Marijuana Abuse in Incarcerated Adolescents: Reasons for Use and Implications for Treatment

Research is needed to better understand marijuana use and effective treatment among the incarcerated adolescent population. Preliminary pilot work in detention indicates that a substantial number (83%) of adolescents are reporting daily marijuana use (Stein et al., 2006). This research attempted to gain a better understanding of the reasons for marijuana use among incarcerated adolescents. The design and intent of this study was to extend basic knowledge of the reasons for marijuana use, establish factors related to marijuana use, and examine ifrelaxation therapy reduces marijuana use in those who use marijuana as a way to regulate emotions. Participants were 189 incarcerated adolescents age 14-19 (M= 17 years). Participants adjudicated between January 2001 and September 2005 were included in the study if they reported marijuana use prior to their incarceration. Of the 189 participants, 86% were boys, 32.8% White, 29.1 % Hispanic /Latino, 28% African American, and 10.1 % other. Analyses included a principal components analysis (PCA) to investigate variables that correlate to explain reasons for marijuana use . The PCA did not reveal separate factors for marijuana use. In order to continue testing the hypotheses for the study, items from the Brief Situational Confidence Questionnaire-Marijuana (BSCQ-M) were calculated to form a negative affect and a positive affect variable. Although most of the analyses did not confirm the hypotheses, a few interesting and significant findings were revealed. Results indicated that incarcerated adolescents who were in the negative affect group report higher averages of weekly marijuana use and a higher number of days using marijuana at 3-month follow-up than did those who were not in the negative affect group. While those in the positive affect group also reported significant days of use in the past 3-months, there was no difference in the average number of "joints" smoked per week in the past 3 months. Additionally , gender differences were found with females (vs. males) reporting a more significant family history of drug use and were more likely to report that they perceive that marijuana has bad effects on a person. Further research is needed to examine the reasons for marijuana use among incarcerated adolescents to inform and streamline treatment needs. In addition ethnicity and gender should be included in further investigations into the motivations, quantity, and frequency of marijuana use. These findings can assist in the development and implementation of future marijuana treatment efforts with incarcerated adolescents . Acknowledgements First, I'd like to thank my major professor Kat Quina for believing that I could actually do this even when I had begun to lose faith. Kat's positive energy and constant support were instrumental to my progress I can not thank her enough. I also want to thank Lyn Stein, my accidental mentor, a job that seemed to land in her lap whether she wanted it or not. Lyn gave me opportunities to learn and discover in this field that have been invaluable. For this and so much more I thank her. I am also grateful to my amazing committee members, who have been patient and encouraging through this process; Ginette Ferszt; Ellen FlannerySchroeder, John Stevenson and Jerome Adams. I want to send a special thank you to the Rhode Island Training School (RITS), in particular Warren Hurlbut, the superintendent of the facility who is dedicated to supporting research projects that will enhance treatment for juvenile detainees; Charles Golumbeske; Louis Como; and the administrative, school, and correctional staff without whom the project could not have been completed. I would also like to thank the RITS detainees for their participation over the years. Last but not least, a great big, heartfelt thank you to my family for unwavering support from the beginning of this long, difficult journey. They are the reason that I had the courage to even try. I am blessed with a supportive family, friends, and colleagues, all of whom have helped me to keep pushing and to stay positive. From the bottom of my heart, thank you all!


Acknowledgements
First, I'd like to thank my major professor Kat Quina for believing that I could actually do this even when I had begun to lose faith. Kat's positive energy and constant support were instrumental to my progress I can not thank her enough .
I also want to thank Lyn Stein, my accidental mentor, a job that seemed to land in her lap whether she wanted it or not. Lyn gave me opportunities to learn and discover in this field that have been invaluable. For this and so much more I thank her.
I am also grateful to my amazing committee members, who have been patient and encouraging through this process; Ginette Ferszt; Ellen Flannery-Schroeder, John Stevenson and Jerome Adams.
I want to send a special thank you to the Rhode Island Training School (RITS), in particular Warren Hurlbut, the superintendent of the facility who is dedicated to supporting research projects that will enhance treatment for juvenile detainees; Charles Golumbeske; Louis Como; and the administrative, school, and correctional staff without whom the project could not have been completed. I would also like to thank the RITS detainees for their participation over the years.
Last but not least, a great big, heartfelt thank you to my family for unwavering support from the beginning of this long, difficult journey. They are the reason that I had the courage to even try. I am blessed with a supportive family, friends, and colleagues, all of whom have helped me to keep pushing and to stay positive. From the bottom of my heart, thank you all!   Monitoring the Future study (Johnston, O ' Malley, & Bachman, 2003) also reported high rates of marijuana use among middle and high school students across the nation.
The NHSDA survey demonstrated that 18% of 8th-graders reported having tried marijuana at least once; and by 10th grade , 17% reported being "current" users ( defined as having used at least once within the past month; Johnston, Bachman, O'Malley, & Shulenberg, 2001). Among 12th-graders, 46% had tried marijuana at least once, and roughly 21 % were current users.
Retrospective adult research has shown that 90% of the adults with marijuana dependence in the United States began using before the age of 18, half of whom began using before age 15 (Dennis, Titus, Diamond , Donaldson, Godley, Tims, et al., 2002).
Moreover, in the past decade, increasing numbers of children under the age of 15 began to experiment with marijuana, illustrating a trend in earlier age of first time use (Dennis, et al., 2002). Treatment admissions for marijuana follow a similar trend , such that those aged 20 and younger account for half of all these admissions in the United States (SAMSHA, 2001 ).
The negative effects of marijuana use and misuse among adolescents have also increased. Adolescent marijuana users in treatment show an association between higher rates of marijuana use and higher rates of mental disorders (Arseneault , Moffitt, Caspi, Taylor, & Silva, 2000), health problems (Hall, Johnston, & Donnelly, 1999), and arrests (Webb, Burelson, & Ungemack, 2002). In fact, most adolescents seeking treatment are mandated through the criminal justice system, arrests ranged from possession of an illegal substance, property, and violent offenses (Dennis, et al., 2002). Researchers have noted that adolescents who use marijuana frequently are more likely to have conduct disorder problems and participate in illegal activites than adolescents who do not use marijuana (Farrell, Danish, & Howard, 1992;Rob, Reynolds, & Finlayson, 1990). In the following section the literature on marijuana use among incarcerated adolescents will be reviewed.

