I think what we see is a transcription by the enumerator of the original householder schedule, so if the handwriting isn't great to start with, it's quite possible there's some guesswork by the enumerator in their version (just as Ancestry's modern indexers then make errors)
