type_1_sdt: Type 1 SDT for a 2AFC design

In craddm/metaSDT: Calculate Type 1 and Type 2 Signal Detection Measures

Type 1 SDT for a 2AFC design

Description

Calculate standard type 1 SDT measures for 2AFC.

Usage

type_1_sdt(
  df,
  stimulus = stimulus,
  response = response,
  counts = total,
  s = 1,
  add_constant = TRUE
)

Arguments

df	Data frame. See notes.
stimulus	Column name for levels of the stimulus. Should be bare, unquoted. Default is 'stimulus'
response	Column name for responses. Should be bare, unquoted. Default is 'response'.
counts	Column name for totals. Should be bare, unquoted. Defaults to "total", as this is the column name output by sdt_counts.
s	Ratio of standard deviations of stimulus types. Defaults to 1 (equal variance).
add_constant	Adds a small constant to every cell to account for boundaries - i.e. log-linear correction. Default = TRUE.

Details

The expected data frame format is one column indicating the stimulus (note that the first level of this factor will be treated as stimulus A), one column indicating the response (note - this should be coded with 1 as response A), and one column indicating the total number of responses of that type. Thus, there should be one row per combination of stimulus and response. If your data is in long format (i.e. one row per trial), you can use the sdt_counts function first to get the data into the expected format. The type_1_sdt function assumes that 1 = responded with first level of stimulus factor (e.g. 1 = stimulus A), and will calculate d-prime on that basis. Note that by default it adds a small constant to all cells to avoid boundary issues.

Author(s)

Matt Craddock, [e-mail]

Examples

type_1_test <- data.frame(
                 expand.grid(
                   stimulus = c("A", "B"),
                   response = c(1, 2)
                 ),
                 total = c(30, 10, 8, 32)
                 )
type_1_sdt(type_1_test)

craddm/metaSDT documentation built on Feb. 10, 2024, 6:57 a.m.