Marijuana Use Among Incarcerated Adolescents
Research assessing marijuana use among incarcerated adolescents is scarce, and the research that has been conducted is somewhat limited in scope by its reliance on the small sample sizes available within the juvenile justice system. The studies investigating basic prevalence data from convenience samples have demonstrated that marijuana use rates are higher among incarcerated adolescents when compared to adolescents in the general population (Dryfoos, 1990). Additionally, prevalence studies have reported that between 50% (n=91) and 97% (n= 178) of incarcerated adolescents used marijuana within the six months prior to their current incarceration (Farrow & French, 1986;Lebeau-Craven, Stein, Barnett, Colby, Smith, & Canto, 2000). Thus, it is not surprising that McManus ( 1984), in a study of 71 incarcerated male adolescents, found that 44% not only used marijuana daily but had a diagnosis of marijuana abuse.
The high rate of use among adolescents in the juvenile justice system precipitated a change in the climate of juvenile justice in terms of its attitude toward increased, as did the demand for treatment (SAMSHA, 2001;Webb, et al., 2002).
Increasingly, adolescents are court ordered to substance abuse treatment while incarcerated, or are court ordered to in-patient treatment in lieu of incarceration  .
There is an increased need for assessing marijuana use among this population because the reasons for initiating and continuing use are still unclear. This study will explore factors (e.g., family history, peer history, reason for use) that relate ·to marijuana use among incarcerated adolescents and attempt to gain a better understanding of the reasons for marijuana use among this population in order to better inform treatment efforts.

Research Examining Variables Related to Adolescent Marijuana Use
A number of variables have been implicated as risk factors that influence initial marijuana use among adolescents (Bailey & Hubbard, 1990). These risk factors include having a family history of substance use, peer influence, age of first use, expectancies of marijuana use, and situations in which teens anticipate difficulty in not using man Juana.
Two of the most widely recognized risk factors are family history /parental use and peer influences due to the exposure to others' attitudes toward marijuana use, substance use behaviors in their environment, and reinforcements and punishments for their marijuana use (Bailey & Hubbard , 1990). There are specific examples within the framework of a family history including parenting styles (reinforcements and punishments), parental discord, negative communication, lack of closeness and family violence factors that have been shown to be associated with initiation of marijuana use among younger adolescents (Carvalho, Pinsky, DeSouza, & Silva, 1995;Isohanni , Moilanen, & Rantakallio, 1991;Stoker & Swadi, 1990, Swadi, 1994. Wills and Cleary (1996) reviewed risk factors for marijuana use and suggested that lower parental support and more parental substance use are strong indicators of initiation as well as the continued use of marijuana among adolescents. Similarly, adolescents with marijuana using peers are more likely to use marijuana themselves, since peer approval is an important factor during adolescence. In fact, peer influence is often labeled as having the largest impact on adolescent substance using behavior (Swadi, 1992). Drug use may serve a social purpose for adolescents as well as providing a setting that is socially acceptable and favorable to drug users. Moreover , Chassin , Curran , Hussong, and Colder (1996) found that adolescents with drug using peers have a much faster trajectory toward abusing substances than adolescents who do not have drug using peers.
Age of onset has also been shown to be a risk factor for long term marijuana use. Babor, et al. (2002) investigated subtypes for classifying adolescent marijuana users who presented for marijuana treatment at a number of in-patient and out-patient facilities. Six hundred adolescents (82.2% of them males) participated in the study.
Among the 600 adolescents, 85% reported initiating marijuana use before the age of 15, with 15% initiating use at age 15 or older. Age of initiation, above all other variables, yielded unique variance when examining variables that classify those who are marijuana abusers or are dependent on marijuana . Age of initiation has also been shown to be a strong predictor of escalated use and long term abuse and consequences (Kaplan, Martin, & Robbins, 1986;Swadi , 1999).
In addition, sexual abuse has been positively correlated with subsequent substance abuse (Bulik, Prescott, & Kendler, 2001;Dube, Felitti, Dong, Chapman, Giles, & Anda, 2003;Holmes, 1997;Johnson, Ross, Taylor, Williams, Carvajal, & Peters, 2005;Kendler , Bulik, Silberg, Hettema , Myers, & Prescott, 2000) . Studies have demonstrated that even when controlling for family and peer factors, sexual abuse history is a significant predictor of marijuana, as well as other drug use (Bulik , et al., 2001). Much of the research in this area focuses on childhood victimization , defined as mental abuse, physical abuse, and sexual abuse . Sexual abuse is generally categorized in terms of severity (i.e., inappropriate sexual talk , touching, or penetration) .
Researchers have concluded that the risk of later substance use increases monotonically based on the severity of abuse (i.e., as the severity of abuse increases so does the , severity of substance abuse) (Dube, et al., 2003;Ferguson, et al., 1996;Fondacaro & Holt, 1999;Johnson, et al., 2005;Kendler , et al., 2000). As a caveat, having a parent or parents who abuse drugs is not only a risk factor for marijuana abuse, but is also a risk factor for childhood sexual abuse (Simpson & Miller, 2002) . Thus, when there is a history of sexual abuse, determining causality between drug abusing parents and adolescent marijuana use is extremely difficult. Taken together , both parental history of drug abuse and sexual abuse may substantially increase the risk of later drug use.
The association between sexual abuse history and substance abuse in an incarcerated population was demonstrated by Harlow ( 1999), utilizing data from the Bureau of Justice Statistics. Additionally , Harlow conducted personal interviews with incarcerated adult males and females . Results indicated that past sexual abuse was not only associated with substance abuse but with violent crime as well. The data confirmed that those who reported sexual abuse in childhood were more likely to use a number of different drugs at a more frequent rate than those who did not report sexual abuse. In 2005 , given the limited research in this area, Johnson et al. attempted to replicate Harlow's work with an incarcerated population in a neighboring state run jail.
One hundred male inmates were recruited (mean age of 34 years) and data were collected that included sexual abuse history and alcohol and drug use , The relationship between childhood sexual abuse and drug use was examined using the Fisher ' s exact test. Results were consistent with those of Harlow in that there was a clear association between childhood sexual abuse and later drug use (p<.05). Interestingly , when examining each drug separately , among the drugs related to childhood sexual abuse, marijuana was the drug used most frequently (p< .04). Although it may not be possible to find a causal link between childhood sexual abuse and marijuana use, this research will add to the scarce literature examining sexual abuse history and marijuana use in an incarcerated population.
Along with specific background variables a number of researchers have also discussed different attitudes, intentions of use, and behavior as playing distinct roles in the initiation of marijuana use in adolescents (Bailey et al., 1990;Comeau, Stewart, & Loba, 2001;O'Callaghan & Hannon, 2003) . For example, Gorman and Derzon (2002) analyzed data from a meta-analytic archive of prospective longitudinal studies ( 40 studies; N=3206). The researchers used the data to examine marijuana use in relation to three behavioral traits. These included negative affect ( defined as depression, anxiety and self-derogation, emotionality ( defined as a susceptibility to becoming easily and intensely distressed), and unconventionality (the propensity toward standards of behavior that are not seen as prevailing values or behaviors of general society, such as drug use). A strong association was found between unconventionality and marijuana abuse (M =.332) and between emotionality and marijuana abuse (M = .337). Babor, et al. (2002) found that low self-esteem, sadness and depression were associated with adolescent marijuana use . Similarly, self-efficacy , defined as a belief in ones abilities to function in a number of different situations, has been found to be correlated with an increase in marijuana use in that those who have low self-efficacy report more marijuana use and more difficulty quitting than those who have high self-efficacy (Stephens , Wertz, & Roffman, 2004). Loeber, Stouthamer-Loeber , and White (1999) examined longitudinal data on boys aged 7-18 for the co-occurrence of persistent substance use with other problem behaviors (n=1500). Logistic regressions showed that mood dysregulation (defined as depressive mood and anxiety) may lead to substance use as self medication. The hypothesis of using substances for the purpose of selfmedicating has been advocated by a number of authors (Aharonvich, Nguyen, & Nunes, 2001;Green, Kavanagh, & Young, 2004;Khantzian;, Loeber et al., 1999Mueser, Drake, & Wallech, 1998). In fact, Chassin, Pillow, Curran, Molina, and Barrera (1993) claim that there is evidence that affective states (i.e., negative emotions) may have a general, yet important, role in substance use.
Additionally, life events have been recognized as playing a role in emotional states of adolescents and subsequent substance use. Some examples of life events that may increase the risk for using substances includes unwanted pregnancy, high rates of bereavement, and sexual victimization (Brook, et al., 1998;Hernandez, 1992, Swadi, et al, 1999. Understanding the role of marijuana for adolescents (e.g., smoking marijuana to relieve depressive symptoms) and the risk factors that relate to initial and continued use will potentially enhance treatment efforts among this population. As stated previously, family and peer history of use, age of onset, and other personal expectancy variables, and life experiences have been established as risk factors.

