Self-reported delinquency among Alberta's youth: findings from a survey of 2,001 junior and senior high school students.
The present paper attempts to further examine, through the use of self-report data, relationships between extent of past-year delinquency and gender, grade level, psychosocial problems, and extent of past-year victimization. By including a range of both noncriminal and criminal acts in analyzing delinquency and victimization, the findings are intended to portray a fuller and more accurate description of the experiences of Canadian youth.
Findings presented in this paper draw on data from a 1999 study on youth victimization, crime and delinquency in Alberta conducted by the Canadian Research Institute for Law and the Family (CRILF) in collaboration with researchers from the University of Alberta (Gomes, Bertrand, Paetsch, & Hornick, 2000). The province of Alberta is located in western Canada and has a population of approximately 3 million, 80% of whom reside in urban areas.
A self-report questionnaire was distributed to junior (Grades 7, 8, and 9) and senior (Grades 10, 11, and 12) high school students aged 12 to 18. The survey was conducted in the spring and fall of 1999. The self-report method utilized in the study is recognized by researchers studying offending behavior as an effective means to obtain information about delinquency and victimization not available from official data sources (Bowling, Graham, & Ross, 1994; Creechan & Silverman, 1995; Tanner, 2001). This method offers the advantage of providing information about incidents not reported by the police as well as information about incidents not reported to the police or possibly to anyone at all.
In total, 67 junior and senior high schools from the public and separate (Catholic) systems agreed to participate in the survey. These schools were located in 15 cities, towns and rural areas. Selection of schools (excluding those with special or alternative curriculums) was made randomly where possible. In larger cities, one public senior high school and two public junior high schools, and one Catholic senior high school and two Catholic junior high schools were randomly chosen from each geographic quadrant of the city. This method was also utilized in smaller cities where there were sufficient numbers of schools. Random selection was not possible in towns and rural areas that had only one public and/or one Catholic school.
Selection of students to survey was also made randomly where possible. Based on provincial enrollments by grade, a proportionate stratification sampling technique was utilized to determine target sample sizes for each of the grade levels. With the help of school officials, student names and addresses were randomly generated from enrollment records. This procedure was not possible for all schools, and adjustments had to be made for selection of student names and distribution of consent forms.
A course-based selection method was used for 38 schools because school administrators were unable to generate the required mailing lists of student names or because class enrollments were too small and the sampling technique was not practical. For these schools, a course that was mandatory for all students was chosen, and students who were enrolled in the course were selected to be contacted about the survey. Consent forms were mailed, except for 14 schools where the students had to take the forms home to their parent/guardian.
In all, 6,656 students were contacted about the survey. Each student's parent/guardian received a letter describing the survey, along with a consent form, which was to be signed and returned to CRILF (a business-reply envelope was provided). Signed consent forms were received for 2,675 students (a response rate of 40%). School principals assisted with notifying students about the survey's date, time, and location. The researchers administered the survey questionnaire in the schools during school hours, and confidentiality, anonymity and the voluntary nature of participation were explained. After completing the questionnaire, the student inserted it into a blank envelope, then sealed it and dropped the package into a collection box.
A total of 2,001 students completed the questionnaire. Two types of response rates are meaningful: based on the 6,656 students contacted about the survey, the response rate was 30%; based on the 2,675 students for whom signed consent forms were received, the response rate was 75%.
Limitations in Representativeness of the Sample
Generalizability of the responses is limited by a number of factors even though procedures were adopted, where possible, to maximize representativeness of the sample to the population of adolescents in Alberta. First, a more systematic selection of schools from throughout the province was not feasible in the research project. Second, the nature of school-based surveys is such that they are limited to sampling students enrolled in school and, therefore, exclude students not in school for various reasons, including skipping classes (truancy), dropping out, attending private school, and being home schooled. Third, it should be noted that the experience of youth living on First Nations (aboriginal) reserves is not addressed in this research. Fourth, because the written materials were available only in English, it is possible that the sample underrepresents youth whose parents have limited English-language comprehension. Finally, voluntary participation may produce a self-selection bias such that those individuals who are most motivated to participate are also the ones at a lower risk of engaging in delinquent activities or being victimized. For these reasons, the rates of delinquency and victimization obtained from this survey should be considered as conservative estimates.
A final caution should be made regarding interpretation of findings. As with all cross-sectional surveys, measures that are taken only at a single point in time limit conclusions about cause-and-effect relationships between variables. Thus, discussion of the relationships between delinquency and psychosocial problems, and between delinquency and victimization, is based on correlations.
