Personality and Individual Differences 30 (2001) 627±640
www.elsevier.com/locate/paid
Sensation seeking as a predictor of positive and negative risk behaviour among adolescents Ellen Beate Hansen, Gunnar Breivik * Norwegian University of Sports and Physical Education, Box 4014 Ullevaal Stadion, 0806, Oslo, Norway Received 15 February 1998; received in revised form 1 February 2000
Abstract The present paper examines the relationship between sensation seeking and risk-taking behaviour among adolescents. Risk behaviour is de®ned as positive risk behaviour (activities like climbing, kayaking, rafting etc.) and negative risk behaviour (crime and socially unacceptable activities like shoplifting, drug use etc.) Perceived challenges and in¯uences from school, parents, friends and social background are examined as contributing factors. Three hundred and sixty adolescents between 12 and 16 years of age from a school in Trondheim, Norway, answered the tests. The test consisted of My Opinion II, a Swedish version of the sensation seeking scale, which measures sensation seeking among adolescents around the age of 14. Three other questionnaires developed especially for this study were also used; one measuring risk behaviour, one measuring challenges from school, parents and friends, and one measuring social background. The results indicate a strong relationship between sensation seeking and both types of risk behaviour. Negative risk behaviour correlates negatively with challenges from both school and parents, and a similar relationship exists between negative risk behaviour and social background. It seems that few challenges and a poor social background could result in more negative risk behaviour. # 2001 Elsevier Science Ltd. All rights reserved. Keywords: Sensation seeking; Risk behaviour; Adolescents
1. Introduction The purpose of this study was to see whether there is reason to believe that risk sports could prevent problem behaviour and juvenile delinquency. Earlier research indicates that risk activities and adventure programs are successful methods in rehabilitation of criminals and drug abusers
* Corresponding author. Tel.: +47-22-185600; fax: +47-22-185765. E-mail address:
[email protected] (G. Breivik). 0191-8869/01/$ - see front matter # 2001 Elsevier Science Ltd. All rights reserved. PII: S0191-8869(00)00061-1
628
E.B. Hansen, G. Breivik / Personality and Individual Dierences 30 (2001) 627±640
(Durgin & McEwen, 1993). Interviews of criminals have shown that criminal behaviour is an alternative when there is a lack of opportunity to do exciting and thrilling activities (Robertson, 1994). Since risk sports are often used in rehabilitation, it would be interesting to try such interventions at an earlier point of time in the delinquent's life. This could possibly prevent him or her from entering the criminal path. The personality trait sensation seeking (Zuckerman, Kolin, Price & Zoob, 1964) characterises people who have a tendency to seek strong sensations and experiences by taking dierent kinds of risks. The various versions of the sensation seeking scale (SSS) measure this trait, and several former studies demonstrate that both participants in high risk sports and criminals get a high score on this scale (Zuckerman, Bone, Neary, Mangelsdorf & Brustman, 1972; Zuckerman, 1994). In this study, a sensation seeking scale developed by BjoÈrck-AÊkesson (1990) has been used along with measures of participation in dierent types of risk behaviour, challenges in the social environment and social background. The respondents were adolescents, 12±16 years of age. Our ®rst objective was to look at the relationship between sensation seeking and participation in different kinds of risk behaviour. The second objective was to examine the respondents' own experiences of challenges in the environment, and to see if there is a relationship between experience of challenges and the kind of risk behaviour they engage in. 2. Earlier research 2.1. Sensation seeking The theoretical basis for the present study has been Zuckerman's sensation seeking theory (Zuckerman, 1979, 1994). His de®nition (1994) serves as an eective basis also for a study of adolescents: ``Sensation seeking is a trait de®ned by the seeking of varied, novel, complex, and intense situations and experiences, and the willingness to take physical, social, and ®nancial risks for the sake of such experience'' (Zuckerman, 1994). The study was designed to determine how adolescents seek stimulation and take risks in relation to these parameters: parents, social background, school, friends and sports. Zuckerman's theory makes it possible to see how individual dierences play a role in how the interaction pattern of the above mentioned factors is realised also among youth. 