Our own study as reported here focuses on drug-using conduct and on some social and psychological correlates among students in five Western colleges or universities. There is also information on drug use among students in four California high schools. Our data on students include information on bad outcomes; this means ill effects reported by students and bad outcomes as identified by official school case finding—the records of school administrative and health authorities. This book also includes an assessment of the psychiatric aspects associated with bad outcomes, an appraisal for school administrators of the implications of drug use, a commentary on the relationship between drug use and one set of predominant religious values, and an overview of the findings from ours and other on-going or recently completed student and youth studies.

The methods we employed were several. The study, which we later called Utopiates (Blum and Associates, 1964), really began in 1962 with our observations on LSD users, some of whom were faculty members, teaching assistants, research workers experimenting on students, and graduate students. These observations were completed in 1964 but began again in 1965 through 1966 as we prepared the ground for the college survey. The method of preparation was to be with student drug mers so as to get to know them well. More formally, this is called "participant observation." Much of that work was done by Funkhouser-Balbaky (1966). A second simultaneous set of observations, extending from 1962 to the present, were those of the intensive clinical case study. Some of these observations appear in Utopiates, a few are presented in Chapter Fifteen, but others unreported constituted a source for the questions asked during the surveys and for the speculations advanced as we pondered upon what we had found.

Once the interview schedule for the survey work itself was constructed—a process of pretesting which took over a year—the business of its administration was relatively straightforward (a statement which takes no notice of the difficulties involved in running a survey in five different colleges). In each of the schools, a sample of undergraduates was randomly drawn from the registrar's lists of matriculated (day) students. In one school a small graduate sample was also drawn. interviewers were trained in the use of the schedule, including its pre-coded answer-recording sectors. In two schools, following the administration of the interview, a reliability check was conducted.

All the results of the interviewing phase represent the reports of students about themselves. Although we presume that skillful interviewing reduced the likelihood either of exaggeration or denial of illicit-drug use per se, we must assume that some such distortion occurred. How much we cannot estimate. The reader will see that our data conform to the range of reports from other schools as cited in Chapter One and to subsidiary samples drawn at the same time on one campus (see Chapter Eleven). All this may imply is that the errors of underreporting and overreporting remain fairly constant across campuses. We do have some reasonably heartening evidence, cited in Chapter Sixteen, to the effect that what students say they will do in regard to drug use in general is, in fact, what they will do when given an opportunity to use illicit drugs in particular, r = .78.

Another method employed involved the selection of special samples as used in the students described in Chapter Eleven. Chapters Twelve through Fourteen report on groups of students with intense interests in illicit-exotic drugs, left-wing politics, right-wing politics, and religion who were identified on the basis of membership in organizations, visible involvement in issues, and nomination by faculty, college ministers, and peers. These students were interviewed and given psychological tests. In Chapter Sixteen, two fraternity houses constituted the sample and the endeavor was to predict individual drug use—specifically "turning on" or being initiated into illicit-exotic use—over the course of a year. The methods employed here were rating scales of experience and attitudes, a brief questionnaire, and sociometric devices. A fourth method employed, reported in Chapter Ten, was the panel. Panelists in two schools comprised randomly drawn subsamples selected from the initial, larger random sample. In a third school—this time intentionally—sampling differed and was based upon the selection of identified users of illicit-exotic drugs. The device used in the panel study was a weekly and monthly drug-use diary kept on a daily basis by students and collected from them monthly. A fifth method employed was use of psychiatric interviews and observation, as reported in Chapter Seventeen. A sixth device, though hardly a method in any elaborate sense, is found in Chapter Nine and consisted of gathering case-finding data from the official—and Unofficial but recorded—records of institutions, school administrators, school health services, and local and state police agencies. To varying degrees all of the foregoing devices for identifying and selecting students and for assessing their conduct were processed and elaborated by means of coding, quantification, computer processing, and, when deemed appropriate, the application of statistical tests.

