Drug Abuse

Chapter 3. Research Methods: The Crime and the Criminal

 

 

Research Methods: The Crime and the Criminal

The object of the research was to determine the social effects of punishment on cannabis users. It is difficult to locate a representative sample of persons who have engaged in an illegal activity, and who are not already "captive" in an institution such as school or prison. Drug researchers (e.g. Goode, 1970; Plant, 1975) have noted that it is practically impossible to obtain a truly random sample of marijuana users in the population-at-large.

To identify criminalized marijuana users presents a somewhat less formidable task, to the extent that such a group has at least been the subject of official recordkeeping in some archive, and must make a public appearance in court. The method chosen for contacting subjects was a direct approach to those found guilty of possessing cannabis. A piloting phase at court proved this to be a feasible means for locating willing subjects. Other potential sources which were considered included official police records - unavailable for reasons of confidentiality - and self selection through advertisements, likely to be highly unrepresentative.

This study focused on internal comparison groups of similar offenders who received different sentences. While the inclusion of an external comparison group of unlabeled cannabis users might have been desirable to expand the explanatory value of the study, obtaining a truly matched group would have been highly problematic. Issues of confidentiality preclude access to official record sources of those arrested but not prosecuted successfully. Another possibility was the "snowball" method used in field studies of drug users (Goode, 1970; Plant, 1975) in which initial contacts volunteer to lead the researcher to other subjects. Reservations based on the pilot period, which were later borne out by the actual sampling, were that naive users and the older more "respectable" ones preferred not to tell their friends about the court experience. This would have limited the willingness of such subjects to identify friends with no official record. At the other extreme, some of the heaviest users reported having no friends who had not been arrested for cannabis. Thus, a truly representative comparison group would have been difficult, perhaps impossible, to achieve by the snowball method. Another possibility would have been to select a sub-sample from a general population survey, matching the offender sample on the basis of demographic factors. No Toronto survey was available at that time, and even if there had been, the identities of respondents would have had to be known in order for repeated contact, thus breaching the usual conditions of anonymity. However, while the lack of an external comparison group may impose some limitations on the conclusions, examination of the social effects of punishment associated with differential sentencing does provide insight into the criminalization process.

This chapter introduces the offender group and provides a backdrop for the discussion, in the following chapters, of becoming a cannabis criminal and the individual consequences of criminalization. The opening section of the chapter describes the operational definition of a "first offender," and the manner in which such persons were selected for inclusion in the sample. The representativeness of the sample is addressed next, both with respect to cannabis offenders in Toronto and elsewhere and with regard to the sentence received. Finally, details about the sorts of persons who are sentenced for this offense, their drug use patterns, and the legal features of their cases are presented.

SAMPLE SELECTION "First Offenders

The sample consists of 95 persons sentenced for the offense of simple possession during an 11-week period from July to September of 1974. The site for sample selection was the Provincial Court of Metropolitan Toronto. At that time, virtually the entire daily caseload of one courtroom was devoted to cannabis matters. Since a total of 433 cases received sentences for possession during the field period, the 95 cases represent a 22% sample of this group of offenders.

Daily court sessions were attended by two interviewers, one male and one female, both in their late twenties. They alternately followed the first and then next potential candidate out of court immediately after final disposition had occurred. The selection was in a random manner without reference to sentence received. A case was excluded where a prior record of conviction for an adult criminal offense had been submitted in court, or the candidate acknowledged such a record when first approached.' In the event of an unsuccessful contact, the interviewer returned to court and pursued the next possible eligible subject. Of those approached, there were 26 refusals, 23 were considered ineligible (due to a prior conviction or in one case a prior interview with the other field worker), and 95 interviews were completed. Thus out of 144 attempts, two in three were successful; the refusal rate was 21% of those who qualified.