2.2. Problem behaviour and sensation seeking Sensation seeking and impulsiveness are factors that have been found to in¯uence problem behaviour, especially drug use and criminality (Petersen et al., 1993). Zuckerman et al. (1972) found that there was a connection between sensation seeking and heterosexual activity, drug- and alcohol use. Zuckerman (1994) has reviewed a number of studies examining sensation seeking and drug abuse. The studies show that the use of drugs correlates positively with high scores on the sensation seeking scales. It seems that the reasons to begin drug use are not necessarily the same as the reasons to continue using them. Sensation seeking is crucial in the early phase where the search for a new experience is often an important factor for trying drugs. The reason that some
E.B. Hansen, G. Breivik / Personality and Individual Dierences 30 (2001) 627±640
629
people continue is often to avoid the withdrawal eects and to relieve pain. High sensation seekers have a greater potential for being multi-drug users, because they constantly seek new experiences. It also seems that high sensation seekers prefer more stimulating drugs such as cocaine and amphetamine than drugs that are more relaxing such as sedatives and hypnotics (Breivik, 1998). Pedersen, Clausen and Lavik (1988) looked at the relationship between sensation seeking and Norwegian adolescents' use of alcohol, tobacco and other psychoactive drugs. The study concluded that the disinhibition and the experience seeking subscales were strongly and moderately correlated, respectively, with some of the drugs, especially tobacco and alcohol. Thrill and adventure seeking did not show any relationship with the use of drugs. Iso-Ahola and Crowley (1991) found that drug abusers were especially susceptible to experience boredom in their leisure time. They used drugs and various kinds of antisocial behaviour to meet the need for activity and stimulation. School and other organised activities failed to satisfy their needs. Iso-Ahola and Crowley (1991) also pointed out that the drug abusers were among those who favoured activities like football, gymnastics, skateboard and roller-skates. The fact that they still did not feel satis®ed showed their high sensation seeking needs. Zuckerman (1994) reviews the studies, which compare scores on the sensation seeking scales with criminality and problem behaviour. The studies show that sensation seeking in early adolescence is a strong predictor of the lack of law-obeying behaviour and of maladjusted social behaviour later in life. Criminals have approximately the same score on the sensation seeking scales as non-criminals who engage in high-risk sports or have high-risk professions such as ®remen, police and lifeguards. As far as criminals in jail are concerned, high sensation seeking shows a positive correlation with rebellious behaviour. It is also found that American high school students, who report high incidences of illegal conduct, have higher scores on total sensation seeking and on all the subscales except the thrill and adventure scale, compared with students who reported no illegal conduct. Boredom susceptibility also correlates positively with problem behaviour in school. In a study by Robertson (1994), eight male criminals explained why they behaved criminally. The majority stated that they committed crimes because they did not have the opportunity to join more socially acceptable and stimulating activities due to poverty and to the poor recreational resources in the neighbourhood. All of them said they broke the law mainly because of the adrenaline rush it gave them. They would all have liked to engage in rafting, climbing and other outdoor activities if they had had the opportunity, and they thought this would give them the same good feeling of rush they sought through criminality. 2.3. Prevention and rehabilitation According to Iso-Ahola and Crowley (1991), it is more important to take easily bored adolescents out to experience stimulating activities, rather than telling them what to do through passive didactic and cognitive methods. For young people who like to take risks, challenging sports and adventure programs can be a great success. Studies show that criminals and rehabilitated drug abusers often end up on their former tracks because they lack the opportunity to continue with stimulating new activities (Durgin & McEwen, 1993). They often end up, either without a job, or with monotonous and boring jobs, due to
