This item is only available as the following downloads:
DO WE FIT THE BILL? A Quantitative Study of Substance Use at a Small Liberal Arts C ollege BY GENEVIEVE VAIDA A Thesis Submitted to the Division of Sociology New College of Florida in partial fulfillment of the requirements for the degree Bachelor of Arts Under the sponsorship of Dr. Chavella Pittman Sarasota, Florida April 2010
For any student who has felt that they could not belong. For any student who loved where they were with a whole heart. For the rest of us, stuck so mewhere in between.
! ! "" Acknowledgements I want to thank my parents, all three, for forever being my support system, and of course, for everything else they have ever done, it's a long list. I want to express my grati tude to my siblings. To my brother who will and should always be the "smartest" of us all. To my baby sister for opening my eyes to the world, both beautiful and terrifying. And my baby brother Jaxson who could not have come at a better time. Thanks t o the roommate who got me through the year I thought would never end. Rochelle, you are a true friend, forever and always. To my GG you are a best friend, for understanding, listening, and being one of the most courageous and sensitive persons I have m et, or yet to meet. Jody Mailer, you always came correct. Somehow you've always known exactly what it meant to be my friend, and for that you are a great one. A great thanks to Dr. Chavella Pittman for her guidance on this thesis. I could not have chosen a sponsor better suited to my needs as a student. Thank you for your patience, knowledge, smiles, and your personal concern for my life outside academic pursuits. It means the world that I was more than a thesis student. Also, thank you to Dr. Sarah Hernandez, who helped me write the first three chapters of my thesis, and for guiding me toward better and bigger endeavors. To Dr. Joseph Mink, one of the most outstanding professors I have had the pleasure of meeting. Thank you for making my sen ior year interesting academically. Without your classes I think I might have fallen into the "pit of despair".
! ! """ And of course, to the students and friends of New College of Florida 2009 2010, without whom, I might not have had a thesis, or a final yea r full of smiles, exasperated sighs, uproarious laughter, and happy tears of joy. ! ! ! !
! ! "# Table of Contents Dedication ................................ ................................ ................................ .......................... i Acknowledgements ................................ ................................ ................................ ........... ii Table of Contents ................................ ................................ ................................ ............ iv List of Tables ................................ ................................ ................................ ..................... v Abstract ................................ ................................ ................................ ............................ v i Introduction ... 1 Literature Review..4 Views of Substance Use Initial Discovery of Social Norms and Referent Groups Social Influence: Normative Influence Theory Normative Influence Theory, as it relates to Substance Use Sense of Belonging and Substance Use Methods ....21 Sample and Procedures Measures Research Questions Results ..28 Sample Descriptives B ivariate Analysis Conclusions ..35 Appendic es ...42 Appendix A: Survey Appendix B: Results Work Cited ...67
! ! # List of Tables All tables are in Appendix B (pgs. 47 64 ) $%&'(! One: Summary Statistics for Demographic Information Table Two: Summary Statistics for Living Situation Table Three : Residence of Respondent Table Four : Housemate of Respondent Table Five : Reliability for Responsibility on Campus Table Six: Reliability for Involvement on Campus Table Seven: Reliability for Feelings f riends might have on individual substance use Table Eight: Reliability for Perceived student use on campu s Table Nine: Reliability for Perceived friend use on campus Table Ten: Reliability for Social atmosphere promote substance use Table Eleven: Reliabili ty for Individual Substance use in the last month Table Twelve: Reliability for Group Use on Campus Table Thirteen: Summary of Individual Substance Use Table Fourteen: Individual substance use in last month Table Fifteen: Individual Drug use in the last month Table Sixteen: Individual alcohol use in the last month Table Seventeen: Alcohol Consumption changes since attendance at school Table Eighteen: Drug Consumption changes since attendance at school Table Nineteen: Five Drinks in One Sitting Table T wenty: Age when respondents first used substances Summary Table Twenty One: Perceived Student Use on Campus Table Twenty Two: Perceived Friend Use on Campus Table Twenty Three : Social Atmosphere Promoting Substance Use on Campus Table Twenty Four: Summary Statistics for feeling friends might have on substance use Table Twenty Five: Summary Statistics for variables of Responsibility on Campus Table Twenty Six: Summary Statistics for Responsibility on Campus Table Twenty Seven: Responsibility on Campus Tabl e Twenty Eight: Pearson Correlations between all variables
! ! #" Abstract Social normative theorists often express the importance of understanding interpersonal influence, in order to understand personal change and individual behaviors. R esearch has often focused on the power of d escriptive and injunctive norms amongst different referent groups, and how these can predict how and why p eople behave in specific ways. The current thesis studies the issues of perceived substance use and it's a ffect on individual student behavior at a small, liberal, arts college, as well as examining the relationship between sense of belonging and substance use. Research was conducted on campus through an online self reporting survey in the January interim of 2010. Major findings from the research show how the school under study is indeed similar to the literature on substance use at colleges when it comes to perceived substance use and individual behaviors. However, the current school differed from previous findings on sense of belonging, as it was a not a protective factor for substance use.
!"#$" % & % Chapter One: Introduction Substance use amongst peer groups is of great concern for colleges and universities. Administrators, faculty, and even parents and students continually develop preventative programs to lower rates of substance use on campuses. Though much sociological research has focused on the topic of peer influence and normative behavior regarding substance use amongst college students, few have explored their relationship with sense of belonging. The current study describes social norms at a small, liberal arts college and examines the relationship between sense of belonging and individual substance use. Although the literature discusses the meaning of these terms it is important to note their meanings in the current thesis. Perceived soc ial norms are broken into two categories. The first, i njunctive norms are the perception an individual may have of what most people in his/her community approve or disapprove of in any given social situation. This norm specifies what is perceived to be m orally sanctionable by the group, and therefore motivates behavior. The second social norm is a descriptive norm, which denote s the perception of what other people do in any given situation. This norm relies on the perception of appropriate behavior in a n individual's community when "normal" behavior is ambiguous. Proximal and global norms are us ed to denote normative referent groups. Proximal represents close friends at school and global represents students at school in general. Sense of belonging is akin to how comfortable a student is on campus. Sense of belonging in the current study is measured by feelings about school, responsibility on campus, involvement on campus, and safety on campus.
!"#$" % % This thesis draws mainly from the literature of Baer, Wec hsler, and other social normative theorists (Maddock & Glanz 2005, Buunk & Bakker 1995, Cialdini & Trost 1998). This theory has been chosen over other theo ries because of the implication that states the environment and those within the environment as a va stly important component of behavior Unlike social network theory and psychological theories ( e.g. Bohnert et al 1991; Ennett et al 2006), social influence theory mainly discusses substance use patterns of behavior, as they relate to others in the enviro nment and environmental factors themselves. The sample for the study consists of undergraduates, aged 18 25, who are registered as full time students in the Fall Semester of 2009. The sample excludes only those registered students who are not within the age range. The study is being cond ucted through an online survey measuring injunctive and descriptive norms on substance use, respondent's individual sub stance use, and feelings of belonging at the particular institution The current study on substance use, and the affect of perceived social norms on the relationship between substance use and sense of belonging, diverges from the literature. R esearch on sense of belonging tends to discuss it as a protective factor against risk taking behavior 1 In the case of substance use, evidence suggests that sense of belonging is a protective factor because as a social variable it protects individuals from committing a behavior dee med risky. %%%%%%%%%%%%%%%%%%%%%%%%%%%%%% %%%%%%%%%%%%%%%%%%%%%%%%% % & %()%*+#,%-",./%"% pr otective factor is any social or psychological variable that could benefit an individual. % %
!"#$" % 0 % In the literature on sense of belonging little to no mention is made o f normative influences. Yet, sense of belonging, in light of social norms, can perhaps be both a buffer against drug use or can encourage drug use The following thesis is separated into five chapters. The current chapter discusses the researcher's reas oning for the study, as well as a brief introduction to the topic and terminology. Chapter two is a short, but comprehensive literature review in all three of the topics discussed in the current research: social norms and behaviors, substance use at the c ollege level, and sense of belonging. A short summary is provided at the end of the chapter, as well as an overview of the research questions. Chapter three lays out the method for this thesis, beginning with sample description and procedure choice. Rel iability results for scaled variables are included in Chapter three, as well as detailed research questions. The results of the research, sample descriptives and bivariate analysis, can be found in Chapter four. The final chapter discusses the results of the research, as well as the weaknesses of the research, encouragement for further study, and implications for policy based on results.