Our plan called for a sample of 300 undergraduates to be drawn randomly only from registrars' lists in each of four schools and a sample of 200 from the junior college in our study. The plan also called for professionals in each school to be in charge of sampling, interviewer training, and interviewer supervision, with all of them using the same interview schedule and receiving the same instructions in sampling, interviewing, and coding procedures. The plan, however, did not work out quite that way. In School I, 2 out of the sample of 300 refused to participate and were replaced by further random drawing. In School II, a sample of 275 was drawn out of which there were five refusals, so that the final sample was 270. In School III, local personnel were unable to perform the work, so that interviewers from School I did so instead. There, a sample of 207 was drawn; because there were six refusals, the final sample was 201. School IV drew an initial sample of 300 but failed adequately to supervise the work so that only 192 were interviewed; a second sample of 300 was drawn, but work was not completed in time so that a final sample of 250, heavily weighted with upperclassmen, was the outcome. School V, because of local needs, required a sample of graduate students as well; 75 graduate students and 218 undergraduates comprised the final selection. These 218 were drawn from a list of 251. There were thirty-three refusals.

In each school all information was collected from students who were personally contacted and privately interviewed. Students were assured of anonymity if they were concerned, and they were shown the coding procedure designed to guarantee that anonymity. Interviews were conducted over a period of three to six months in each school; all work was done in the 1966 to 1967 academic year. The interview schedule, which standardized question format, question order, and recording and coding had been developed and pretested on several campuses over the prior eighteen-month period. The average interview time was one hour; it was rarely less than forty-five minutes, rarely more than ninety minutes.

The accuracy of the information collected depended upon what the student was able to remember of his own lifetime drug use and what he was willing to report to the interviewer. There was only one student who accepted the interview and then refused to answer some of the questions. We have seen that the rate of refusal varied from less than 1 per cent (the private university) to 34 per cent (the state college) . During interviews the interviewer could raise additional queries if a student appeared to be giving inconsistent or unclear information about his drug-use history; however, there was no way for the interviewer to know whether the student was lying. It was the impression of the interviewers—all of whom were either graduate students themselves or older persons who enjoyed and were sympathetic to students—that most students were interested in the interview discussions and were open about their own views and experiences with mind-altering drugs.

The 39 inquiry areas and 215 questions or index items which constituted the interview were derived either from the need for specific descriptive data about the students and their history of drug use or from expectations that particular characteristics of students would be associated with varying styles of drug use—that is, that there would be regular psychological relationships. We do not claim that these expected relationships, ones which emerged as expected differences among students in their drug use as a function of some other personal or social characteristics, verify anything as elegant as "hypotheses." Rather, these relationships, most of which were found to operate as anticipated, are consistent with what any thoughtful observer would expect. The fact that drug use does turn out to be reasonably predictable simply confirms that people—their experiences, personalities, and actions—are not composed of unjointed and disparate parts but rather that they are of a piece, articulated, coordinated, or otherwise a systematic whole.

In presenting these findings, we have been faced with a typical problem of our technological age, that of computer output amounting to many thousands of tables. We have taken it as our task to identify those findings which were linked to our expectations or which were otherwise meaningful and to translate them into expository prose. We have not used the shotgun or random method of discussing—or even paying attention to—much of the inevitable computer output that was unrelated to our general expectations. We have tried to avoid tables, for the sake of readability as well as of publishing costs. Any reader who wishes to have more details on numerical distribution is invited to write to us Finally, we have strongly limited the use of statistical tests. We feel that, for the most part, interview data which consist of hundreds of interrelated subitems interpreted for the most part in terms of trends and major links are not to be gilded with statistical testing. The latter are procedures we prefer to apply to the rigorous test of hypotheses under controlled experimental conditions. Such statistical tests as we apply are demonstrative or, occasionally, helpful to inference.