630
E.B. Hansen, G. Breivik / Personality and Individual Dierences 30 (2001) 627±640
missed education. Furthermore, they have a poor social network, since most of their old friends are still drug abusers (Zuckerman, 1994). Prevention is, however, important when it comes to criminality and drug abuse among adolescents. Most of the adventure programs are based on the rehabilitation of individuals who have already ended up on the wrong side of the law. It would be useful to ®nd out whether it is possible to start programs earlier in life to avoid some of the seemingly unnecessary youth involvement in drugs, criminality and problem behaviour. School is one of the interesting arenas for action plans. BjoÈrck-AÊkesson (1990) found that pupils scoring high on the sensation seeking scale (My Opinion I and II) preferred to develop their own creativity and activity and did not want strong control from the teacher. Lively and dynamic teaching with room for the student's own thinking and discovery, and a high activity level with variation and unpredictability, suits a high sensation seeking person. If the teaching situation is too passive, the high sensation seeker will begin with other activities than schoolwork, and soon get a reputation as a problem pupil. According to BjoÈrck-AÊkesson (1990) schools should have more challenging environments and oer learning possibilities that involve stimulating experiences, even for high sensation seekers. The teaching should be adjusted to meet the needs of these pupils, as well as the needs of low sensation seekers. 3. Methodology 3.1. Subjects The respondents in this study were pupils at a large school in Trondheim, Norway. Not all students were in attendance. Six pupils did not ®nish the questionnaire, while 360 completed all the scales. They were pupils from 16 classes on the 7th, 8th and 9th grade levels. Their ages were from 12 to 16. The mean was 13.8 years. The respondents were 184 boys (51.1%) and 172 girls (48.9%). 58.4% of the respondents lived in the centre or close to the centre of the town, while 40.6% lived outside the centre of the town. The data was collected throughout 1 week in October. Questionnaires were ®lled out by the pupils in the classrooms during one of their lessons. No questions were answered by the teacher while they ®lled out the questionnaires, except for practical ones. The respondents used from 30 to 45 min to ®ll out the questionnaires. The scales were administered anonymously. 3.2. Tests Sensation seeking was measured with the standardised questionnaire My Opinion II (BjoÈrckAÊkesson, 1990) which is a Swedish scale inspired by Zuckerman's scales. This scale was chosen instead of the Norwegian sensation seeking scale developed by Pedersen et al. (1988) because the Swedish scale was developed for adolescents around age 14 whereas the Pedersen scale was developed for adolescents 17±19 years old. The same applies to the Norwegian version of Zuckerman's sensation seeking scale form V by Breivik (1983) which is not well suited to investigate attitudes among adolescents, since the scales contain items referring to adult behaviour. The Swedish language is very similar to Norwegian and the translation of BjoÈrck-AÊkesson's
E.B. Hansen, G. Breivik / Personality and Individual Dierences 30 (2001) 627±640
631
scale therefore should not aect validity or reliability in any considerable way. BjoÈrck-AÊkesson obtained a measure of reliability close to Cronbachs alpha by using a lisrel analysis and the algorithm by Reuterberg and Gustafsson. This gave an estimate of 0.82 (BjoÈrck-AÊkesson, 1990). My Opinion II consists of 62 pairs of statements where the respondent has to choose the one that best describes him or her. One of the statements describes a high sensation seeking attitude, while the other describes a low sensation seeking attitude. For example: (A) I often dream about being a mountain climber; (B) I don't understand people that risk their lives just to climb a mountain. The scale is divided into four subscales, but they have slightly dierent de®nitions than those in the Zuckerman's SSS. Thrill and adventure seeking (TAS) measures the need for physical risks and is similar to Zuckerman's TAS-scale. New experience seeking (NES) measures the desire to engage in new activities, to be impulsive and not have control. NES has much in common with Zuckerman's ESscale. Activity (Act) measures the urge to do wild things, preferably with the gang. This