mean_impute_items(df, FearAf_items, thresh=max_impute)
if(fearsoma) df <- mean_impute_items(df, FearSoma_items, thresh=max_impute)
if(friend) df <- mean_impute_items(df, Friend_items, thresh=max_impute)
if(instrsup) df <- mean_impute_items(df, InstrSup_items, thresh=max_impute)
if(lone) df <- mean_impute_items(df, Lone_items, thresh=max_impute)
if(perchost) df <- mean_impute_items(df, PercHost_items, thresh=max_impute)
if(percrej) df <- mean_impute_items(df, PercRej_items, thresh=max_impute)
if(apathy) df <- mean_impute_items(df, Apathy_items, thresh=max_impute)
}
#compute row sums
if(percstrs) df <- df %>% mutate(PercStrs_Raw = rowSums(across(all_of(PercStrs_items))))
if(self) df <- df %>% mutate(Self_Raw = rowSums(across(all_of(Self_items))))
if(mp) df <- df %>% mutate(MP_Raw = rowSums(across(all_of(MP_items))))
if(sad) df <- df %>% mutate(Sad_Raw = rowSums(across(all_of(Sad_items))))
if(gls) df <- df %>% mutate(GLS_Raw = rowSums(across(all_of(GLS_items))))
if(posaf) df <- df %>% mutate(PosAf_Raw = rowSums(across(all_of(PosAf_items))))
if(angaf) df <- df %>% mutate(AngAf_Raw = rowSums(across(all_of(AngAf_items))))
if(angpa) df <- df %>% mutate(AngPA_Raw = rowSums(across(all_of(AngPA_items))))
if(anghost) df <- df %>% mutate(AngHost_Raw = rowSums(across(all_of(AngHost_items))))
if(emosup) df <- df %>% mutate(EmoSup_Raw = rowSums(across(all_of(EmoSup_items))))
if(fearaf) df <- df %>% mutate(FearAf_Raw = rowSums(across(all_of(FearAf_items))))
if(fearsoma) df <- df %>% mutate(FearSoma_Raw = rowSums(across(all_of(FearSoma_items))))
if(friend) df <- df %>% mutate(Friend_Raw = rowSums(across(all_of(Friend_items))))
if(instrsup) df <- df %>% mutate(InstrSup_Raw = rowSums(across(all_of(InstrSup_items))))
if(lone) df <- df %>% mutate(Lone_Raw = rowSums(across(all_of(Lone_items))))
if(perchost) df <- df %>% mutate(PercHost_Raw = rowSums(across(all_of(PercHost_items))))
if(percrej) df <- df %>% mutate(PercRej_Raw = rowSums(across(all_of(PercRej_items))))
if(apathy) df <- df %>% mutate(Apathy_Raw = rowSums(across(all_of(Apathy_items))))
#compute Theta scores
if(percstrs) df <- df %>% mutate(PercStrs_Theta = case_when(PercStrs_Raw%%1==0~percstrs_adult(PercStrs_Raw), PercStrs_Raw%%1!=0~percstrs_adult_approx(PercStrs_Raw)))
if(self) df <- df %>% mutate(Self_Theta = case_when(Self_Raw%%1==0~self_adult(Self_Raw),Self_Raw%%1!=0~self_adult_approx(Self_Raw)))
if(mp) df <- df %>% mutate(MP_Theta = case_when(MP_Raw%%1==0~mp_adult(MP_Raw),MP_Raw%%1!=0~mp_adult_approx(MP_Raw)))
if(sad) df <- df %>% mutate(Sad_Theta = case_when(Sad_Raw%%1==0~sad_adult(Sad_Raw),Sad_Raw%%1!=0~sad_adult_approx(Sad_Raw)))
if(gls) df <- df %>% mutate(GLS_Theta = case_when(GLS_Raw%%1==0~gls_adult(GLS_Raw),GLS_Raw%%1!=0~gls_adult_approx(GLS_Raw)))
if(posaf) df <- df %>% mutate(PosAf_Theta = case_when(PosAf_Raw%%1==0~posaf_adult(PosAf_Raw),PosAf_Raw%%1!=0~posaf_adult_approx(PosAf_Raw)))
if(angaf) df <- df %>% mutate(AngAf_Theta = case_when(AngAf_Raw%%1==0~angaf_adult(AngAf_Raw),AngAf_Raw%%1!=0~angaf_adult_approx(AngAf_Raw)))
