Academic journal article Adolescence

Can Higher Grades Result in Fewer Friends? A Reexamination of the Relation between Academic and Social Competence

Academic journal article Adolescence

Can Higher Grades Result in Fewer Friends? A Reexamination of the Relation between Academic and Social Competence

Article excerpt

In contradiction to the assumption that social competence is independent of academic intelligence, significant positive correlations have been found between the two (Cauce, 1987; Keating, 1978; Walker & Foley, 1973). Cauce (1987) reported a correlation of .16 between grade point average (GPA) and peer judgments of popularity in a sample of secondary school students. Green et al. (1980) found a correlation of .33 between sociometric peer judgments of "best friend" and grade point average for primary school children. Pelligrini (1985) found a correlation of .38 between academic achievement and peer judgments of positive aspects of social conduct (such as leadership qualities) with primary school children. Fork and Tisak (1983) reported a positive correlation of .52 between peer judgments of social competence and grade point average with secondary school children.

A possible explanation for these positive correlations is that social and academic competence have global competence as a common element. Ford and Tisak (1983) have suggested that social behavior forms a separate domain, while cognitive social competencies have more similarities with academic competencies. Children who have better reasoning ability are better at both tuning their conduct in complex social situations and solving academic problems. Higher social competence could be explained by better cognitive abilities to choose social goals and to determine how to reach those goals (Ford, 1982).

The hypothesis of global competence contradicts the common notion that children with high learning capabilities are sometimes unpopular. Neisser (1976) states that "academically intelligent people do often behave stupidly. The existing evidence does not suggest that they are markedly more successful than the unintelligent" (p. 139). A possible explanation for a nonlinear relationship is offered by Coleman (1961): "Those who are seen as the intellectuals and who come to think of themselves in this way are not really those of highest intelligence, but are only the ones who are willing to work hard at a relatively unrewarded activity" (p. 265). Similarly, Hartup (1970) points out that "research shows that bright children who achieve at a realistic level have preferred status in the peer group; the overachieving child, however, may have a less certain position in his peer group" (p. 392).

Thus, there are two competing theories: the global-competence theory predicts a positive correlation between school grades and popularity, and the overachiever theory predicts a nonlinear relationship. Coleman's statement also suggests that the reward for a specific activity may play some role. It is assumed that efforts to achieve in some school subjects lead to lesser rewards than efforts in other subjects. Efforts directed at more abstract and less attractive school subjects may lead to decreased popularity, while efforts directed at subjects with everyday relevance, such as learning languages (e.g., English and Dutch), do not necessarily lead to decreased popularity.



Subjects were drawn from three secondary schools in Utrecht, a city in the Netherlands. A total of 157 students (81 females and 76 males, ages 16-17 years) in seven third-grade classes participated. The classes consisted of students of moderate academic level.


End-of-year grade points for mathematics, physics, Dutch, and English were used as indicators of academic performance.

Social competence was measured by having each student judge whether classmates were more, the same, or less competent in making friends than he or she was. Social competence was calculated by averaging the responses. Earlier research (Landsheer, 1991) has shown that using at least three judges provides reasonable test-retest reliability (r = .79). Intergroup reliability is also adequate provided groups of at least three judges are used (r = .61). All students received the judgments of at least seven peers. …

Search by... Author
Show... All Results Primary Sources Peer-reviewed


An unknown error has occurred. Please click the button below to reload the page. If the problem persists, please try again in a little while.