Repeat interviews were conducted in two schools on a small sample of twenty-two students (ten on one campus, twelve on the other) randomly drawn from the basic sample. The second interview was conducted by a different interviewer. The repeat interviews took place from one to five months after the initial interview and were designed to test the reliability of the information gained. One set of interviews was conducted in the school where interviewer supervision had been poorest, although the interviewers there had also been trained and were considered competent. The other school was one where interviewer supervision was considered satisfactory.

The 215 replies or indices were coded for computer analysis. That meant for the ten students reinterviewed in one school 2,150 agreements were possible between the original and the repeat interview; for the twelve students in the other school, there were 2,580 opportunities for agreement or disagreement. The actual rate of agreement in the first school was 86.4 per cent and in the second 84.4 per cent. The agreement between the two rates of consistency was such as to indicate that our assessment of poorer interview supervision in the second school was not actually related to any important difference in what the interviews yielded. A check was made of the items which were the basis for the greatest inconsistency between the initial interview and the repeat interview. There were twenty-five items where more than seven of the twenty-two students gave a different reply in each of the two interviews. Two items were shifts within one question (q. 12) which required the ranking of several potential activities as to their current importance to the student. Here, his political and "other" activities changed rank over time. Another item (q. 17) was related for it inquired about the direction of the student's political beliefs and whether they were, in fact, shifting. The implication -here is that more than 30 per cent of those who made up the reliability sample experienced some shift in the importance and direction of their political activities during the school year. In the lifetime drug-history sector, inconsistency showed up in replies to inquiries about events associated with stopping or decreasing use of tobacco (which could be related to starting or stopping use in the between-interview interval) and about major events associated with increasing or decreasing alcohol consumption and with the source of alcohol when first consumed (the latter is either clear error or represents students beginning to drink during the interval between interviews). In the lifetime drug history, there were also inconsistencies in sedative reports, both in intentions to use again and in reports of parental use; with tranquilizers there was also inconsistency in reports of parental use. No major inconsistencies existed on any items regarding lifetime drug histories of amphetamines, marijuana, hallucinogens, opiate-narcotics, or special substances used for kicks. With no drug were there major inconsistencies, as defined here by 30 per cent or more disagreement, on any feature or age of onset of use, on amount of use during any year of life, on beneficial or ill effects, on dependency worries, or on reports of friends' or relatives' reactions to a student's using a drug. Further inconsistency did show up, however, in the area of common drugs in regard to "yes-no" reports of experience with strong (opiate) painkillers and characteristics of the present amount of use of mild stimulants (coffee, tea, chocolate and cocoa, Coca Cola and other cola drinks, and No-Doze and other over-the-counter stay-awakes). This latter item—amount of present use of any one of these subclasses of minor stimulants—was the least consistent of all, with fifteen of twenty-two students reporting a shift in amount of use between the first interview and the repeat interview.

In reference to the functions or aims for which drugs were used, changes took place in the "yes-no" reporting of having taken something to satisfy cravings or compulsions, or to relieve boredom or monotony, or to make the student feel healthier or stronger. There were also changes in "other" reasons (spontaneous) given to account for the use of drugs by the student. There were inconsistencies in elaborations or replies to an inquiry as to beliefs which guided the student in regard to his personal use of illicit drugs. These guiding principles, complexly coded because of the open-ended question, accounted for three unreliable subitems (out of a set of eighteen such subitems). Inconsistency also appeared in responses to the question as to hoW ready the student's parents were to give medication when he was ill as a child, in the evaluation of advantages to being sick as a child, on the final score given mother-father-student consonance on a list of nineteen issues of potential disagreement (this subitem, in fact, made up nineteen questions, each with three sectors; reliability scoring provided a "disagreement" score whenever there were three or more shifts in reply out of the fifty-seven possible). The final inconsistencies were in self-ratings of satisfaction/dissatisfaction with present teachers and in the degree of optimism the student had about how likely he and people of similar belief could bring about the changes in the world (technological, social, and so on) that he would like to see happen.