Ninety of the sample members were interviewed immediately, in the environs of the court. Interviews with the remaining five subjects took place elsewhere within one week. The duration of the interview ranged from half an hour to nearly two hours, but on the average about one hour. At the completion of the post-trial interview, the topic of a follow-up Iinterview was broached and additional personal information requested for tracing purposes (Erickson and Salutin, 1976; 1977). Ninety percent of the original 95 subjects were re-interviewed after one year.

Since the aim of the study was to identify the nature and extent of any social consequences following from the designation of a criminal label, it was necessary to control for the effects of prior record. This was done by excluding from the sample all those who said they had previously been convicted of an adult criminal offense. But, as the difficulties in establishing a standard indicator of "recidivism" have illustrated (Hawkins et al., 1977), the definition of a "first offender" likewise is problematic.

There is a whole spectrum of potential justice system involvement that may occur without the registering of an official finding of guilt. A person may be questioned, arrested, released, charged, prosecuted, even acquitted - all steps falling short of conviction. The term "criminal record" has been used in reference to a police record of arrest only (e.g. Grupp, 1971). But the more conventional and classical definition restricts usage of the term to that following conviction for a criminal offense (Martin and Webster, 1971).2 In this study, the decision was made to draw the line at conviction. This was because the adverse effects of criminalization are thought to increase in proportion to the intensity of the official response, with the greatest impact being the public designation of the criminal label (Lemert, 1951:320; Becker, 1963:31; Garfinkel, 1956). This is not to deny that negative consequences may result from any form of legal history antecedent to conviction (Schwartz and Skolnick, 1962; Palys, 1976).

In refining the concept of a first offender, a self-admitted juvenile record was not considered to affect one's criminal or noncriminal status as an adult. Since the juvenile court record is confidential and proceedings are not public, with some exceptions:' a finding of delinquency does not constitute a finding of guilt equivalent to that of conviction for an adult criminal offense (see Albright and Dent, 1978). Twenty-four percent of the sample acknowledged appearing in juvenile court on at least one occasion.-'

As an adult, various types of contacts with the criminal justice system are possible. The computerization of criminal history information in the Canadian Police Information Centre (CPIC) provides rapid retrieval of such information. Charges pending, acquittal, or charges being withdrawn or dismissed can all produce an official record of involvement but not of guilt. The disposition option of discharge is ambiguous since it is not technically a conviction, but does register a finding of guilt and creates a criminal record that, in some aspects, is indistinguishable from that obtained by conviction (Swabey, 1973; Leon, 1977b). Persons who reported all these variants of legal history short of conviction were included in the sample. As adults, 13% had prior contacts for cannabis offenses, 2% for other illicit drugs, and 14% for non-drug related criminal matters. In combination, 74% reported no kind of record, while 10% had received discharges, and 16% had been charged before but not found guilty. Thus, some first offenders in our sample were more "pure" (with less serious records) than others, but all 95 shared the property of reporting no prior convictions for an adult criminal offense.

Tabel 5. Distribution of sentences for the sample and for all cases in the sampling period

REPRESENTATIVENESS OF SAMPLE

Was the Sample Representative of Cannabis Offenders Generally?

Since the sample was Chosen over an 11-week period in the summer of 1974, it is important to know how typical this period was and how well the sample reflected the basic characteristics of all Metropolitan Toronto cannabis offenders. Police statistics (see Appendix A) show a decline in total possession offenses from 1973 to 1975, then a small rise in 1976. The total number of cannabis offenses in 1974 was 5,475, of which 4,949 (90%) were for simple possession. In looking at monthly variation, the decrease in the number of offenses appears to have begun towards the end of 1974, after the field period was complete. This was the appropriate year for comparison since all but one of the 95 were charged as well as tried in 1974.

The sex and age distribution for all cannabis possession offenders in Metropolitan Toronto remained predominantly male and youthful from 1973 to 1976. Of the total 1974 figure of 4,949, 87% were male and 13% female. This compares well with the 90% male and 10% female composition of the sample. Of all Toronto offenders (excluding juveniles), 68% were age 21 or younger, compared with 65% of the sample. The sample selected, therefore, had a sex and age distribution in agreement with all the subjects of cannabis possession cases prosecuted in this jurisdiction during 1974.

