Third, the variability was similar for composite and psychomotor tests. Fourth, all but two validity coefficients were shown to be statistically non-significant, when really the effect size is moderate for both the Rotary Pursuit Test (r = .26, p [less than] .01) and the composite (r = .38, p [less than] .001).
Moreover, since the psychomotor tests are highly specific and they seem to test several abilities (Dunnette, 1976; Fleishman, 1972) the observed variance for these tests should be larger than the variance for the composite test, but the findings do not support this prediction.
At the initial testing of psychomotor tests (Table no.3), low levels are due largely to the sensibility A.E.
Initial psychomotor tests reveal the deficiencies of subjects in terms of psychomotor qualities so necessary for obtaining a motor independence, and the final tests show the progress recorded over the 7 and a half months of training, in which the operational systems proposed by us were strictly applied every time.