paper 34 THE AIMS AND OBJECTIVES OF THE MONITORING THE FUTURE STUDY AND PROGRESS TOWARD FULFILLING THEM Second Edition Lloyd D. Johnston Patrick M. O’Malley John Schulenberg Jerald G. Bachman THE AIMS AND OBJECTIVES OF THE MONITORING THE FUTURE STUDY AND PROGRESS TOWARD FULFILLING THEM Monitoring the Future Occasional Paper 34 Second Edition Lloyd D. Johnston Patrick M. O’Malley John Schulenberg Jerald G. Bachman Institute for Social Research University of Michigan Ann Arbor, Michigan 1996 TABLE OF CONTENTS INTRODUCTION.........................................................................................................................1 Specific Aims of the Project................................................................................................1 Overview of the Ten Objectives..........................................................................................1 Part 1: THEORETICAL BACKGROUND AND CONCEPTUAL FRAMEWORK.............5 Domain of Variables and General Theoretical Grounding.................................................7 The Basic Conceptual Framework......................................................................................8 The Immediate Social Environment..................................................................................10 Person Characteristics.......................................................................................................15 Attitudes and Beliefs about Drugs.........................................................................18 Comparison with another Theoretical Approach..............................................................20 Indexes vs. Individual Variables.......................................................................................20 Part 2: AN ELABORATION OF THE OBJECTIVES, THEIR RATIONALE, AND A SYNOPSIS OF PROGRESS TO DATE...................................................................................22 Objective 1 Monitoring drug use and potential explanators in grades 8, 10, and 12; including....................................................................................................22 Objective 1a Self-reported use............................................................................22 Objective 1b Patterns of multiple drug use.........................................................23 Objective 1c Beliefs about the harmfulness of drugs.........................................28 Objective 1d Personal disapproval of drugs.......................................................30 Objective 1e Perceived disapproval by peers.....................................................31 Objective 1f Stereotypes of cigarette smokers and marijuana users..................32 Objective 1g Friends use and exposure to use....................................................34 Objective 1h Perceived availability....................................................................36 Objective 1i Contexts in which drugs are used..................................................37 Objective 1j Reason for use, abstention, and discontinuation...........................38 Objective 2 Monitoring drug use and potential explanators among young adults (ages 19-32), including college students...................................................39 Objective 3 Monitoring and studying use at ages 35 and 40........................................42 Objective 4 Distinguishing period, age, and cohort effects..........................................45 Objective 5 Explaining secular trends and cohort differences......................................47 Objective 6 Studying the natural history of drug use and explaining age-effects........50 Objective 6a Individual characteristics and social environment during secondary school...................................................................................................58 Objective 6b Changes in social environments and social roles after high school63 Objective 7 Assessing the consequences of drug use...................................................67 Objective 8 Studying frequent or heavy-users..............................................................72 Objective 9 Studying various under-represented groups..............................................77 TABLE OF CONTENTS (continued) Objective 10 Methodological, policy, and other contributions to the field....................79 Objective 10a Methodological contributions........................................................79 Stability and Reliability.................................................................79 Validity..........................................................................................81 Representativeness........................................................................82 Objective 10b Providing measures and national norms for others.......................82 Objective 10c Providing a data resource for others..............................................84 Objective 10d Providing the potential for additional policy-relevant research....86 Evaluating Natural Experiments...............................................................86 Evaluating Planned Experiments...............................................................87 Extended Study of Select Subgroups........................................................87 Monitoring Additional Populations...........................................................88 Other Research Potentials.........................................................................88 Objective 10e Providing measurement of the achievement of national goals......89 In Closing.....................................................................................................................................89 References....................................................................................................................................91 INTRODUCTION Specific Aims of the Project Monitoring the Future is an ongoing program of research designed to accomplish three broad and complementary aims. (1) To monitor drug use and potential explanatory factors among American secondary school students, college students, other young adults, and selected age groups beyond young adulthood (detailed