Treating Incarcerated Marijuana Users
Ever since Martinsen (1974) asserted that any treatment, even at its best, could not deter the behavior of a criminal, there has been .reluctance among practitioners, researchers, and policy makers to treat incarcerated populations. Changing the thinking about substance abuse treatment for incarcerated populations has been challenging, especially since there has not yet been a treatment that has been proven successful for long term outcome (Grietens & Hellinckx , 2003) .
Only approximately 40% of juvenile facilities across the United States offer treatment for drug offenders (Thornberry , Tolnay, Flannigan, & Glynn , 1991) . Most of these treatments are modeled after those conducted with non-adjudicated adolescents, or incarcerated adults  . Many of the treatments provided to incarcerated, substance-abusing adolescents have different focuses and modalities, and lack specific drug abuse elements . Some of the more common substance abuse treatments for incarcerated juveniles include Residential Treatment , Therapeutic Communities , Boot Camp programs, Milieu Therapy, Relapse Prevention, and self-help programs (SAMSHA , 1995). Each of these treatments will be briefly discussed .
Residential Treatment (RT) designed for offenders incorporates several different models and philosophies with variable lengths of stay . RT incorporates individual and group components with differing intensities. The RT programs designed for offenders often use a group treatment approach to create an environment that duplicates certain aspects of functional family relationships. Creating this environment for the residents is . meant to foster collaboration and communication among the group . Parts of most residential treatment programs incorporate a segment on subs tances of abuse , and can be tailored to the audience being treated (Epstein , 2004;Grietens & Hellinckx , 2003;Jainchill , Hawke , & Messina , 2005;Orlando , Chan , & Morral , 2003 ;Rosenheck & Seibyl , 1998;SAMSHA , 1995) .
Therapeutic Communities (TC) are a type of residential program meant to help individuals phase into independent living. The intention of TC is to promote global changes in life-style, attitudes, and values. The treatment is highly structured and incorporates a number of life services in a single setting. Services include education, medical, family, legal, and social services as well as counseling, individual therapy, and group therapy (California Dept. of Corrections, 1999;Jainchill, Hawke, & Messina, 2005;Lipton,1996;SAMSHA, 1995). Therapeutic communities are one of the more common therapies used in juvenile correctional settings for substance abuse treatment (SAMSHA, 1995).
Boot Camp Programs (BCP) were created as a response to mandatory drug sentencing. Currently 27 states have BCP camps with 47 camps total. The BCP is highly regimented, military-styled discipline, combined with confrontation, and behavior modification (Pearson & Lipton, 1999;SAMSHA, 1995).
Milieu Therapy (MT) is similar to but less intensive than therapeutic communities. MT is generally carried out in an isolated, drug-free living area within the prison. It incorporates group and individual counseling, and often uses confrontational group sessions and peer interaction (LeCuyer, 1992;SAMSHA, 1995;Weid & Lutova, 2002).
Relapse Prevention (RP) and Self Help (SH) are generally given in conjunction with another form of treatment, for example TC. The focus of RP is behavior modification. RP teaches individuals to recognize their triggers for substance use, and provides participants with situation-specific tools to avoid using. SH groups are commonly referred to as 12 step programs (i.e., AA, NA). These forms of treatment bring the individual through a series of 12 steps to change their behavior emphasizing the importance of seeking guidance from religion and/or a higher power (Parks & Marlatt, 1995;SAMSHA, 1995).
Researchers have shown that effective programs share elements that include targeting the offender's cognition, self-evaluation and expectations, and focus on changing the participant's view of the world and societal values (Izzo & Ross, 1990).
Overwhelmingly what has emerged from this research is that a program's effectiveness depends not only on the elements of the program, but also on the experience I training of the treatment providers, how the program is targeted, the treatment setting, by whom it is administered (i.e. treatment providers within the residential setting vs. an outside vendor ;Izzo & Ross, 1990;Parks & Marlatt, 1995;Pearson & Lipton, 1999). More often than not, treatment in prison facilities is often dictated not by clinical research recommendations, but by funding agencies. This leaves large gaps in the organization and management of critical elements of the treatment program with respect to fidelity or implementation, quality of treatment being offered, and understanding of the clinical phenomena of interest (Lipton , 1999). Programs cannot be duplicated and retested for long-term usefulness without placing more emphasis on experimental research to evaluate the effectiveness of the program controlling for the specifics of the treatment models, the training and credentials of staff members, treatment delivery, and supervising procedures . Juvenile detention facilities may provide a useful venue for doing just that, and more specifically, for furthering extant knowledge of adolescent marijuana use. This could lead to more insight into the reasons for initiation and continued use, and would monitor the adherence to treatment programs for a more indepth understanding of the components that contribute to long-term change . The expectation is that this research will guide understanding of the factors that influence marijuana abuse among incarcerated adolescents and subsequently pave the way for more tailored treatments. This is important particularly given the high rates of marijuana use and misuse among incarcerated adolescents.
The larger clinical trial from which these data are taken utilized randomized treatment assignment to investigate the effects of Motivational Intervention (Ml) and Relaxation Therapy (RT) on group participation rates within the training facility, described elsewhere (Stein , Colby, Barnett , Monti , Golembeski , Lebeau-Craven , & Miranda, 2006). The larger study investigates which treatment option (Ml vs. RT) is more effective at reducing marijuana use among juveniles three months after release from incarceration. Further, secondary analysis of this data will be explored to determine which treatment (Ml vs . RT) is more useful for adolescents who use marijuana to regulate their emotions (i.e., to reduce tension). The proposed study will inform the results of the main analyses of the larger clinical trials .
The proposed study will also examine several of these background variables (family and peer history of use, age of onset, and sexual victimization) as predictors of more serious marijuana use ( defined as abuse by the SCID) among a group of incarcerated adolescents. Additionally , reasons for marijuana use , marijuana expectancies and use of marijuana for emotional regulation will be examined, utilizing the Marijuana Effects Expectancy Questionnaire and the Brief Situational Confidence Scale . Using marijuana to regulate negative affect, defined as having high expectancies for the effects of marijuana to help one relax and reduce tension, and a low confidence score for not smoking in situations where negative feelings will be alleviated will form a negative affect factor. A distinct factor should emerge identifying those who use marijuana for positive affect, defined as having: a high expectancy for marijuana to make parties with friends more fun and make one more social, and a low confidence for not smoking in situations that involve celebrating with friends and enjoying oneself at a party.