!"#$" % 1 % Chapter Two: Literature Review In this literature review, discussion will mainly focus on the three main a reas of study that the current research utilizes. Before delving into social normative research and norm transmission, the text of this chapter explains the reasoning for using the lens of social normative theory, as opposed to other well developed and va lid theories on substance use. Following this brief explanation, the literature review covers a number of older studies, which have been widely influential in norm transmission (Sherif, 1935) and the effect of referent groups on attitudes and behaviors (F endrich, 1967). The discussion then confers on the topics of social normative theory and model network theory, focusing on Bohnert et al and Buunk & Bakker. Following the discussion of these two very divergent theories, the review converses about substan ce use and its interaction with social norms. All of the research discussed in this section is focused on college campuses. Research in this section is mainly from authors Baer (1991, 1994), and Wechsler et al (1995, 2008). Lastly, this literature revie w covers sense of belonging at schools, both generally and how it relates to substance use ( Battistich and Hom 1997, Napoli et al 2003). Views of Substance Use Two main schools of thought dominate the study of substance use and abuse. The first fo llows Freudian thought, which builds itself on the concept of drives and their vicissitudes. The second, most ardently purported by Sullivan and Fairbanks, gives primary importance to the influence of the external environment, including other people, on i ndividual's substance use. The Freudian theory of motivation views substance abuse as stemming from the "needs a person has and an absence of awareness of these needs
!"#$" % 2 % and of their significance" (Rotgers et al, 1996:69). This conflict is created by either internal conflict and outward experience, or internal need and external limitations. According to psychoanalytic theory, this conflict creates a "deep uncertainty about one's right to exist[one's] identityor control" (1996:75). Though different paradi gms posit psychological problems as predating or being secondary to substance use, the overarching belief is that substance use and abuse is a psychosomatic disorder. While psychoanalytic theory predominated the discussion of substance use and abuse for an extended period of time, the last 20 years has shown a growing field which views substance use and abuse as a function of relating to others and the social environment. Leading this discussion are social network theorists and social normative influence theorists. Social network theory begins to answer how and why individuals in certain networks are more or less likely to use substances. This theoretical framework focuses on social distance and network size as it relates to habits (in this case substan ce use). The other side of the spectrum, normative influence theory, explores why and how the transmission of certain social norms lead to the use of substances. Initial Discovery of Social Norms and Referent Groups Evidence of the role of norm trans mission and the external environment 's e ffect on individual decision making first began with M. Sherif's studies on the interpretation of autokintetic phenomenon. In the study, participants placed in an ambiguous in group setting are influenced by the gro up definition of a given stimulus (Cialdini & Trost, 1998:154). Cialdini and Trost explore Sherif's findings, as related to norm transmission and formation. In Sherif's study it was determined that the group did have a strong influence over the definitio n of stimulus, allowing him to conclude, "that unstable
!"#$" % 3 % situations evoke confusion and uncertainty" causing people to look to the group to establish a common definition or norm (1998:154). Since then much research has been conducted on norms as they af fect attitudes and behaviors. James Fe ndrich's (1967) study examined the affect of perceived reference group support on racial attitudes and overt behaviors. His research engages college students about racial attitudes and a willingness to participate in interracia l activities. His study focuses on perceived reference group support as it affects overt behavior; i.e. the research further explored the relationship between group influences on the behaviors of individuals. Fendrich (1967) proposes four hypo thetical models to explain the causal relationship between overt behavior, racial attitudes, and reference group support The study (1967:963) measures reference group support using a scale that "measured the degree to which perceived reference group memb ers would be willing to participate in interracial activities". The scale asks participants to rate each question in relation to five perceived reference groups: closest friends of the same sex, closest friends of the opposite sex, parents, roommates (or husband or wife), and some older person whom you respect. The findings suggest that the fourth model, which states, reference group support determines racial attitude and overt behavior, but also that racial attitude is a partially independent determin ant of overt behavior, is the best of the four models. Though attitudes were a determinant, reference group support was proven to be an antecedent to racial attitudes and overt behavior, and participants seemed to act in accordance with reference group su pport more often than their own racial attitudes (1967:970). Furthermore, Fendrich posits, "patterns of behavior initially develop in response to significant others or reference groups" (1967:968).
!"#$" % 4 % Social Influence: Normative Influence Theory Though Fendrich establishes that reference groups affect individual behavior, it is unclear why social interactions and/or groups influence individual behavior. The body of literature on normative influences and network models has attempted to discuss this why Bohnert, Bradshaw, and Latkin, in their research on adult heroine users, test two social network models. One is social selection, where by a person who is using drugs changes networks in order to spend more time with others who are also using drugs ( 1991 : 2011). In defending this particular theory the authors reference Kandel, who "found that having peers who use marijuana was a strong predictor of initiating use of marijuana or other illicit drugs among adolescents" (1991:2011). The other theory that th e authors test is social influence, which states "the observation of substance use by members leads to changes in the substance use among others in the network, such that the other members' substance use becomes more like their friend's behavior" (1991:201 0 11). Bohnert et al describe the research as being integral in determining causes of substance use/abuse, because "it is not clear whether individuals are influenced by their network members or if their own drug use influences their choice of friends" (2010). Though this theory diverges significantly from normative influence models, as it focuses on social distance it is important to note that findings for social influence network theory share similar discussion outcomes with normative influence theor y. For instance, the results of Bohnert et al suggest, "network drug use on personal drug use operates through long term processes" (1991:1216). In normative influence models it is also posited that amongst college students norm influence is associated w ith those outcomes where individual alcohol use is influenced by use and perceived use of close friends. However,
!"#$" % 5 % unlike normative influence models, the authors state that this does not occur through friends, but instead, was "accounted for by the total a mount of substance use around an index or the embeddedness in the drug use network" (1991:2016). Normative influence models, unlike network theories, also often discuss the interaction between attitudes, behaviors, and norms. For instance, in 1995 Bu unk and Bakker conducted a study, which focuses on the willingness of adults to engage in extradyadic sexual behavior. The researchers examine how attitudes, descriptive norms, and injunctive norms influence participants' willingness to participate in ext radyadic sexual behaviors. Buunk and Bakker's work continue to build on the assumptions of Fendrich and Sherif. The study further discusses how behaviors can stem from refe rence groups, but also includes norms. The argument of the study drew from the ba se of literature that defines social norms "in terms of social pressure and approval [injunctive norms] of others who are important to the individual" (Buunk and Bakker, 1995:313). Though they believe that the above definition is true for those who are e ither dependent or are of a lower status than the referent group 2 the researchers find it plausible "that when it comes to the influence of others with equal status and to behavior that these persons may perform themselveswhat others are perceived to do [i.e. descriptive norms]" become extremely influential and important (1995:313). Buunk and Bakker support the validity of injunctive norms 3 when status and dependency are prominent in the rela tionship between an individual and their referent group. However, when the individual and referent group are of equal status, or when the individual is not dependent on the referent group, descriptive norms may play a larger %%%%%%%%%%%%%%%%%%%%%%%%%%%%%% %%%%%%%%%%%%%%%%%%%%%%%%% % 2 Referent group in this context refers to others who are important to the individual (1995:313). 3 For a more extensive definition of this in the context of this article see Buunk & Bakker (1995:313 14)
!"#$" % 6 % role than injunctive norms. To prove this theory, the researchers conduct two studies i n a 15 year period. The data from both studies reflects participants' attitudes on extradyadic sexual behavior, how friends would respond to the participant engaging in the behavior, their own past behavior, and willingness to participate in extradyadic s exual behavior. While the researchers did make slight modifications to the above variables, they remained relatively constant. Only descriptive norms were measured differently in each study. In the first study, participants were asked if friends had be en involved in extradyadic sexual behaviors (1995:314). In the second study, participants were asked to rate how likely a friend would engage in said behavior if the opportunity arose (1994:316). Results of the study proved that attitudes, descriptive norms, and in study one, injunctive norms and past behavior, were individual predictors of willingness to engage in extradyadic sex. Past behaviors, attitudes, and descriptive norms all had an additive effect on willingness to be involved in extradyadic sex (1995:317). Based on the findings, it was concluded, it is "not only injunctive norms that imply social pressurebut also descriptive norms that specify what is most common in a specific situationseem to be associated with the willingness to engage i n extradyadic sex" (1995:317). From this study, it can be inferred that injunctive norms are not the only way, nor the most beneficial tool in defining social norms affecting sexual behavior (1995:317). Besides the prominent findings, the researchers als o note the overwhelming importance of habit and familiar activity with respect to engagement in specific behaviors or activities. This can be seen as growing support for the conception that increased substance use and influence occurs through long term pr ocesses. Yet it can also be
!"#$" % &7 % assumed if taking Fendrich's results into consideration that habit is merely a reflection of the internalization of norms adopted from the reference group. Normative Influence Theory, as it relates to Substance Use Genera lly the initialization of a particular behavior by an individual is most commonly associated with the introduction of a new group with their own set of social pressures. Research on substance use at college campuses best exemplifies the heavy focus on in itialization of substance use due to new social influences and the environment. A key researcher in the discussion of normative influence theory Baer (1991) focuses on the relationship between norms and drinking motives, and social affiliation His r esearch has made obvious the power that interpersonal influence has on an individual's actions, beliefs, and attitudes. In general, research has shown "the extent to which these behaviors could be interpreted as purposive attempts to achieve a relatively small set of goals: to behave effectively, to build and maintain relationships, and to manage self concept" (Cialdini & Trost, 1998:151). Baer's literature argues that college students' alcohol use is a reflection of social norms that reaffirm the need to achieve this small set of goals. The research makes an attempt to measure the "ratings students make about the acceptability and typicality of v arious drinking behaviors[and] measures students' understanding of the social support and acceptance of drin king practices" (Baer, 2002:44). In his overall findings, Baer determines that "peer use isthe strongest predictor of adolescent alcohol use" (2002:46). Again, this is a nod to other research in the field, which argues that substance use is largely infl uence d by friend use. In his 1991 study, Biases in the Perception of Drinking Norms among College Students Baer finds that perceived friend drinking was positively related to individual's
!"#$" % && % drinking. New findings from Baer suggest that this correlation was strongest amongst reference groups who participated in more social activities together, and the correlation was weaker in dormitory samples 4 According to results, "beliefs may serve to excuse or exacerbate risky drinking problems and pose an order of resistance to prevention efforts. False perceptions of behavioral norms may be one mechanism of peer influence that allows intact living groups to ignore signs of risk" (1991:585). Baer further provided evidence for this theory in his 1994 article on c ollege drinking during the first year of college. The research focused on living situations/dormitory setting s affecting commonality of the drinking model and the availability to drink. Social settings, such as fraternities, were more likely to encourage and maintain heavy drinking because of social processes that allowed for t he extremes of drinking (1994 ). Findings also suggest that norms about drinking were m ore extreme in the Greek system than the population at large. Though p erceived norms for drin king were independently related to drinking rates 5 Baer found in his review that this held true mostly when perceived norms justified and exacerbated those heavy drinkers, who were surrounded by more accepting social attitudes already in existence (2002). However, Baer was unable to support the hypotheses that living in the Greek s ystem creates different norms or facilitates a different course of norm development. Relatively extreme views of fraternity drinking were present before subjects' entry into the Greek system" (1994:47). His research shows that approval of heavy drinking decreases over time, though less so in the Greek system, both for weekend and weeknight %%%%%%%%%%%%%%%%%%%%%%%%%%%%%% %%%%%%%%%%%%%%%%%%%%%%%%% % 4 Again this provides support for the concept of long term processes and referent groups (people individuals spend more time with) having an affect on an individual's behaviors. 5 These findings further provid e support for refe rence groups affecting behavior, a nod and confirmation of earlier research, such as Fendrich's fourth model.