Excluding these items with relatively high inconsistency, we find the overall consistency for the remaining majority of (190) items to be high. These included nearly all aspects of lifetime drug use, self-descriptions, background data, relationships with others, and so forth. With regard to the inconsistent items, it is possible to speculate, but not to know, as to what contributed to variability. In each instance the repeat interviewer was a different person from the initial interviewer, so that some direct interpersonal influence may have operated. Several of the items were sensitive to any shift over time in what the student was doing politically or how particular patterns of current drug use were changing. Some items were possibly sensitive to mood changes or to intervening events, as in relations with parents. If lies about drug use were told, we would expect them to have centered about the illicit substances. Since no great inconsistencies occurred in reporting any aspect of illicit-drug use, we may at least conclude that if lying did occur, it was done with consistent sophistication.

Table 1 above sets forth major characteristics of the student samples randomly drawn (with the exception of School IV as noted) for each school.

The information in Table 1 allows the inference that the student populations vary from one school to another in reputation and structure of the college. Students from School I, the private university, come from the wealthiest families. Social sciences predominate as a major along with a large undecided or general-studies group. Most of the students are either Protestant or deny any current religious affiliation. Most students at School II, the Catholic university, are Catholic and come from middle-income families. Social sciences and the hard sciences (including technology) are the most popular majors. The senior class at School II is small; there is a consistent attrition from early to later school years. School III, the junior college, has the youngest population and the greatest number of students in nonacademic majors (vocational programs). Most of its students are Protestants or deny current religious affiliation. School IV, the state college, has the largest number of older undergraduates in its sample, the largest proportion of Jewish students, more students from lower-income families, and the highest proportion in arts and humanities. The large number of upperclassmen in the sample represents, we believe, a sampling error. School V, the state university, located in a small city in a rural area, includes the graduate-student sample. Social and hard sciences, plus technology, are the most popular majors. It has the highest proportion of Protestant students. These student characteristics are consistent with differences in the reputations, traditions, and locations of each of the schools.

Table 2 below presents gross data on drug experience among the samples in each school. More detailed information on how drugs are used is presented in later chapters; our purpose here is simply to compare experience with each class of mind-altering substances by school. Experience in this instance is defined by the statement on the part of the student that he has had one or more experiences with any drug in the class of drugs inquired about; thus, any experience with wine, beer, or hard liquor constitutes an affirmative statements for alcohol. Similarly, any experience with LSD, DMT, STP, mescaline, peyote, psilocybin, and so on would classify the student as having had experience with hallucinogens.

Table 2, with its lifetime or one-or-more drug experiences, shows that in each school alcohol is the drug most often cited, nine out of ten or more students in all schools (except 89 per cent in one) reporting use. Alcohol use is greatest in the state colleges. Tobacco is consistently second with from two thirds to over four fifths reporting use. T'obacco has been used by fewest students in the state university, while the junior college has the greatest number of students reporting its use. The medically available compounds—the amphetamines, sedatives, and tranquilizers—show more between-school diversity than either alcohol or tobacco. The state college is highest on amphetamine experience and tranquilizers; the Catholic university is lowest on both of these and shares the lowest frequence of tranquilizer experience with the private university. The junior college students report the greatest experience with sedatives. In terms of numbers of students reporting experience, marijuana ranks about the same as the medically available stimulants, sedatives, and tranquilizers, with from one tenth to one third of the students reporting experience. The state college is highest on marijuana use (perhaps reflecting the sample bias with its greater number of upperclassmen), while the state university and the Catholic university are lowest. Hallucinogen use is markedly less than marijuana experience for all schools; again, the state university and the Catholic university are lowest and the stage college is highest. Hallucinogen experience is, in terms of ranked averages, slightly less than use of special substances, although with both groups less than 10 per cent of the students (with one exception, 11 per cent in the junior college reporting special-substance experience) stating any experience. Least often used are the opiates with 2 per cent as the highest figure—that, too, in the junior college. For opiates, Table 2 also shows actual numbers of students reporting since our rounding-off procedure for presenting percentages does not allow exact calculations of numbers involved. It is to be kept in mind that the opiate classification is exclusively illicit for it excludes any opiates (for example, morphine) administered under medical supervision.