Rationale for Current Study and Significance
The proposed research project has a number of important potential implications for furthering basic knowledge and assisting in the implementation and development of future marijuana treatment efforts with incarcerated adolescents. Moreover, intervening with incarcerated adolescents who use marijuana could help to reduce recidivism and improve the overall health of this underserved population.

Potential Implications of this Research
Preliminary pilot work in detention indicates that a substantial number (83%) of adolescents are reporting daily marijuana use (Stein et al., 2006). This illustrates a need for more specialized treatment and appropriately trained intervention specialists to provide substance use treatment to these teens. The proposed research study may help to identify factors that make teens more likely to be marijuana users and the type of treatment that is more effective for their individual needs. This could potentially decrease the amount resources needed to treat incarcerated adolescents by providing a more efficient and cost effective way to reduce marijuana use among this population .
A great deal of research is needed to better understand marijuana use and effective treatment among the incarcerated adolescent population. This research will improve the efficacy of treatment for both male and female adjudicated adolescent populations by gaining a better understanding of the reasons for marijuana use, and subsequently adding to the knowledge base of marijuana treatment for these adolescents.
The Current Study The current study was designed to extend basic knowledge of the reasons for marijuana use, establish factors related to marijuana use, and examine ifrelaxation therapy reduces marijuana use in those who use marijuana as a way to regulate emotions.

1.
It is hypothesized that, upon entrance to a juvenile training facility, youth who abuse marijuana will report a related set of background variables ( e.g., family history of substance use, peer history of substance use, and personal use for emotional regulation) that differs from youth who do not have a marijuana abuse diagnosis.
a. It is hypothesized that for teens who present with negative affectivity, relaxation therapy will be more effective than motivation intervention for reducing marijuana use 3 months after release.

2.
Based on previous research (Stephens, Wertz, & Roffman, 2004), it is hypothesized that upon entrance to a juvenile training facility, youth with negative affect ( defined as having high expectancies for the effects of marijuana to help one relax and reduce tension, and a low confidence score for not smoking in situations where negative feelings will be alleviated) will have elevated marijuana use compared to those with positive affect (high expectancy for marijuana to make parties with friends more fun and make one more social, and a low confidence for not smoking in situations that involve celebrating with friends and enjoying oneself at a party).

3.
It is hypothesized that adolescents with a record of childhood sexual abuse will report more marijuana use upon entrance to the training facility.

4.
It is hypothesized that adolescents with a record of childhood sexual abuse will be more likely to use marijuana for negative affectivity than for positive affectivity.

Participants
One hundred and ninety-one incarcerated adolescents aged 14-19 (M= 17) were recruited for a larger treatment study described elsewhere (Stein, et al., 2006), in which participants and their guardians consented to participate. Potential participants adjudicated between January 2001 and September 2005 were included in the larger study if they met any of the following criteria: a) in the year prior to incarceration they used marijuana or alcohol at least monthly; b) they used marijuana or drank in the month preceding the offense for which they were committed; or c) they used marijuana or drank in the month preceding commitment. A total of 417 potentially eligible adolescents were identified . For 170 youth, parents/guardians could not be reached for consent, 45 could not be recruited due to project staff not having enough time to complete the assessment and treatment at baseline, and 7 were not interested in participating. Of the 195 who consented and participated in the study, 6 dropped out prior to completion of baseline assessment due to early release or loss of interest in participating.
The population at the training school is diverse. Consent for adolescents to participate was obtained from guardians and an assent was signed by the adolescents themselves . Guardians and teens were presented with a description of the study including procedures, potential risks and benefits, and confidentiality, and asked to sign a consent form. Assessments were conducted in private rooms with Bachelor's-or Master ' s-level research interviewers . Condition assignment was revealed at the end of the assessment and a time to complete the baseline treatment was selected . The treatment was completed within one week of completing the baseline assessment by the same researcher that completed the assessment. Each of the research interviewers were extensively trained over a threemonth period in the methods of treatment used and the project protocol.

Treatment conditions.
Following assessment, participants were randomly assigned to one of two treatment conditions, Relaxation Therapy (RT; 48%) or Motivational Intervention (MI; 52%), both designed to help prepare them for the facility's standard care treatment. This Motivational Intervention. In MI the intervention is modified for each participant based on the answers given during the baseline assessment and his or her interest in changing. The intervention consisted of four components: establishing rapport, assessing motivation for change, enhancing motivation, and establishing goals to aid in making changes. Establishing rapport aims to present the counselor as empathic, concerned, non-authoritarian, and non-judgmental, elements presented by Miller (1995) as essential to MI. In order to assess a participant's motivation to change, Relaxation Therapy. In RT participants were instructed in using muscle relaxation and meditation as a behavior management tool. Participants received feedback on their use of the techniques as well as handouts on relaxation methods. The treatment providers maintained rapport and provided generalized advice to stop use of marijuana, criminal behavior, and risky activities .