Of 2,001 respondents, 54% were female and 46% were male. The slightly higher proportion of females in the survey as compared to the population is not uncommon in social science surveys. About 60% of the participants were junior high school students and about 40% were senior high school students. The somewhat lower proportion of senior high students may be attributed to the fact that a considerable number of surveys had to be scheduled late in the school year when these students were busy preparing for examinations. It should also be noted that participation by older students tends to be lower for in-school surveys. Almost all of the students (97%) viewed themselves as "Canadian" and 74% indicated "White" when asked about ethnic/racial identity. Similar to the provincial population of adolescents, about 80% of the students were from urban centers and 20% from rural areas.
Psychosocial factors were examined via three scales: conduct problems, hyperactivity, and emotional problems. Using a problem behavior checklist describing the more common childhood psychiatric disorders, students were asked to indicate the extent to which each item was self-descriptive ("never/not true," "sometimes/somewhat true," or "often/very true"). The checklist is a modified version of the behavior/personality scale originally developed by Boyle and Offord (1994). Of the 34 items on the list, 15 measure conduct problems, 6 measure hyperactivity problems, and 13 measure emotional problems. Internal consistency measures (Cronbach's alpha) were determined to be within acceptable ranges for all three scales. The reliability coefficients were .83 for conduct problems, .72 for hyperactivity, and .82 for emotional problems.
The conduct problems scale includes a range of behaviors (e.g., being mean to animals, skipping school). Seven of the 15 items describe behaviors related to damaging property, stealing things, threatening people, and being physically violent. These behaviors are also included among the 13 delinquent acts asked about in the questionnaire. Taken as a single measure, however, level of conduct problems was judged to be intrinsically different enough from the measure of delinquency to warrant including the conduct problems scale in the analysis. More detailed examination confirmed that level of conduct problems was strongly and consistently related to each delinquent act (Gomes et al., 2000). A subscale of conduct problems consisting of only the eight items not comparable to any of the 13 delinquent acts was also tested and found to be significantly related to extent of delinquency and to property- and violence-related delinquency (all ps < .001; Cronbach's alpha = .67).
Delinquency was defined as engaging in one or more of 13 delinquent acts at least one time in the past year. Property-related delinquency and violence-related delinquency were analyzed separately. Property-related delinquency included the following: (1) damage/destroyed someone else's property on purpose; (2) stolen something worth less than $50; (3) stolen something worth $50 or more; (4) stolen something with a group of friends; (5) broken into a house; and (6) taken a car/motorcycle for a ride without the owner's permission. Violence-related delinquency included the following: (1) taken/tried to take something from someone using force/threat of force; (2) threatened to hurt someone/cause that person harm; (3) slapped, punched, or kicked someone in anger; (4) thrown something at someone to hurt that person; (5) threatened someone with a weapon (or object used as a weapon); (6) together with a group of friends, fought with others; and (7) touched someone in a sexual way against that person's will.
Extent of delinquency was measured by students' self-reported frequency of engaging in the different kinds of delinquent behaviors over the past year ("never," "once," "twice," or "three or more times"). Use of a one-year reporting time frame provided a way to "equalize" age differences; that is, that older individuals may report more delinquent behaviors simply because they have had more time and opportunity to do so (Bowling, Graham, & Ross, 1994). A delinquency score was computed by summing the number of incidents (ranging from 0 to 39). The scale was then recoded into the following levels of past-year delinquency: "none," indicating the student did not carry out any of the delinquent acts; "low," indicating at least one type of delinquent act was carried out one or two times; "moderate," indicating at least one type of delinquent behavior occurred three to seven times; and "high," indicating at least one type of delinquent act was carried out eight or more times.
Extent of victimization was measured by students' experience of a range of victimization incidents over the past year. They were asked to indicate whether they had been victimized (yes or no) in the following 11 ways: (1) something damaged/destroyed on purpose; (2) something stolen; (3) something taken by force/threat of force; (4) someone threatened harm; (5) slapped, punched or kicked; (6) something thrown; (7) threatened with a weapon; (8) attacked by a group/gang; (9) someone sexually exposed himself or herself; (10) someone said something sexually offensive; and (11) unwanted sexual touching. A victimization score was computed by summing the number of items to which the student responded yes (ranging from 0 to 11). The scale was then recoded into the following levels: "none," indicating the student was not victimized; "low," indicating one or two types of incidents occurred; "moderate," indicating three or four types of incidents; and "high," indicating at leave five types of incidents.
The chi-square test of association was used to assess statistically significant relationships. Discussion of the results is based on significant findings (ps < .001) and clearly specifies when this is not the case.
