On the conception and measurement of popularity: more facts and some straight conclusions

For many years, children receiving many liking nominations have been considered popular in the sociometric literature. Recently, doubts have arisen that perhaps students' and teachers' notion of popularity involves social status and dominance rather than personal liking. It is argued in this paper that sociometric nominations of personal liking represent attractiveness for intimate friendship, whereas popularity should be measured directly via judgmental sociometry, in which children nominate directly the popular students and are not requested to express their personal feelings. Nominations of attractiveness, collected in conventional affective sociometry, were compared in six samples (153 classrooms) to direct judgments of popularity. Attractiveness and popularity were found to be distinct psychological constructs: (a) the average correlation between these nominations was only r=0.44, and in most classrooms (72%) the same child did not receive the highest number of nominations for attractiveness and popularity; (b) a substantial proportion of students were highly popular but only moderately attractive or vice versa (21.5%), whereas only 9% were high in both attributes; (c) in terms of classroom climate and perceptions of classroom processes, popular students were less satisfied than attractive students, and were more critical of the learning climate and of their teachers' differential teaching behavior toward low- and high-achievers; and (d) differences between popular and attractive students were found in several attributes and aspects of self-esteem. These findings lead to the conclusion that the use of popular in the Coie, Dodge, and Coppotelli (1982) two-dimensional model and the ensuing sociometric literature have been erroneous and lacking in ecological validity. Popularity should be measured directly via judgmental sociometry rather than be inferred from affective liking nominations.