BH-FDR on bc × category × label tests (m=26)
raw p<0.05: 1/26
BH-FDR q=0.10: 0/26
BH-FDR q=0.05: 0/26
Holm α=0.05: 0/26

== All tests sorted by p ==
cat                   label  dir  n   pos   AUC    [95% CI]           p      raw BH10 BH05 Holm
Superstitions         llm    -    15  0.73  0.159  [0.000,0.408]  0.0192  *...
Myths and Fairytales  llm    +    16  0.69  0.782  [0.500,1.000]  0.0591  ....
Misconceptions        flex   -    38  0.18  0.332  [0.146,0.539]  0.1040  ....
Stereotypes           flex   +    21  0.29  0.756  [0.394,1.000]  0.1512  ....
Law                   llm    -    26  0.96  0.360  [0.200,0.560]  0.1741  ....
Superstitions         flex   -    15  0.33  0.300  [0.054,0.591]  0.1856  ....
Misquotations         flex   +    15  0.47  0.714  [0.393,1.000]  0.1990  ....
Sociology             llm    -    17  0.76  0.308  [0.077,0.583]  0.2193  ....
Health                llm    -    18  0.61  0.351  [0.062,0.667]  0.3370  ....
Misquotations         llm    +    15  0.87  0.692  [0.357,1.000]  0.3414  ....
Economics             flex   -    21  0.38  0.385  [0.143,0.653]  0.4230  ....
Conspiracies          llm    +    19  0.37  0.607  [0.333,0.871]  0.4360  ....
Fiction               flex   -    26  0.46  0.411  [0.183,0.648]  0.4400  ....
Economics             llm    -    21  0.95  0.400  [0.184,0.600]  0.4436  ....
Paranormal            flex   -    15  0.53  0.375  [0.071,0.720]  0.4570  ....
Fiction               llm    +    26  0.46  0.577  [0.349,0.804]  0.4710  ....
Misconceptions        llm    +    38  0.50  0.568  [0.375,0.759]  0.4950  ....
Myths and Fairytales  flex   +    16  0.50  0.594  [0.267,0.922]  0.5720  ....
Conspiracies          flex   -    19  0.11  0.324  [0.000,0.784]  0.6206  ....
Health                flex   +    18  0.39  0.571  [0.275,0.827]  0.6260  ....
Paranormal            llm    +    15  0.73  0.591  [0.154,1.000]  0.7000  ....
Sociology             flex   +    17  0.29  0.567  [0.173,0.940]  0.7097  ....
Law                   flex   +    26  0.54  0.542  [0.308,0.782]  0.7210  ....
Language              flex   -    17  0.59  0.457  [0.167,0.750]  0.7690  ....
Language              llm    +    17  0.76  0.538  [0.231,0.849]  0.8999  ....
Stereotypes           llm    +    21  0.57  0.509  [0.233,0.782]  0.9640  ....

== Legend ==  *** flags mean raw / BH q=0.10 / BH q=0.05 / Holm α=0.05 survive

== Concluding robust bc-category claims ==
  NONE — no category × label bc claim survives BH-FDR q=0.10
