SetMethods (version 4.0)

rob.calibrange: Function for identifying the calibration threshold ranges for a condition within which the Boolean formula for the solution does not change.

Description

Function for identifying the calibration threshold ranges for a conditio within which the Boolean formula for the solution does not change. The function gradually increases and, then, decreases each of the three (0, 0,5, 1) calibration thresholds of a condition by the value specifyied in the step argument and checks whether the solution formula changes. The function performs this iteration for the number of times specified in the max.runs argument. If the solution formula does not change given the number of runs specified, it will return an NA, meaning that it could not find a limit to the range.

Usage

rob.calibrange(raw.data,
    calib.data,
    test.cond.raw,
    test.cond.calib,
    test.thresholds,
    type = "fuzzy",
    step = 1,
    max.runs = 20,
    outcome,
    conditions,
    incl.cut = 1,
    n.cut = 1,
    include = "",
    ...)

Arguments

raw.data

A data frame containing the raw data used to test the calibration thresholds for the chosen condition.

calib.data

A data frame containing the calibrated data for the sufficient solution.

test.cond.raw

A character string specifying the name of the condition in the RAW dataset for which to test the calibration ranges.

test.cond.calib

A character string specifying the name of the condition in the CALIBRATED dataset for which to test the calibration ranges.

test.thresholds

The initial qualitative anchors used for calibrating the chosen condition.

type

The calibration type, "fuzzy" is the default, but it can also be changed to "crisp".

step

The value to be gradually added and subtracted from the threshold tested.

max.runs

The maximum number of times the step value gets gradually added and subtracted.

outcome

A character string with the name of the outcome in capital letters. For the negated outcome a tilde "~" should be used. This had the same usage as the outcome argument in the minimize function.

conditions

A vector of character strings containing the names of the conditions.This had the same usage as the conditions argument in the minimize function.

incl.cut

The raw consistency threshold for the truth table rows.

n.cut

The frequency threshold for the truth table rows.

include

A vector of other output values (for example "?" for logical remainders) to include in the minimization. This had the same usage as the include argument in the minimize function.

...

Other options that the minimize function in the QCA package accepts. Check them out using ?minimize.

Author

Ioana-Elena Oana

References

Oana, Ioana-Elena, and Carsten Q. Schneider. 2020. Robustness tests in QCA: A fit-oriented and case-oriented perspective using R. Unpublished Manuscript.

Oana, Ioana-Elena, Carsten Q. Schneider, and Eva Thomann (forthcoming). Qualitative Comparative Analysis (QCA) using R: A Gentle Introduction. Cambridge: Cambridge University Press.

Examples

# Load the calibrated data:
data(LIPF)

# Load the raw data:
data(LIPR)

# Get calibration ranges for condition DEVELOPED:

rob.calibrange(
  raw.data = LIPR,
  calib.data = LIPF,
  test.cond.raw = "DEVELOPED",
  test.cond.calib = "DEVELOPED",
  test.thresholds = c(400, 550, 900),
  step = 500,
  max.runs = 2,
  outcome  = "SURVIVED",
  conditions = c("URBAN","LITERATE","INDUSTRIAL","STABLE"),
  incl.cut = 0.8,
  n.cut = 1,
  include = "?"
)