AB resistance is not close to calibrated reading and measured Rw is higher than max spec

I have several AD5235bruz25 ICs each in a very simple circuit.

I have VSS, B1, and B2 tied to GND, and I have PR/ and WP/ tied to VDD which is 5V. The A1 and A2 contacts are open and W1 and W2 are only connected to an open test point.

Here are the results of readings from 18 ICs. "Cal" is the calibrated AB resistance (ohms) computed from the % variance read from the IC. As you can see in the table, the AB measured resistance (ohms) is always greater than the factory calibrated reading, on average by 130 ohms. If I'm reading the data sheet right, should the Rab measured value actually be within 0.1% of the calibrated value? Also, with wiper setting at 0, the W1 and W2 resistances are much higher than the max 50 ohms (typ 30) indicated in the datasheet for wiper resistance, Rw.

I'll assume I'm maybe doing something wrong here. Any ideas? Again, the circuit is dead simple with nothing else connected beyond what's described above except the SPI signals connected to an MCU.