Follow-up.
Two follow-ups were completed . An in-facility follow-up was conducted two months after the initial interview. This follow-up consisted of assessments and a treatment booster. The treatment booster was given to participants in both treatment groups and essentially repeated the original discussion with the adolescent. Upon completion of the in-facility follow-up, research staff obtained extensive contact information to complete the second follow-up which occurred at 3-months post-release.
The 3-months post-release follow-up was completed on average 61 days from the baseline assessment. Assessments were conducted by a trained research assistant who was blind to treatment condition. Participants were compensated for their time in the study by receiving a meal for participation while still in facility and $50-$60 in gift cards to the local mall for the 3 month post-release participation (see Table 1).

Measures
A copy of all of the questionnaires is found in Appendix B.
Background Questionnaire. Sociodemographic information was recorded at baseline, including: parent alcohol/drug history; peer alcohol/drug involvement; age of onset of alcohol /drug use; and age youth began to use regularly. This questionnaire was The BSCQ first asks people to imagine themselves as they are "right now" in each of the eight situations noted above. They are then asked to indicate on a scale from 0% ("not at all confident") to 100% ("totally confident") how confident they are to resist using marijuana in each of the eight situations . The BSCQ has been tested with high-risk drinkers and has shown to correlate well with the subscales from the original 100 item version it has also shown to have a high level of internal consistency (alpha=.85). Psychometric validation has not been completed on the marijuana version of the assessment.

CES-D. The Center for Epidemiological Studies Depression
Scale is a 20-item selfreport scale designed by Radloff (1977) to measure depressive symptoms in the general population (Barlow, 1998) shown to have a sensitivity of 89% and a specificity of 70% at a cutoff score of 27 (Mulrow,I 995). In addition, the adult CES-D has been tested on adolescents with a sensitivity of 84% and a specificity of 75% at a cutoff score of 24 (Schulberg , McClelland, Gangul, Christy, & Frank , 1985).   (Kranzler , Kadden , Babor, et al., 1996).

TLFB-Marijuana .
Time-Line Followback (Sobell , Maisto , Sobell , & Cooper , 1979) is a calendar-assisted measure based on a subject's retrospective account of his/her drinking and drug use behavior over a specified time period. TLFB has been shown to have excellent reliability (.79-.98 ; Sobell et al., 1979;Sobell , Sobell, Leo, & Cancilla, 1988) and high content, criterion , and construct validity in clinical and non clinical populations (Connors, Watson , & Maisto , 1985;Sobell , Sobell, Klajner, & Pavan , 1986). The TLFB was adapted for marijuana use for this study and asks adolescents to recall their marijuana use based on amount of marijuana used and how often they use each day for a 90-day period. These data were collected for the 90 days prior to sentencing and for the 90-day s after release. Teens were assessed within one month of sentencing .

Data Analyses
The data are stored off training school grounds at the University of Rhode Island. The software system used to analyze the data was SPSS (Statistical Package for the Social Sciences, version 11.0 SPSS, Inc., Chicago, IL). The present study utilized this archival data to conduct data analysis in two phases. Phase 1 addresses data cleaning including the handling of missing data. Phase 2 directly tests the aforementioned hypotheses.

Phase I: Preliminary Data Analyses
Data cleaning and handling missing data was based on the recommendations of Tabachnick and Fidel (2001). Descriptive analyses were conducted to be sure that all of the assumptions of normality and linearity were met and missing data were handled using the Expectation Maximization (EM) procedure. However, there were not large amounts of missing data due to the placement of the participants (i.e., teens are incarcerated and easily accessible) and the way data were collected via an interviewer.

Phase 2: Data Analyses Addressing Dissertation Hypotheses
Prior to investigating the hypotheses, descriptive statistics were computed on all variables to be examined in the subsequent analyses.  Exploratory analy ses will also be conducted to gain needed information on differences between male and female juvenile detainees in terms of reasons for use, relevant background variables (e.g., family history , peer use history), and treatment effects . These analyses are meant to be exploratory with the expectation of promoting further research in this area, adding to the knowledge base, and ultimately expanding the literature in this area.

Results
Before examining the hypotheses , distributions of all variables used in this study were checked. There were two Timeline follow-back variables ( average number of joints smoked per week in the past three months and frequency of marijuana use in the past three months) that did not have normal distributions. As a result, outliers were coded as one unit higher than the highest non-outlier data point. This transformation brought skewness and kurtosis of these variables within the normal range .
The test of the first hypothesis was conducted using principal components analysis (PCA) with varimax rotation on the item intercorrelation matrices . The number of components to retain was determined using the minimum average partial procedure (MAP;Velicer , 1976) and parallel analysis (Horn, 1965). These two methods have been found to be accurate for determining the number of components to retain (Zwick & Velicer, 1986).
Fifteen items were included in the exploratory factor analysis (6 MEEQ items & 9 BSCQ items; see Table 2). PCA with varimax rotation on the 15 X 15 matrix of item intercorrelations was conducted to determine the factor structure of the scale . MAP The 2-factor solution suggested by the parallel analysis criterion separated the BSCQ items from the MEEQ items . Two of the MEEQ items "loaded" on the BSCQ factor. Actually, loadings less than .40 are not considered strong enough loadings to be considered on that factor (see Table 3).
Since the exploratory factor analyses did not reveal a negative affect /positive affect factor structure, negative affect was derived from three items on the BSCQ scale.
These items include : negative emotional states (Item 1 ); negative physical states (Item 2); and interpersonal conflict (Item 6). If a participant endorsed having little to no confidence (:::25%) on any two of these three items then they would score positive for negative affect.
Next, a one-way Analysis of Variance (AN OVA) was run to examine differences in depression as measured by the CES-D scale among those with greater negative affect (M= 21.3 5) versus those with less negative affect (M=20. 6). There were no significant differences on the CES-D scale between the groups with differing levels of the negative affect variable, F(l, 187)=.208,p>.05, eta-squared=. 001 (see Figure 1).
Hypothesis la was first analyzed using analysis of covariance (ANCOVA).
AN COVA was chosen in order to include baseline levels of marijuana use as a covariate. Only those who had been classified as having negative affect on the calculated variable were included in these analyses. Two separate ANCOV As were run.
The first assessed average number of joints per week at 3 month follow-up as the DV with treatment condition (MI vs. RT) as the IV. The second used the number of days smoked at 3 month follow-up as the DV and treatment group as the IV. There were no significant differences in either ANCOVA,p>.05.
Since no significant differences were found in the previous analyses, two additional 2 x.2 ANCOVAs were run, for treatment group (MI vs. RT) X affect level.
Number of joints per week at three month follow-up was the DV for the first analysis; number of days smoked for the second, and the corresponding baseline level (i.e., number of joints per week and number of days smoked at baseline assessment) were the covariates. There were no significant differences found in these new analyses.
However, the covariate in these analyses was significant, showing the importance of using the baseline variable of marijuana use as a covariate, as it was significantly related to marijuana use at the follow up.   (SAMSHA, 2001;Webb, et al., 2002). Increasingly , adolescents are courtordered to substance abuse treatment while incarcerated, or are court-ordered to inpatient treatment in lieu of incarceration .
The purpose of this project was to identify reasons for marijuana use among incarcerated adolescents. It was hypothesized that reasons for use would be separated into two groups , those who use marijuana for reducing personal issues (negative affect) and those who use marijuana to celebrate and have more fun with their friends. These groups, separated by the reasons that one uses marijuana, would then result in differences in quantity and frequency of use. An attempt was made to show that adolescents giving different reasons for use (positive versus negative affect) would differ significantly on marijuana use variables at baseline, and that the two groups of adolescents would show further differences in response to the two different treatments (MI versus RT) at the 3-month follow-up. Additionally, sexual abuse history was investigated as a predictor of marijuana use. Finally, gender differences among this population to examine variation in marijuana use, expectancies and confidences , as well as family history and peer substance use between incarcerated male and female adolescents were explored.
A principal components analysis on the Brief Situation Confidence