!"#$" % &' % drinking. Still, Baer was able to show that norm facilitation or creation were dif ferent between groups (1994:47) 6 Besides his own work, Baer draws on earlier literature where students who had moderate attitudes about heavy drinking, "perceived the community norm of alcohol use as much more liberal than their own" (2002:46). Students therefore thought that "normative drinking rates and drinking consequences were higher than their own [experiences and]higher than they actually were", when gathered by independent studies (2002:46). Literature on this topic generally views perceived dr inking rates and norms as having a vast impact on individual substance use, as well as perceived understanding of community environments. Maddock and Glanz's (2005) research on alcohol use and perceived normative influences has shown similar findings to Ba er. In order to build on working methods and interventions for alcohol use and prevention, the researchers studied the misperception and overestimation that college students often make about alcohol consumption in relation to their peers. Beginning their study by examining the "mixed results" of interventions across campuses to reduce drinking, the researchers initial impression states, the "misperception of the social norm is more often a function of individual's perceptions than the actual college envir onmentalthoughthe magnitude of overestimation differs by housing situation" (2005:316). Based on these impressions, the researchers hypothesize "normative perceptions of proximal referent individualsmediate the relationship between alcohol use and c ampus wide social norms" (2005:317). %%%%%%%%%%%%%%%%%%%%%%%%%%%%%% %%%%%%%%%%%%%%%%%%%%%%%%% % 6 G roups in this setting are Greek housing and normal dorm residential settings. (1994:45 ). %% %
!"#$" % &0 % For instance, members of sororities and dormitory living students often perceived that others drank more than they did. However, fraternity members perceived that others drank less than themselves. Individual alco hol consumption can then be seen to correlate to the perceived use of alcohol by proximal normative beliefs (2005:316). For heavy drinkers, or environments that encourage heavy drinking, perceptions of others drinking are lower than self reports. Howeve r, for those who are moderate, light, or abstain perceived drinking rates are higher than self reported consumption. The findings of the research state that proximal normative beliefs are significant mediators of global social norms. Based on the result s, the researchers conclude, "proximal normative beliefs about alcohol use are a possible salient pathway that effects consumption and related problems" (2005:321). Overall, Maddock and Glanz suggest that interventions turn from focusing merely on global social norms and spend more efforts focusing on perceived proximal norm beliefs of close friends. Maddock and Glanz's research approaches alcohol use differently than some of the previous researchers discussed. Unlike Baer, who does not state the differe nce between proximal and global norms, Maddock and Glanz focus on defining the different social norms across reference groups. W hile Buunk and Bakker focus on the kinds of norms, descriptive versus injunctive, which affect behavior, Maddock and Glanz are more focused on the affect of norms across different reference groups. Though Baer's research does not define the difference in norms, Maddock and Glanz's research reaffirms Baer's research on peer use affecting individual substance use. Both find that social settings and different social reference groups affect how perceived norms for drinking have an effect on drinking rates.
!"#$" % &1 % Wechsler and Nelson (2008) have also found that drinking is correlated with where a student lives, as well as how involved a student is on campus, and the kind of affiliations a student is involved with 7 Wechsler in the last twenty years has focused on drinking practices and patterns on college campuses. Wechsler's research is based on evidence from a national survey, the CA S 8 which mostly measures binge drinking 9 Although there is disagreement over the figure and definition of binge drinking, Wechsler and Nelson state that the number used in the CAS results is adequate in capturing the full picture of drinking on US colle ge campuses. Their assumption is based on the fact that binge drinking has remained at a constant rate although they admit that a polarization 10 pattern has emerged (2008 ). The CAS shows that drinking rates vary by college implying that environmental fa ctors (such as availability of alcohol, also referenced by Baer) may influence student binge drinking (2008:484). The following factors are considered important determinants in binge drinking rates according to Wechsler's research: 1. Student affiliations a nd the surrounding environment. These factors are especially pertinent for initiation of alcohol use (2008:484). 2. Where a student lives, especially in relation to the amount of supervision associated with each living situation (2008:484 85). %%%%%%%%%%%%%%%%%%%%%%%%%%%%%% %%%%%%%%%%%%%%%%%%%%%%%%% % 7 This is similar to the conce pt of availability of drinking mentioned in Baer's research. 8 CAS the Harvard School of Public Health College Alcohol Study, started in 1993. The CAS (2008:481) conducted from 1993 2001, with participation from 120 colleges, "provide[s] the first n ationally representative picture of college student alcohol use and describe[s] the drinking behavior of this high risk group". 9 Binge drinking is defined as 5 or more drinks in a row for men, or 4 or more for women, as well as one or more drinks on occ asion in a two week period immediately before the survey is administered. 10 Polarization has occurred as a result of a simultaneous increase in the number of binge drinkers and abstainers ( 2008:483 ) % %
!"#$" % &2 % 3. Demographic composition of the student body influences student drinking. The CAS finds that "greater racial and ethnic diversity on campus is associated with lower binge drinking rates among the white majority students" (2008:485). 4. High amount of social capitol, me asured by student volunteerism, decreases the risk of participation in binge drinking. It is interesting to note that the CAS and Baer's research both place student affiliations and student activities above actual living situation. Though both do note the importance of a new social environment in affecting individual drinking rates. Graham, Mark, and Hansen's (1991) research delineate s different forms of social influence focusing on misperception and modeling. The authors label social influence as ex plicit offers (active social pressure), modeling behavior (passive social pressure), or misperception (passive social pressure). Misperception according to Graham et al. is caused by one of two things; ambiguous or misleading statements made by friends, o r the individual's own use, changing the way he/she sees friends' use (Graham et al, 1991:292) 11 The other social influence, modeling behavior, is similar to the findings from Buunk and Bakker. Like descriptive norms which influence individuals based on the assumption or actuality of what others are doing in any given situation, Graham et al. found that peers' modeling behaviors had a great e ffect on adolescents' own use. %%%%%%%%%%%%%%%%%%%%%%%%%%%%%% %%%%%%%%%%%%%%%%%%%%%%%%% % 11 The difference between misperception and perception (perceived view) should not be conflated. Mispercepti on implies the individual's change in use as changing the perception they have of reference group use. Perceived v iew is whereby an individual's use changes because of their perception of reference gr oup use, not vice versa. Though Baer's research does somewhat account for misperception, as evidence by the 1994 and 1991 articles on fratern ity drinking, it is important to make a distinction between individual use changing because of a perception versus perception changing because of individual use.
!"#$" % &3 % Generally, the authors believe that all three of the social influences have an aff ect on substance use among adolescents. "The results provide good support for the hypothesized model positing three distinct social influence processes underlying adolescent" substance use, "namely, active pressure involving explicit offers to try alcohol and passive pressure involving social modeling and overestimation of peer use of a substance" (1991:295) 12 Though the authors are hesitant to apply their findings to older adolescents, it seems that the findings in this study are very similar to Baer and other researchers of the time, who are studying college enrolled young adults. In general, all seem to find proximal passive forms of social influence as important predictors of substance use initiation. Sense of Belonging and Substance Use The r eview of literature thus far has discussed the consumption of substances, mainly alcohol, in relation to social influences and reference groups. Y et the current study also focuses on the relationship between sense of belonging and substance use, which is a field of literature that is often separate, though closely related, to the previous topics discussed While there is limited evidence that social norms are an important factor in the relationship between sense of belonging and substance use there is ext ensive research on sense of belonging at schools, and its affect on the initiation or continued use of substances. Napoli, Marsiglia, and Kulis (2003) carry out one such study, on sense of belonging as a protective factor against drug abuse among an urban adolescent Native American population. Since the population is considered at high risk for drug use and %%%%%%%%%%%%%%%%%%%%%%%%%%%%%% %%%%%%%%%%%%%%%%%%%%%%%%% % 12 Italics are author's own.
!"#$" % &4 % abuse, the researchers studied sense of belonging at school delaying the onset of drug use and/or fostering a lower use rate amongst this population (2003:29). The aut hors define sense of belonging as being a part of a school community, defined by personal membership, respect, and support students felt (2003:28). A high sense of belonging was determined for those students who had better grades, spen t more time on homework, were more motivated for educational achievement, and had an affiliation to a subgroup (2003:28). Sense of belonging was also self reported with participants being asked how do you feel you belong?' Napoli et al found that a stro ng sense of belonging in school buffered drug use behaviors and norms of Native American students (2003:38). Oppositely, school isolation and disconnectedness contributed to higher rates of drug use and earlier drug initiation (2003:35). However, the res earchers note that the correlation between increased academic interest and/or achievement and lower drug use was weak (2003:36). Overall, "sense of belonging played a protective role against drug use, even after accounting for differences in school achiev ement" (2003:36). Battistich and Hom discuss the hypothesis that "there would be less problem behavior among students at schools where the social context promotes bonding" (1997:1). The researchers examine the relationship between students' sense of the school as a community and problem behaviors. Students in the study are from elementary schools in six school districts, and are asked a 38 item scale that determines a students' sense of the school as a community. Individual substance use is gauged with individual questions on cigarettes, alcohol, and marijuana. Students are also asked about 10 other
!"#$" % &5 % delinquent behaviors and six victimization behaviors, and how often the student has experienced or committed any of the behaviors in the last year. The r esearchers find that "students' sense of school as a community was negatively associated with drug use, delinquency, and victimization" (1997:3). Between the six schools, increases in the sense of community were associated with lower problem behaviors. T he authors believe "that the social context of the school is related to a wide range of student attitudes, motives, and behaviors, and thus merits increased attention in future research as an important determinant of children's developmental outcomes" (199 7:4). In the discussion of their results, the authors reference Coleman who suggests that socialization amongst children within a "functional community" occurs "through the salient normative consensus among community members presumably owning to both incr eased clarity about appropriate and inappropriate behaviors and increased monitoring, and enforcement of community norms" (1997:4). The authors admit the limitations of their study, stating that their data is cross sectional and therefore evidence of caus ation does not exist. However, the researchers provide sufficient evidence to state that sense of community, especially in schools, is important in the developmental outcomes of young people. Other research on sense of belonging has mainly focused on t he integration of adolescents into schools 13 The PISA, Programme for International Student Assessment, gathers data on students' involvement in school, as the measurement for sense of belonging. The survey conducted across 43 countries, defines sense of belonging similarly to Napoli et al. Feelings of belonging' were rated by students' "participation in %%%%%%%%%%%%%%%%%%%%%%%%%%%%%% %%%%%%%%%%%%%%%%%%%%%%%%% % 13 Much research is also focused on minorities or those at a disadvantage in the community, although they are not mentioned in this review.