Profile of Delinquent Youth and Extent of Delinquency
Table 1 displays prevalence rates for total delinquency. The rates for property-related delinquency and violence-related delinquency are shown in Tables 2 and 3, respectively. Data on extent of delinquency are distributed by gender, school grade level, and level of psychosocial problems. Overall, 44.3% of the sample reported no involvement in past-year delinquency and 55.7% reported carrying out at least one delinquent act in the past year. Almost two-thirds (63.1%) of the students reported not engaging in any of the property-related delinquent behaviors, while over half (56.3%) reported no violence-related delinquency in the past year.
Proportionately more males than females indicated they had carried out one or more delinquent acts in the past year. At the low/moderate levels of extent of delinquency, rates between males and females were quite similar; however, females reported a slightly higher rate of violence-related acts (35.5%, as compared to 34.2% for males). Gender differences were particularly apparent at the high level of delinquency, where males (13.4%) were almost twice as likely as females (7.3%) to report engaging in delinquent acts. Similarly large differences were found for high levels of property-related and violence-related acts.
Younger students tended to report lower levels of delinquency as compared to older students. Grade 7 respondents were the least likely (41.6%) to report carrying out delinquent activities, and Grade 9 students were the most likely to report engaging in at least one type of delinquent behavior in the past year (64.2%). While Grade 9 students had the largest proportion reporting a high level of violence-related delinquency (14.3%), Grade 12 students had the largest proportion who reported a high level of property-related delinquency (15.2%).
Junior high school students reported higher rates for violent acts than for property-related ones. For example, among Grade 9 students, 14.3% had a high level of engagement in violent acts as compared to 12.5% for property-related delinquency. The opposite pattern was found for senior high students, and differences were larger. For example, for Grade 10 students, 6.7% had a high level of violence-related delinquency while 14.9% reported carrying out a high level of property-related delinquent acts.
Overall, students who had higher levels of psychosocial problems also reported a higher level of engagement in delinquent behavior over the past year. Notably, students who scored high on conduct problems were more likely to report a high level of delinquency than were students with high scores on hyperactivity or emotional problems. For total delinquency, 42.7% of students who had a high level of conduct problems also reported a high level of engagement in delinquent activities, as compared to only 24.8% of students with high hyperactivity problems and 16.7% with high emotional problems. Comparison of property-related and violence-related delinquency for psychosocial factors indicated that for low/moderate extent of past-year delinquency, students were consistently more likely to report engaging in violent acts than in property-related delinquent behaviors. For instance, of the students with a high level of conduct problems, 45% had engaged in a low/moderate level of violence-related acts as compared to 37% who reported the same level for property-related delinquency. A smaller difference was found for students with high hyperactivity levels; 39.1% had a low/moderate level of violence-related delinquency as compared to 33.5% for property-related acts. The largest difference between the two types of delinquency occurred for students who had a high level of emotional problems; 41.9% carried out a low/moderate level of violence-related acts, while only 28.3% reported low/moderate property-related delinquency. In general, the pattern appeared to be reversed for a high level of delinquent activity. Differences between types of delinquency for students with high levels of hyperactivity and emotional problems were too small, however, to allow any definitive conclusions to be drawn. For students with a high score on conduct problems, a greater proportion reported a high level of property-related delinquency (39.3%) as compared to violence-related acts (32.8%).
Correlation Between Extent of Delinquency and Extent of Victimization
As shown in Table 4, students who reported lower levels of engaging in delinquent behavior in the past year were less likely to report that they were victimized in the past year. For total delinquency, 48% of the students who indicated no delinquent behavior in the past year also reported that they were not victimized in the past year. Students who reported some level of delinquency were more likely to report being victimized. For example, 41.6% of students reporting a high level of delinquency also reported a high level of victimization, while only 6.4% of students reporting a high level of delinquency reported no victimization.
In comparing the two types of delinquency, the pattern of findings indicated that a higher level of engagement in violent behaviors was related to a higher level of victimization, more so than was the case for property-related delinquency and victimization. While 34% of the students who reported they had carried out a high level of property-related acts also reported a high level of victimization, 44.6% who indicated a high level of violence-related delinquency also reported they had been highly victimized. Further, 11.7% of the students with a high level of property-related delinquency reported no victimization, while only 2.3% of those with a high level of violence-related delinquency reported no victimization in the past year.
The delinquency-victimization relationship was further examined (in separate analysis) by grade level, and some notable differences were found between younger and older students (all ps < .001 except for Grade 10 students, where p < .05). Compared to senior high students, junior high students who reported moderate/high levels of delinquency were more likely to also report a high level of victimization. This difference was more evident for property-related delinquency than for violence-related acts. For those who had carried out one or more property-related delinquent acts, senior high students were more likely than junior high students to also report they had not been victimized. This pattern was weaker for violence-related delinquency and victimization (however, conclusions are limited by the small number who reported a high level of delinquency and no victimization).