In 1974, there were 27,202 cannabis possession "convictions" recorded in Canada, of which 10,290 were in Ontario.' The official record source, the Bureau of Dangerous Drugs, does not prepare a detailed provincial breakdown of offense type by sex and age. However, the Narcotic Control Act total, of which the bulk were cannabis possessioncases, showed at least 90% of offenders to be male. Nationally, 53% of cannabis possession offenders were under 21 years, not unlike the 58% of the sample. Thus, the sample was similar to the provincial and national picture with regard to sex and age composition.

How Representative Was the Sample with Respect to Sentence?

Potential subjects were chosen in a random way without reference to outcome. The sentences awarded for all possession cases during the ll-week sampling period were extracted from a public file kept at the courthouse. When the outcomes of the 95 sampled cases were compared to the sentences of the 433 total court cases of cannabis possession offenders receiving final disposition in the study period, the result was a very high degree of correspondence, as shown in Table 5. The difference between the two distributions for each of the outcomes was always less than 3%. Thus despite its restriction to first offenders, the sample may be taken as representative of all those sentenced for simple possession of cannabis in that interval.

It may perhaps seem surprising that greater sentencing discrepancy between first offenders and a total group including recidivists was not displayed. But observation of the court proceedings had suggested that prior record would not weigh heavily in sentencing for two reasons. The first was that the existence of such a record was frequently not submitted in court by the crown prosecutor, for reasons not apparent to the observers. This was illustrated by the 22 persons (and others in the pilot period) for whom no record had been entered in court but who freely admitted to one when questioned afterwards. Secondly, even when a prior record was entered by the crown, it was observed that some judges disregarded any record while others considered only a drug record relevant (Erickson, 1975). No consistent policy of more severe sentencing for those with a record was observed, although some judges appeared to adopt such a stance on occasion.

Sentences awarded to persons in the Toronto sample were considerably more lenient than those given on a national scale in 1974. During that year in Canada, 8% of those found guilty of simple possession of cannabis received an absolute discharge, 12% a conditional one, while 69% were fined and the balance received a suspended sentence or imprisonment, whereas three-quarters of the Toronto sample received some form of discharge. The tradition was for Ontario judges to be less severe than judges elsewhere in Canada in their drugsentencing behavior (Cook, 1970; Whealy, 1970) and less punitive in their attitudes to drug offenses (Leon, 1976). Nor could Toronto patterns be said to have been typical of Ontario, where slightly over half of such offenders received discharges in 1974. The suggestion is that this group of offenders, having been processed in the more "lenient" judicial atmosphere of Toronto, represents the lower or more minimal point of a scale of criminalization consequences associated with sentence, when compared to offenders in the rest of the provinces or in the nation as a whole.

Table 6 sentence by personal attributes

SAMPLE CHARACTERISTICS

Personal Attributes: What sort of people go to court?

The sample members were predominantly young single males with incomplete secondary schooling. The mean age of the 85 men and 10 women was 20.8 years. Two-thirds of the sample fell in the 16 to 21 age range, and five subjects were over 30. The large majority (72%) of those interviewed were employed and 17% were students. The balance of 12% were unemployed but looking for work, or doing "nothing" (e.g. staying home, planning to travel). In Ontario, the legal schoolleaving age is 16 and either grade 12 or grade 13 marks the graduation from secondary education, depending on the student's program. Of the sample, 45% had attained grade 11 or less and 34% had grade 12 or 13. For the remainder who had undertaken extra-secondary schooling, 8% had participated in a community college or other technical training program and 13% had attended university. Of those who were employed or had ever had a regular job (N=71), 25% had held their most recent position for a year or more, 18% for seven months to a year, and 56% had worked six months or less in their most recent one. The most common living situation was for the subject to live at home with parents (43%), followed by shared accomodation (22%), living alone (18%), and with a spouse or partner (17%). Seven sample members had children. Fathers' occupational category (Pineo and Porter, 1967) was classified as professionalmanagerial for 40% of respondents, clerical-sales for 12%, skilled or semi-skilled for 40%, and the balance unskilled or unknown.7

