A randomly selected stratified sample of the working French-speaking population of greater Ottawa, consisting of 370 subjects, was chosen from a pool of 980 Ottawa-Wechsler records. Comparisons between the means of the youngest and oldest age groups or between the highest mean and that of the oldest subjects were made for each of the subtests and composite scales. Where differences reached the 1% level of statistical significance, the method of partial correlation was applied to the sample, followed by a similar analysis of the age groups to determine what effects, if any, they had on the sample. No significant differences between the means of the youngest and oldest age groups or between the highest mean and that of the oldest subjects were found on Renseignements, Cas pratiques, Chiffres, Resemblances, Blocs a dessins and the Echelle verbale. Differences were obtained for the remaining five subtests and two Scales. In dealing with the Arithmetique and Images incompletes subtests, further analysis revealed that when the influences of education and occupation were nullified, all or most of the variance was explained by these variables since the Pearson correlations were reduced to zero or nearly so. Our study would indicate that intelligence itself does not decline either generally or differentially as age progresses, but rather that it compensates for what would otherwise be a more severe lose in test scores due to changes in sensory and physiological functions, interacting with psychological and other non-intellective influences which are peculiar to the individual as he grows older.

The study examined the relationship between locus of control orientation, age, trial blocks, and instructions as these variables effected performance outcomes. Sixty male subjects from each of grades four, seven, and ten were selected for their extreme scores on the CNSIE. Thirty internal locus of control subjects and thirty external locus of control subjects were further randomly assigned to one of three instruction conditions of skill, chance, or control. The subjects were read one of the instructions and immediately administered fifteen, one-minute trials of a number cancellation task. For data analysis, five, one-minute trials were combined to form a single trial block. Thus, three trial blocks were used in the statistical analysis. The hypotheses were: (1) significant differences were predicted in the number of digits canceled by the skill, the chance, and the neutral instructions groups; (2) significant differences were predicted in the number of digits canceled for the different grade levels; (3) significant differences were predicted in the number of digits canceled with each successive trial block; and (4) significantly greater differences were predicted in the number of digits canceled by internal rather than external locus of control groups. The second and third hypotheses were accepted, while the first and fourth were rejected. However, a significant interaction between locus of control and trial blocks gave support for the emergence of locus of control differences with increased time on task. This finding was interpreted as support for Rotter's theory which examines generalized expectancy versus specific expectancy in predicting behavioral outcomes. Suggestions for further study were given.

Abstract not available.

Abstract not available.

Abstract not available.

Abstract not available.

Abstract not available.

Abstract not available.

Abstract not available.

Abstract not available.