The birth places are as quoted.
Census data are problematical. As in today's census, no effort was ever made to verify the accuracy of the information submitted.
We have no way of knowing who submitted the information. A large proportion of the population were illiterate so the schedule could have been completed by the enumerator himself - the head of the house no dobut absent at work when the form was collected.
The Schedules were then copied into the Census Books and handwriting evidence would suggest that the copying was not always done by the actual enumerator.
There is evidence that the copying procedure was by columns not persons. So, especially when ditto marks were used, items sometimes "slip" by a row.
Even baptizmal records can be "wrong". The baptism could take place in the most conveient church, not necessarily the Parish of residence, or the mother could have been staying with her own family at the time of birth.