On the basis of the data in Table 2, one may conclude that genuine differences exist among schools in the (reported) drug experience of students. There are also across-the-board or "cross-campus" consistencies in average ranking of one-time-or-more drug use, ranging from alcohol and tobacco as everywhere most common to hallucinogens and opiates as least common. It would not be wise to generalize about kinds of institutions as such on the basis of our one-of -a-kind samples. It is interesting simply to see that a state college locâted in a metropolitan center is highest or shares the highest frequencies for alcohol, amphetamines, tranquilizers, marijuana, and hallucinogens and that a suburban junior college ranks highest on tobacco, sedatives, opiates, and special substances. These rankings raise questions as to the relationships of drug experience to the characteristics of the student body and, specifically, of the relationship of kind of drug experience to the characteristics of students. The rankings also pose the possibility that there will be correlations among drug experiences—that is, persons reporting use of one unusual drug might also be more likely to report use of another unusual drug. We shall explore these likely possibilities in later chapters.

Another way of comparing drug use by schools was to examine the intensity of drug use as measured by the distribution of individual scores in contrast simply to experience. An intensity score was developed for each student on the basis of his statements about his lifetime history of the use of each drug in the nine categories. Scored intensity of use would increase for each year the student had taken the drug and by the amount taken, classified as "intermittent," "occasional," "regular," or "considerable" (definitions of these classifications were specific for each drug). Most of the intensity comparisons we made were based on the score equivalent of the "mean drug-use score."

That score was obtained by first summing up all the entries in the drug-use table (individual drug profile) for a given drug, then dividing by the number of age categories which applied to that individual. For example, a total of fourteen categories is applicable to an individual who is twenty-eight years old or more; therefore, the sum of entries would be divided by 14. For a person eighteen years old, the divisor for each drug-use score would be 8. By dividing in this fashion, one prevents overweighting of the drug use of older students when developing use-intensity scores for students of different ages.

Table 3 above gives the distribution of total drug-use scores on each campus. From Table 3 we see that there are considerable differences among schools in the distribution of scores. School IV has the fewest students in the zero-score category and by far the greatest number in the over 26-, over 51-, and over 100-score categories. School II has the greatest proportion of students in the range of scores of 25 and under. Except for School IV (which suffered from sampling inadequacy and bias), we see that the majority of students are in the very low range of drug-intensity scores.

To refine the scores for drug-use intensity for each campus, the two drugs which most students on every campus have used, alcohol and tobacco, can be excluded. Thus, what is left are the prescribed drugs (amphetamines, sedatives, and tranquilizers) and the illicit drugs (marijuana, hallucinogens, illicit opiates, special substances—and, again, amphetamines') . Table 4 below presents the distribution of drug-intensity scores when such an exclusion is made.

The data in Table 4 above are consistent with Table 3 in placing School II as the lowest in intensity of use, for only one third of its students report the use of any psychoactive drug (in our-major categories) except for alcohol and tobacco. School II is closely followed by School V in this respect. We may infer that most of the drug experience which constitutes the drug-history intensity score for students in Schools II and V was with alcohol and tobacco, for with these drugs removed from consideration, the majority in both schools have zero-intensity scores. School IV is in the first rank in terms of the percentage of students with intensive drug-use histories, but now we see that School I ranks a close second in intensity and that School III is not far behind. Combined with data from Table 3, this implies that students in these three schools are considerably greater users of both prescribed and illicit drugs. Such is exactly the case, of course, as shown in Table 2. Recall that School IV ranks first in marijuana use, followed by Schools III and I; a similar ranking obtains for hallucinogens. These are the schools that are also highest in reported experience with the amphetamines, sedatives, and, with one reversal, tranquilizers. What this shows is that the experience measure is related to the intensity measure,2 a relationship which will be considered in further detail in later chapters. In the meantime, we are assured that there are differences among schools whether drug use is measured in terms of reported student experience—"yes" or "no"—with each class of drugs or whether overall intensity of drug use is recorded over a lifetime and includes all the major psychoactive-drug classes or, alternatively, limits the measure of intensity to illicit and prescribed drugs.