It is important to know whether any of these personal or "extra-legal" attributes were differentially associated with a sentence category. Since the researcher had no experimental control over case outcome, it was possible that judges could have tended to favor a certain type of person for a particular sentence. Such a bias would have created a problem in the analysis of subsequent consequences. Any exhibited differences in social effects could have been a result of variation in preexisting personal attributes and only spuriously related to sentence.

In Table 6, some personal characteristics of the sample are shown in relation to sentence received. A chi-square test was done for each personal attribute compared to sentence outcome; none was significant. Any of the observed differences in sentence, in relation to the extra-legal variables, could have occurred by chance.8

Of course, judges were often not in possession of all this personal information about each case before them. Sex was the only readily observed characteristic, and to some extent, age. The other items could have been introduced by the lawyer, if one appeared on behalf of the accused, or as a result of the judge's questioning of the accused, or not at all. The observation in court indicated that personal details such as living situation and employment were routinely elicited by some of the 13 presiding judges. Due to this variation in approach by different judges, any sentencing differences that might have been accounted for by the personal attributes of offenders tended to be minimized. The result was not unlike random assignment of sentences on these extra-legal dimensions.

Table 7: Sentence by Legal Attributes

*Chi-square statistically significant at .01 level

Legal Attributes: What kinds of cases are prosecuted?

Judges are more consistently aware of the legal characteristics of the case before them than the extra-legal ones. These are presented during the trial by the crown prosecutor on the basis of the police information. The 95 cases in the sample were typified by a plea of guilty to a single charge of possessing a small amount of marijuana, and were resolved within one or two appearances without representation by private defense counsel.

To consider the legal profile in more detail, the only charge was of possession of one form of cannabis - e.g. marijuana or hashish - in 80 cases. In the remaining 15 cases, added charges such as possession of a second form of cannabis (five cases) or another illicit drug (three cases), possession for the purpose of trafficking (six cases), or cultivation (one case), were withdrawn by the crown and the finding of guilt was on a simple possession charge. Only two of the 95 accused pleaded "not guilty." The accused was charged alone in 68% of the cases and jointly in 32%. The cannabis substance which formed the basis of the charge was solely marijuana in 81%, hashish in 14%, and a Cannabis sativa plant or plants in 5%.' The weight of the substance was 14 grams or less (half an ounce) in 74% of the cases, and 29 grams or less in 89% of the total (excluding plants).

Thirty-two percent of the sample had retained their own lawyer. These tended to be the cases in which a more serious additional charge and a larger amount of cannabis were involved."' Others had no legal representation or were represented by duty counsel." Performance of the duty counsel role varied considerably. Only those cases in which the duty counsel was observed to make some submission on behalf of the subject were counted as "represented" by duty counsel; this occurred in 27% of the cases. The balance, 41%, were categorized as unrepresented in court.

The accused appearing for the first time on a particular charge may plead, be tried, and sentenced that day, or be remanded one or more times. Of the sample, 48% received their sentence on first appearance, 32% on their second, and 20% made three or more trips to court before final disposition (the largest number of visits being eight). These delays were generally for arrangements to be made for a lawyer or to wait for a certificate of analysis;" no pre-sentence report was requested by the judge for any sample member. The preliminary field work had shown that each day's court list was a mixture of new cases, remands, and trials, heard in an interchangeable order. Thus any tendency towards selection of sample cases from the earlier portion of the day's proceedings did not lead to any systematic bias in the legal nature of the case.

