FACTORS RELATED TO PHYSICAL ACTIVITY: A STUDY OF ADOLESCENTS RUNAR VILHJALMSSON 1 * and THOROLFUR THORLINDSSON 2 1 Department of Nursing, University of Iceland, Eiriksgotu 34, 101 Reykjavik, Iceland and 2 Faculty of Social Sciences, University of Iceland, Sturlugotu, 101 Reykjavik, Iceland AbstractÐAlthough the consequences of physical activity have been carefully documented, less is known about its correlates, particularly among children and youth. Based on a representative national survey of 1131 Icelandic adolescents, the study examined various physical, psychological, social and demographic factors related to physical activity. Male sex, signi®cant others' involvement in physical activity (father, friend and older brother), sociability, perceived importance of sport and of health improvement and satisfaction with mandatory gym classes in school, were all related to more involve- ment, whereas hours of paid work and TV-viewing were related to less. Furthermore, the data suggested that the in¯uence of friend's participation in physical activity depends on his or her emotion- al signi®cance. In¯uential others appeared to aect males and females in the same way. The meaning of the results and their implications for future research are discussed. # 1998 Elsevier Science Ltd. All rights reserved Key wordsÐsport, exercise, socialisation, attitudes, life-style, Iceland INTRODUCTION Physical activity in the form of sport and exercise is a vast enterprise in contemporary Western society. It involves major social institutions, such as the family, schools and the media and occupies rapidly increasing numbers of participants, workers and consumers (Leonard, 1993, pp. 3±8). Its broad sig- ni®cance has drawn attention from scholars in var- ious ®elds. Despite exceptions (Hafsteinsson, 1989; Martens et al., 1990; Begel, 1992), the picture of sport and exercise is generally favorable. Psychological and physiological bene®ts have been repeatedly documented in adults (e.g. Taylor et al., 1985; Bouchard et al., 1990) and recent studies of adolescents suggest the same (Mechanic and Hansell, 1987; Thorlindsson et al., 1990; Vilhjalmsson and Thorlindsson, 1992). Sport and exercise provides avenues to autonomy, indepen- dence and self-esteem (Gruber, 1986; Coakley and White, 1992; Thorlindsson et al., 1994). Participants are less depressed and anxious, have fewer aches and pains and are more energetic and satis®ed (Vilhjalmsson and Thorlindsson, 1992; Biddle, 1993). They are also less likely to smoke cigarettes and drink alcoholic beverages (Buhrmann, 1977; Hastad et al., 1984; Thorlindsson, 1989; Thorlindsson and Vilhjalmsson, 1991), they assess their health more positively than others (Mechanic and Hansell, 1987; Thorlindsson et al., 1990; Vilhjalmsson, 1994) and get better academic results (Marsh, 1993; Thorlindsson et al., 1994). Also, through sport and exercise, participants interact with others and maintain social relationships (Snyder and Spreitzer, 1983, pp. 288±289; Mechanic and Hansell, 1987). The generally favorable research results are re¯ected in policy statements stressing the import- ance of physical activity for the health of young people and the need to sustain the activity into adulthood (American College of Sports Medicine, 1988; The Sports Council, 1988). Despite all the bene®ts physical activity may have, researchers and educators repeatedly report that many young people are physically inactive (Williams, 1988; Armstrong et al., 1990; Bjarnason, 1991; Biddle and Armstrong, 1992) and that eorts to mobilise them are hampered by a limited understanding of the process of getting and staying involved (Dishman and Dunn, 1988; Biddle and Armstrong, 1992; Coakley and White, 1992). Factors related to physical activity Recent theoretical and empirical work pertaining to adolescent and adult physical activity suggests that it is a complex phenomenon related to a host of dierent personal and social variables (Nutbeam et al., 1989; Anderssen and Wold, 1992; Coakley and White, 1992). The literature identi®es or impli- cates at least ®ve sets of correlates, i.e. personal attitudes and beliefs, socialisation, lifestyle, medical condition and sociodemographic background. Attitudes/beliefs. Attitudes are widely believed to be related to human behaviour in general and sport and exercise involvement in particular. Attitude- behaviour researchers emphasise cognitive and Soc. Sci. Med. Vol. 47, No. 5, pp. 665±675, 1998 # 1998 Elsevier Science Ltd. All rights reserved Printed in Great Britain 0277-9536/98 $19.00 + 0.00 PII: S0277-9536(98)00143-9 *Author for correspondence. 665