Table 5 below expands on the information in Table 2 to give data elicited from inquiries supplemental to the detailed life histories of the nine major classes of drugs. These less comprehensive inquiries covered experience with common other drugs, including general anesthetics.

Examination of Table 5 shows a consistent pattern across campuses. Nearly all students currently employ mild painkillers and/or tranquilizers and drink cola beverages. Over three quarters in all schools drink coffee, tea and chocolate/cocoa beverages, and two thirds to three quarters in all schools have undergone general anesthesia. About half in each school, except for the junior college, have had medicall5r prescribed or administered opiates, and from one quarter to two fifths are presently employing over-the-counter stay-awakes.

Compared with these quite constant across-campus figures, marijuana and amphetamine experience shows in Table 2 a very considerable variation from campus to campus; there is also a variation —though slightly less wide—in tranquilizer experience. Proportionately, very considerable differences exist in special-substance and hallucinogen experiences. One campus, for example, has almost four times as many students as another whose students have taken special substances; one campus has almost five times as many students as another reporting the use of hallucinogens. These comparisons lead to the proposition that student behavior, regardless of campus, is very similar with regard to the culturally sanctioned and freely available minor stimulants, painkillers, and social drugs. On the other hand, the campus "culture"—that is, the school's special environment composed of different kinds of student populations, traditions, location, and so on—is reflected in the use of the illicit-exotic substances and in presumed patterns of medical care as inferred from the differential prescription of tranquilizers. What emerges, on the one hand, is an overriding general culture or style of drug conduct which determines the level of use of painkillers, mild stimulants, and social drugs, and also some prescribed psychoactive drugs (sedatives) compared with a much more specific and variable sensitivity by campus to illicit-exotic drug use and to some kinds of prescribed drugs—the latter reflecting, we presume, medical-care variation.

This chapter has presented an overview of the methods employed in the college study, as well as gross characteristics of the student populations in the five samples studied, information on the prevalence of experience with each of the major classes of drugs, and data on intensity of use. The samples and, inferentially, the colleges, show considerable differences in student-body characteristics, differences which are presumably associated with differences in illicit-drug use. Differences among campuses are slight with regard to the use of the culturally sanctioned and easily available minor drugs, are greater for the approved social and prescription drugs, and are greatest, proportionatelY, for illicit substances. What is implied is that across-campus conduct with regard to over-the-counter analgesics, minor stimulants, and so forth is much the same as that of our national drug-use culture, of which it is a part. Furthermore, differences in student characteristics and campus milieu are beginning to play a stronger role in the use of alcohol, tobacco, and prescription psychoactives, and student characteristic and campus milieu are emerging most strongly as determinants of the use and non-use of illicit-exotic drugs. These features may be identified generally as cultural, social-environmental, and family-individual factors, all of which affect drug conduct.

1 Amphetamines are obtained by students through both legal and illegal channels.

2 That there should be a relationship between reported experience with a drug or lifetime prevalence of use and measures of intensity is not surprising; experience as such contributes to the intensity score (although one-time experience would contribute very little by way of magnitude), and we have already shown, in our pilot study of normal-population drug use (see companion vol-urne Drugs I) that past experience is related to present use. Given both past and present use the intensity score necessarily increases.