Questionnaire (BSCQ) and the Marijuana Effects Expectancies Questionnaire (MEEQ)
did not reveal separate factors that represent reasons for marijuana use. In fact, the scales remained almost completely intact. In order to test the hypotheses of this study , items from the BSCQ that corresponded with low confidence to avoid marijuana use in situations were included in a negative (either mentally or physically) affect variable. A positive affect variable was also calculated , using the items from the same scale that corresponded with low confidence to avoid marijuana in situations that relate to using marijuana for enjoyment or celebrating (i.e. , recreational use).
Most of the analyses did not reveal significant differences between marijuana users with negative affect and those users with positive affect, with respect to reasons for marijuana use , family history , depression , peer history , or quantity of marijuana use at 3 month follow-up .
However, there were significant findings that support the thinking behind the original hypotheses. While incarcerated adolescents who were in the negative affect group (negative affect yes) and the positive affect groups (positive affect yes) had higher averages of weekly marijuana use. Only those in the negative affect group (negative affect yes) had a higher number of days using marijuana at 3-month follow-up than did those who were not in the negative affect group (negative affect no). This was true regardless of treatment group, and when baseline levels of use were covaried out.
Those in the positive affect group reported no difference in the average number of "joints" smoked per week in the past 3 months. This may support the hypothesis that those with negative affect will use marijuana more frequently than those using for recreation. As mentioned, other researchers have examined and advocated for the hypothesis that teens use substances for the purpose of self-medication (Aharonvich, et al., 2001;Green, et al., 2004;Khantzian;, Loeber et al., 1999Mueser, et al., 1998). In fact, there is some evidence that affective states (i.e., negative emotions) may have a general, yet important, role in substance use in general (Chassin, et al., 1993).
Past research also indicates that low self-esteem, sadness and depression, as well as low self-efficacy, were associated with adolescent marijuana use (Babor, 2002;Stephens, et al., 2004). Further , research indicates that boys aged 7-18 who reported mood disregulation (e.g., depressive mood and anxiety) were more likely to use marijuana as a form of self-medication (Loeber, et al., 1999).
In addition significant gender differences were found, with females reporting a more considerable family history of drug use and reporting that they perceive that marijuana has bad effects on a person, compared to males. One explanation for these findings may be that males and females react differently, with males tending to exhibit externalizing behaviors and females tending to exhibit internalizing behaviors (Rhodes & Fischer, 1993;Storvoll, Wichstrom, & Pape, 2003;Webster-Stratton, 1996). Reasons for referral to a detention facility, for example, differed between males and females, with males more likely to be referred for violations of the law, including selling drugs, while females were more likely to be referred for status offenses such as running away from home (Rhodes, et al., 1993). Specifically of interest, researchers also found a correlation between what they termed predictor variables (self-reported drug use and delinquency) and later arrest. For females, physical abuse in the home was associated with status offenses, while this was not the case for males (Dembo, et al., 1995). The finding that females report more history of drug use in the home may suggest an alternate hypothesis: the experience of living with such abuse may lead to an aversion to use and may also lead to viewing drug use as more detrimental to a person, more so than for males. Relating this back to the previous rationale, females may be more likely to see the negative side of marijuana use than males, as indicated by the higher rate of family history reported.
Marijuana use is extremely common among incarcerated adolescents (Lebeau-Craven, et al., 2003). Gaining knowledge about the reasons for marijuana use, and about effectiveness of treatment approaches for incarcerated youth, could potentially have a powerful impact on marijuana users post-incarceration .
Several of the hypotheses did not have significant results. There are various limitations that may have contributed to these findings. This was an exploratory study based on archival data that were self-reported and therefore could be inaccurate. It is As a result, further research is needed to examine the hypothesis that incarcerated adolescents who begin using marijuana as a way to neutralize emotions can be distinguished from those who use marijuana for recreation or social experience.
More research on this issue will allow for a better understanding of how to shape treatment efforts in this population . It is important to understand the perceived benefits of using marijuana among incarcerated adolescents, and how those perceptions may mediate continued use. This could inform not only methods to' encourage cessation but could potential inform prevention efforts as well. In terms of the experimental treatments tested in this protocol, self-medication may be best treated by utilizing techniques which do not involve drugs to facilitate new ways of dealing with difficult situations.
This population is also ethnically diverse . In addition, females are often not included as part of the sample when researching incarcerated adolescents due to the significantly smaller number of incarcerated adolescent females. As a result, both ethnicity and gender should be included in further investigations into the motivations for marijuana use, quantity and frequency of use, as well as treatment implications.
There is an increased need for assessing marijuana use among this population because the reasons for initiating and continuing use are still unclear. Future research will need to explore factors (e.g., family history, peer history, reason for use) that relate to marijuana l:se among incarcerated adolescents in terms of using marijuana as a way to regulate emotions versus using marijuana as a recreational tool. This will help to better inform treatment efforts in training facilities and among the incarcerated adolescent population. Further, previous research does promote the value of future studies that focus on reasons for marijuana use and treatment in this population. Table 1 Project flow chart:     Teens of the Rhode Island Training School for Youth (RITS) are being asked to take part in a Brown University research project. This form will be explained to you and you can also read the form yourself. Please ask any questions you might have. Then if you decide to allow your teen to be in the project, please sign this form in front of the Researcher and a witness.

1.
Nature and Purpose of the Project We are trying to determine if talking to teenagers at the RITS will increase their treatment participation and reduce or prevent risky health behaviors (like being injured or drinking alcohol) after release .

