In StatisMike/stenR: Standardization of Raw Discrete Questionnaire Scores

knitr::opts_chunk$set(
  collapse = TRUE,
  comment = "#>",
  fig.path = "man/figures/README-",
  fig.width = 6,
  fig.height = 4
)
Sys.setenv(LANGUAGE='en')
library(stenR)

stenR

stenR is a package tailored mainly for users and creators of psychological questionnaires, though other social science researchers and survey authors can benefit greatly from it.

It provides tools to help with processes necessary for conducting such studies:

1. processing data from raw item scores to raw factor/scale scores
2. standardization of the raw scores into standard scale of your choosing, either by:
3. normalization of the raw scores using frequency table (if no norms have been developed before). Usually for authors of questionnaires or their adaptations.
4. importing scoring table developed by questionnaire authors - for researchers only using the measure

Furthermore, tools for developing or using norms on grouped basis are also provided (up to two intertwined grouping conditions are supported).

Most in-depth information is provided within vignette Tour from data to results (for basic and verbose explanation) and less verbose but more complete in stenR usage vignette.

Installation

You can install the current version from GitHub with:

# install.packages("devtools")
devtools::install_github("StatisMike/stenR")

Usage

Data processing

Process data from item raw scores to factor/scales scores

# Exemplary data provided within package
str(SLCS)

# create scale specifications
SL_spec <- ScaleSpec(
  name = "Self-Liking",
  item_names = c("SLCS_1", "SLCS_3", "SLCS_5", "SLCS_6", "SLCS_7", 
                 "SLCS_9", "SLCS_11", "SLCS_15"),
  min = 1,
  max = 5,
  reverse = c("SLCS_1", "SLCS_6", "SLCS_7", "SLCS_15")
)

SC_spec <- ScaleSpec(
  name = "Self-Competence",
  item_names = c("SLCS_2", "SLCS_4", "SLCS_8", "SLCS_10", "SLCS_12",
                 "SLCS_13", "SLCS_14", "SLCS_16"),
  min = 1,
  max = 5,
  reverse = c("SLCS_8", "SLCS_10", "SLCS_13")
)

GS_spec <- CombScaleSpec(
  name = "General Score",
  SL_spec,
  SC_spec
)

# summarize data into factors/scales
summed_data <- sum_items_to_scale(
  data = SLCS,
  SL_spec,
  SC_spec,
  GS_spec,
  retain = c("user_id", "sex")
)

str(summed_data)

Create FrequencyTable, ScoreTable for normalization

Generate norms from raw data to normalize and standardize results

GS_ft <- FrequencyTable(summed_data$`General Score`)
plot(GS_ft)

GS_st <- ScoreTable(GS_ft, scale = STEN)
plot(GS_st)

normalized_GS <- normalize_score(
  summed_data$`General Score`,
  table = GS_st,
  what = "sten"
)

normalized_data <- normalize_scores_df(
  data = summed_data,
  vars = "General Score",
  GS_st,
  what = "sten",
  retain = c("user_id", "sex")
)

str(normalized_GS)
str(normalized_data)

Create GroupedFrequencyTable and GroupedScoreTable

Generate norms for different groups on basis of up to two GroupConditions objects

sex_grouping <- GroupConditions(
  conditions_category = "Sex",
  "M" ~ sex == "M",
  "F" ~ sex == "F",
  "O" ~ sex == "O"
)

GS_gft <- GroupedFrequencyTable(
  data = summed_data,
  conditions = sex_grouping,
  var = "General Score",
  .all = FALSE
)
plot(GS_gft)

GS_gst <- GroupedScoreTable(GS_gft, scale = STEN)
plot(GS_gst)

grouping_normalized <- normalize_scores_grouped(
  data = summed_data,
  vars = "General Score",
  GS_gst,
  retain = c("user_id", "sex"),
  what = "sten",
  group_col = "Group"
)

str(grouping_normalized)

Create and export ScoringTable

Export generated norms in universal format

ST_csv <- tempfile(fileext = ".csv")
GS_scoring <- to_ScoringTable(
  table = GS_gst,
  min_raw = 16,
  max_raw = 80
)

export_ScoringTable(
  table = GS_scoring,
  out_file = ST_csv,
  method = "csv"
)

Create ScoringTable from csv or json file

Import ScoringTable from universally readable formats (eg. create csv on basis of published norms)

"sten","M","F","O"
1,NA,"16-22",NA
2,"16-32","23-26",NA
3,"33-36","27-29","16-35"
4,"37-40","30-37",NA
5,"41-47","38-41","36-63"
6,"48-51","42-47","64-68"
7,"52-58","48-49",NA
8,"59-71","50-55","69-80"
9,"72-80","56-59",NA
10,NA,"60-80",NA

imported <- import_ScoringTable(
  source = ST_csv,
  method = "csv",
  conditions = sex_grouping
)

scoring_normalized <- normalize_scores_scoring(
  data = summed_data,
  vars = "General Score",
  imported,
  retain = c("user_id", "sex"),
  group_col = "Group"
)

str(scoring_normalized)

StatisMike/stenR documentation built on Aug. 20, 2022, 9:30 a.m.