National Academies Press: OpenBook

Research on Sentencing: The Search for Reform, Volume II (1983)

Chapter: 1 MAKING SENSE OF SENTENCING: A REVIEW AND CRITIQUE OF SENTENCING RESEARCH

« Previous: FRONT MATTER
Suggested Citation:"1 MAKING SENSE OF SENTENCING: A REVIEW AND CRITIQUE OF SENTENCING RESEARCH." National Research Council. 1983. Research on Sentencing: The Search for Reform, Volume II. Washington, DC: The National Academies Press. doi: 10.17226/101.
×
Page 1
Suggested Citation:"1 MAKING SENSE OF SENTENCING: A REVIEW AND CRITIQUE OF SENTENCING RESEARCH." National Research Council. 1983. Research on Sentencing: The Search for Reform, Volume II. Washington, DC: The National Academies Press. doi: 10.17226/101.
×
Page 2
Suggested Citation:"1 MAKING SENSE OF SENTENCING: A REVIEW AND CRITIQUE OF SENTENCING RESEARCH." National Research Council. 1983. Research on Sentencing: The Search for Reform, Volume II. Washington, DC: The National Academies Press. doi: 10.17226/101.
×
Page 3
Suggested Citation:"1 MAKING SENSE OF SENTENCING: A REVIEW AND CRITIQUE OF SENTENCING RESEARCH." National Research Council. 1983. Research on Sentencing: The Search for Reform, Volume II. Washington, DC: The National Academies Press. doi: 10.17226/101.
×
Page 4
Suggested Citation:"1 MAKING SENSE OF SENTENCING: A REVIEW AND CRITIQUE OF SENTENCING RESEARCH." National Research Council. 1983. Research on Sentencing: The Search for Reform, Volume II. Washington, DC: The National Academies Press. doi: 10.17226/101.
×
Page 5
Suggested Citation:"1 MAKING SENSE OF SENTENCING: A REVIEW AND CRITIQUE OF SENTENCING RESEARCH." National Research Council. 1983. Research on Sentencing: The Search for Reform, Volume II. Washington, DC: The National Academies Press. doi: 10.17226/101.
×
Page 6
Suggested Citation:"1 MAKING SENSE OF SENTENCING: A REVIEW AND CRITIQUE OF SENTENCING RESEARCH." National Research Council. 1983. Research on Sentencing: The Search for Reform, Volume II. Washington, DC: The National Academies Press. doi: 10.17226/101.
×
Page 7
Suggested Citation:"1 MAKING SENSE OF SENTENCING: A REVIEW AND CRITIQUE OF SENTENCING RESEARCH." National Research Council. 1983. Research on Sentencing: The Search for Reform, Volume II. Washington, DC: The National Academies Press. doi: 10.17226/101.
×
Page 8
Suggested Citation:"1 MAKING SENSE OF SENTENCING: A REVIEW AND CRITIQUE OF SENTENCING RESEARCH." National Research Council. 1983. Research on Sentencing: The Search for Reform, Volume II. Washington, DC: The National Academies Press. doi: 10.17226/101.
×
Page 9
Suggested Citation:"1 MAKING SENSE OF SENTENCING: A REVIEW AND CRITIQUE OF SENTENCING RESEARCH." National Research Council. 1983. Research on Sentencing: The Search for Reform, Volume II. Washington, DC: The National Academies Press. doi: 10.17226/101.
×
Page 10
Suggested Citation:"1 MAKING SENSE OF SENTENCING: A REVIEW AND CRITIQUE OF SENTENCING RESEARCH." National Research Council. 1983. Research on Sentencing: The Search for Reform, Volume II. Washington, DC: The National Academies Press. doi: 10.17226/101.
×
Page 11
Suggested Citation:"1 MAKING SENSE OF SENTENCING: A REVIEW AND CRITIQUE OF SENTENCING RESEARCH." National Research Council. 1983. Research on Sentencing: The Search for Reform, Volume II. Washington, DC: The National Academies Press. doi: 10.17226/101.
×
Page 12
Suggested Citation:"1 MAKING SENSE OF SENTENCING: A REVIEW AND CRITIQUE OF SENTENCING RESEARCH." National Research Council. 1983. Research on Sentencing: The Search for Reform, Volume II. Washington, DC: The National Academies Press. doi: 10.17226/101.
