A Statistical Investigation of the Synoptic Problem · supplementary table
Corrected p-values
Every ordered pair of the nineteen category codes, after correction — 340 comparisons in all. 186 of them were corrected to 1 and carry no signal, so those are greyed and the surviving results stand out. A further 15 came out too small for the sheet to hold and are shown as ~0. 18 results are marked as significant at some level.
Columns → predictors
| 222 | 211 | 112 | 212 | 221 | 122 | 121 | 022 | 012 | 021 | 220 | 120 | 210 | 020 | 202 | 201 | 102 | 200 | 002 | |
|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|
| 222 | 1 | 1 | 1 | <0.001 | 0.008 | 1 | 0.006 | 1 | 1 | <0.001 | 1 | 1 | 1 | 0.894 | 1 | 1 | 1 | 1 | |
| 211 | 1 | 1 | <0.001 | 0.041 | 1 | 0.788 | 0.968 | 1 | 0.477 | 1 | 1 | ~0 | 1 | 1 | 0.098 | 1 | <0.001 | 1 | |
| 112 | 1 | 1 | 0.328 | 1 | 0.019 | 1 | 1 | ~0 | 0.296 | 1 | 1 | 0.309 | 0.985 | 1 | 1 | 0.026 | 1 | ~0 | |
| 212 | 1 | <0.001 | 0.494 | 1 | 1 | 1 | 0.137 | 0.731 | 0.776 | 1 | 1 | <0.001 | 1 | 1 | 1 | 0.361 | 1 | 0.501 | |
| 221 | <0.001 | 0.060 | 1 | 1 | 1 | ~0 | 1 | 1 | 0.001 | <0.001 | 0.019 | 0.097 | 0.006 | 1 | 1 | 1 | 1 | 1 | |
| 122 | 0.137 | 1 | 0.177 | 1 | 1 | 0.015 | 0.074 | 0.973 | 0.062 | 0.868 | 0.009 | 1 | 1 | 1 | 1 | 1 | 1 | 0.217 | |
| 121 | 1 | 0.694 | 1 | 0.803 | ~0 | <0.001 | 0.020 | 0.363 | ~0 | 0.010 | ~0 | 0.983 | <0.001 | 1 | 1 | 1 | 1 | 1 | |
| 022 | 0.043 | 1 | 1 | 0.145 | 1 | 0.051 | 0.085 | 0.606 | 0.007 | 0.827 | 0.037 | 1 | 0.093 | 0.964 | 1 | 1 | 1 | 1 | |
| 012 | 1 | 1 | ~0 | 0.840 | 1 | 0.946 | 0.412 | 0.579 | 0.500 | 1 | 0.917 | 0.608 | 0.833 | 1 | 1 | 1 | 1 | 0.012 | |
| 021 | 1 | 0.639 | 0.588 | 0.797 | 0.002 | 0.045 | ~0 | 0.008 | 0.422 | 1 | <0.001 | 0.909 | <0.001 | 1 | 1 | 1 | 1 | 1 | |
| 220 | <0.001 | 1 | 1 | 1 | <0.001 | 0.805 | 0.094 | 0.669 | 1 | 1 | 0.013 | 0.342 | 0.012 | 0.892 | 1 | 1 | 1 | 1 | |
| 120 | 0.738 | 1 | 1 | 1 | 0.015 | <0.001 | ~0 | 0.024 | 0.902 | <0.001 | 0.011 | 1 | <0.001 | 1 | 1 | 1 | 1 | 1 | |
| 210 | 1 | <0.001 | 0.736 | <0.001 | 0.132 | 1 | 0.977 | 0.650 | 0.487 | 0.976 | 0.612 | 1 | 1 | 1 | 0.187 | 1 | 0.528 | 1 | |
| 020 | 1 | 1 | 1 | 1 | 0.010 | 1 | <0.001 | 0.121 | 0.906 | <0.001 | 0.027 | <0.001 | 1 | 0.231 | 1 | 0.484 | 1 | 1 | |
| 202 | 0.736 | 1 | 1 | 1 | 1 | 1 | 1 | 0.871 | 1 | 1 | 0.548 | 1 | 0.698 | 0.042 | <0.001 | <0.001 | 1 | ||
| 201 | 1 | 0.148 | 1 | 1 | 1 | 1 | 1 | 1 | 1 | 1 | 1 | 1 | 0.248 | 1 | <0.001 | 1 | ~0 | 1 | |
| 102 | 1 | 1 | 0.125 | 0.221 | 1 | 1 | 1 | 1 | 1 | 1 | 1 | 1 | 1 | 0.495 | 0.001 | 1 | 1 | <0.001 | |
| 200 | 0.242 | <0.001 | 1 | 1 | 1 | 1 | 1 | 1 | 1 | 1 | 1 | 1 | 0.186 | 1 | ~0 | 0.795 | 1 | ||
| 002 | 0.409 | 1 | ~0 | 1 | 1 | <0.001 | 1 | 0.724 | ~0 | 1 | 1 | 1 | 1 | 1 | 0.962 | 1 | ~0 | 1 |
Rows ↓ predicted categories
Notes
The matrix is not symmetric: the value for a pair read one way differs from the same pair read the other way, so each of the cells is its own test.
~0 marks a p-value so small the spreadsheet overflowed rather than returning a figure. These are the strongest results in the table, not missing ones.
Blacked-out cells are pairs disqualified from the analysis. The hatched diagonal is each category against itself.
Dispersion correction applied: 2.17.
Values are shown to three decimal places, and as <0.001 below that. Point at
any cell for the figure as computed.