if(angpa) df <- df %>% mutate(AngPA_Theta = case_when(AngPA_Raw%%1==0~angpa_adult(AngPA_Raw),AngPA_Raw%%1!=0~angpa_adult_approx(AngPA_Raw)))
if(anghost) df <- df %>% mutate(AngHost_Theta = case_when(AngHost_Raw%%1==0~anghost_adult(AngHost_Raw),AngHost_Raw%%1!=0~anghost_adult_approx(AngHost_Raw)))
if(emosup) df <- df %>% mutate(EmoSup_Theta = case_when(EmoSup_Raw%%1==0~emosup_adult(EmoSup_Raw),EmoSup_Raw%%1!=0~emosup_adult_approx(EmoSup_Raw)))
if(fearaf) df <- df %>% mutate(FearAf_Theta = case_when(FearAf_Raw%%1==0~fearaf_adult(FearAf_Raw),FearAf_Raw%%1!=0~fearaf_adult_approx(FearAf_Raw)))
if(fearsoma) df <- df %>% mutate(FearSoma_Theta = case_when(FearSoma_Raw%%1==0~fearsoma_adult(FearSoma_Raw),FearSoma_Raw%%1!=0~fearsoma_adult_approx(FearSoma_Raw)))
if(friend) df <- df %>% mutate(Friend_Theta = case_when(Friend_Raw%%1==0~friend_adult(Friend_Raw),Friend_Raw%%1!=0~friend_adult_approx(Friend_Raw)))
if(instrsup) df <- df %>% mutate(InstrSup_Theta = case_when(InstrSup_Raw%%1==0~instrsup_adult(InstrSup_Raw),InstrSup_Raw%%1!=0~instrsup_adult_approx(InstrSup_Raw)))
if(lone) df <- df %>% mutate(Lone_Theta = case_when(Lone_Raw%%1==0~lone_adult(Lone_Raw),Lone_Raw%%1!=0~lone_adult_approx(Lone_Raw)))
if(perchost) df <- df %>% mutate(PercHost_Theta = case_when(PercHost_Raw%%1==0~perchost_adult(PercHost_Raw),PercHost_Raw%%1!=0~perchost_adult_approx(PercHost_Raw)))
if(percrej) df <- df %>% mutate(PercRej_Theta = case_when(PercRej_Raw%%1==0~percrej_adult(PercRej_Raw),PercRej_Raw%%1!=0~percrej_adult_approx(PercRej_Raw)))
if(apathy) df <- df %>% mutate(Apathy_Theta = case_when(Apathy_Raw%%1==0~apathy_adult(Apathy_Raw),Apathy_Raw%%1!=0~apathy_adult_approx(Apathy_Raw)))
#compute T scores
theta_to_t <- function(x) { 50 + 10*x }
df <- df %>% mutate(across(ends_with("_Theta"), .fns = list(T=~theta_to_t(.)), .names = "{gsub('_Theta','_T',{col})}"))
#compute alphas
if (add_alphas) {
if(percstrs) attr(df[["PercStrs_Raw"]],"alpha") <- psych::alpha(df[,PercStrs_items],max=100,warnings = F)$total
if(self) attr(df[["Self_Raw"]],"alpha") <- psych::alpha(df[,Self_items],max=100,warnings = F)$total
if(mp) attr(df[["MP_Raw"]],"alpha") <- psych::alpha(df[,MP_items],max=100,warnings = F)$total
if(sad) attr(df[["Sad_Raw"]],"alpha") <- psych::alpha(df[,Sad_items],max=100,warnings = F)$total
if(gls) attr(df[["GLS_Raw"]],"alpha") <- psych::alpha(df[,GLS_items],max=100,warnings = F)$total
if(posaf) attr(df[["PosAf_Raw"]],"alpha") <- psych::alpha(df[,PosAf_items],max=100,warnings = F)$total
if(angaf) attr(df[["AngAf_Raw"]],"alpha") <- psych::alpha(df[,AngAf_items],max=100,warnings = F)$total
if(angpa) attr(df[["AngPA_Raw"]],"alpha") <- psych::alpha(df[,AngPA_items],max=100,warnings = F)$total
if(anghost) attr(df[["AngHost_Raw"]],"alpha") <- psych::alpha(df[,AngHost_items],max=100,warnings = F)$total
if(emosup) attr(df[["EmoSup_Raw"]],"alpha") <- psych::alpha(df[,EmoSup_items],max=100,warnings = F)$total
if(fearaf) attr(df[["FearAf_Raw"]],"alpha") <- psych::alpha(df[,FearAf_items],max=100,warnings = F)$total
if(fearsoma) attr(df[["FearSoma_Raw"]],"alpha") <- psych::alpha(df[,FearSoma_items],max=100,warnings = F)$total
