Abstract In order to compare analytical methods with data obtained during exposure to space radiation, a phantom head instrumented with a large number of radiation detectors was flown on the Space Shuttle on three occasions: 8 August 1989 (STS-28), 28 February 1990 (STS-36), and 24 April 1990 (STS-31). The objective of this experiment was to obtain a measurement of the inhomogeneity in the dose distribution within a phantom head volume. The orbits of these missions were complimentary—STS-28 and STS-36 had high inclination and low altitude, while STS-31 had a low inclination and high altitude. In the cases of STS-28 and STS-36, the main contribution to the radiation dose comes from galactic cosmic rays (GCR) with a minor to negligible part supplied by the inner belt through the South Atlantic Anomaly (SAA), and for STS-28 an even smaller one from a proton enhancement during a solar flare-associated proton event. For STS-31, the inner belt protons dominate and the GCR contribution is almost negligible. The internal dose distribution is consistent with the mass distribution of the orbiter and the self-shielding and physical location of the phantom head.