!"#$" % &6 % academic and non academic pursuits" and if "students tend to have good relations with school staff and other students" (2000:8). In the survey, partici pants were asked to rate "responses to six items describing their personal feelings about being accepted by peers, and whether or not they felt lonely", considering themselves an outsider or out of place (2000:11). The study of social normative theory in relation to substance use and abuse on school campuses has led to important and interesting discourses about how and why individuals behave in certain ways. Most importantly, Baer has shown support for peer influence being the best predictor of individua l substance use. However, other research has both supported and sown the conditionality of this work. Firstly, descriptive norms can play a larger role in influencing behavior than injunctive norms when members of the reference group are of equal status to the individual. Also, social affiliation and residential setting seem to play a large role in determining behavior. Though this role was often mitigated by values and attitudes in existence before college entrance, especially moderate attitudes towar d substance use, which also play a large role in determining individual substance use and norm perception. Interestingly the heterogeneity of a school and social capital of an individual also affect substance use. Sense of belonging research continues t o support social norm research in that sense of belonging plays an integral role in the onset of substance use. Studies on sense of belonging generally view high sense of belonging in school as a protective factor against risk taking behaviors, especially for groups at high risk. Yet, research fails to encompass the impact of a school's culture and normative influences The literature
!"#$" % '7 % informs readers that sense of belonging provides a buffer to the initiation of drug and alcohol use. The current study ab out sense of belonging's affect on substance use, and th e strength of normative influences diverges from the literature The following thesis attempts to describe social norm influence on substance use at a small, liberal arts college, and the relationshi p between sense of belonging and substance use, when considering social norms. The following research attempts to show that social norms have the same effect on a small, liberal arts college, as they would at a larger university. The researcher anticipat es perceived peer use to be higher than individual substance use reports, or in the case of this analysis have lower correlation. Oppositely, perceived friend use should report a higher correlation with individual substance use. However, all should be po sitively correlated to individual substance use, despite some relationships being weaker. It is also hypothesized that sense of belonging does not have the same kind of relationship at a small, liberal arts college, as it does at a larger university (i.e. that sense of belonging does not act as a protective factor against substance use).
!"#$" % '& % Chapter Three: Methods SAMPLE AND PROCEDURES The sample for this study ( N = 300 ) consist s of undergraduates, aged 18 25, who attend a l iberal arts college Th e sample excludes only those registered students who are not within the age range. Those under 18 are excluded due to being a vulnerable population. Individuals over 25 were excluded because the researcher felt they are not representative of the majority of the student body at this institution Potential participants were found using the College's enrollment list, provided by the school's registrar. In order to have a sufficiently large number for adequate statistical analysis, all students included on the Registrar list were invited to participate in the survey. The data was collected by an anonymous online survey, conducted through Survey Monkey. The online source kept track of which email addresses/potential participants filled out the survey. How ever, the re searcher is blind to the data. The survey wa s emailed to all students on the enrollment list provided by the College's registrar. This s urvey does not colle ct any identifiable information To increase return rates, potential participants who had not yet completed the survey were sent one follow up email, relaying the same information as the first email. Procedures for the research were approved by the College's IRB. A self report survey is the best option for a variety of reasons. First, sel f reports about drug use can threaten an individual's sense of safety and may also influence the reliability of responses. A survey facilitates anonymity to all participants, reducing the risk associated with participating, and also improves the reliabili ty of responses. Second,
!"#$" % '' % the sample population is chosen from a small, tight knit community. This community atmosphere may make risks mentioned above particularly acute; therefore, providing anonymity is especially integral to obtaining accurate data. Th ird, a survey allows for more precise questions compared to open ended questions. Lastly, a quantitative approach allows for a larger sample to be included in the study This increases the ability to generalize findings to the entirety of the college M EASURES All of the questions used in the study are well known instruments in the research literature on social normative influences on substance use. Survey questions are developed from four surveys previously used by other scholars and schools, but modif ied for the current study. Questions are pulled from th e Core Alcohol and Drug Survey Long Form (), College Alcohol Survey Short Form (1995) and the ACHA National College Health Assessment () The indexes have been tested at over 120 schools in an 8 year period. Sense of Belonging Measures Sense of belonging on the college campus was to be c alculated using four variables: feelings about school, responsibility on campus, involvement on campus, and safety on campus. Safety on campus and feelings a bout school are each measured with one question 14 Responsibility on campus, which was measured via three questions, are all given on a Likert type response scale where 1= "strongly disagree" and 5= "strongly agree" and the scale score is the sum of these three questions. Involvement on campus is measured via 6 components, asking respondents to rate how active they were in each of %%%%%%%%%%%%%%%%%%%%%%%%%%%%%% %%%%%%%%%%%%%%%%%%%%%%%%% % 14 See Appendix A for a copy of the su rvey
!"#$" % '0 % the six components on a Likert type scale where 1= "not involved" and 4= "leadership position". Measures for responsibility o n campus 15 evidenced strong reliability Reliability estimates based on Cronbach's alpha were .738 for RoC. Involvement on campus evidenced weak reliability. The reliability estimate was .509 It is evident that involvement on campus was not a reliable scale in measuring a unidimensional concept. For these reasons, results about sense of belonging are based on only three variables, responsibility on campus, safety on campus, and feelings about school. Social Norm Measures Social norms are measured in two categories, injunctive and descriptive. Questions related to injunctive and descriptive norms are asked twice, using two different reference groups; close friends and students at the college in general. For instance, participants are asked, How many of your close friends at your school do you think engage in drug use?" and "How many of the students at your school do you think engage in drug use?" These two reference groups are most pertinent because previous research has shown that the influence of descriptive and injunctive norms has varying magnitudes depending on the source. For this particular college, the researcher feels it is best to focus on close friends at school, or proximal norms, and students at school in general, or global social norms Focusing on global social norms fits best with the college, because party activities are most often campus wide and the campus lacks social groups that have a particular focus on partying, such as the Greek system. Using close friends at school is %%%%%%%%%%%%%%%%%%%%%%%%%%%%%% %%%%%%%%%%%%%%%%%%%%%%%%% % 15 Referred to henceforth as RoC.
!"#$" % '1 % the best fit because of a lack of organized social groups on campus, and a majority of students reside on campus Descriptive norms are measured with three variables, group use on campus (global), perceived student use on campus (global), and perceived frien d use on campus (proximal). For global descriptive norms, respondents are asked to rate whether drinking is a central part of 5 different groups, rating either a 1= "yes" or 0= "no". Respondents are also asked two questions relating to the perception of student use, and the simple average is calculated to create one variable. Both questions are on a Likert scale asking respondents how many students at school engage in drug or alcohol use, ranging from 1= "none of them" to 5= "all of them". Proximal desc riptive norms are measured similarly; with respondents rating how many close friends on campus engage in drug or alcohol use, on the same scale. Injunctive norms are measured with two variables, social atmosphere promoting substance use (global ref. gro up) and feelings friends might have about a respondents' individual substance use (proximal). The global injunctive norm variable, social atmosphere, is measured by averaging two questions which asked respondents whether the social atmosphere promoted alc ohol or drug use, rating either 1= "yes" or 0= "no". Proximal injunctive norms are measured on a 10 component question asking respondents how a close friend feels (or would feel) about the individual doing a number of behaviors rating the answers from 1= "strongly approve" to 5= "strongly disapprove", with 3= "neither approve nor disapprove". For each of the four variables (proximal and global descriptive and proximal and global injunctive), a single index based on averages of the multiple measures is
!"#$" % '2 % developed. Measures for all variables evidenced strong reliability. Reliability estimates based on Cronbach's alpha ranged from .911 for FIS, .783 for PFU, .618 for PSU, .788 for SA, and .803 for GU 16 A ll had high (< .600) alpha's evidencing that indexe s were reliable. Individual Substance Use Measures Individual substance use in the last month for respondents is measured with two questions, which averaged together to create a single scale ranging from 1= "I did not drink/use drugs at all" to 5= "Once a day or more" 17 Individual Substance use in the l ast month had an alpha of .557 Though reliability for this scale ranged lower than expected, the scale is still high enough to state that the index is useful in predicting substance use for individuals. Three other questions gauge individual substance use. The first ask respondents to rate how their individual alcohol use and drug use had changed since attendance at their current school. Options are increased, decreased, stayed consistent, and abstain ed from alcoho l/drugs throughout attendance. Another question asks respondents to state the age when they first used each of the following substance s : alcohol (beer, wine, liquor), marijuana (pot, hash, hash oil), amphetamines (diet polls, speed, adderall ), hallucinogens (LSD, PCP, psilocybin, mushrooms), and designer drugs (ecstasy, MDMA). Finally, r espondents rate how often they had five or more drinks in one sitting, attempting to gauge how many respondents could be considered binge drinkers 18 The %%%%%%%%%%%%%%%%%%%%%%%%%%%%%% %%%%%%%%%%%%%%%%%%%%%%%%% % 16 FIS feelings friends might have on individual substance use. PFU perceived friend use at school. PSU perceived student use at school. SA social atmosphere promote substance use. GU group use on campus. 17 To see each question see Appendix A, numbers 21 22 %% % 18 Definition of binge drinking pulled from the CAS, as well as the question itself.
!"#$" % '3 % que stion asks respondents to note how many times they have had five or more drinks in a sitting, with answers ranging from 1 = "None" to 6 = "10 or more time". RESEARCH QUESTIONS The researcher of the study asks the following questions : 1. Is there a relat ionship between descriptive norms at a small liberal arts college and individual substance use? 2. Is there a relationship between injunctive norms at a small liberal arts college and individual substance use? 3. Is there a relationship between sense of belong ing and substance use? HYPOTHESES 1. Social norms will have the same affect on substance use at a small, liberal arts college, as they would at a larger university. 2. Sense of belonging will not have the same relationship on substance use at a small, liberal arts college, as it does in the literature at large. Research questions are answered through b ivariate analysis, using SPSS software. Correlations are determined using Pearson with a significance level of .05 (2 tailed). Each sense of belonging variabl e, feelings about school, sense of responsibility on campus, and feeling safe on campus, are run with the variable individual substance use in the last month 19 For descriptive global norms two tests using group use on campus and perceived student use on c ampus are run with the variable individual substance use in the %%%%%%%%%%%%%%%%%%%%%%%%%%%%%% %%%%%%%%%%%%%%%%%%%%%%%%% % &6 The remaining three measures for individual substance use are not used in the vivariate analysis, and merely a dd to the descriptive analysis. %
!"#$" % '4 % last month. Descriptive proximal norms (perceived friend use at school), and both global (social atmosphere promote substance use) and proximal injunctive norms (feelings friends might have o n individual substance use), are tested once each with individual substance use in the last month.