×
Page 13
Suggested Citation:"1 MAKING SENSE OF SENTENCING: A REVIEW AND CRITIQUE OF SENTENCING RESEARCH." National Research Council. 1983. Research on Sentencing: The Search for Reform, Volume II. Washington, DC: The National Academies Press. doi: 10.17226/101.
×
Page 14
Suggested Citation:"1 MAKING SENSE OF SENTENCING: A REVIEW AND CRITIQUE OF SENTENCING RESEARCH." National Research Council. 1983. Research on Sentencing: The Search for Reform, Volume II. Washington, DC: The National Academies Press. doi: 10.17226/101.
×
Page 15
Suggested Citation:"1 MAKING SENSE OF SENTENCING: A REVIEW AND CRITIQUE OF SENTENCING RESEARCH." National Research Council. 1983. Research on Sentencing: The Search for Reform, Volume II. Washington, DC: The National Academies Press. doi: 10.17226/101.
×
Page 16
Suggested Citation:"1 MAKING SENSE OF SENTENCING: A REVIEW AND CRITIQUE OF SENTENCING RESEARCH." National Research Council. 1983. Research on Sentencing: The Search for Reform, Volume II. Washington, DC: The National Academies Press. doi: 10.17226/101.
×
Page 17
Suggested Citation:"1 MAKING SENSE OF SENTENCING: A REVIEW AND CRITIQUE OF SENTENCING RESEARCH." National Research Council. 1983. Research on Sentencing: The Search for Reform, Volume II. Washington, DC: The National Academies Press. doi: 10.17226/101.
×
Page 18
Suggested Citation:"1 MAKING SENSE OF SENTENCING: A REVIEW AND CRITIQUE OF SENTENCING RESEARCH." National Research Council. 1983. Research on Sentencing: The Search for Reform, Volume II. Washington, DC: The National Academies Press. doi: 10.17226/101.
×
Page 19
Suggested Citation:"1 MAKING SENSE OF SENTENCING: A REVIEW AND CRITIQUE OF SENTENCING RESEARCH." National Research Council. 1983. Research on Sentencing: The Search for Reform, Volume II. Washington, DC: The National Academies Press. doi: 10.17226/101.
×
Page 20
Suggested Citation:"1 MAKING SENSE OF SENTENCING: A REVIEW AND CRITIQUE OF SENTENCING RESEARCH." National Research Council. 1983. Research on Sentencing: The Search for Reform, Volume II. Washington, DC: The National Academies Press. doi: 10.17226/101.
×
Page 21
Suggested Citation:"1 MAKING SENSE OF SENTENCING: A REVIEW AND CRITIQUE OF SENTENCING RESEARCH." National Research Council. 1983. Research on Sentencing: The Search for Reform, Volume II. Washington, DC: The National Academies Press. doi: 10.17226/101.
×
Page 22
Suggested Citation:"1 MAKING SENSE OF SENTENCING: A REVIEW AND CRITIQUE OF SENTENCING RESEARCH." National Research Council. 1983. Research on Sentencing: The Search for Reform, Volume II. Washington, DC: The National Academies Press. doi: 10.17226/101.
×
Page 23
Suggested Citation:"1 MAKING SENSE OF SENTENCING: A REVIEW AND CRITIQUE OF SENTENCING RESEARCH." National Research Council. 1983. Research on Sentencing: The Search for Reform, Volume II. Washington, DC: The National Academies Press. doi: 10.17226/101.
×
Page 24
Suggested Citation:"1 MAKING SENSE OF SENTENCING: A REVIEW AND CRITIQUE OF SENTENCING RESEARCH." National Research Council. 1983. Research on Sentencing: The Search for Reform, Volume II. Washington, DC: The National Academies Press. doi: 10.17226/101.
×
Page 25
Suggested Citation:"1 MAKING SENSE OF SENTENCING: A REVIEW AND CRITIQUE OF SENTENCING RESEARCH." National Research Council. 1983. Research on Sentencing: The Search for Reform, Volume II. Washington, DC: The National Academies Press. doi: 10.17226/101.
×
Page 26
Suggested Citation:"1 MAKING SENSE OF SENTENCING: A REVIEW AND CRITIQUE OF SENTENCING RESEARCH." National Research Council. 1983. Research on Sentencing: The Search for Reform, Volume II. Washington, DC: The National Academies Press. doi: 10.17226/101.