The distribution of sentences within categories of legal attributes was examined (see Table 7) and tested for significance with the chi-square statistic. Only the relationship between legal representation and sentencing outcome was significant. Those who were represented either by their own lawyer or spoken for by duty counsel were much more likely to receive an absolute discharge than those who went unrepresented; individuals in the latter group were most likely to be convicted and fined. While other trends may be observed notably that those with more serious added charges, possessing larger amounts of cannabis, and appearing more than once on the charge tended to receive more severe sentences - no relationships were of such a magnitude that they could not have occurred by chance. Although 10% of the sample had prior records of discharge, these were not consistently introduced in court and therefore this attribute has been excluded from the presentation in Table 7.

The presiding judge may, too, be considered a legal feature of the case (Hogarth, 1971; Perry, 1977). Thirteen judges sat in the drug court during the sampling period, and at least one case from each is found in the sample. The distribution of sentence by judge for all cases in the field period and for the sample is shown in Table 8.

Large sentencing differences for the offense of simple possession were evident among judges, and the sample cases reflected the typical pattern for each judge. These data reflect the general trend observed by Hogarth (1971) for judges to vary greatly in their practices while being individually uniform. The data also lend empirical support for the inconsistency of the judicial reaction to hypothetical drug cases presented in Le Dain (1972:246).

Variation in sentences according to judge contributed to an approximation of randomness in the distribution of both personal and legal attributes among sentence categories in the sample of cannabis offenders. The judicial discretion exercised in sentencing them illustrates the point that "it is quite possible to have a widespread disparity without systematic bias (Scheingold, 1977:266; emphasis in original).

Table 8: Sentence by Judge for Total Cases in sample period and for sample cases

Sentence - Total Cases

Sentence - Sample Cases

Cannabis Use Attributes: What kind of users get caught?

The sample members were asked about their prior involvement with the drug. Employing the Le Dain classification of cannabis use levels (1972:194), most of the sample members (87%) were in the regular use category of smoking more than once a month on the average in the year preceding the court appearance. The other 13% of the sample were considered irregular users of various types. These included occasional users (no more than once a month on the average: six cases), experimental users (once or twice: four cases), and nonusers (never or not at all in the past year: one case of each).13 The regular users were more likely to be "heavy" users" (twice a week to daily - 73% of the 95) than "moderate" users (once a week - 14% of the total).

When asked their age when they first tried cannabis, 27% of the sample responded that they had been 13 years or younger, 44% that they were aged 14 to 16, and the balance reported 17 or older at the time of first use. Seventy-six percent had commenced use more than three years before their official labeling as an offender.

The need to obtain a steady supply of the drug also characterizes the regular user (Le Dain, 1972:195). Nine subjects said they had never purchased cannabis and another nine reported only one buying experience. Of the remainder of the sample who had purchased the drug more often (N=77), 71% reported buying it at least once a month. Thus the general profile of the sample is of a group of fairly heavy and regular cannabis users with a minority of irregular and more naive users.

Summary

In review, the sample consisted of 95 "first offenders" who were sentenced for simple possession of cannabis in 1974. Grounds for inclusion were the self-report of no prior convictions, and a later official record check of reliability showed a 7% under-reporting of this form of criminal history.15 Respondents were first interviewed at the time of court appearance and sentencing, and 90% were successfully re-interviewed one year later. In regard to age and sex composition predominantly young and male - the sample was found to be representative of offenders generally in Metropolitan Toronto, Ontario, and Canada. The sentences imposed, while corresponding highly with all those awarded for simple possession in Toronto during the sampling period, were less severe than the national pattern for the same year. The personal attributes of the offenders and their legal case characteristics were described in some detail and, except for legal representation, were found to be unrelated to sentence received. The presiding judge appeared to be the most important determinant of the outcome: absolute discharge, conditional discharge, or conviction with fine. The profile of the respondents as cannabis consumers revealed a majority were fairly regular, long-term users of twice weekly or more.