under Objectives 1 through 3, below). (2) To distinguish which of three fundamentally different kinds of change— maturational, historical, and/or cohort—are occurring for various types of drug use, including the use of tobacco and alcohol (Objective 4 below). (3) To study the causes, the consequences, and (where relevant) the maturational patterns associated with these different types of change in drug use (detailed under Objectives 5 through 9, below). A number of additional methodological, policy, and other benefits are listed under Objective 10. In pursuit of these three interrelated aims, the project employs a cohort-sequential research design consisting of (a) annual cross-sectional surveys of large, nationally representative samples of high school seniors (beginning with the class of 1975) and 8th and 10th graders (beginning in 1991), and (b) follow-up surveys of each senior class annually through age 32, and at half-decade intervals thereafter. Additionally, follow-up surveys of panels drawn from those who participated in eighth and tenth grade in the years 1991, 1992, and 1993 have been carried out, in part for the purpose of generating good national samples of high school dropouts. At present, some 50,000 secondary school students located in approximately 435 secondary schools are surveyed annually (for example, the 1995 survey had 16,000 seniors in 144 schools, 17,000 tenth graders in 139 schools, and 18,000 eighth graders in 152 schools). In addition, a follow-up survey of previously graduating classes yields a young adult sample of about 11,000 respondents in the age range of 19 to 32, which includes about 1500 young adults one to four years past high school who are actively enrolled in a two- or four-year college and comprise the college student sample.1 As the panels have become older, a follow-up at age 35 has been introduced, and we hope to follow these cohorts at five year intervals thereafter. The first of the broad aims—monitoring drug use and related factors in order to provide social indicators of historical change, as well as to explain those changes—clearly implies an 1For an in-depth description of the study’s research design, the reader should consult Bachman, J. G., Johnston, L. D., O’Malley, P. M., & Schulenberg, J. E. (1996). Monitoring the Future project after twenty-two years: Design and procedures. Monitoring the Future Occasional Paper No. 38. Ann Arbor, MI: Institute for Social Research. A less detailed description can be found in Johnston, L. D., O’Malley, P. M., & Bachman, J. G. (1995). National survey results on drug use from the Monitoring the Future study, 1975-1994. NIH Publication No. 95-4026. Rockville, MD: National Institute on Drug Abuse. 1 ongoing effort. The same is true for the second broad aim of distinguishing the three types of change—period, age, and cohort. Although it may be less obvious, the third broad aim— exploring causes, consequences, and maturational patterns—also is an ongoing effort, involving analyses at both individual and aggregate levels. At the aggregate level we continue to document the emergence of new secular (and recently cohort) changes, as well as the emergence of new drugs. At the individual level of analysis, the process is ongoing in part because new developmental stages are being added to the study, and in part because the underlying relational patterns are themselves subject to change (e.g., see Bachman, Johnston, & O’Malley, 1981, 1989; Bachman, O’Malley, & Johnston, 1986; Johnston, O’Malley, & Bachman, 1995). Further, new policies and programs are constantly being introduced (e.g., decriminalization, raising the drinking age, the parent group movement, the Partnership for a Drug Free America advertising program), which means that having this “social observatory” in place permits a more timely, and often a prospective, assessment of their impacts. We believe that the multiple aims of the study are interconnected and mutually facilitating. Thus, a substantial efficiency results in addressing them in a single project rather than a number of separate ones, and in some important ways, the total product is greater than would be possible under a more fractionated approach. Overview of the Ten Objectives To provide an overview, the ten specific objectives of the study are listed below without commentary. Then, after a section on the theoretical background and conceptual framework which gave rise to the many variables included in the study to address these objectives, each of the ten objectives will be discussed separately. Each will be explained more fully, its logic and rationale presented, relevant literature will be cited, and the study’s progress will be briefly summarized. Objective 1: To continue monitoring a broad range of drug-related behaviors, as well as explanators of change—including attitudes, beliefs, and perceptions—across nationally representative samples of eighth, tenth, and twelfth grade students. This annual monitoring began in 1975 for twelfth grade students, and in 1991 for eighth and tenth grade students. The specific characteristics to be monitored are detailed under the ten sub-objectives below. a. Self-reported use of over 30 classes and subclasses of drugs, both licit and illicit. b. Patterns of multiple drug use, both concurrent and non-concurrent. c. Beliefs about the harmfulness of various types of drugs at various levels of use. d. Personal disapproval of various types of drugs at various levels of use. e. Perceptions of the extent of disapproval by peers of the use of various drugs (i.e., perceived peer norms). f. Beliefs (or stereotypes) regarding cigarette smokers and frequent marijuana users. g. Extent of direct exposure to various forms of drug use, and proportions of friends using various drugs. h. Perceived availability of the various illicit drugs. i. Contexts in which drugs are used (when, where, and with whom). 