Post- and re-test data
As there were pre-existing differences in the number of kicks made during shearing, a repeated measures ANCOVA was performed on the post- and re- test number of kicks. This analysis revealed no significant differences due to source, training condition, or test session. Three-way ANOVAs for unequal subjects were conducted on the depen- dent measures that were not affected by pre-existing differences. Handling treatments affected the latencies to approach the trainer during the approach- ability test. Gently handled sheep had significantly lower latencies to ap- proach than either the control or restrained groups (P< 0.05 ). The mean times for approach were 144.95_+37.74 s for GH, 269.95_+20.63 s for RH, and 270.60_+ 17.43 s for UH. A main effect (P