rateServer: Actions for likert scales

In abbey-thomas/speechcollectr: R Tools for Collecting Speech Production & Perception Data

Actions for likert scales

Description

Actions for likert scales

Usage

rateServer(
  id = "rate",
  trigger = NULL,
  wait = 1,
  type = c("button", "slider"),
  n_scales = 1,
  answer_all = FALSE,
  choices = NULL,
  instructions = "",
  pretext = "",
  scale_labs = NULL,
  posttext = "",
  direction = "horizontal",
  sliderInit = 50,
  sliderMin = 0,
  sliderMax = 100,
  pips = NULL,
  step = 0.01
)

Arguments

id	The id of the module. Must be the same as the ID of rateUI().
trigger	If not NULL (the default), a reactive expression returning the event that should trigger the appearance of the instructions.
wait	Integer. How long should we wait to display the scale(s) after the trigger event occurs? In milliseconds.
type	One of "button" or "slider". Must be the same as the type of rateUI().
n_scales	Integer. The number of scales to be displayed on the page. Must match the value given in the corresponding call to rateUI()
answer_all	If n_scales > 1, does the participant need to give a response on each scale or only one of them?
choices	A list with length n_scales. The answer choices for each scale should be a character vector in this list. If type="slider", each scale must have EXACTLY 2 answers (the first representing the lower extreme and the second representing the higher extreme). If n_scales > 1 and the options are different for each scale, the character vectors should be wrapped in a list.
instructions	Instructions that will appear at the top of the page.
pretext	Character. The text that will appear above the likert scale.
scale_labs	Character. The individual label that will appear above each scale if n_scales > 1.
posttext	Character. The text that will appear below the likert scale.
direction	Either "horizontal" or "vertical". Describes the position of the scale options (NOT the scales themselves) relative to each other.
sliderInit	Numeric. The starting value of the slider. Must be between sliderMin and sliderMax. Default is in the middle of the scale.
sliderMin	Numeric. The lower of the slider. Default is 0.
sliderMax	Numeric. The upper limit of the slider. Default is 100.
pips	List with two elements. Where to place numeric markers on the slider scale. See Details for more information.
step	Numeric. The distance between consecutive points (i.e., potential answers) on the slider. Default is .01,

Details

Notes on sliders:

• Consider carefully whether you need a slider. Do you expect participant responses to fall on a continuous scale in meaningful ways, or do you expect answers to fall in a number of blocks along the slider? If the latter, consider using a button scale as they are easier for participants to use (See Toepoel & Funke, 2018 and Chyung et al., 2018 (references below) for more information.).

• Placing numeric markers along the scale and displaying the numeric response values are best for scales where the answer is truly numeric. (i.e., "How tall (in inches) do you think the talker in this recording is?") (See Gummer & Kunz, 2021 (reference below) for more information.)

• The argument pips should be a list with 2 values: The first value, "density", indicates how far apart the minor divisions on the scale should be (e.g., density=5 will place a marker every 5 integers between sliderMin and sliderMax). The second element in the list, "values", is a vector denoting where the major breaks of the scale and their numeric labels should go (e.g., if values=c(0, 50, 100) markers and numeric labels will be placed at 0, 50, and 100.) All values in values must fall between sliderMin and sliderMax.

• The default starting answer is in the center of the scale, as initializing the handle at the left of the scale has been shown to bias the answers towards the lower values. (CITE)

• Currently, the ⁠rate*()⁠ functions only support text values at the beginning and end of the slider. These will be placed inline with the slider and should represent the two extreme values (low and high, in that order). To place text at only one or the other extreme, just use an empty text string ("") for the end of the scale where you do not wish to show a value.

Value

Returns a reactive vector of the answers selected from each scale when the participant clicks submit.

References

Chyung, S. Y., Swanson, I., Roberts, K., & Hankinson, A. (2018). Evidence‐based survey design: The use of continuous rating scales in surveys. Performance Improvement, 57(5), 38-48.
Gummer, T., & Kunz, T. (2021). Using only numeric labels instead of verbal labels: Stripping rating scales to their bare minimum in web surveys. Social Science Computer Review, 39(5), 1003-1029.
Toepoel, V., & Funke, F. (2018). Sliders, visual analogue scales, or buttons: Influence of formats and scales in mobile and desktop surveys. Mathematical Population Studies, 25(2), 112-122.

Examples

if (interactive()){
  library(shiny)
  library(shinyjs)
  ui <- fluidPage(
    actionButton("btn", "Click me"),
    rateUI(id = "example",
           type = "button"),
    actionButton("submit", "SUBMIT"),
    textOutput("confirmation")
  )

  server <- function(input, output, session) {
  rvs <- reactiveValues(rating = NULL)
    observeEvent(input$btn, {
      disable("btn")

      rvs$ans <- rateServer(id = "example",
                               type = "button",
                               instructions = "What do you think?",
                               answers = c("Strongly disagree", "Disagree",
                                           "Neutral", "Agree", "Strongly agree"),
                               pretext = "The vowels 'aw' and 'ah' sound exactly the same.")
    })

    observeEvent(input$submit, {
    enable("btn")
        output$confirmation <- renderText({
          paste0("You selected ", rvs$ans$ratings[1],".")})

    })
  }
  shinyApp(ui = ui, server = server)
}

# An example with 2 scales....
if (interactive()){
   library(shiny)
   library(shinyjs)
   ui <- fluidPage(
     actionButton("btn", "Click me"),
     rateUI(id = "example",
            type = "button",
            n_scales = 2),
     actionButton("submit", "SUBMIT"),
     textOutput("confirmation")
   )

   server <- function(input, output, session) {
     rvs <- reactiveValues(rating = NULL)

     observeEvent(input$btn, {
       disable("btn")

       rvs$ans <- rateServer(id = "example",
                                trigger = NULL,
                                type = "button",
                                instructions = "Finish the sentence:",
                                answers = list(c("Sound completely the same",
                                                 "Sound similar, but not totally alike",
                                                 "Sound pretty different",
                                                 "Sound totally different"),
                                               c("Are produced in the exact same way",
                                                 "Are produced similarly",
                                                 "Are produced pretty distinctly",
                                                 "Are produced in totally distinct ways")),
                                pretext = "The vowels 'aw' and 'ah'...",
                                n_scales = 2,
                                answer_all = TRUE,
                                direction = "vertical",
                                scale_labs = c("perception", "production"))
     })

     observeEvent(input$submit, {
       enable("btn")
       output$confirmation <- renderText({
         paste0("You selected '", rvs$ans$ratings[1],"' and '",  rvs$ans$ratings[2], "'.")})
     })
   }
   shinyApp(ui = ui, server = server)
 }

abbey-thomas/speechcollectr documentation built on Nov. 19, 2024, 7:09 p.m.