TSQCA
Threshold Sweep Extensions for Qualitative Comparative Analysis

Package index
Search the TSQCA package
Vignettes
Functions
108
Source code
18
Man pages
47
Home
/
CRAN
/
TSQCA
/
README.md

README.md
In TSQCA: Threshold Sweep Extensions for Qualitative Comparative Analysis

TSQCA

DOI

TSQCA is an R package implementing Threshold-Sweep QCA (TS-QCA), a framework for systematically varying the thresholds used to binarize the outcome and conditions in crisp-set QCA.

After calibration, QCA results may change depending on how thresholds are set. TS-QCA evaluates this sensitivity by automatically:

Implemented sweep types:

Scope: TSQCA focuses on sufficiency analysis. Necessity analysis is planned for future versions.

Features

Bug Fix for Intermediate Solutions (v1.2.0)

Important: v1.2.0 fixes a critical bug where intermediate solutions were incorrectly extracted when using dir.exp. If you used intermediate solutions in previous versions, please re-run your analyses.

Verification Recommendation for Publications

For academic publications, we strongly recommend verifying TSQCA results directly with the QCA package:

library(QCA)

# Run QCA directly to verify TSQCA results
tt <- truthTable(dat, outcome = "Y", 
                 conditions = c("X1", "X2", "X3"),
                 incl.cut = 0.8)
sol <- minimize(tt, include = "?", dir.exp = c(1, 1, 1))
print(sol)  # Compare with TSQCA report output

TSQCA reports include a "QCA Package Output" section for this purpose. Always ensure the solution expressions match before publishing.

Solution Type Display and Verification (New in v1.2.0)

Reports now explicitly show the Solution Type (Complex/Parsimonious/Intermediate) and include optional QCA package output for verification:

# Reports now show Solution Type in Analysis Overview
generate_report(result, "report.md", dat = mydata)

# Disable QCA raw output if not needed
generate_report(result, "report.md", dat = mydata, include_raw_output = FALSE)

Three Types of QCA Solutions (New in v1.1.0)

As of v1.1.0, TSQCA uses the same defaults as QCA::minimize():

| Solution Type | include | dir.exp | Logical Remainders | Description | |--------------|-----------|-----------|-------------------|-------------| | Complex (default) | "" | NULL | Not used | Most conservative | | Parsimonious | "?" | NULL | All used | Most simplified | | Intermediate | "?" | c(1,1,...) | Theory-guided | Most common in publications |

data(sample_data)
thrX <- c(X1 = 7, X2 = 7, X3 = 7)

# 1. Complex Solution (default)
result_comp <- otSweep(dat = sample_data, outcome = "Y", 
                       conditions = c("X1", "X2", "X3"),
                       sweep_range = 7, thrX = thrX)

# 2. Parsimonious Solution
result_pars <- otSweep(dat = sample_data, outcome = "Y",
                       conditions = c("X1", "X2", "X3"),
                       sweep_range = 7, thrX = thrX,
                       include = "?")

# 3. Intermediate Solution (most common in publications)
result_int <- otSweep(dat = sample_data, outcome = "Y",
                      conditions = c("X1", "X2", "X3"),
                      sweep_range = 7, thrX = thrX,
                      include = "?", dir.exp = c(1, 1, 1))

Migration from v1.0.0: The default has changed from intermediate to complex solution. To reproduce v1.0.0 behavior, explicitly specify include = "?" and dir.exp = c(1, 1, 1).

Multiple Solution Detection

QCA minimization can produce multiple equivalent intermediate solutions. TSQCA detects and reports these cases, allowing researchers to identify robust essential prime implicants versus solution-specific selective prime implicants.