if(friend) attr(df[["Friend_Raw"]],"alpha") <- psych::alpha(df[,Friend_items],max=100,warnings = F)$total
if(instrsup) attr(df[["InstrSup_Raw"]],"alpha") <- psych::alpha(df[,InstrSup_items],max=100,warnings = F)$total
if(lone) attr(df[["Lone_Raw"]],"alpha") <- psych::alpha(df[,Lone_items],max=100,warnings = F)$total
if(perchost) attr(df[["PercHost_Raw"]],"alpha") <- psych::alpha(df[,PercHost_items],max=100,warnings = F)$total
if(percrej) attr(df[["PercRej_Raw"]],"alpha") <- psych::alpha(df[,PercRej_items],max=100,warnings = F)$total
if(apathy) attr(df[["Apathy_Raw"]],"alpha") <- psych::alpha(df[,Apathy_items],max=100,warnings = F)$total
}
#drop reverse codes and item-level data
if (!keep_reverse_codes & percstrs) { df <- df %>% select(-all_of(reverse_items_recode)) }
if (drop_items) { df <- df %>% select(-starts_with("TB_")) }
return(df)
}
#######################
## ORIGINAL FUNCTION ##
#######################
# score_emotb <- function(df, drop_items=FALSE) {
# df <- df %>% mutate(Apathy_Theta = if_else(Age >= 18, apathy_adult(Apathy_Raw), NA_real_),
# PercStrs_Theta = if_else(Age >= 18, percstrs_adult(PercStrs_Raw), percstrs_teen(PercStrs_Raw)),
# Self_Theta = if_else(Age >= 18, self_adult(Self_Raw), if_else(Age <= 12, self_child(Self_Raw), self_teen(Self_Raw))),
# MP_Theta = if_else(Age >= 18, mp_adult(MP_Raw), NA_real_),
# Sad_Theta = if_else(Age >= 18, sad_adult(Sad_Raw), sad_teen(Sad_Raw)),
# GLS_Theta = if_else(Age >= 18, gls_adult(GLS_Raw), if_else(Age <= 12, gls_child(GLS_Raw), gls_teen(GLS_Raw))),
# PosAf_Theta = if_else(Age >= 18, posaf_adult(PosAf_Raw), if_else(Age <= 12, posaf_child(PosAf_Raw), posaf_teen(PosAf_Raw))),
# AngAf_Theta = if_else(Age >= 18, angaf_adult(AngAf_Raw), angaf_teen(AngAf_Raw)),
# AngPA_Theta = if_else(Age >= 18, angpa_adult(AngPA_Raw), NA_real_),
# AngHost_Theta = if_else(Age >= 18, anghost_adult(AngHost_Raw), NA_real_),
# EmoSup_Theta = if_else(Age >= 18, emosup_adult(EmoSup_Raw), emosup_teen(EmoSup_Raw)),
# FearAf_Theta = if_else(Age >= 18, fearaf_adult(FearAf_Raw), fearaf_teen(FearAf_Raw)),
# FearSoma_Theta = if_else(Age >= 18, fearsoma_adult(FearSoma_Raw), NA_real_),
# Friend_Theta = if_else(Age >= 18, friend_adult(Friend_Raw), friend_teen(Friend_Raw)),
# InstrSup_Theta = if_else(Age >= 18, instrsup_adult(InstrSup_Raw), NA_real_),
# Lone_Theta = if_else(Age >= 18, lone_adult(Lone_Raw), lone_teen(Lone_Raw)),
# PercHost_Theta = if_else(Age >= 18, perchost_adult(PercHost_Raw), perchost_teen(PercHost_Raw)),
# PercRej_Theta = if_else(Age >= 18, percrej_adult(PercRej_Raw), percrej_teen(PercRej_Raw)))
# theta_to_t <- function(x) { 50 + 10*x }
# df <- df %>% mutate_at(vars(ends_with("_Theta")), list(T=~theta_to_t(.)))
#
# if (drop_items) { df <- df %>% select(-starts_with("TB_")) }
# return(df)
# }
R Package Documentation
Browse R Packages
We want your feedback!
Note that we can't provide technical support on individual packages. You should contact the package authors for that.
Embedding an R snippet on your website
Add the following code to your website.
For more information on customizing the embed code, read Embedding Snippets.