                           

Notes CHAPTER 3: Research Methods

1 For the purposes of sample selection in this study, a reported conviction for an offense involving driving or alcohol was regarded as "non-criminal." This was done because related offenses are found in Canadian federal, provincial, and municipal statute books, and those interviewed were often uncertain as to the actual legal infraction they had committed. Some precedent was found in West's (1973) exclusion of traffic offenses in defining the nature of conviction, and the labeling literature generally emphasizes, the importance of criminal conviction rather than more minor law-breaking.

2 An international survey of the statuatory legal consequences of conviction makes the point that the U.S. is unusual in attaching significance to an arrest record (Damaska, 1968).

3 Morris v. the Queen (1978), 43 C.C.C. (2d) 129.

4 A juvenile in Ontario is someone aged 15 or less. The sample of adult cannabis offenders does not appear to have been unusually delinquent when compared to West's cohort study in which 24.8% had made a juvenile court appearance (1973:3).

5 Source: Bureau of Dangerous Drugs, Health Protection Branch, Health and Welfare Canada. Discharges are included in its conviction figures.

6 Source: Bureau of Dangerous Drugs, Health Protection Branch, Health and Welfare Canada.

7 The socioeconomic status of respondents' family backgrounds, a relatively stable characteristic, was ascertained at the final interview. See Chapter 6, Table 28. (For characteristics of cannabis offenders elsewhere, see Crow, 1973; Morton et aL, 1968; Shafer, 1972; Johnson et al., 1977.)

8 The analysis was repeated for males only and again no sign ficant differences emerged. The extra-legal attribute of race was not included since less than 5% of the sample was "non-white" (see Hagan, 1974).

9 While five of the 95 cases involved plants, in only one had an actual charge of cultivation been laid.

10 Further analysis of the interrelationships between legal representation, other case attributes, and sentence has been conducted for a forthcoming paper (Erickson and Leon). Judges were categorized as "less punitive" or "more punitive" according to whether an absolute discharge or other penalty was predominant in their sentencing patterns over the entire 11-week field period. Multi-variate analysis showed that for "less punitive" judges, having a lawyer or duty counsel speak on behalf of the accused was the most important single factor affecting outcome when all other legal and extra-legal variables were controlled. However, for "more punitive" judges, the weight of the cannabis substance had the greatest effect on sentence when all other factors were held constant.

11 The duty counsel role is a rotating position occupied by a lawyer paid by Legal Aid, who advises unrepresented defendants (Legal Aid Act, Stats. Ont., 1966, c8O).

 

12 Federal analysts, appointed under the Food and Drugs Act and the Narcotic Control Act, provide certificates of proof that the substance forming the basis of the charge is actually that prohibited drug.

13 The "never" user said the drug had been planted on him and the other nonuser, who admitted previous use but not in the year before arrest, was in possession of one "decorative" plant.

14 More recent definitions of the "heavy" level of cannabis use apply it to daily or almost daily use (Johnston, 1980, Fehr et al., 1980).

 

15 After the follow-up phase was completed, the adult criminal record history reported by the subjects at the time of the outcome of their cannabis possession case was compared to that shown in the official record for each individual. It was found that two subjects had a prior conviction for "theft under" and two others had been convicted for "causing a disturbance." The latter is a minor Criminal Code infraction which includes drunkenness, and these two subjects had reported having a record for drinking (see footnote I above). These incidents had occurred two years or more before the court appearance for cannabis possession. Two other respondents, both men over 30, had prior convictions at least seven years earlier for "victimless" crimes (one involving gambling and the other, homosexual acts). In addition, one subject had received an unreported discharge for attempted theft. The accounts of the others corresponded to their official files. The 7% under-reporting of criminal history compares favorably with that of narcot Iic addicts (Ball, 1967), and enhances confidence in the data on measures that cannot be verified from other sources. Moreover, the integrity of the sample - i.e. it was shown to contain only those without a prior conviction for a drug offense was sustained.

< Prev		Next >