In this paper, the focus has been on the association of self-reported extent of delinquency with several personal characteristics (gender, grade level, and psychosocial problems) and with victimization. The results provided fairly strong and consistent support for findings in the general research literature regarding these relationships. There were some interesting findings, however, when different levels and forms of delinquent behavior were taken into account.
Over half (56%) of the students reported that they had carried out at least one delinquent act in the past year. While evaluation of the prevalence rates is limited by the lack of comparable Canadian findings, our results appear to be in line with data from the International Self-report Delinquency Study, one of the few comparative studies about youth delinquency in various western cultures. Junger-Tas's (1994) past-year delinquency prevalence rates for five western countries range from 44% to about 72% (representing the Netherlands, England and Wales, Portugal, Switzerland, and Spain). A comparison of the findings with prevalence rates from a survey of 962 Calgary youth conducted in 1994/95 by the Canadian Research Institute for Law and the Family indicates that the level of past-year delinquency has stayed fairly constant, if not decreasing slightly, during a recent five-year period (Paetsch & Bertrand, 1999; Smith, Bertrand, Arnold, & Hornick, 1995).
The finding that males were more likely than females to report being involved in delinquency is also consistent with previous studies (Corsaro & Eder, 1990; Paetsch & Bertrand, 1999; Tanner, 2001). Gender differences increased for high levels of delinquency. While a larger proportion of students reported being involved in violence-related delinquency than property-related delinquency (44% as compared to 37%), the slightly higher reported rate for females than males on low/moderate violence-related delinquency warrants further investigation.
Older students were more likely to report engaging in delinquent behavior than younger ones. This pattern is comparable to that observed by Smith et al. (1995). Grade 9 students were noteworthy because of their relatively higher rates of delinquency as compared to other age groups. While reasons can possibly be located in developmental theories, it would also be useful to consider other factors such as school climate and peer-group culture in different school systems. In their study of the effect of peer group, social and school variables on level of delinquency, Paetsch and Bertrand (1999) found consistently strong correlations.
Delinquency rates for Grade 7 students appear to be lower than those reported in other studies. Sprott, Doob, and Jenkins (2001) draw on findings about delinquency in youth aged 12 to 13 from the National Longitudinal Survey of Children and Youth (NLSCY) to report past-year prevalence rates for property-related delinquency and aggressive behavior. They found that 30% of females and 40% of males had engaged in property-related delinquent acts, and 29% of females and 56% of males had engaged in some form of aggressive behavior. In contrast, our findings indicated that for Grade 7 students (the majority of whom would be about 12-13 years old), 19% reported property-related delinquency and 36% reported violence-related delinquency. More detailed analysis of the Grade 7 students indicated that 19% of females and 18% of males reported they had engaged in property-related delinquency, and 35% of females and 37% of males reported violence-related acts. (Gender differences were not statistically significant for proper ty-related delinquent behavior, but were significant at p <.001 for violence-related delinquency.) Reasons for the higher overall rates found in the NLSCY data are partly attributable to differences in survey methodologies and measurement of delinquency. NLSCY data on self-reported engagement in delinquent behavior are obtained from forms completed by the youths and sent back to Statistics Canada (Sprott et al., 2001). Possibly, this method would allow respondents to more easily exaggerate their activities. In addition, the NLSCY measure of delinquency includes five more items than our measure of delinquency.
For high levels of delinquency, younger students were found to report higher rates for violent acts than for property-related ones, with Grade 9 students reporting the largest differences between types of delinquency. Senior high students who reported high levels of delinquent behavior, however, were considerably more likely to report engaging in property-related delinquency than in violent acts. These differences between younger and older adolescents should be further explored.
Conduct, hyperactivity and emotional problems are recognized in the research literature as important risk factors in delinquent behavior. In the present study, students scoring high on psychosocial problems were consistently more likely to report a higher frequency of involvement in delinquency. The relationship between a high level of psychosocial problems and high extent of delinquency was found to be more prevalent for conduct problems than for hyperactivity or emotional problems. This pattern was generally consistent for both property-related and violence-related behaviors, and the results are comparable to 1994/95 Calgary findings by Smith et al. (1995). Again, these results point to the need to recognize and address differences in types of adolescent delinquency.
