I have seen a process where the tape measures are calibrated using another tape measure. All tape measures were of the same brand and length. One tape measure was kept as a "master" and sent out annually to an accredited lab. All other tape measures would be compared to this tape measure. Is this an acceptable method to calibrate tape measures? The part that I found questionable is that the master tape measure listed its tolerance as +/- 1/32". The tape measures calibrated using this master tape measure were also stated to be at +/- 1/32"
This depends on your definition of "acceptable".
You can compare a tape measure to a "master", it is done. In this case you would be accepting a measurement uncertainty that is larger than the tolerance of your tape measure, so that will determine if the calibration is "acceptable" to your requirements.
I have attached a NIST SOP that uses this method to calibrate a tape measure.