2.
Explanation of Procedures RITS teens will be asked to complete several questionnaires regarding their education , health behaviors (for example, accidental injuries and fighting), misbehaviors, feelings of sadness, and treatment participation. All questionnaires will be read to teens. These questionnaires will take approximately 2 hours on one day. Your teen will be given breaks with snacks the questionnaires. Teens will be randomly assigned (using a procedure like flipping a coin) to one of two treatment groups. We do not know whether one of these treatments is more helpful than the other. Half of the teens will be trained in meditation and relaxation to reduce stress. Teens in the other half will discuss risky health behaviors . Your teen will receive about 2 hours of treatment on one day. Snack breaks will be provided during the session and after completing the treatment session, your teen will receive a larger food item (such as McDonalds, pizza , or Chinese food). Teens will be interviewed two more times while in the RITS about attitudes towards treatment and towards various health behaviors (like using marijuana or alcohol). Teens who received training in meditation and relaxation will receive further training in these skills. Again, the remaining half of teens will continue discussing the negative impact of risky health behaviors on their lives. Each of these two sessions will be about 90 minutes. Again, teens will receive snacks and a larger food item. RITS staff will rate how active teens are in treatment. Project staff will review the teen's record to determine, for example, number of prior incarcerations. Teens will be contacted at 3 months after release from the RITS. At this time they will again complete several questionnaires regarding health behaviors and misbehaviors, and all questionnaires will be read to teens. They will also be asked to provide a urine specimen . This final contact will last for about 2 hours and your teen will receive a snack break as well as a gift certificate at the end. This is a $50 gift certificate for movies , groceries, music or clothes. Should your teen complete this follow-up within one week of its scheduled date, he/she will receive a $10 bonus gift certificate. Questions about this study should be directed to Dr. Lynda Stein at ( 401) 444-1824.
We need to know whether our intervention was helpful. Therefore , we will make every effort to contact your teen for the one time follow-up interview at 3 months after release . Your teen will be asked to give information on two people (family or friends) who are likely to know where he/she is after release. We will use these two people only if other ways of contacting your teen don ' t work. We may ask family , a case-worker , or probation officer where the teen is, but at no time will we share any information about your teen . No one will know what the study is about and no information about your teen will be shared with other people.

Discomforts and Risks
The risks in this study are considered minimal. The questionnaires used in this study are commonly used in research and clinical practice. Teens may experience some emotional discomfort in discussing their health behaviors.

4.
Benefits By answering these questions , teens have the opportunity to learn more about the dangers of risky health behaviors. This may prevent or reduce the chances ~hat they are injured after release . This may reduce the chances that they will be in trouble with the law again .

5.
Alternati ve Treatment If you decide that your teen should not participate in this study ( or if your teen decides not to participate) , we offer no alternative treatments.

6.
Confidentiality Each teen ' s records from the project will be treated as confidential and private records. Your teen's name will not be on the questionnaires. Records will be kept in a locked filing cabinet at Brown University. Only researchers will have access to the information provided by teens. No information will be shared with other parties such as legal authorities , parents , school personnel , case-workers, parole officers , employers , or the RITS without the teen's approval. We have a Certificate of Confidentiality , granted by the Secretary of Health and Human Services (HHS) . The Secretary does not encourage or discourage this research. The Certificate does not apply if HHS wishes to examine the researchers or research project activities. The Certificate is a legal document to protect information that the teens provide from being shared in any Federal, State or local civil, criminal, administrative , legislative, or other proceedings . The purpose of the Certificate is to protect researchers from being forced to release information on teens. Instances when information will not be kept confidential involve reports of escape plans , child abuse , or threats to harm self or others . While the results of the research project will probably be shared with other people and may be published in scientific reports , the names of teens and the fact that any particular teen was in the project will be kept confidential by the researchers .

Refusal/Withdrawal
The decision whether to allow RITS teens to be in the study is entirely up to parents/legal guardians and RITS teens. Participation is voluntary. There will be no effect on the care provided by the RITS if teens refuse to participate or if parents/legal guardians refuse to let teens participate . There will be no effect on the care provided by the RITS if teens drop out of the project or if teens are withdrawn by parents /legal guardians.

8.
Medical Treatment /Compensation in Case of Injury We do not expect any unusual risk in this research project. Risks involved in this project are considered minimal and should be similar to those experienced in daily life or the performance of routine physical or psychological tests.

9.
Rights and Complaints If you or your teen have any questions about a teen's participation in this project , or would like more information about the rules for research projects , or the rights of people who take part in research , please contact Dorinda Williams at Brown University, telephone number (401)  Researcher Signature Date ASSENT-N Teens at the Rhode Island Training School for Youth (RITS) are being asked to take part in a Brown University research project. This form will be explained to you, and you will also have an opportunity to read the form to yourself and ask questions. If you decide to fill out the brief questionnaire, please sign this form in front of the researcher.

1.
Nature and Purpose of the Brief Questionnaire You decided not to participate in a Brown University study on reducing alcohol and marijuana use in RITS teens. Instead you will be asked to fill out a brief , anonymous questionnaire containing background information (like your age), and a few items concerning your alcohol and marijuana use and the number of times you have been incarcerated. This is so we can have a better sense of the people who come through the RITS and how we can help them with our research .

2.
Explanation of Procedures You will be asked to complete just one brief questionnaire regarding use of substances like alcohol and marijuana, and reasons why you did not want to be in the research project. We will read the questionnaire to you. It is only one page in length and will take just a few minutes of your time. Questions about this study should be directed to Dr. Lynda Stein at (401) 444-1824.

3.
Benefits /Discomforts and Risks /Treatment Alternatives There are no known benefits to filling out this questionnaire. The risks in filling out this questionnaire are considered minimal. The questions asked are commonly used in research and when you see a doctor. If you decide not to fill out this questionnaire (or if your parent/guardian decides you should not fill it out), we offer no alternative questionnaires to complete .

4.
Confidentiality This questionnaire will be treated as confidential and private. Your name will not be on this questionnaire. This brief questionnaire will be kept in a locked filing cabinet at Brown University. Only researchers will have access to the information you provide. No information will be shared with others such as legal authorities, parents, school personnel, case-workers, parole officers, employers, or the RITS without your written approval. We have a Certificate of Confidentiality, granted by the Secretary of Health and Human Services (HHS). The Secretary does not encourage or discourage this research. The Certificate does not apply if HHS wishes to examine the researchers or research project activities. The Certificate is a legal document to protect information that you give us from being shared in any Federal, State or local civil , criminal , administrative, legislative, or other proceedings. The purpose of the Certificate is to protect researchers from being forced to release information about teens . All information will be confidential and private unless it involves reports of child abuse, escape plans, or threats to harm yourself or others. The results of the research project will probably be published, but your name and the fact that you were in the project will be kept confidential by the researchers.