!"#$" % '5 % Chapter Four: Results SAMPLE DESCRIPTIVES Demographics The sample consists of 190 females and 105 males, with 5 respondents not reporti ng gender. The largest section of the sample population is 20 years old (79, 26.3%), while the remainder of respondents in the sample population are as follows: 18(14.3%), 19(22.7%), 21(21.7%), 22(11%), 23(3.7%), and 25(.3%). For housing statistics valid respondent numbers are higher than the population because respondents are able to choose more than one option on either or both questions relating to the topic. Respondents stated current residence ( N = 310 ) based on a list of eleven choices, with 28.4% living in what are traditionally considered first year dorms. Off campus housing garnered 21% of respondents, follo wed by an on campus suite style dorm with 11% of respondents. Respondents also stated with whom they lived ( N = 325 ). Most respondents (76 .3%) lived with a room/housemate, followed by students who lived alone (8.9%). Individual Substance Use Based on the mean ( 2.7766 ), most respondents fell between using substances 2 3 times a month in the last month (3 on the scale). However, the stand ard deviation for the dependent variable was 1.44 If the substances are broken down, the mean remains relatively consistent for drug use in the last month ( 2.53 ), though the standard deviation is slightly larger ( 1.9 ). The mean for alcohol use in th e last month rises substantially to 3.03 with a standard deviation of 1.437 It is important to note that of 300 respondents,
!"#$" % '6 % 30 were above 5 on the subscale, meaning respondents used substances 3 4 times a week or more in the last month. The largest percentage of respondents did not use substances at all in the last month (20%). When differentiating between drugs and alcohol this number changed significantly, with 50.3% of students not using drugs at all in the last month. However, 27% of responden ts use drugs between 1 2 times a week to once a day or more. The percentage for once a day or more was 6.3%. 22% abstained from consuming alcohol in the last month similar to the overall substance use variable, and the largest percentage (50%) fell betw een 3 4 on the subscale (or 2 3 times a month to 1 2 a week). Consuming alcohol everyday (.7%) was a much smaller percentage than drug use everyday Overall, students overwhelming (49%) stated that their personal alcohol use had increased. 13.7% stated a decrease in use, 20.7% stated consistence, and 12.3% stated abstention. 40.7% of respondents noted an increase in drug use. 11.3% of respondents stated a decrease, 16.3% stated consistency, and 27.7% of respondents abstained. Ages of first use were exceptionally varied. The following are the age ranges for each substance in which all respondents who answered the question fell, and how many noted that they had abstained : 1. Alcohol range 4 26; 21 abstained 20 2. Marijuana range 11 23; 70 abstained 3. Amp hetamines range 8 21; 170 abstained 21 4. Hallucinogens range 14 21; 138 abstained %%%%%%%%%%%%%%%%%%%%%%%%%%%%%% %%%%%%%%%%%%%%%%%%%%%%%%% % '7 % 8%9.:%,*;$.)*,%)<*.$%=><:#)=%;?%#)%@;>.%:+.>.%*+.%$>#)A#)=%"=.%B">#.,/%C;*%#,%<9*.)%D<:.>%*+")% #*%#,%#)%*+.%E)#*.$%F*"*.,G%% % '& %F
!"#$" % 07 % 5. Designer drugs range 14 22; 166 abstained The question gauging binge drinking had an overwhelming majority ( 62.7% ) of respondents choosing none, and zero percent of respo ndents chose 10+ times a week. 19.3% of respondents chose once 22 The standard deviation in alcohol use is 1.117 R esults for drug use are similar with a standard deviation of 1.284 Descriptive Norms Global According to respondents, male and fema le students, as well as alums all have drinking as a central part of their social lives. Respondents also overwhelmingly felt that faculty and staff did not have drinking as central to their social life. Male students had the highest mean at .73 (on a 0 1 scale), and the smallest standard deviation ( .445 ) amongst groups who had drinking as central to their social life. The mean for female students and all students was the same, .68 and had similar deviations at .467 and .468 respectively. Though Alums had a mean of .42 which suggests many respondents believed that drinking was central to their social life, it is not high enough to suggest that most students believe this statement. When all of these groups were combined to form a single variable, 33. 3% of respondents were over .80 on the scale, 25% were at .60, and 21% were at .00 on the scale. This result suggests that respondents varied greatly on the scale, and the combined statistic seems less useful then the statistics of each group on campus. Summing the two separate questions about perceived student drug use and drinking on campus created the results for perceived student substance use on campus. %%%%%%%%%%%%%%%%%%%%%%%%%%%%%% %%%%%%%%%%%%%%%%%%%%%%%%% % 22 For the remaining statistics on binge drinki ng see Table Eleven and Nineteen in Appendix B.
!"#$" % 0& % On a scale of 1 5, with five being all students and 1 being none, the mean for perceived substan ce use on campus was 3.7620 with a standard deviation of .39632 This suggests that respondents perceived that most students on campus engaged in substance use of some kind. 58.3% of respondents scaled at 4.00 or above, or 58.3% respondents marked tha t most or all students engage in substance use. Proximal Perceived use of substances by friends on campus shows similar results to global descriptive norms (perceived substance use by students). The mean for perceived substance use by friends was 3.7397 only slightly below the mean for perceived substance use by students. However, the standard deviation, .95287 is much greater suggesting that respondents perceived a greater variation in fr iend use than student use. 58 % of respondents scaled at 4.00 or above on this variable, yet 25% marked between 3 3.5 on the scale, and a number of respondents (4.7%) marked below 2.0 on the scale. Again, t his suggests a greater variation in the perception of substance use by friends, than by students generally, on campus. Perceived drinking and drug use by friends on campus differed significantly. Perceived friend drinking had a mean of 4.04 while perceived drug use by friends was 3.42 However, both had significant standard deviations ( .9 or above ). 49.7 % of respondents marked 4.00 or above on the scale for perceived drinking by friends. The percent age for drug use was similar with 58% of respondents choosing four or above However, more respondents (frequency wise) perceived that most friends drank, th ough they did not necessarily perceive them as engaging in drug use. Higher variation in
!"#$" % 0' % respondent choice existed on the question regarding drug use amongst friends than on the alcohol use amongst friends. Injunctive Norms Proximal Respondents' cho ices for feelings friends might have had about the individual ranged on a scale from 1 5, with 1 being strongly agree and 5 being strongly disapprove. The sum of 10 questions determined the variable that gauged how a respondent's close friend would feel a bout them using substances. The mean for this variable was 3.1994 with a standard deviation of .86497 If each question is broken down, respondents' answers seem to indicate that close friends would approve of using marijuana occasionally and regularl y (means 2.34, 2.94 ), as well as trying LSD once or twice (mean 2.46 ), but were less likely to approve of drinking every day (mean 4.40 ) or taking amphetamines regularly (mean 4.16 ). Respondents' answers also indicated that close friends were likely to ne ither approve nor disapprove of using LSD regularly (mean 3.51 ), trying amphetamines (mean 3.30 ), and having four or five drinks in one sitting or having one or two drinks a day (mean 3.43, 3.29 ). However it is important to note that the standard deviatio ns for all of these means was above 1.05, except having four or five drinks nearly every day ( .873 ). Global Social atmosphere promoting substance use had a mean of .8598 on a scale of 0 1. The standard deviation for this mean was .31586 and of t he 296 respondents 80.7% were a 1.00 on the scale and 9.7% with 0.0, with rest of respondents either missing or in
!"#$" % 00 % the middle. Broken down, each of the questions asking about drug use and alcohol being promoted showed similar results. Alcohol use being p romoted had a mean of .85 with a standard deviation of .353 while drug use being promoted had a mean of .87 and a standard deviation of .342 84.3% of respondents agreed that the social atmosphere promoted alcohol use. Similarly, 85.7% of responden ts believed that social atmosphere promoted drug use. Sense of Belonging Respondents were asked if they felt safe on campus, and of the 294 responses 97% chose yes. The mean for this variable was .99 with a standard deviation of .101 Another vari able gauged how respondents felt about going to school overall. The scale for this variable was 1 5, with 1 being that respondent hated going to school, and 5 being respondent loved going to school. The mean for this variable was 4.2114 with a standard deviation of .80336 Observing the frequencies for this question, most respondents (82%) fell between 4 5. Only 2.6% of respondents feel between 1 2, with the remainder (14.7%) falling at 3.00. Respondents' sense of responsibility on campus were com puted through three separate questions; campus encouragement to help others, a responsibility to contribute to the wellbeing of others, and feeling like a valued person on campus. The scales ranged from 1 5, strongly disagree to strongly agree. The mean for help to others was 4.0671 similar to the mean for the wellbeing of others was 4.0269 The mean for feelings of value was 4.0468 All three had standard deviations between 0.8 0.9. Over 70% of respondents answered 4 or 5 on all three of the questi ons, which formed the sense of responsibility variable. When these three questions were computed to form the variable
!"#$" % 01 % the mean was 4.0473 with a standard deviation of .70029 The scale for this variable was also 1 5, from strongly disagree to strongly a gree, and 66.1% of respondents averaged between a 4 or 5. 26.7% of respondents fell between 3 4, leaving only 5.9% falling under 3 on the scale. BIVARIATE ANALYSIS Pearson Correlations were run between the various variables, with a significance level of .05 (2 tailed), unless otherwise noted. The relationship between feelings about school and individual substance use were negatively correlated, but was highly insignificant (r = .005, N = 291, p = .926). Individual substance use and involvement on camp us, safety on campus and individual substance, social atmosphere promoting substance use and individual substance use, perceived group use on campus and individual substance use, individual substance use and perceived student use on campus all had n.s. cor relations 23 R espondents' sense of responsibility to campus was positively correlated with individual substance use, r = .126, with p = .032. The significance and correlation levels imply a small amount of positive effect on each other. Individual subst ance use and perceived friend use on campus were positively correlated (r = .543, N = 285, p = .000). These two variables were correlated at a .01 level (2 tailed). Individual substance use and feelings friends might have on individual substance use wer e positively correlated (r = .461, N = 284, p = .000). %%%%%%%%%%%%%%%%%%%%%%%%%%%%%% %%%%%%%%%%%%%%%%%%%%%%%%% % 23 For full statistical picture, see Table Twenty Eight in Appendix B. %
!"#$" % 02 % Chapter Five: Conclusion and Discussion Overall, the results of this study show that proximal norms, both descriptive and injunctive, are significantly, positively related to individual substance use. However, results were unable to replicate any kind of relationship from previous research between global norms and individual substance use. Though the data supports some of the hypothesized reactions, and lends itself to the growing body of literat ure, some results prove to be divergent from expectations. Congruent with the literature, students of this study generally believe that the social atmosphere promotes substance use. Perception of high substance use amongst students generally on campus is wide spread, and this is greater for alcohol use than drug use. No student in the study believed that substance use on campus was at none, suggesting that students perceive no one at their institution as not participating and engaging in substance use. Respondents did however perceive that students engaged in alcohol use more than drug use. Despite high perceptions of substance use, a large percentage of students do not actually engage in substance use at all, supporting Baer's research at larger univers ities. The results also support Wechsler et al's findings on polarization in substance use, with most students falling at either extreme of the scale (not using substances at all or using over 3 4 times a week). Though polarization does occur when lookin g at substance use generally, this is much more concentrated in drug use. It appears that within this population many students use drugs at a level that could be considered heavy, and alcohol use does not come close in matching this statistic. In fact, f ew students based on self reports are considered heavy drinkers. While students perceive high substance use, especially drinking, few students actually engage in substance use. Many students, based
!"#$" % 03 % on self reports, abstain from alcohol use of any kind, s upporting the inflation of perception of use on campus. Though this is in line with the literature, relationships between both global injunctive and descriptive norms and substance use are non existent in the current study. Perceived student use, group use, and the social atmosphere promoting substance use are not correlated at all to individual substance use in the last month. The lack of relationship between these variables may be due to a number of factors. It seems most likely that students' substa nce use levels or values regarding substance use were developed before college entry, which could explain the lack of relationship between global norms and individual substance use 24 The lack of relationship could also be the result of campus size. With an enrollment of about 800, the school environment might limit the power of global norms because students view the entire campus as friends (proximal level). This supposition about campus size is plausible based on the results of the relationship between proximal descriptive norms and individual substance use. As hypothesized, perceived friend use on campus is positively correlated with individual substance use. Though this does not prove that there is affect or causation, it does support that the two v ariables move in the same direction. Like previous research (Baer, Wechsler et al), proximal descriptive norms about the perception of drinking are associated with individual use. Proximal injunctive norms are also correlated with individual substance us e, such that as approval ratings increased, so did individual %%%%%%%%%%%%%%%%%%%%%%%%%%%%%% %%%%%%%%%%%%%%%%%%%%%%%%% % '1 % Though a majority of respondents did state an increase or decrease in alcohol use, the majority of students maintained or abstained from drug use throu ghout college.
