The package ships one CAIS normative sample, the child sample. A second,
drawn from the source's adult respondents, was shipped until version 2.0.0
and has been withdrawn. The CAIS is rated on a 5-point scale, but three of
the octant means the source publishes for those respondents fall above 5, so
that sample cannot be on the same metric as the scores it would standardize:
norm_standardize() refused it, which left it as data no call could use.
The transcription was faithful and the source's own table is the origin of
the discrepancy, so the values are not lost — they are recorded, with the
evidence about what went wrong in the source and what a reply from its
authors would reopen, in the package's source repository under
cairn/references/sodano2006.md, which also records the state of the query
to the source's authors.