J. R. Harris's (1995) theory that peer influence plays a central role
in shaping personality was tested from the data of the National Merit
twin study (J. C. Loehlin & R. C. Nichols, 1976), by examining whether
members of 839 late-adolescent twin pairs who shared more friends wer
e more similar in personality than those who shared fewer friends. Amb
iguity regarding the direction of effects was partially controlled by
comparing monozygotic and dizygotic twin pairs. Comparisons were made
between the effect of peer overlap and similarity of parental treatmen
t. These were made both for personality traits, as measured by the Cal
ifornia Psychological Inventory, and for academic achievement, as meas
ured by the subscale and total scores of the National Merit Scholarshi
p Qualifying Test. A number of the results were consistent with Harris
's theory, although quantitatively the effects were not strong.