Use the extract_mode parameter to control output:

| Mode | Description | Use Case | |------|-------------|----------| | "first" | Returns only the first solution (M1) | Default, backward compatible | | "all" | Returns all solutions concatenated | See all equivalent solutions | | "essential" | Returns essential prime implicants common to all solutions | Identify robust findings |

# Detect and show all solutions
result <- otSweep(
  dat = mydata,
  outcome = "Y",
  conditions = c("X1", "X2", "X3"),
  sweep_range = 6:8,
  thrX = c(X1 = 7, X2 = 7, X3 = 7),
  extract_mode = "all"  # Show all solutions
)

# Extract essential prime implicants only
result_essential <- otSweep(
  dat = mydata,
  outcome = "Y",
  conditions = c("X1", "X2", "X3"),
  sweep_range = 6:8,
  thrX = c(X1 = 7, X2 = 7, X3 = 7),
  extract_mode = "essential"  # Show essential prime implicants
)

Automatic Report Generation

The generate_report() function creates comprehensive markdown reports:

# Run analysis with return_details = TRUE (the default)
result <- otSweep(
  dat = mydata,
  outcome = "Y",
  conditions = c("X1", "X2", "X3"),
  sweep_range = 6:8,
  thrX = c(X1 = 7, X2 = 7, X3 = 7)
)

# Generate full report
generate_report(result, "my_analysis.md", dat = mydata, format = "full")

# Generate simple report (for journal manuscripts)
generate_report(result, "my_analysis_simple.md", dat = mydata, format = "simple")

Reports include: - Analysis settings (for reproducibility) - Solution formulas with essential/selective prime implicants - Fit measures (consistency, coverage, PRI) - Cross-threshold comparison tables

Configuration Charts

Configuration charts are automatically included in reports. Generate Fiss-style configuration charts (Table 5 format):

# Reports include configuration charts by default
generate_report(result, "my_report.md", dat = mydata, format = "full")

# Disable charts if needed
generate_report(result, "my_report.md", dat = mydata, include_chart = FALSE)

# Use LaTeX symbols for academic papers
generate_report(result, "my_report.md", dat = mydata, chart_symbol_set = "latex")

Standalone chart functions are also available:

# From path strings
paths <- c("A*B*~C", "A*D")
chart <- config_chart_from_paths(paths)
cat(chart)

Output:

| Condition | M1 | M2 |
|:--:|:--:|:--:|
| A | ● | ● |
| B | ● |   |
| C | ⊗ |   |
| D |   | ● |

Three symbol sets available: "unicode" (● / ⊗), "ascii" (O / X), "latex" ($\bullet$ / $\otimes$)

S3 Methods

All sweep functions return objects supporting S3 class methods:

result <- otSweep(...)
print(result)    # Quick overview
summary(result)  # Full details

Negated Outcome Support

Analyze conditions sufficient for the absence of an outcome:

# Analyze when Y is low (below threshold)
result <- otSweep(
  dat = mydata,
  outcome = "~Y",  # Tilde prefix for negation
  conditions = c("X1", "X2", "X3"),
  sweep_range = 6:9,
  thrX = c(X1 = 7, X2 = 7, X3 = 7)
)

Terminology Note

TSQCA uses precise Boolean algebra terminology:

| Term | Meaning | |------|---------| | Essential Prime Implicants (EPI) | Terms in ALL equivalent solutions (M1, M2...) | | Selective Prime Implicants (SPI) | Terms in SOME solutions only |

Note: This is distinct from Fiss (2011) "Core Conditions" which compares parsimonious vs. intermediate solutions. See docs/TSQCA_Terminology_Guide_EN.md for details.

Installation

install.packages("devtools")
devtools::install_github("im-research-yt/TSQCA")

Relationship with QCA Package

TSQCA is built on top of the QCA package (Duşa, 2019). All function arguments follow QCA conventions:

Why This Matters

Quick Reference

# Check QCA parameter documentation
?QCA::truthTable  # For incl.cut, n.cut, pri.cut
?QCA::minimize    # For include, dir.exp

# TSQCA uses these same parameters
result <- dtSweep(
  dat = sample_data,
  outcome = "Y",
  conditions = c("X1", "X2"),
  sweep_list_X = list(X1 = 6:7, X2 = 6:7),
  sweep_range_Y = 6:7,
  incl.cut = 0.8,   # QCA parameter
  n.cut = 1,        # QCA parameter
  pri.cut = 0       # QCA parameter
)

Basic Setup

library(QCA)
library(TSQCA)

dat <- read.csv("sample_data.csv", fileEncoding = "UTF-8")

outcome  <- "Y"
conditions <- c("X1", "X2", "X3")

str(dat)

Working with Mixed Data Types

Handling Binary Variables (0/1)

When your dataset contains both continuous and binary (0/1) variables, special attention is needed:

Why Threshold = 1 for Binary Variables?