!"#$" % 04 % substance use. This further supports the literature, which suggests that injunctive norms at the proximal level are just as, or more powerful than descriptive norms. Injunctive norms at the p roximal level have a stronger relationship to substance use than descriptive norms at the proximal level. This result comes in direct conflict with Buunk and Bakker's results about individuals not dependent on each other. The researcher of the current st udy assumed that the results would reflect findings similar to Buunk and Bakker because college students should feel that they have relatively equal status. However, the results prove that injunctive matters more than descriptive norms. Approval of use is tied in some way to individual substance use, such that it is a bigger factor than what others are actually doing. Approval mattering more might have to do with a number of factors. It seems most plausible that the homogeneity of the campus creates an environment where approval matters most, even more than merely what friends are doing. Wechsler found that greater diversity led to lower drinking rates among majority white students. Perhaps, proximal injunctive norms were the most salient due to this factor. Also, perhaps a greater distinction between class levels makes injunctive norms more salient then descriptive norms. Perception of behaviors relationship to individual substance use was only a piece of this study. The discussion now turns to th e relationship between sense of belonging and substance use. Previous literature has stated that sense of belonging acts as a buffer to substance use. However, the results of the current study deviate greatly from these findings. Though most of the vari ables representing sense of belonging had no relationship to individual substance use, one relationship did exist. Responsibility on campus and individual substance use had a positive relationship, whereby an increase in
!"#$" % 05 % substance use also meant an increa se in responsibility on campus. Students who felt valued, felt the need to contribute to the wellbeing of others, and believed that the campus encouraged helping others, had higher levels of substance use, than those students who did not agree with the ab ove statements. This relationship can be explained in two possible ways. First, a high sense of belonging might be the result of more accepting attitudes and perceptions toward substance use, and those who use them. This is plausible based on global desc riptive and proximal descriptive findings at this institution. Baer discussed how beliefs serve to exacerbate heavy drinking problems, and that perceptions allow for individuals to hold to their beliefs because accepting social attitudes exist for use. I f we take this to be true than, individuals who enter into the college environment with accepting attitudes and beliefs about substance use, might feel a higher sense of belonging to the campus. For instance, students who use substances more than their peers have a higher sense of belonging, due to the moderate acceptance of use on campus and the perception that everyone on campus uses substances. This is not to say that an acceptance or approval of substance use promotes increased drug use 25 but that individuals who already use substances feel a higher sense of belonging than those who do not use substances 26 Another possible explanation for this result is that a third factor, not tested here, is affecting both variables. The researcher posits that attitudes and values about the school environment might shed light into this relationship, based on the above stated %%%%%%%%%%%%%%%%%%%%%%%%%%%%%% %%%%%%%%%%%%%%%%%%%%%%%%% % 25 As results proved, most students believed that many students used substances but did not use substances themselves. % 26 This supposition makes sense if considering Baer's research on the Greek system. %
!"#$" % 06 % assumptions. Research examining how students on campus felt about substance use, and students' beliefs about the campus environment and ca mpus culture, might be helpful in determining why the relationship between sense of belonging and substance use is positively correlated. Based on the relationship of sense of belonging and individual substance use in this study perhaps sense of belo nging does not always function as a protective factor against substance use. Still, the results of this study show, a small, liberal arts college may have similar patterns to larger universities in regards to the relationship between proximal norms and su bstance use. The results also conclusively support earlier findings about polarization and perceived use on college campuses. However, the relationship between global norms and individual substance use is not conclusive in this study. It is undeterminab le if this relationship on small, liberal arts college campuses is similar in any way to larger universities. Limits of the current study and suggestions for further research Though the current study has hinted at some intriguing patterns, the study is limited in a number of ways. Data on substance use is self reported, and therefore the margin of error could be large, and data could be falsified. Another limit is the sample size of the current study. A larger sample size, garnered either through more students from the particular institution studied or a more diverse sample from multiple liberal arts schools, would make the research stronger. Given more time future research should incorporate longitudinal data, as well as a more in depth analysis of d ata. Conducting more complex tests, such as a Sobel testing, could do this. This might help to clarify some of the relationships between the variables, especially social norms and sense of
!"#$" % 17 % belonging both affecting substance use. Future studies might als o incorporate more questions on individual substance use, as well as include questions on values and/or attitudes about substance use on campus. These questions might help explain the relationship between injunctive norms, individual substance use, and a high sense of belonging. Implications The polarization of substance use, perception of high substance use on campus, and the power of proximal reference groups should be taken into consideration when discussing programming for substance use preventi on. As previous researchers have stated, an alternative to prescrip tive programming and prevention should be considered on college campuses. Administrators should focus on lowering perceived prevalence of substance use through modeling behavior, and chan ging perception of substance use. Campuses might have better luck in changing perceptions at the global level, though attempts should also be made at the proximal level. Encouraging more small group social events, in which substance use is not allowed, or the main focus, could help in lowering the perception of substance use by close friends. More importantly, encouraging small group interactions could affect perceptions of approval ratings, and weaken the relationship between close friend approval and individual substance use. P rograms more closely associated with creating alternate saf e spaces on campus, and creating alternatives to dominant norms could also counteract the positive relationship between sense of belonging and individual substance use f ound on this campus. By providing a physical space where attention is not focused
!"#$" % 1& % on substances or activities and behaviors that are often found in tandem with substance use, schools can promote alternate places to belong and also discourage substance use The current study also notes that administrators and other supervisory roles have the ability to lower perception of substance use merely by not continuing a discourse concerning high levels substance use on campus. A move away from this discourse pl aces less value and importance on substance use generally, and might allow students to realize the insignificance or unimportance of engaging in substance use and its relationship to their involvement in the campus environment.
! "# Appendix A Survey 1. Please specify your age: 2. Please specify your gender: 3. What is your current contract number? a. 1 b. 2 c. 3 d. 4 e. 5 f. 6 g. 7 h. 8 + 4. Did you transfer to this institution in the Fall of 2009? a. Yes b. No 5. Where do you currently live ? ( If your current living situation has been for one month or less, please mark both your current living situation and the living situation previous) a. B Dorm b. Dort c. Goldstein d. Pei first court e. Pei second court f. Pei third court g. V Dorm h. W Dorm i. X Dorm j. Y Dorm k. Z Dorm l. Parent/guardian's home m. Other Off Campus Housing 6. With whom do you currently live? ( Mark all that apply) a. alone
! "$ b. roommate(s) or housemate(s) c. spouse d. parent(s) or other relative(s) e. significant other f. children 7. Based on your time at this institutio n, which of the following statements best describes how you felt overall about going to school ? a. You loved going to school b. You liked going to school a lot c. You kind of liked going to school d. You didn't like going to school very much e. You hated goin g to school 8. To w hat extent do you agree with the following statements? a. I feel valued as a person on this campus b. I have a responsibility to contribute to the well being of other students c. My campus encourages me to help others in need d. I abide by the university policy and regulations that concern alcohol and other drug use. Answers: strongly agree, agree, neither agree or disagree, disagree, strongly disagree 9. Within the last year to what extent have you participated in any of the following activ ities mark one for each line a. C lub sports b. Religious and interfaith groups c. Minority and ethnic organizations d. Political and social action groups e. Community Service f. Music and other visual/performing arts groups g. Student newspaper, radio, TV, magazine, etc. h. Ot her: _________________ Answers: not involved, attended, active involvement non leader, leadership position not available 10. Do you feel safe on this campus? a. Yes b. No 11. Does the social atmosphere on this campus promote alcohol use?