×
Page 27
Suggested Citation:"1 MAKING SENSE OF SENTENCING: A REVIEW AND CRITIQUE OF SENTENCING RESEARCH." National Research Council. 1983. Research on Sentencing: The Search for Reform, Volume II. Washington, DC: The National Academies Press. doi: 10.17226/101.
×
Page 28
Suggested Citation:"1 MAKING SENSE OF SENTENCING: A REVIEW AND CRITIQUE OF SENTENCING RESEARCH." National Research Council. 1983. Research on Sentencing: The Search for Reform, Volume II. Washington, DC: The National Academies Press. doi: 10.17226/101.
×
Page 29
Suggested Citation:"1 MAKING SENSE OF SENTENCING: A REVIEW AND CRITIQUE OF SENTENCING RESEARCH." National Research Council. 1983. Research on Sentencing: The Search for Reform, Volume II. Washington, DC: The National Academies Press. doi: 10.17226/101.
×
Page 30
Suggested Citation:"1 MAKING SENSE OF SENTENCING: A REVIEW AND CRITIQUE OF SENTENCING RESEARCH." National Research Council. 1983. Research on Sentencing: The Search for Reform, Volume II. Washington, DC: The National Academies Press. doi: 10.17226/101.
×
Page 31
Suggested Citation:"1 MAKING SENSE OF SENTENCING: A REVIEW AND CRITIQUE OF SENTENCING RESEARCH." National Research Council. 1983. Research on Sentencing: The Search for Reform, Volume II. Washington, DC: The National Academies Press. doi: 10.17226/101.
×
Page 32
Suggested Citation:"1 MAKING SENSE OF SENTENCING: A REVIEW AND CRITIQUE OF SENTENCING RESEARCH." National Research Council. 1983. Research on Sentencing: The Search for Reform, Volume II. Washington, DC: The National Academies Press. doi: 10.17226/101.
×
Page 33
Suggested Citation:"1 MAKING SENSE OF SENTENCING: A REVIEW AND CRITIQUE OF SENTENCING RESEARCH." National Research Council. 1983. Research on Sentencing: The Search for Reform, Volume II. Washington, DC: The National Academies Press. doi: 10.17226/101.
×
Page 34
Suggested Citation:"1 MAKING SENSE OF SENTENCING: A REVIEW AND CRITIQUE OF SENTENCING RESEARCH." National Research Council. 1983. Research on Sentencing: The Search for Reform, Volume II. Washington, DC: The National Academies Press. doi: 10.17226/101.
×
Page 35
Suggested Citation:"1 MAKING SENSE OF SENTENCING: A REVIEW AND CRITIQUE OF SENTENCING RESEARCH." National Research Council. 1983. Research on Sentencing: The Search for Reform, Volume II. Washington, DC: The National Academies Press. doi: 10.17226/101.
×
Page 36
Suggested Citation:"1 MAKING SENSE OF SENTENCING: A REVIEW AND CRITIQUE OF SENTENCING RESEARCH." National Research Council. 1983. Research on Sentencing: The Search for Reform, Volume II. Washington, DC: The National Academies Press. doi: 10.17226/101.
×
Page 37
Suggested Citation:"1 MAKING SENSE OF SENTENCING: A REVIEW AND CRITIQUE OF SENTENCING RESEARCH." National Research Council. 1983. Research on Sentencing: The Search for Reform, Volume II. Washington, DC: The National Academies Press. doi: 10.17226/101.
×
Page 38
Suggested Citation:"1 MAKING SENSE OF SENTENCING: A REVIEW AND CRITIQUE OF SENTENCING RESEARCH." National Research Council. 1983. Research on Sentencing: The Search for Reform, Volume II. Washington, DC: The National Academies Press. doi: 10.17226/101.
×
Page 39
Suggested Citation:"1 MAKING SENSE OF SENTENCING: A REVIEW AND CRITIQUE OF SENTENCING RESEARCH." National Research Council. 1983. Research on Sentencing: The Search for Reform, Volume II. Washington, DC: The National Academies Press. doi: 10.17226/101.
×
Page 40
Suggested Citation:"1 MAKING SENSE OF SENTENCING: A REVIEW AND CRITIQUE OF SENTENCING RESEARCH." National Research Council. 1983. Research on Sentencing: The Search for Reform, Volume II. Washington, DC: The National Academies Press. doi: 10.17226/101.