2 j. Reasons young people give for their use of the various drugs, for abstention, and for discontinuation. Objective 2: To continue to monitor and study these same drug-using behaviors and potential explanatory variables among nationally representative samples of young adult high school graduates (modal ages 19 through 32), including that critical segment which constitutes a nationally representative sample of American college students. Objective 3: To monitor and study longer-term patterns and consequences of drug use beyond young adulthood (a) by continuing to conduct follow-up surveys at age 35 and (b) by initiating follow-up surveys at age 40. Objective 4: To attempt to distinguish among three basic types of change in drug use and related factors at the aggregate level: age, period, and cohort.2 Objective 5: To attempt to explain, at the aggregate level of analysis, secular trends and lasting cohort differences in drug use, emphasizing changes in cultural influences, attitudes, beliefs, and value orientations as possible explanators. Objective 6: To examine at the individual level of analysis the natural history of drug use and related factors from early adolescence through middle adulthood, and to attempt to explain any “age effects” observed in the initiation, maintenance, and cessation of drug use. a. To assess during the secondary school years the impact of individual characteristics (e.g., values, beliefs, lifestyles, and other behaviors such as delinquency, school performance, and religiosity) and social environments (e.g., part-time work, sports and other extra-curricular activities, activities outside of school, peer groups) on drug use and related factors, with particular emphasis on the specification of risk and protective factors. b. To assess during the post-high school years the impact of individual characteristics, changes in major social environments (college, military service, civilian employment, homemaking, unemployment), as well as changes in roles (marriage, pregnancy, parenthood, divorce) on drug use and related variables. Objective 7: To assess both the short and longer-term consequences of various types of drug use—particularly heavy use—on a number of outcomes in the domains of physical health, psychological well-being, status attainment, role performance, driving performance, deviant behavior, and social alienation. Objective 8: To give special emphasis throughout to the more frequent or heavier users of the different drugs, i.e., individuals most likely to be characterized as “abusers.” (This objective crosscuts most of those above.) 2Age effects are consistent changes with age observed across different birth cohorts (or in this case, across graduating class cohorts). Period effects are consistent changes over an historical period observed across various age groups. Cohort effects are enduring differences among cohorts compared at equivalent ages. 3 Objective 9: To continue to study drug use and drug-related attitudes and beliefs among a number of subgroups which historically have been under-represented in drug abuse research. These include women, ethnic minorities, and young adults who do not attend college, as well as those in military service, civilian employment, or homemaking after high school. (This objective also crosscuts those above.) Objective 10: To continue to make methodological, substantive, and policy-relevant contributions to the larger fields of social, behavioral, educational, and medical research dealing with drugs and/or youth. a. To refine methodologies for the analysis and interpretation of self-report measures of drug use, including documenting the reliability and validity of such measures. b. To continue to provide measures for, and to stimulate the comparability of measurement in, drug research at the local, state, national, and international levels; and to provide national norms for comparison. c. To continue to provide a national data resource for use by other investigators by placing data into a national data archive, publishing comprehensive annual volumes of descriptive results, and providing analytic support to facilitate other investigators’ access to the study’s panel data. d. To continue to develop a potential for research of policy importance, particularly the evaluation of “natural experiments” which can build upon the main study with great economies in cost and time. Two past examples are the evaluation of laws decriminalizing the use of marijuana at the state level, and the evaluation of changes in the drinking-age laws. e. To continue to provide measures of progress toward the accomplishment of national goals relevant to drug use and/or youth more generally. These include the National Education Goals, the National Health Objectives for the Year 2000, the President’s National Drug Control Strategy goals, and the President’s new goals for reducing teen smoking. Even this comprehensive listing of objectives does not provide a complete enumeration of the questions and hypotheses which the study can be used to address, or has addressed already. These are spelled out in greater detail below under a separate discussion of each objective. 4 Part 1 THEORETICAL BACKGROUND AND CONCEPTUAL FRAMEWORK In 1980, Lettieri, Sayers, and Pearson edited a volume presenting fully 43 different theoretical perspectives on drug use, and since then there has continued to be a great deal of development of new and existing theoretical approaches (for example, Akers & Cochran, 1985; Baumrind & Moselle, 1985; Brook, Brook, Gordon, Whiteman, & Cohen, 1990; Donovan, Jessor, & Costa, 1991; Elliott, Huizinga, & Ageton, 1985; Elliott, Huizinga, & Menard, 1989; Flay & Petraitis, 1994; Hawkins, Lishner, Catalano, & Howard, 1986; Johnston, 1991; Kaplan, 1985; Kumpfer & Turner, 1990-1991; Marcos, Bahr, & Johnson, 1986; Massey & Krohn, 1986; Newcomb & Bentler, 1988; Oetting & Beauvais, 1987; Rhodes & Jason, 1990; Tarter & Mezzich, 1992; Zucker, 1989).3 The various perspectives run a wide gamut in terms of (a) the classes of substances encompassed (e.g., heroin only, alcohol only, any illicit substance, etc.), (b) the stage of involvement being explained (e.g., initiation, continuation, transition to “abuse,” cessation, or relapse), (c) the classes of determinants under consideration (e.g., social, psychological, physiological, etc.), (d) the more general theory of human behavior, if any, upon which the specific drug use theory is grounded (e.g., psychoanalytic, behavioral, social field, social learning, etc.) and (e) the level at which the phenomenon to be