Abstract Objective. Reference dosimetry on an MRI-linac requires a chamber specific magnetic field correction factor, k B ⃗ . This work aims to measure the correction factor for a parallel plate chamber on a parallel MRI-linac. Approach. k B ⃗ is defined as the ratio of the absorbed dose to water calibration coefficient in the presence of the magnetic field, N D , w B ⃗ relative to that under 0 T conditions, N D , w 0 T . k B ⃗ was measured via a N D , w transfer to a field chamber at each magnetic field strength from a chamber with known N D , w and k B ⃗ . This was achieved on the parallel MRI-linac by moving the measurement set-up between a high magnetic field strength region at the MRI-isocentre and a low magnetic field strength region at the end of the bore whilst maintaining consistent set-up and scatter conditions. Three PTW 34001 Roos chambers were investigated as well as a PTW 30013 Farmer used to validate methodology. Main Results. The beam quality used for the measurements of k B ⃗ was TPR _20/10 = 0.632. The k B ⃗ for the PTW Farmer chamber at 1 T on a parallel MRI-linac was 0.993 ± 0.013 (k = 1). The average k B ⃗ factor measured for the three Roos chambers on a 1 T parallel MRI-linac was 0.999 ± 0.014 (k = 1). Significance. The results presented are the first measurements of k B ⃗ for a Roos chamber on a parallel MRI-linac. The Roos chamber results demonstrate the potential for the chamber as a reference dosimeter in parallel MRI-linacs.