×
Page 41
Suggested Citation:"1 MAKING SENSE OF SENTENCING: A REVIEW AND CRITIQUE OF SENTENCING RESEARCH." National Research Council. 1983. Research on Sentencing: The Search for Reform, Volume II. Washington, DC: The National Academies Press. doi: 10.17226/101.
×
Page 42
Suggested Citation:"1 MAKING SENSE OF SENTENCING: A REVIEW AND CRITIQUE OF SENTENCING RESEARCH." National Research Council. 1983. Research on Sentencing: The Search for Reform, Volume II. Washington, DC: The National Academies Press. doi: 10.17226/101.
×
Page 43
Suggested Citation:"1 MAKING SENSE OF SENTENCING: A REVIEW AND CRITIQUE OF SENTENCING RESEARCH." National Research Council. 1983. Research on Sentencing: The Search for Reform, Volume II. Washington, DC: The National Academies Press. doi: 10.17226/101.
×
Page 44
Suggested Citation:"1 MAKING SENSE OF SENTENCING: A REVIEW AND CRITIQUE OF SENTENCING RESEARCH." National Research Council. 1983. Research on Sentencing: The Search for Reform, Volume II. Washington, DC: The National Academies Press. doi: 10.17226/101.
×
Page 45
Suggested Citation:"1 MAKING SENSE OF SENTENCING: A REVIEW AND CRITIQUE OF SENTENCING RESEARCH." National Research Council. 1983. Research on Sentencing: The Search for Reform, Volume II. Washington, DC: The National Academies Press. doi: 10.17226/101.
×
Page 46
Suggested Citation:"1 MAKING SENSE OF SENTENCING: A REVIEW AND CRITIQUE OF SENTENCING RESEARCH." National Research Council. 1983. Research on Sentencing: The Search for Reform, Volume II. Washington, DC: The National Academies Press. doi: 10.17226/101.
×
Page 47
Suggested Citation:"1 MAKING SENSE OF SENTENCING: A REVIEW AND CRITIQUE OF SENTENCING RESEARCH." National Research Council. 1983. Research on Sentencing: The Search for Reform, Volume II. Washington, DC: The National Academies Press. doi: 10.17226/101.
×
Page 48
Suggested Citation:"1 MAKING SENSE OF SENTENCING: A REVIEW AND CRITIQUE OF SENTENCING RESEARCH." National Research Council. 1983. Research on Sentencing: The Search for Reform, Volume II. Washington, DC: The National Academies Press. doi: 10.17226/101.
×
Page 49
Suggested Citation:"1 MAKING SENSE OF SENTENCING: A REVIEW AND CRITIQUE OF SENTENCING RESEARCH." National Research Council. 1983. Research on Sentencing: The Search for Reform, Volume II. Washington, DC: The National Academies Press. doi: 10.17226/101.
×
Page 50
Suggested Citation:"1 MAKING SENSE OF SENTENCING: A REVIEW AND CRITIQUE OF SENTENCING RESEARCH." National Research Council. 1983. Research on Sentencing: The Search for Reform, Volume II. Washington, DC: The National Academies Press. doi: 10.17226/101.
×
Page 51
Suggested Citation:"1 MAKING SENSE OF SENTENCING: A REVIEW AND CRITIQUE OF SENTENCING RESEARCH." National Research Council. 1983. Research on Sentencing: The Search for Reform, Volume II. Washington, DC: The National Academies Press. doi: 10.17226/101.
×
Page 52
Suggested Citation:"1 MAKING SENSE OF SENTENCING: A REVIEW AND CRITIQUE OF SENTENCING RESEARCH." National Research Council. 1983. Research on Sentencing: The Search for Reform, Volume II. Washington, DC: The National Academies Press. doi: 10.17226/101.
×
Page 53
Suggested Citation:"1 MAKING SENSE OF SENTENCING: A REVIEW AND CRITIQUE OF SENTENCING RESEARCH." National Research Council. 1983. Research on Sentencing: The Search for Reform, Volume II. Washington, DC: The National Academies Press. doi: 10.17226/101.
×
Page 54

Below is the uncorrected machine-read text of this chapter, intended to provide our own search engines and external engines with highly rich, chapter-representative searchable text of each book. Because it is UNCORRECTED material, please consider the following text as a useful but insufficient proxy for the authoritative book pages.

