Inter- and intra-examiner reliability of the upper cervical X-ray marking system: a second look.

To determine the degree of reliability (stability over time) for six Pettibon practitioners, the scores resulting from the reading and re-reading of 30 X rays were analyzed using bivariate scattergrams, Pearson Product-moment correlation coefficient estimates and correlated samples t tests. To examine reliability (equivalence over experts) across the practitioners, a repeated measures analysis of variance approach was used. Liberal and conservative reliability coefficients for the upper angle and lower angle were computed. Examination of the data suggest that the reliability (stability over time) for the practitioners is very good. The data on reliability (equivalence over experts) across the practitioners also suggests reliability is very good.