New to QCA? This assumes your data is already
calibrated — each column already a crisp 0/1 or fuzzy [0,1]
set-membership score, whichever you used. Enter your data, outcome and
conditions, and get a working truthTable()
→ minimize() script ready to paste into R.
No calibration step, and no need to say which columns are crisp and
which are fuzzy — truthTable() handles
both from the values themselves.
1Data, outcome & conditions
Assumes every column below (outcome and conditions alike) already
holds set-membership scores, either crisp (0/1) or fuzzy ([0, 1]).
If any of yours are still raw values, calibrate them first with
calibrate() — this tool starts from
truthTable().
2Analysis settings
3Directional expectations dir.exp, one per condition
For each condition: does its presence/high level (Favors outcome) or its absence/low level (Disfavors outcome) help produce the outcome, or is there No expectation either way? This is what turns the parsimonious solution into the intermediate solution — the type most QCA guides (Ragin 2008; Schneider & Wagemann 2012) recommend reporting.
Why this shape: truthTable() and
minimize() are QCA's own functions, used
exactly as documented — this page only assembles the call
for you, it does not compute anything itself.