1 Making Sense of Sentencing: A Review and Critique of Sentencing Research John Hagan and Kristin Bumiller One of the few certain things about criminal sentencing is that it is an increasingly common subject of empirical research. The bibliography of this paper lists more than 40 studies of sentencing published in the past decade. These studies are notable not only for number but also for their diversity of methods and results. Although early studies of sentencing relied heavily on the use of contingency tables, a striking feature of the past decade has been the widespread application of multivariate techniques, including the development of structural equation models and log linear analyses of the sentencing process. The results have often been provocative, touching most sensitively on issues of racial discrimination in sentencing. The results of recent studies are provocative not only because they raise important issues of equality before the law but also because they frequently appear to contradict one another. To cite only one recent example, while Eisenstein and Jacob (1977:v) conclude from a study of sentencing in Baltimore, Chicago, and Detroit that "blacks are not treated worse than whites . . . ," Lizotte (1978:577) uses some of the same data from Chicago to calculate that ". . . the 'cost' of being a black laborer is an additional 8.06 months of prison sentence. . . ." The purpose of this paper is to critically review these and other findings of sentencing research and to outline directions this research literature might usefully take in the future. 1

2 TWO INCIPIENT THEORIES OF SENTENCING The literature on criminal sentencing is not guided by a dominant theory or set of theories. There have been attempts by sociologists to tie this literature to the debate between consensus and conflict perspectives (see Hagan et al., 1979; Chiricos and Waldo, 1975; Lizotte, 1978) and to link it to a labeling perspective on crime and deviance (Bernstein et al., 1977a). These perspec tives do not have wide currency outside sociology, however, so only a small part of the literature can be tied directly to these theoretical frameworks. We argue in this paper that there are two incipient theoretical orientations implicit in the assumptions that sociolegal researchers bring to this area of work. We believe that an awareness of these two orientations--the individual- processual approach and the structural-contextual approach--is helpful to understanding developments in this research literature. Early sentencing research observed bivariate relation- ships between attributes like race and sentencing outcomes (i.e., type and length of sentence). These studies (e.g., E. Johnson, 1957) were particularly concerned with demonstrating the differential use of the death penalty against blacks in the southern United States. These studies are important today as a signi- ficant source of historical-comparative data; however, legitimate questions have been raised about their tendency to equate correlation with cause in imputing sentencing differentials to discrimination, without controlling relevant "legal" variables (see Green, 1961; Wolfgang and Riedel, 1973). As subsequent studies began to take additional variables into account, initially with tabular techniques, what we call an individual-processual approach to sentencing research began to take form. In the 1960s this research largely used contingency tables to test whether attributes like race remained significantly correlated with sentence outcome when type of offense and/or prior record were held constant. Although this research labored under the inherent liabilities of tabular techniques, particularly problems of controlling for more than one or two variables simultaneously, it served the important function (at least implicitly) of encouraging researchers to develop models of the sentencing process. For example, the burden of Edward Green's (1961, 1964) early and important

3 work on sentencing was to argue that when "legal" variables like offense type and prior record are taken into account, the relationship between race and sentence disappears. Implicit in this argument is the assumption that prior record and current offense mediate (in a causal and sequential sense) the race-sentence relation- ship. Later arguments have focused on whether race linked patterns of offense type and prior record should be taken as reflecting differences in criminal behavior or as reflecting earlier experiences of differential treatment by legal authorities Afar "Y=m~l~ =~^ = and ~wlamr~ 1 IRAQ=` ~ ~-Or ~` I ~-- _~___, ~ ,. Both positions could be correct; what is important for our immediate purposes is that in either case it is assumed that offense type and prior record play a causally intervening role in the process by which judges reach sentencing decisions. Two different types of processes are involved, but each is an example of an individual-processual approach to the Anal v.~i .