The victim-offender link among youth has been less widely studied; however, it has been suggested that there is a correlation between victimization and delinquency such that adolescents who are at greatest risk of being victimized are also those who engage in delinquent behavior (Lauritsen, Sampson, & Laub, 1991). The strong relationship between extent of delinquency and extent of victimization reported in this paper supports findings by Paetsch and Bertrand (1999). A higher level of involvement in delinquent behavior was associated with increased experience of victimization, and conversely, students with lower levels of engagement in delinquency were more likely to report they were not victimized at all in the past year. Students highly involved in violent behaviors were more likely to report being highly victimized than was the case for those highly involved in property-related delinquent acts. The relationship between high extent of victimization and type of delinquency differed by age group as well. For m oderate/high extent of property-related delinquency, junior high students had a higher prevalence rate of victimization than did senior high students. One possible reason for the differences between the types of delinquency is that engagement in violence-related delinquency often involves altercations with others and thus the individual takes on the roles of both the attacker and the attacked. In a separate qualitative analysis of descriptions of the most serious incidents, it was found that a considerable proportion of comments indicated that the respondent was involved in some type of altercation with a sibling, boyfriend or girlfriend. On the other hand, property-related delinquent acts such as theft or vandalism are less likely to occur in the presence of the persons being affected. The delinquency-victimization relationship described in this paper is limited by the kinds of activities asked about in the questionnaire. There is some evidence that this relationship does not hold for more extreme types of i ncidents (Regoeczi, 2000).
While the results presented are mostly consistent with previous research, generalizability of these findings is limited by several factors. Given the survey sample--youth in the public and Catholic school systems--the findings should be considered conservative. Marshall and Webb (1994) note that one characteristic of self-report studies of delinquent behavior is that they tend to exclude the extreme cases, that is, the most delinquent or the most serious criminal offenders. Tanner (2001) states that the self-report method, by its nature, "decontextualizes delinquent acts" and is unable to reflect the collective nature of such acts. Analyses of the extent of delinquency and victimization presented in this paper include a wide range of activities and do not distinguish between criminal and noncriminal acts in order to present a more realistic portrayal of young people's experiences. It should be kept in mind that a high level of delinquency does not necessarily translate into highly criminal behavior and charge able offenses.
Results of this study also support wider findings pointing to a general decline in youth delinquency and violence over the past several years. Nevertheless, the 56% self-reported delinquency rate must be taken seriously by those involved in policy development and by researchers. The findings suggest that we need to recognize demographic and personality differences in extent of delinquency and type of delinquent behavior. Policies must address these differences in order to be effective. Not only is youth delinquency related to gender, age, and psychosocial problems, its interrelationship with victimization indicates the need for research methodologies which would enable researchers to more clearly and fully describe these complex relationships.
