Abstract The Minnesota Rate of Manipulation Test (MRMT) placing and turning tasks were studied for suitability for repeated measures applications: training evaluation, selection and environmental research. Twenty male volunteers performed four runs of placing and turning in each of 15 daily trials (Monday to Friday). The tasks were evaluated in terms of the amount of practice required to yield scores with homogeneous variances and intertrial correlations, and means with a linear trend across days. Results revealed that placing and turning means increased linearly in parallel fashion, and variances were homogeneous after the third day. Intertrial correlations were homogeneous after day 3 for turning and day 10 for placing; with respective stabilized intertrial correlations of 0-64 and 0-68. Correlations between the tasks at each stage of practice were described. The results were discussed in terms of Jones' two-process model of skill acquisition. Implications for personnel measurements and intervention exp...