! "" a. Yes b. No 12. Doe s the social atmosphere promote drug use? a. Yes b. No 13. How many of the students at your school do you think engage in drug use ? a. None of them b. A few of them c. Some of them d. Most of them e. All of them f. Don' t Know 14. How many of your close friends a t your school do you think engage in drug use? a. None of them b. A few of them c. Some of them d. Most of them e. All of them f. Don't Know 15. How many of the students at your school do you think drink alcoholic beverages? a. None of them b. A few of them c. Some of them d. Most of them e. All of them f. Don't know 16. How many o f your close friends at school do you think drink alcoholic beverages? a. None of them b. A few of them c. Some of them d. Most of them e. All of them
! "% f. Don't Know 17. How do you think your close friend feel s (or would feel) about you ( mark one for each line) a. smoking marijuana once or twice b. smoking marijuana occasionally c. smoking marijuana regularly d. Trying LSD once or twice e. Trying LSD regularly f. Trying amphetamines once or twice g. Taking amphetamines regularly h. Having one or two drinks of an alcoholic beverage (beer, wine, liquor) nearly every day i. Having four or five drinks nearly every day j. Having five or more drinks in one sitting Answers: strongly approve somewhat approve neither ap prove nor disapprove somewhat d isapprove, strongly disapprove don't know 18. At this particular institution drinking is a central part in the social life of the following groups: ( answer yes or no ) a. M ale students b. F emale students c. A ll students d. F aculty/sta ff e. A lumni f. Religious clubs g. Sports clubs h. Academic groups or cliques i. other (please specify): 19. Please specify the age when you first used ( specify for each option. If you have not used one or any of the following write N/A) a. Alcohol (beer, wine, liquor) b. Marijuana (pot, hash, hash oil) c. Amphetamines (diet pills, speed adderall )
! "& d. Hallucinogens (LSD, PCP psilocybin mushrooms ) e. Designer Drugs (ecstasy, MDMA) f. Other substance ( please specify drug, as well as age): 20. During the past month to what ex tent have you engaged in any of the following behaviors? ( mark all that apply) a. refused an offer of alcohol or other drugs b. bragged about your alcohol or other drug use c. heard someone else brag about his/her alcohol or other drug use d. held a drink to have p eople stop bothering you about why you were not drinking Answers: zero time, one time, two times, 3 5 tim es, 6 9 times, 10 or more times 21. How often did you drink during the last month ? ( mark one) a. I did not drink at all. b. About once a month. c. Two to three times a month. d. Once or twice a week. e. Three to four times a week. f. Nearly every day. g. Once a day or more. 22. How often did you engage in drug use during the last month ? ( mark one) a. I did not use drugs at all. b. About onc e a month. c. Two to three times a month. d. Once or twice a week. e. Three to four times a week. f. Nearly every day. g. Once a day or more. 23. Think back over the last two weeks. How many times have you had five or more drinks at a sitting? ( mark one ). None Once Twice 3 5 times 6 9 times 10 or more times
! "' 24. H as your alcohol use changed since your attendance at this school ? a. Yes (if yes, answer 1.b) b. N o 1.b a. In creased b. D ecreased 25. Has your drug use changed since your attendance a t this school ? a. Yes ( if yes, answer 2.b) b. No 2.b a. Increased b. D ecreased
47 Appendix B Demographic Information Table One: Summary Statistics for Demographic Information Table Two: Summary Statistics for Living Situation Respondent number is higher then in the normal population becaus e respondents were able to choose more than one answer, if they had lived in their current location for one month or less. Age Contract Number Transfer Stud. Gender Valid 300 300 300 300 N Missing 0 0 0 5 Mean 20.05 4.65 0.07 Std. Deviation 1.369 2.367 0.261 Vari ance 1.874 5.6 0.068 Minimum 18 1 0 Maximum 25 8 1 Residence House Mates Valid 310* 326* N Missing 0 1
48 Table Three : Residence of Respondent Frequency Percent Valid Percent Cumulative Percent B Dorm 10 3.2 3.2 3.2 Dort 29 9.4 9.4 12.6 Goldstein 27 8.7 8.7 21.3 Pei Total 88 28.5 28.4 49.8 1st Court 33 10.6 10.6 2nd Court 28 9 9 3rd Court 27 8.7 8.7 V Dorm 6 1.9 1.9 51.7 W Dorm 14 4.5 4.5 56.2 X Dorm 11 3.5 3.5 59.7 Y Dorm 15 4.8 4.8 64.5 Z D orm 36 11.6 11.6 76.1 Parent/ Guardian's Home 9 2.9 2.9 79 Other Off Campus Housing 65 21 21 100 Total* 310 100 99.8
49 Table Four : Housemate of Respondent Frequency Percent Valid Percent Cumulative Percent Missing 1 0.3 0.31 0.3 Alone 33 10.1 10.1 10.4 Roommate(s) or Housemate(s) 248 76.1 76.1 86.5 Spouse 0 0 0 Parent(s) or other relative(s) 14 4.3 4.3 90.8 Significant Other 28 8.6 8.6 99.4 Children 1 0.3 0.31 99.7 Other 1 0.3 0.31 100 Total* 326 100 100.3
50 App endix B Reliabi lity Tests for Scaled Variables Table Five : Reliability for Responsibility on Campus Cronbach's Alpha Cronbach's Alpha Based on Standardized Items N of Items 0.738 0.739 3 Table Six: Reliability for Involvement on Campus Cronbach's Alp ha Cronbach's Alpha Based on Standardized Items N of Items .509 .516 6 Table Seven: Reliability for Feelings friends might have on individual substance use Cronbach's Alpha Cronbach's Alpha Based on Standardized Items N of Items .911 .909 10 Table Ei ght: Reliability for Perceived student use on campu s Cronbach's Alpha Cronbach's Alpha Based on Standardized Items N of Items .618 .629 2
51 Table Nine: Reliability for Perceived friend use on campus Cronbach's Alpha Cronbach's Alpha Based on Standardized Items N of Items .783 .793 2 Table Ten: Reliability for Social atmosphere promote substance use Cronbach's Alpha Cronbach's Alpha Based on Standardized Items N of Items .788 .788 2 Table Eleven: Reliability for Individual Substance use in the last m onth Cronbach's Alpha Cronbach's Alpha Based on Standardized Items N of Items .557 .577 2 Table Twelve: Reliability for Group Use on Campus Cronbach's Alpha Cronbach's Alpha Based on Standardized Items N of Items .803 .793 5
52 Appendix B In dividual Substance Use Descriptive Statistics Table Thirteen: Summary of Individual Substance Use Individual substance use in last month Drug use in the last month Alcohol Use in the last month Valid 291 291 292 N Missing 9 9 8 Mean 2.7766 2.53 3.03 Medi an 2.5000 1.00 3.00 Std. Deviation 1.44016 1.980 1.437 Variance 2.074 3.919 2.064 Minimum 1.00 1 1 Maximum 7.00 7 7
53 Table Fourteen: Individual substance use in last month Frequency Percent Valid Percent Cumulative Percent 1.00 6 0 20.0 20.6 20.6 1.50 22 7.3 7.6 28.2 2.00 41 13.7 14.1 42.3 2.50 37 12.3 12.7 55.0 3.00 26 8.7 8.9 63.9 3.50 25 8.3 8.6 72.5 4.00 30 10.0 10.3 82.8 4.50 20 6.7 6.9 89.7 5.00 12 4.0 4.1 93.8 5.50 8 2.7 2.7 96.6 6.00 9 3.0 3.1 99.7 7.0 0 1 .3 .3 100.0 Valid Total 291 97.0 100.0 Missing System 9 3.0 Total 300 100.0
54 Table Fifteen: Individual Drug use in the last month Frequency Percent Valid Percent Cumulative Percent 1 151 50.3 51.9 51.9 2 29 9.7 10.0 61.9 3 30 10. 0 10.3 72.2 4 26 8.7 8.9 81.1 5 17 5.7 5.8 86.9 6 19 6.3 6.5 93.5 7 19 6.3 6.5 100.0 Valid Total 291 97.0 100.0 Missing System 9 3.0 Total 300 100.0
55 Table Sixteen: Individual alcohol use in the last month Frequency Percent Valid Pe rcent Cumulative Percent 1 66 22.0 22.6 22.6 2 32 10.7 11.0 33.6 3 75 25.0 25.7 59.2 4 75 25.0 25.7 84.9 5 36 12.0 12.3 97.3 6 6 2.0 2.1 99.3 7 2 .7 .7 100.0 Valid Total 292 97.3 100.0 Missing System 8 2.7 Total 300 100.0 Table Sev enteen: Alcohol Consumption changes since attendance at school Frequency Percent Valid Percent Cumulative Percent Increased 147 49 51.2 51.2 Decreased 41 13.7 14.3 65.5 Stayed Consistent 62 20.7 21.6 87.1 Abstained 37 12.3 12.9 100 Valid Total 287 95.7 100 Missing System 13 4.3 Total 300 100
56 Table Eighteen: Drug Consumption changes since attendance at school Frequency Percent Valid Percent Cumulative Percent Increased 122 40.7 42.4 42.4 Decreased 34 11.3 11.8 54.2 Stayed Consistent 49 16.3 17 71.2 Abstained 83 27.7 28.8 100 Valid Total 288 96 100 Missing System 12 4 Total 300 100 Table Nineteen: Five Drinks in One Sitting Frequency Percent Valid Percent Cumulative Percent None 188 62.7 65. 5 65.5 Once 58 19.3 20.2 85.7 Twice 24 8 8.4 94.1 3 5 Times 13 4.3 4.5 98.6 6 9 Times 4 1.3 1.4 100 Valid Total 287 95.7 100 Missing System 13 4.3 Total 300 100 No respondents chose the last option (10 or more times). Table Twen ty: Age when respondents first used substances Summary