Table 1 Total Self-Reported Deliquency (Past Year) Extent of Total Delinquency Personal None Low (1-2 times) n % n % Total Sample (N = 2,001) 887 44.3 458 22.9 Gender (1) Male (n = 915) 373 40.8 210 23.0 Female (n = 1,084) 514 47.4 247 22.8 Grade (2) 7 (n = 454) 265 58.4 86 18.9 8 (n = 399) 175 43.9 105 26.3 9 (n = 335) 120 35.8 81 24.2 10 (n = 282) 104 36.9 74 26.2 11 (n = 288) 122 42.4 54 18.8 12 (n = 237) 99 41.8 57 24.1 Psychosocial Factors Conduct Problems (3) Low (n = 805) 546 67.8 172 21.4 Moderate (n = 934) 306 32.8 259 27.7 High (n = 262) 35 13.4 27 10.3 Hyperactivity Problems (4) Low (n = 969) 535 55.2 228 23.5 Moderate (n = 766) 275 35.9 186 24.3 High (n = 266) 77 28.9 44 16.5 Emotional Problems (5) Low (n = 950) 487 51.3 218 22.9 Moderate (n = 793) 313 39.5 185 23.3 High (n = 258) 87 33.7 55 21.3 Extent of Total Delinquency Personal Moderate High (3-7 times) (8+ times) n % n Total Sample (N = 2,001) 454 22.7 202 Gender (1) Male (n = 915) 209 22.8 123 Female (n = 1,084) 244 22.5 79 Grade (2) 7 (n = 454) 88 19.4 15 8 (n = 399) 80 20.1 39 9 (n = 335) 89 26.6 45 10 (n = 282) 65 23.0 39 11 (n = 288) 81 28.1 31 12 (n = 237) 49 20.7 32 Psychosocial Factors Conduct Problems (3) Low (n = 805) 81 10.1 6 Moderate (n = 934) 285 30.5 84 High (n = 262) 88 33.6 112 Hyperactivity Problems (4) Low (n = 969) 165 17.0 41 Moderate (n = 766) 210 27.4 95 High (n = 266) 79 29.7 66 Emotional Problems (5) Low (n = 950) 174 18.3 71 Moderate (n = 793) 207 26.1 88 High (n = 258) 73 28.3 43 Extent of Total Delinquenc y Personal High (8+ times) % Total Sample (N = 2,001) 10.1 Gender (1) Male (n = 915) 13.4 Female (n = 1,084) 7.3 Grade (2) 7 (n = 454) 3.3 8 (n = 399) 9.8 9 (n = 335) 13.4 10 (n = 282) 13.8 11 (n = 288) 10.8 12 (n = 237) 13.5 Psychosocial Factors Conduct Problems (3) Low (n = 805) 0.7 Moderate (n = 934) 9.0 High (n = 262) 42.7 Hyperactivity Problems (4) Low (n = 969) 4.2 Moderate (n = 766) 12.4 High (n = 266) 24.8 Emotional Problems (5) Low (n = 950) 7.5 Moderate (n = 793) 11.1 High (n = 258) 16.7 (1) Missing data for 2 students. Statistical results for total delinquency: [chi square](3) = 23.6, p < .001. (2) Missing data for 6 students. Statistical results for total delinquency: [chi square](15) = 80.0, p < .001 (3)Statistical results for total delinquency: [chi square](6) = 656.7, p < .001. (4)Statistical results for total delinquency: [chi square](6) = 179.0, p < .001. (5)Statistical results for total delinquency: [chi square](6) = 55.4, p < .001. Table 2 Self-Reported Property-Related Delinquence (Past Year) Extent of Property-Related Delinquency Personal None Low/Moderate Characteristic (1-3 times) n % n % Total Sample (N = 2,001) 1,262 63.1 533 26.6 Gender (1) Male (n = 915) 542 59.2 252 27.5 Female (n = 1,084) 720 66.4 279 25.7 Grade (2) 7 (n = 454) 369 81.3 69 15.2 8 (n = 399) 262 65.7 105 26.3 9 (n = 335) 194 57.9 99 29.6 10 (n = 282) 149 52.8 91 32.3 11 (n = 288) 157 54.5 94 32.6 12 (n = 237) 129 54.4 72 30.4 Psychosocial Factors Conduct Problems (3) Low (n = 805) 675 83.9 122 15.2 Moderate (n = 934) 525 56.2 314 33.6 High (n = 262) 62 23.7 97 37.0 Hyperactivity Problems (4) Low (n = 969) 700 72.2 218 22.5 Moderate (n = 766) 441 57.6 226 29.5 High (n = 266) 121 45.5 89 33.5 Emotional Problems (5) Low (n = 950) 648 68.2 224 23.6 Moderate (n = 793) 472 59.5 236 29.8 High (n = 258) 142 55.0 73 28.3 Extent of Property-Related Delinquency Personal High Characteristic (4 + times) n % Total Sample (N = 2,001) 206 10.3 Gender (1) Male (n = 915) 121 13.2 Female (n = 1,084) 85 7.8 Grade (2) 7 (n = 454) 16 3.5 8 (n = 399) 32 8.0 9 (n = 335) 42 12.5 10 (n = 282) 42 14.9 11 (n = 288) 37 12.8 12 (n = 237) 36 15.2 Psychosocial Factors Conduct Problems (3) Low (n = 805) 8 1.0 Moderate (n = 934) 95 10.2 High (n = 262) 103 39.3 Hyperactivity Problems (4) Low (n = 969) 51 5.3 Moderate (n = 766) 99 12.9 High (n = 266) 56 21.1 Emotional Problems (5) Low (n = 950) 78 8.2 Moderate (n = 793) 85 10.7 High (n = 258) 43 16.7 (1) Missing data for 2 students. Statistical results for property-related delinquency: [chi square](2) = 18.6, p < .001. (2) Missing data for 6 students. Statistical results for property-related delinquency: [chi square](10) = 105.5, p < .001. (3) Statistical results for property-related delinquency: [chi square](4) = 476.1, p < .001. (4) Statistical results for property-related delinquency: [chi square](4) = 101.7, p < .001. (5) Statistical results for property-related delinquency: [chi square](4) = 29.0, p < .001. Table 3 Self-Reported Violence-Related Delinquency (Past Year) Extent of Violence-Related Delinquency Personal None Low/Moderate Characteristic (1-3 times) n % n % Total Sample (N = 2,001) 1,127 56.3 669 34.9 Gender (1) Male (n = 915) 492 53.8 313 34.2 Female (n = 1,084) 634 58.5 385 35.5 Grade (2) 7 (n = 454) 292 64.3 140 30.8 8 (n = 399) 213 53.4 149 37.3 9 (n = 335) 162 48.4 125 37.3 10 (n = 282) 143 50.7 120 42.6 11 (n = 288) 173 60.1 93 32.3 12 (n = 237) 140 59.1 72 30.4 Psychosocial Factors Conduct Problems (3) Low (n = 805) 619 76.9 177 22.0 Moderate (n = 934) 450 48.2 404 43.3 High (n = 262) 58 22.1 118 45.0 Hyperactivity Problems (4) Low (n = 969) 652 67.3 278 28.7 Moderate (n = 766) 371 48.4 317 41.4 High (n = 266) 104 39.1 104 39.1 Emotional Problems (5) Low (n = 950) 602 63.4 289 30.4 Moderate (n = 793) 415 52.3 302 38.1 High(n = 258) 110 42.6 108 41.9 Extent of Violence-Related Delinquency Personal High Characteristic (4 + times) n % Total Sample (N = 2,001) 175 8.7 Gender (1) Male (n = 915) 110 12.0 Female (n = 1,084) 65 6.0 Grade (2) 7 (n = 454) 22 4.8 8 (n = 399) 37 9.3 9 (n = 335) 48 14.3 10 (n = 282) 19 6.7 11 (n = 288) 22 7.6 12 (n = 237) 25 10.5 Psychosocial Factors Conduct Problems (3) Low (n = 805) 9 1.1 Moderate (n = 934) 80 8.6 High (n = 262) 86 32.8 Hyperactivity Problems (4) Low (n = 969) 39 4.0 Moderate (n = 766) 78 10.2 High (n = 266) 58 21.8 Emotional Problems (5) Low (n = 950) 59 6.2 Moderate (n = 793) 76 9.6 High(n = 258) 40 15.5 (1) Missing data for 2 students. Statistical results for violence-related delinquency: [chi square](2) = 22.8, p < .001. (2) Missing data for 6 students. Statistical results for violence-related delinquency: [chi square](10) = 44.9, p < .001 (3) Statistical results for violence-related delinquency: [chi square](4) = 417.9, p < .001 (4) Statistical results for violence-related delinquency: [chi square](4) = 142.8, p < .001 (5) Statistical results for violence-related delinquency: [chi square](4) = 51.6, P < .001 Table 4 Relationship Between Self-Reported Delinquency and Victimization (Past Year) (1) Extent of Victimization (2) Extent of Delinquency None Low (1-2 types) n % n Total Delinquency None (n = 887) 426 48.0 307 Low (1-2 times) (n = 458) 109 23.8 201 Moderate (3-7 times) (n = 454) 49 10.8 161 High (8+ times) (n = 202) 13 6.4 50 Property-Related Delinquency None (n = 1,262) 490 38.8 464 Low/Mod. (1-3 times) (n = 533) 83 15.6 201 High (4+ times) (n = 206) 24 11.7 54 Violence-Related Delinquency None (n = 1,127) 492 43.7 417 Low/Mod. (1-3 times) (n = 699) 101 14.4 258 High (4+ times) (n = 175) 4 2.3 44 Extent of Victimization (2) Extent of Delinquency Low Moderate (1-2 (3-4 types) types) % n % Total Delinquency None (n = 887) 34.6 109 12.3 Low (1-2 times) (n = 458) 43.9 103 22.5 Moderate (3-7 times) (n = 454) 35.5 144 31.7 High (8+ times) (n = 202) 24.8 55 27.2 Property-Related Delinquency None (n = 1,262) 36.8 208 16.5 Low/Mod. (1-3 times) (n = 533) 37.7 145 27.2 High (4+ times) (n = 206) 26.2 58 28.2 Violence-Related Delinquency None (n = 1,127) 37.0 153 13.6 Low/Mod. (1-3 times) (n = 699) 36.9 209 29.9 High (4+ times) (n = 175) 25.1 49 28.0 Extent of Victimization (2) Extent of Delinquency High (5+types) n % Total Delinquency None (n = 887) 45 5.1 Low (1-2 times) (n = 458) 45 9.8 Moderate (3-7 times) (n = 454) 100 22.0 High (8+ times) (n = 202) 84 41.6 Property-Related Delinquency None (n = 1,262) 100 7.9 Low/Mod. (1-3 times) (n = 533) 104 19.5 High (4+ times) (n = 206) 70 34.0 Violence-Related Delinquency None (n = 1,127) 65 5.8 Low/Mod. (1-3 times) (n = 699) 131 18.7 High (4+ times) (n = 175) 78 44.6 (1) N = 2,001 (2) Statistical results: total delinquency, [chi square](9) = 465.0, p < .001; property-related delinquency, [chi square](6) = 232.7, p < .001; violence-related delinquency, [chi square](6) = 426.1, p < .001.