5.
Refusal/Withdrawal The decision whether you fill out the brief questionnaire is entirely up to your parent/legal guardian and you. Participation is voluntary. The care you receive at the RITS will not change if you decide not to complete the brief questionnaire or if your parent/legal guardian does not want you to complete the brief questionnaire.

6.
Medical Treatment/Compensation in Case of Injury We do not expect any unusual risk in completing th1s brief questionnaire. Risks involved are considered minimal and should be similar to those experienced in daily life or the performance of routine physical or psychological tests.

7.
Rights and Complaints If you have any questions about filling out the brief questionnaire, or would like more information about the rules for research projects, or the rights of people who take part in research, please contact Dorinda Williams at Brown University, telephone number (401)  Teens of the Rhode Island Training School for Youth (RITS) are being asked to take part in a Brown University research project. This form will be explained to you, and you will also have an opportunity to read the form to yourself and ask any questions you may have. If you decide to be in the project , please sign this form in front of the researcher.

1.
Nature and Purpose of the Project We are trying to determine if talking to teenagers like you will increase treatment participation and reduce or prevent substance use and negative things that can happen when using substances (for example, being injured when drinking alcohol) .

2.
Explanation of Procedures You will be asked to complete several questionnaires regarding your education, use of substances like marijuana and alcohol, misbehaviors and injuries, feelings of sadness, and treatment participation. All questionnaires will be read to you. These questionnaires will take approximately 2 hours on one day. You will be given breaks with snacks during the session . Then you will be assigned (using a procedure like flipping a coin) to one of two treatment groups, and we do not know whether one of these groups is more helpful than the other. You may be trained in meditation and relaxation to reduce stress. Or you may be asked to discuss your use of alcohol and marijuana and what happens when you use these substances . You will receive approximately 2 hours of treatment on one day. Snack breaks will be provided during this session and after completing the treatment, you will receive a larger food item (such as McDonalds , pizza, or Chinese food). You will be interviewed two more times while in the RITS about attitudes towards treatment and towards use of alcohol or marijuana. If you received training in meditation and relaxation you will receive further training in these skills. Or you may continue discussing experiences with alcohol and marijuana. Each of these two sessions will be approximately 90 minutes. Again , you will receive snacks and a larger food item. RITS staff will rate how active you are in treatment. Project staff will review records to determine , for example, number of prior incarcerations. Finally, we will contact you 3 months after release from the RITS. At this time we will again complete several questionnaires regarding use of substances like alcohol and marijuana . All questionnaires will be read to you. You will be asked to provide a urine specimen. This final contact will last for approximately 2 hours and you will receive a snack break as well as a gift certificate at the end. This is a $50 gift certificate for movies , groceries, music or clothes. Should you complete this follow-up within one week of its scheduled date, you will receive a $10 bonus gift certificate . Questions about this study should be directed to Dr. Lynda Stein at (401) 444-1824 .
We need to know whether our intervention was helpful. Therefore , we will make every effort to contact you for the one time follow-up interview at 3 months after release. You '11 be asked to give information on two people ( family or friends) who are likely to know where you are after release . We will use these two people only if other ways of contacting you don't work. We may ask family, a case-worker , or parole officer where you are, but at no time will we share any information you gave us. No one will know what the study is about and no information you give us will be shared with other people .

Discomforts and Risks
The risks in this study are considered minimal. The questions we ask in this study are commonly used in research and when you see a doctor. You may feel a little uncomfortable discussing substance use and some of your behaviors.

4.
Benefits By answering these questions , you have the opportunity to learn more about yourself and others who use alcohol or marijuana. You may be able to lower the chances that you will have problems or injuries related to use of substances like alcohol or marijuana . You may be able to reduce the chances that you will be in trouble with the law again.

5.
Alternative Treatment If you decide that you should not participate in this study , we offer no alternative treatments.

6.
Confidentiality Records from the project will be treated as confidential and private . Your name will not go on the questionnaires. Records will be kept in a locked filing cabinet at Brown University. Only researchers will have access to the information you provide. No information will be shared with others such as legal authorities , parents , school personnel, case-workers , parole officers, employers, or the RITS without your written appro val. We have a Certificate of Confidentiality , granted by the Secretary of Health and Human Services (HHS). The Secretary does not encourage or discourage this research. The Certificate does not apply if HHS wishes to examine the researchers or research project activities. The Certificate is a legal document to protect information that you give us from being shared in any Federal, State or local civil , criminal , administrative , legislative , or other proceedings. The purpose of the Certificate is to protect researchers from being forced to release information about teens. All information will be confidential and private unless it involves reports of child abuse , escape plans , or threats to harm yourself or others. The results of the research project will probably be published , but your name and the fact that you were in the project will be kept confidential by the researchers.

7.
Refusal/Withdrawal The decision whether you will be in the study is entirely up to you. Participation is voluntary . There will be no change in the care provided by the RITS if you decide not to be in the study. There will be no change in the care provided by the RITS if you drop out of the project later.

8.
Medical Treatment/Compensation in Case of Injury We do not expect any unusual risk in this research project. Risks involved in this project are considered minimal and should be similar to those experienced in daily life or the performance of routine physical or psychological tests.

9.
Rights and Complaints If you have any questions about your participation in this project , or would like more information about the rules for research projects, or the rights of people who take part in research , please contact Dorinda Williams at Brown University , telephone number (401)  28. How old were you when you first got arrested? ears old.
---~ Note for next questions: "Once per month or more" refers to a 3 month period or more; "once per week or more" refers to a 4 week period or more; and "daily" refers to a 5 day period or more .

BSCQ-M
Listed below are nine types of situations in which some people experience a problem with marijuana. Imagine yourself as you are right now in each of the following types of situations. Indicate how confident you are right now that you would be able to resist marijuana use in each situation. 0% means you are "Not At All Confident" and 100% means you are "Totally Confident." Right now how confident are you that you can resist marijuana use when ...

1.
... you feel UNPLEASANT EMOTIONS (If you were depressed about things in general , if every thing were going badly for you)?

MEEQ-B
The following pages contain statements about the effects of marijuana. Answer each statement according to your own personal thoughts, feelings, and beliefs about marijuana. We're interested in what you think about marijuana , not what others might think. Whether or not you 've had actual marijuana experience, you should answer in terms of how you think marijuana affects the typical or average user .
Answer according to how much you agree or disagree with each question.

1.
Marijuana makes it harder to think and do things (harder to concentrate or understand; slows you down when you move).