~ And understanding of sentencing data. .. . . . Mucn ot the sentencing research of the 1970S involved variations and elaborations of individual-processual models of the sentencing process. Most significant in the development of this approach was the introduction of a number of important "case-processing variables" into these models and the application of more sophisticated multivariate techniques in the effort to test the fit of these models with actual case data. Among the new variables considered were pretrial bail decisions (e.g., Bernstein et al., 1977a), plea and charging decisions (e.g., Hagan, 1975c), and the presentence recommendations of probation officers and prosecutors (e.g., Hagan, 1975b; Hagan et al., 1979). These studies made increasingly explicit the premise that sentencing is an end result of a decision-making process that involves offenders moving through a series of potentially important stages in a complex criminal justice system. Farrell and Swigert (1978a:442) make this point well: "The highly structured nature of the judicial system lends itself to a systematic analysis of legal - processing. The discrete ordering of events--the social characteristics of the defendants prior to their entry into the system, their accumulated criminal histories, the type of legal representation, pretrial release, the mode of adjudication, and final disposition--constitutes a series of stages that allows the researcher to assert the causal sequence of relationships." Structural

4 equation models and log linear techniques have provided the technology for modeling this complicated Process. Criminal sentencing is not only a matter of processing individuals through a criminal justice system. Both the individuals and the system occupy variable positions or locations within a social structure, so individual processing decisions can vary by social context. This point has been recognized implicitly in some past sentencing research, and it is made increasingly explicit in recent work. For example, the early studies of capital punishment often attempted to measure variation in the use of capital punishment against individuals across periods of time (E. Johnson, 1957), in different jurisdictions (Bedau, 1964, 1965), and according to whether the crime was interor intraracial (Wolfgang and Riedel, 1973) in character (i.e., interracial crimes represent a conflict across assumed status positions in American society). These studies also suffered from the limitations we have associated with the application of tabular techniques, and they were undertaken with little awareness of one another, thus limiting the full develop- ment of their contextual implications. Since 1977, a number of studies have emerged that begin to exploit the possibilities of a structural- contextual approach. Combining data sets from several jurisdictions, Eisenstein and Jacob (1977), Levin (1977) and Balbus (1973) have linked variations in the political environment to sentencing behavior. Lizotte (1978) has identified the class as well as racial positions of individuals in the social structure and linked these to sentencing outcomes. Most recently Hagan et al. (1980) have distinguished proactive and reactive court organi- of zations and considered their impact on the sentencing white-collar offenders in 10 federal district courts, while Hagan (1982) has examined the consequences of corporate entities compared with individuals acting victim-complainants in the criminal justice process. All of these studies add some feature of structural and contextual variation to their consideration of the individual processing that leads to sentencing decisions. It should be emphasized that what we call the individual-processual and structural-contextual approaches are not mutually exclusive. Indeed, each of these approaches is increasingly persuasive as it includes variables emphasized in the other. For example, it is impossible to be sure whether a political

5 environment correlated with sentencing outcomes is a cause of those differences in outcome unless the variables considered in an individual-processual approach are taken into account. Similarly, it is impossible to know the generalizability of individual-processual variables apart from structural and contextual consider- ations. Thus the limitations on combining these approaches are clearly not conceptual but rather have to do with the availability of comparable kinds of data across contexts. We return to this problem several times in the course of this paper, for it is central to the advancement of this area of work. Meanwhile, we proceed to a discussion of a variety of more specific problems that plague the various kinds of sentencing research we have introduced. DEFINITIONAL PROBLEMS IN SENTENCING RESEARCH Confusion of central concepts has often made the collation of findings from sentencing studies difficult. The most important of these problems has involved the attempt to draw distinctions between legal and extralegal factors in sentencing decisions. Much of the sentencing research of the 1960 s and early 1970s was premised on such a distinction (see Green, 1961). The distinction frequently drawn was that offense seriousness (as indicated by the maximum sentence allowed by law) and prior conviction record (often written into the law as a basis for more severe sentences) were "legal" variables and that race, sex, age, and other characteristics not included in the law were "extralegal." Difficulties with this distinction cut in at least two directions. On one hand it has been noted that what are called legal variables vary from jurisdiction to jurisdiction (i.e., rankings of offense seriousness vary among states), that what is legal at one stage of decision making may not be at another (e.g., community ties may be considered relevant for bail decisions and irrelevant at sentencing), and that what is legal at the sentencing stage (e.g., prior record) may be the product of discrimination at earlier stages (e.g., by the police) (see Bernstein et al., 1977b). On the other hand, it can also be noted that what are called extralegal variables are directly or indirectly built into some parts of the criminal law. For example, probation statutes often include consideration of the offender's age, and there