explained is measured (e.g., levels of use at the individual, institutional, or societal level). Yet in spite of—or perhaps in part because of—this breadth and diversity of theorizing in the drug field, the research in this area has at times been described as mostly atheoretical (Kandel, 1980; Radosevich, Lanza- Kaduce, Akers, & Krohn, 1980). This is not actually so surprising, considering that theories in this area—particularly social-psychological theories—are still very much “in process.” Although there have been some efforts made toward developing fairly general theories (Kaplan, et al., 1982, 1986, 1987; and Jessor & Jessor, 1977, to cite just two examples), most of these theoretical perspectives are still rather limited in scope, and are what Merton called “theories of the middle range, theories intermediate to the minor working hypotheses evolved in abundance during day-to-day routines of research, and the all-inclusive speculations comprising a master conceptual scheme.” Such theories consist of a “general orientation toward data, suggesting types of variables which need somehow to be taken into account, rather than clear, verifiable statements of relationships between specified variables” (Merton, 1957, pp. 5,9). Writing much later, Petraitis, Flay, and Miller (1995) commented that “... as social scientists we might be aware of many (if not most) of the constructs that contribute to [adolescent substance use], but we do not yet know how all these constructs ... fit together.” (p. 67) We certainly regard the theoretical approach which has guided much of the development of our work to be middle-range; and it is eclectic, since we did not feel that any single extant theoretical approach was sufficiently comprehensive and specified to serve as the sole basis for selecting measurement and guiding analysis in this multipurpose study. Our theorizing provided a general approach for conceptualizing the data, generating hypotheses, and organizing many analyses. There are quite a number of hypotheses implied in our conceptual framework, 3 The word “theory” is used rather loosely here, as is common practice in this field; we take a broad-based view of the term and recognize that its meaning can range from conceptual frameworks to elaborated and specific systems of testable hypotheses. 5 described below; but there are a great many more we could posit. Thus our approach to theory, like most others mentioned here, has really been an evolving process in which further elaboration and specification have occurred and are to be expected in the future. It is an iterative process, in that the theoretical framework leads to some empirical tests, which in turn yield some revisions and/or elaborations of the framework. This process is consistent with what Cattell (1966) described as the “inductive-hypothetico-deductive spiral.” In the present study a theoretical structure has been evolving to provide explanation of trends in drug use at the societal or aggregate level. Building upon a set of measures of perceived risk, disapproval, and peer norms, which were included in the study from the start (Johnston and Bachman, 1980), we have tested our hypothesis that these are important determinants of changes in drug use. Early in the decline phase for marijuana, which began after 1979, we described changes in perceived risk as a likely determinant because of its correlated upturn (Johnston, 1982; Johnston, Bachman and O’Malley, 1981). More evidence was presented from the reasons quitters and abstainers gave for their not using marijuana and in the trends in the frequency with which they gave these reasons (Johnston, 1982, 1985). The hypothesis was also advanced at that time that changes in perceived risk may be helping to drive changes in disapproval of use (and, derivatively, in peer norms). Trends in both marijuana use and those attitudes and beliefs continued to evolve in the predicted way throughout the 1980s, giving further evidence to the theoretical position. Further, evidence on one major competing hypothesis—that changes in availability were driving the downturn—did not receive empirical support (Johnston, O’Malley, and Bachman 1989). Another alternative hypothesis was offered by Jessor (1985)—that a shift in young people toward a more conservative lifestyle could have caused the downturn—was addressed and found inadequate (Bachman, Johnston, O’Malley, and Humphrey, 1988). The latter article also demonstrated that if one looked separately at trends across high school classes within each level of perceived risk, no downturn in use occurred, but rather some increase. (A subsequent article dealing with the same issues with regard to the downturn in cocaine use leads to much the same conclusions (Bachman, Johnston, & O’Malley, 1990).) More recently, as marijuana use has again risen, changes in both the perceived risk and disapproval proved to be leading indicators of the turnaround in use (Johnston, O’Malley and Bachman, 1995). Based on our original set of hypotheses, the confirmation of a number of those over a number of years, and other correlated evidence, we have offered a first statement of a fairly comprehensive theory of the spread, maintenance, and contraction of drug epidemics (Johnston, 1991). Subsequently, the country entered what we have characterized as a “relapse phase,” for which further theoretical formulations are in development for an expansion of the theory. See Johnston, O’Malley, and Bachman, 1995 for a discussion of the relevancy of “generational replacement” and “generational forgetting,” as well as other factors at the societal level. But, at the individual level of analysis and explanation, there is not as yet a comprehensive theory to deal with all aspects of drug use. We are not unduly discomforted by this fact; nor are we especially uneasy about taking a somewhat eclectic (though certainly not arbitrary) approach to the development of our measurement and analysis. Almost three decades ago, Dudley Duncan, in discussing next steps in social reporting, wrote at some length about the “theorist” versus the “inductivist.” His comments are sufficiently fundamental to theory in the drug area, and particularly to our own approach, that we quote them here. 6