subscale resembles Zuckerman's Dis-scale. Outgoingness (Out) measures the desire to be the centre of attention. This subscale contain a lot from both the ES- and the BS-scale in Zuckerman's version. (BjoÈrck-AÊkesson, 1990). The other questionnaires that were used were designed especially for this study. One questionnaire measured involvement in risk behaviour. It listed 35 dierent activities. Eighteen of them are criminal or socially not well accepted (negative) activities and 17 activities are socially well accepted and legal (positive) activities. The respondents were to choose between the answers `never done', `done once' and `often done'. Cronbach's alpha was 0.77 in the positive scale and 0.83 in the negative scale. The questionnaire that measures challenges in the social environment is divided into three parts. One part is related to the school environment, one concerns parents and one concerns friends. All three parts have statements regarding the challenge that the adolescents experience. The respondents were to answer by grading whether this was right or wrong in their case on a scale from 1 to 5. One statement from the scale that measures challenges from parents goes like this: ``My parents won't let me do exciting and risky activities''. Social background was measured in the same way, with statements and a 5-point grading scale. The statements in this scale gave information about the family's economy, whether both parents worked, whether parents travelled or were away from home a lot, did the adolescent like it at home with their parents, etc. The scales about risk behaviour, challenges and social background were ®rst tested on two classes (7th grade and 9th grade) at a school in Oslo to see if they measured what they were supposed to measure. After a conversation with the respondents and statistical analysis of the pilot test, some changes were made on the scales. All the statistical analyses were done in the program Statistical Package for Social Sciences (SPSS). The analyses most used were frequencies and bi-variate correlations, but also t-tests were used.
632
E.B. Hansen, G. Breivik / Personality and Individual Dierences 30 (2001) 627±640
4. Results 4.1. Sensation seeking The total sensation seeking scores of the respondents showed a normal distribution pattern, with a mean of 88.5. The highest possible score is 112 and the lowest is 56. Most of the respondents scored somewhere between 75 and 100. The sub-scales showed a similar distribution. T-tests (Table 1) indicated that there were some weak, but still signi®cant dierences between boys and girls. Girls scored higher than boys on new experiences and outgoingness, while boys scored higher than girls on thrill and adventure seeking. In the case of Total sensation seeking and the sub-scale activity, there was no dierence between boys and girls. These ®ndings are somewhat unusual and will be discussed later. 4.2. Risk behaviour We were interested in ®nding out how common dierent types of risk behaviour were among youth, both the positive and the negative forms. We therefore looked at frequencies for the different types of behaviour. Distribution of percentages of involvement in risk behaviour is given in Figs. 1 and 2. The distribution of risk behaviour showed dierent patterns for the negative and positive categories. The scores for positive risk behaviour had a mean of 30.3 (SD 4.79), where maximum was 51 and minimum was 17. Negative risk behaviour had a maximum score at 54 and a minimum score at 18, and the mean was 23.1 (SD 5.18). The respondents reported more engagement in positive risk behaviour than in negative, relative to the categories we had chosen. The negative behaviour was concentrated on the most innocent alternatives like ``ring the doorbell and run'' and ``sneak in on places with cover charge''. Still there were also some adolescents who were engaged in more serious negative activities, which can harm themselves as well as others. A similar list was made for positive risk behaviour. Many students had been involved in dierent positive risk activities. The most popular ones were activities like ``riding thrilling rollercoasters'', ``cycle downhill at high speed'' and ``ski downhill at high speed''. These are among the more common activities that Norwegian boys and Table 1 T-test between the sexes on sensation seeking Test
TAS NES Act Out Total SS
Girls (N=176)
Boys (N=184)
M
SD
M
SD
23.73 25.08 22.10 18.34 89.25
2.89 2.50 3.08 2.32 8.06
24.62 24.15 21.57 17.47 87.82
2.98 2.39 3.04 2.43 8.32
t (two-tailed)
p
2.88 ÿ3.62 ÿ1.63 ÿ3.46 ÿ1.66