Bowling, B., Graham, J., & Ross, A. (1994). Self-reported offending among young people in England and Wales. In J. Junger-Tas, G. Terlouw, & M. W. Klein (Eds.), Delinquent behavior among young people in the western world (pp. 42-64). New York: Kugler.
Boyle, M. H., & Offord, D. R. (1994). Psychiatric disorder and substance use in adolescence. Canadian Journal of Psychiatry, 36, 699-705.
Carsaro, W. A., & Eder, D. (1990). Children's peer cultures. Annual Review of Sociology, 16, 197-220.
Creechan, J. H., & Silverman, R. A. (1995). Canadian delinquency. Scarborough: Prentice-Hall.
Dolmage, W. R. (1996). One less brick in the wall: The myths of youth violence and unsafe schools. Education and Law, 7, 185-207.
Doob, A. N., & Sprott, J. B. (1998). Is the "quality" of youth violence becoming more serious? Canadian Journal of Criminology, April, 185-194.
EKOS Research Associates, Inc. (2000, September 12). Canadian attitudes toward the prevention of crime: Overview of study findings. Presented to National Crime Prevention Centre, Ottawa, Canada.
Gomes, J. T., Bertrand, L. D., Paetsch, J. J., & Hornick, J. P. (2000). The extent of youth victimization, crime and delinquency in Alberta, 1999. Calgary, Alberta: Canadian Research Institute for Law and the Family.
Junger-Tas, J. (1994). Delinquency in thirteen western countries: Some preliminary conclusions. In J. Junger-Tas, G. Terlouw, & M. W. Klein (Eds.), Delinquent behavior among young people in the western world (pp. 370-380). New York: Kugler.
Lauritsen, J. L., Sampson, R. J., & Laub, J. H. (1991). The link between offending and victimization among adolescents. Criminology, 29(2), 265-292.
Marshall, I. H., & Webb, V. J. (1994). Self-reported delinquency in a midwestern American city. In J. Junger-Tas, G. Terlouw, & M. W. Klein (Eds.), Delinquent behavior among young people in the western world (pp. 319-338). New York: Kugler.
Paetsch, J. J., & Bertrand, L. D. (1999). Victimization and delinquency among Canadian youth. Adolescence, 34(134), 351-367.
Regoeczi, W. C. (2000). Adolescent violent victimization and offending: Assessing the extent of the link. Canadian Journal of Criminology, 42(4), 493-505.
Smith, R. B., Bertrand, L. D., Arnold, B. L., & Hornick, J. P. (1995). A study of the level and nature of youth crime and violence in Calgary. Calgary: Calgary Police Service.
Sprott, J. B., Doob, A. N., & Jenkins, J. M. (2001). Problem behavior and delinquency in children and youth. Juristat, 21(4). Ottawa: Canadian Centre for Justice Statistics.
Statistics Canada. (2001). Youth court statistics 1999-2000. The Daily, May 30. Ottawa: Statistics Canada.
Stevenson, K., Tufts, J., Hendrick, D., & Kowalski, M. (1999). Youth and crime. Canadian Social Trends, Summer, 17-21.
Sullivan, M. L., & Miller, B. (1999). Adolescent violence, state processes, and the local context of moral panic. In J. M. Heyman (Ed.), States and illegal practices (pp. 261-283). New York: Berg.
Tanner, J. (2001). Teenage troubles--Youth and deviance in Canada (2nd ed.). Toronto: Nelson-Canada.
This article is based on data from a study by the Canadian Research Institute for Law and the Family. Administration of the survey was conducted in collaboration with researchers from the University of Alberta, Canada. Funding for the study was provided by the Alberta Law Foundation and the Alberta Heritage Foundation for Medical Research.
Jeanette T. Gomes, Lame D. Bertrand, Joanne J. Paetsch, and Joseph P. Hornick, Canadian Research Institute for Law and the Family, Calgary, Alberta, Canada.
Requests for reprints should be sent to Jeanette T. Gomes, Canadian Research Institute for Law and the Family, c/o Faculty of Law, University of Calgary, 2500 University Drive N.W., Calgary, Alberta, Canada T2N 1N4. Email: firstname.lastname@example.org
|Printer friendly Cite/link Email Feedback|
|Author:||Gomes, Jeanette T.; Bertrand, Lorne D.; Paetsch, Joanne J.; Hornick, Joseph P.|
|Date:||Mar 22, 2003|
|Previous Article:||The relationship between depression and smoking in adolescents.|
|Next Article:||Understanding depressed mood in the context of a family-oriented culture.|