The qca_bin() function uses x >= thr for binarization: - If x = 0: 0 >= 1 → FALSE → 0 (preserved) - If x = 1: 1 >= 1 → TRUE → 1 (preserved)

Example: Mixed Data

# Suppose X1 is binary (0/1), while X2 and X3 are continuous (0-10)
sweep_list <- list(
  X1 = 1,      # Binary variable: use threshold 1 to preserve values
  X2 = 6:8,    # Continuous: sweep thresholds
  X3 = 6:8     # Continuous: sweep thresholds
)

res_mcts <- ctSweepM(
  dat = dat,
  outcome = "Y",
  conditions = c("X1", "X2", "X3"),
  sweep_list = sweep_list,
  thrY = 7
)

This explores 1 × 3 × 3 = 9 threshold combinations, treating X1 as a fixed binary condition.

Common Mistake

# WRONG: Using sweep range for binary variables
sweep_list <- list(
  X1 = 6:8,    # All values become 0 (since 0 < 6 and 1 < 6)
  X2 = 6:8,
  X3 = 6:8
)

Best Practice: Always specify thresholds explicitly for each variable based on its data type.

Usage Examples

1. CTS-QCA: single-condition X sweep (ctSweepS)

sweep_var <- "X3"      # Condition (X) whose threshold will be varied
sweep_range <- 6:9     # Candidate threshold values

thrY <- 7              # Fixed threshold for Y
thrX_default <- 7      # Fixed thresholds for other X's

res_cts <- ctSweepS(
  dat            = dat,
  outcome        = outcome,
  conditions     = conditions,
  sweep_var      = sweep_var,      # X to sweep
  sweep_range    = sweep_range,    # Threshold candidates for X
  thrY           = thrY,           # Fixed Y threshold
  thrX_default   = thrX_default    # Fixed thresholds for other X's
)

summary(res_cts)

2. MCTS-QCA: multi-condition X sweep (ctSweepM)

# Threshold candidates for each X
sweep_list <- list(
  X1 = 6:7,
  X2 = 6:7,
  X3 = 6:7
)

res_mcts <- ctSweepM(
  dat            = dat,
  outcome        = outcome,
  conditions     = conditions,
  sweep_list     = sweep_list,     # Threshold candidates for each X
  thrY           = 7               # Fixed Y threshold
)

summary(res_mcts)

3. OTS-QCA: outcome Y sweep (otSweep)

thrX <- c(X1 = 7, X2 = 7, X3 = 7)  # Fixed thresholds for X
sweep_range_Y <- 6:8               # Candidate thresholds for Y

res_ots <- otSweep(
  dat            = dat,
  outcome        = outcome,
  conditions     = conditions,
  sweep_range    = sweep_range_Y,  # Y threshold candidates
  thrX           = thrX            # Fixed X thresholds
)

summary(res_ots)

# Generate report for detailed analysis
generate_report(res_ots, "ots_report.md", dat = dat, format = "full")

4. DTS-QCA: 2D sweep of X and Y (dtSweep)

# X-side threshold candidates (multiple conditions)
sweep_list_X <- list(
  X1 = 6:7,
  X2 = 6:7,
  X3 = 6:7
)

# Y-side threshold candidates
sweep_range_Y <- 6:7

res_dts <- dtSweep(
  dat            = dat,
  outcome        = outcome,
  conditions     = conditions,
  sweep_list_X   = sweep_list_X,   # X threshold candidates
  sweep_range_Y  = sweep_range_Y   # Y threshold candidates
)

summary(res_dts)

Sample Data

data(sample_data)
str(sample_data)

References

Core QCA Methodology

QCA R Package

Robustness and Threshold Sensitivity

Acknowledgements

This package implements methods developed in research supported by JSPS KAKENHI Grant Number JP20K01998.

License

MIT License.



Try the TSQCA package in your browser

Any scripts or data that you put into this service are public.

TSQCA documentation built on Feb. 18, 2026, 5:06 p.m.

R Package Documentation

rdrr.io home R language documentation Run R code online

Browse R Packages

CRAN packages Bioconductor packages R-Forge packages GitHub packages

We want your feedback!

Note that we can't provide technical support on individual packages. You should contact the package authors for that.
GitHub issue tracker
ian@mutexlabs.com
Personal blog

 
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.

Close