57 Alcohol (beer, wine, liquor) Marijuana (pot, hash, hash oil) Amphetamins (diet pills, speed, adderall) Hallucinogens (LSD, PCP, psilocybin, mushrooms) Designer Drugs (ecstasy, MDMA) Valid 287 2 88 284 288 288 N Missing 13 12 16 12 12 Mean 14.3 12.43 7.07 9.34 7.75 Std. Deviation 4.859 7.301 8.766 9.036 9.095 Variance 23.611 53.299 76.839 81.646 82.713 Minimum 0 0 0 0 0 Maximum 26 23 21 21 22
58 Appendix B Social Norms (order ed from Descriptive Global and Proximal, Injunctive Global and Proximal) Table Twenty One: Perceived Student Use on Campus Frequency Percent Valid Percent Cumulative Percent 2.00 1 .3 .3 .3 2.50 1 .3 .3 .7 3.00 31 10.3 10.6 11.3 3.50 84 28. 0 28.8 40.1 4.00 162 54.0 55.5 95.5 4.50 12 4.0 4.1 99.7 5.00 1 .3 .3 100.0 Valid Total 292 97.3 100.0 Missing System 8 2.7 Total 300 100.0
59 Table Twenty Two: Perceived Friend Use on Campus Frequency Percent Valid Percent Cu mulative Percent 1.00 3 1.0 1.0 1.0 1.50 11 3.7 3.8 4.8 2.00 16 5.3 5.5 10.3 2.50 13 4.3 4.5 14.7 3.00 34 11.3 11.6 26.4 3.50 41 13.7 14.0 40.4 4.00 76 25.3 26.0 66.4 4.50 63 21.0 21.6 88.0 5.00 35 11.7 12.0 100.0 Valid Total 292 97.3 10 0.0 Missing System 8 2.7 Total 300 100.0
60 Table Twenty Three : Social Atmosphere Promoting Substance Use on Campus Frequency Percent Valid Percent Cumulative Percent .00 29 9.7 9.8 9.8 .50 25 8.3 8.4 18.2 1.00 242 80.7 81.8 100.0 Valid Total 296 98.7 100.0 Missing System 4 1.3 Total 300 100.0
61 Table Twenty Four: Summary Statistics for feeling friends might have on substance use Feelings friends might have on individual substance use S moking marijuana once or twic e S moking marijuana occasional ly Smoking marijuana regularly Trying LSD once or twice Trying LSD regularly Valid 291 293 293 293 288 290 N Missing 9 7 7 7 12 10 Mean 3.1994 2.21 2.34 2.94 2.46 3.51 Median 3.2000 2.00 2.00 3.00 2.00 4.00 Std. Devia tion .86497 1.135 1.167 1.300 1.268 1.265 Variance .748 1.289 1.361 1.691 1.608 1.600 Minimum 1.00 1 1 1 1 1 Maximum 5.00 5 5 5 5 5
62 Table Twenty Four cont: Summary Statistics for feelings friends might have on substance use Trying amph etami nes once or twice Taking Amphetami nes regularly Having one or two drinks nearly every day Having four or five drinks nearly every day Having five or more drinks in one sitting Valid 284 282 291 293 291 N Missing 16 18 9 7 9 Mean 3.30 4.16 3.29 4.40 3.43 Median 3.00 4.00 3.00 5.00 3.00 Std. Deviation 1.197 1.056 1.132 .873 1.177 Variance 1.432 1.114 1.280 .762 1.384 Minimum 1 1 1 1 1 Maximum 5 5 5 5 5
63 Appendix B Sense of Belonging Variables Table Twenty Five: Summary Statisti cs for variables of Responsibility on Campus Help others Wellbeing of Others Feelings of Value Valid 298 297 299 N Missing 2 3 1 Mean 4.0671 4.0269 4.0468 Median 4 4 4 Std. Deviation 0.86585 0.8849 0.83815 Variance 0.75 0.783 0.702 Mini mum 1 1 1 Maximum 5 5 5 Table Twenty Six: Summary Statistics for Responsibility on Campus Valid 296 N Missing 4 Mean 4.0473 Median 4.0000 Std. Deviation .70029 Variance .490 Minimum 1.00 Maximum 5.00
64 Table Twenty Seven: Responsibilit y on Campus Frequency Percent Valid Percent Cumulative Percent 1.00 1 .3 .3 .3 2.33 7 2.3 2.4 2.7 2.67 10 3.3 3.4 6.1 3.00 14 4.7 4.7 10.8 3.33 24 8.0 8.1 18.9 3.67 42 14.0 14.2 33.1 4.00 77 25.7 26.0 59.1 4.33 32 10.7 10.8 69.9 4 .67 41 13.7 13.9 83.8 5.00 48 16.0 16.2 100.0 Valid Total 296 98.7 100.0 Missing System 4 1.3 Total 300 100.0
! "# "# Table Twenty Eight: Pearson Correlations between all variables Felt about school Responsibility on Campus Social Atmosphere promote substance use Involvement on Campus Safety on Campus Group Use on Campus Perceived Student Use on Campus Individual su bstance use in last month Perceived Friend Use on Campus Pearson Correlation .156 ** 0.042 0.061 0.048 Sig. (2 tailed) 0.007 0.475 0.299 0.418 Safety on campus N 292 290 290 289 Involvement on Campus Pearson Correlati on 0.052 .154 ** 0.026 Sig. (2 tailed) 0.377 0.008 0.66 N 293 292 291 Pearson Correlation .375 ** Sig. (2 tailed) 0 Responsibility on Campus N 294 Pearson Correlation 0.068 .121 .291 ** 0.097 0.005 Sig. (2 tailed) 0.259 0.045 0 0.109 0.933 Group Use on Cam pus N 279 277 279 276 275 Pearson Correlation 0.112 .122 .245 ** .118 0.065 .311 ** Perceived Student Use on Campus Sig. (2 tailed) 0.056 0.039 0 0.046 0.272 0 N 290 288 288 287 286 274 Pearson Correlation 0.003 0.04 0.035 0.037 0.027 .141 .298 ** .543 ** Sig. (2 tailed) 0.956 0.497 0.553 0.534 0.653 0.019 0 0 Perceived Friend Use on Campus N 290 288 288 288 286 275 286 285
! "" "" $$!%&''()*+,&-!,.!.,/-,0,1*-+!*+!+2(!3435!)(6()!78 9 +*,)(:;4 $!%&''()*+,&-!,.!.,/-,0,1*-+!*+!+2(!343#!)(6()!78 9 +*,)(:;4 Pearson Correlation 0.02 0.008 Sig. (2 tailed) 0.735 0.894 Social Atmosphere promote substance use N 294 292 Felt about school Responsibility on Campus Social Atmosphere promot e substance use Involvement on Campus Safety on Campus Group Use on Campus Perceived Student Use on Campus Individual substance use in last month Perceived Friend Use on Campus Pearson Correlation 0.005 .126 0.0 44 0.021 0.08 0.053 0.076 Sig. (2 tailed) 0.926 0.032 0.453 0.719 0.178 0.381 0.202 Individual substance use in last month N 289 287 287 286 285 274 283 Pearson Correlation 0.049 0.026 0.016 0.096 0.004 .136 .178 ** .46 1 ** .585 ** Sig. (2 tailed) 0.404 0.659 0.786 0.107 0.95 0.024 0.003 0 0 Feelings friends might have on individual substance use N 290 287 287 286 286 274 283 284 286
! "# Work Cited DDQ R (Daily Drinking Questionnaire Revised) and Standard Drink Conversion. College Health and Safety. Education Development Center, Inc. Modified version of Quantity Frequency Variability Index (Cahalan et al. 1969) Baer, John S., Alan Stacy, a nd Mary Larimer. ( 1991) Biases in the Perception of Drinking Norms among College Students*. Journal of Studies on Alcohol, 52 (6), 580 86. Baer, John. (1994) Effects of College Residence on Perceived Norms for Alcohol Consumption: An Examination of the Fir st Year in College Psychology of Addictive Behaviors, 8 (1), 43 50 Baer, John S. (2002) Student Factors: Understanding Individual Variation in College Drinking Journal of Studies on Alcohol, supp 14 40 53. Battistich, Victor and Hom, Allen. (1997) T he Relationship between Students' Sense of Their School as a Community and Their Involvement in Problem Behaviors. American Journal of Public Health, 87 (12). Bohnert, Amy S. B., Catherine P. Bradshaw, & Carl A. Latkin. (2009). A Social Network Perspectiv e on Heroin and Cocaine use among adults: evidence of bidirectional influences. Addiction, 104, 1210 1218. Buunk, Bram P. and Bakker, Arnold B. (1995) Extradyadic Sex: The Role of Descriptive and Injunctive Norms. The Journal of Sex Research 32 (4), 313 3 18.
! "$ Cialdini, R.B., & Trost, M.R. (1998). Social influence: Social norms, conformity, and compliance. In D. Gilbert, S. Fiske, & G. Lindzey (Eds.) The handbook of social psychology, (4th edition) vol. 2, pp. 151 192. New York: McGraw Hill. Ennett, Susan T., Karl E. Bauman, Andrea Hussong, Robert Faris, Vangie A. Foshie, & Li Cai. (2006). The Peer Context of Adolescent Substance Use: Findings from Social Network Analysis. Journal of research on Adolescence, 16 (2), 159 186. Fendrich, James M. (1967) Perciev ed Reference Group Support: Racial Attitudes and Overt Behavior. American Sociological Review 32 (6), 960 70. Maddock, Jay and Glanz, Karen. (2005) The relationship of proximal normative beliefs and global subjective norms to college students' alcohol co nsumption. Addictive Behaviors, 30 315 23. Napoli, Maria, Marsiglia, Flavio Francisco and Kulis, Stephen. (2003) Sense of Belonging in School as a Protective Factor Against Drug Abuse Among Native American Urban Adolescents. Journal of Social Work Practi ce in the Addictions, 3 (2), 25 41. Rotgers, Frederick, Daniel S Keller, & Jon Morgenstern, eds. (1996) Treating substance abuse: theory and technique. Guilford Substance Abuse Series Guilford, New York. Wechsler Henry Barbara Moeykens Andrea Daven port Sonia Castillo and Jeffery Hansen (1995) The Adverse Impact of Heavy Episodic Drinkers on Other College Students*. Journal of Studies on Alcohol, 56 628 34.
! "% Wechsler, Henry and Toben, Nelson E. (2008) What we have Learned from the Harvard Schoo l of Public Health College Alcohol Study: Focusing Attention on College Student Alcohol Consumption and the Environmental Conditions that Promote it. Journal of Studies on Alcohol and Drugs, 69, 481 490. Westermeyer, Joseph, Thuras, Paul, and Waaijer, Ann et (2004) Size and Complexity of Social Networks among Substance Abusers: Childhood and Current Correlates The American Journal on Addictions, 13 372 380.