6 remain some state statutes (e.g., many prostitution laws) that justify differential treatment by sex. Many statutes encourage judicial consideration of an offender's employment record at various stages of the criminal justice process (see, for example, the language of the criminal code bill that passed the Senate Judiciary Committee in the first session of the 19th Congress), a factor that works disproportionately against black offenders. Even though the Fourteenth Amendment to the U.S. Constitution provides that No state shall . . . deny to any person within its jurisdiction the equal protection of the laws," the law seems to provide plenty of latitude to do just that. In sum, the law is an ambiguous guide as to those factors that may legitimately influence sentencing decisions. This issue of legitimacy is complicated further by the fact that it has empirical and moral as well as legal dimensions. The empirical dimension involves the issue of what the American public thinks should influence sentencing, while the moral dimension is concerned with what in some more ultimate sense actually should influence sentencing. We speak to the former rather than the latter issue in this paper, and we deal with it primarily in a contemporary context, focusing first on contemporary American judgments about influences on sentencing. We also note that such judgments may vary across time and place in the social structure (see Hagan and Albonetti, 1982). To offer a specific example, what are thought to be legitimate influences on sentences by most Americans today may be significantly different from what were thought to be legitimate influences earlier in this century in the South. Conceptions of what consti- tutes criminal justice change. To acknowledge the moral and variable nature of what influences on sentencing are regarded as acceptable, we speak in this paper of legitimized and nonlegitimized, rather than of legal and extralegal influences on sentencing, and we regard the content of these categories as the product of ongoing social and legal processes. Legitimized and nonlegitimized influences are those within a given social structure and context that the public thinks should and should not affect sentence severity. Although there obviously is no method for unambiguously sorting all influences on sentencing into these two categories, social survey techniques are one important source of information on what influences are

7 and are not regarded as legitimate by a surveyed population. The top half of Table l-1 presents data on public attitudes toward nine factors that may influence sentencing decisions. These data come from a national survey of American adults interviewed in 1977 to measure their perceptions of and experiences with local, state, and federal courts as well as their more general attitudes toward the administration of justice (see Inter-University Consortium for Political and Social Research, 1979). These data seem to indicate two rather different kinds of concerns. The first is that offenders with a prior record, offenders previously convicted of the same crime, and offenders convicted of a violent crime should receive tougher sentences than offenders who have done none of these things. Well over 80 percent of the respondents endorsed tougher sentencing of these kinds of offenders. The second concern is that whether an offender is well-to-do, poor, or of minority status should have no influence on sentencing; similar levels of support are apparent for this position. In terms of measured attitudes, it is clear that the American public regards prior record and type of offense as legitimate influences on sentencing and that they do not recard - economic and ethnic characteristics as legitimate influences on sentencing. The bottom half of Table l-l summarizes the responses of elite members of the community--i.e., judges, lawyers, community leaders--who were surveyed separately. A similar pattern is apparent. In the remainder of this paper we speak of the variables in Table 1-1 as legitimized and nonlegitimized influences on sentencing decisions. Of course, as we noted above, these data cannot resolve in any ultimate moral sense the issue of whether the legitimacy of variables such as prior record is or is not deserved. In addition, all relevant influences are not considered in this survey. For example, these data do not inform us as to public attitudes toward the influence on sentencing of an offender's employment record. It is also important to note that the data presented in Table 1-1 indicate that the American public as well as some of its elite members believe that neither of its primary concerns is translated fully into the admin- istration of criminal justice in America. In general these data indicate that both groups believe that legitimized factors do not result in sentences as severe

a) · - - .-l c a, u] o u] o a, ·-t as o a a) H a ·,4 a m E~ ~ r oo r · · I r. 0 u. tD r~ L4 ~ ~ ~ O ~ O ~ O ~ 0 ~ 0 U o.- ~ 0 ~ 0 ~0 o S: .- O C: .- U) ~: .- C . ~ .,~ ~ .,~ ~.- ~ {~ ~ ~, u) ~ ~ 4' ~Q. ~ 01 ~ C) Q. 4 Q. ~ ~ ~, 0~ ~ Q. a~ ~ ~2, )~ a, >1 ~ ~ L. a, ~ i4 Q4 O 0 C) ~ ~ ~ ~ .~ U ~ o' - ~ O ~ 1 Q) ~ · - a, ~ ~ ~ ~J ~ ·~ a) ~ 0) 0 ~ a) L' 0) a 0 54 a) ~ ~ a, ~ ~ Q. a' ~ Q) ~ ~ a) O ~ a., ~ ~ ~ ~ ~ e, ·- ~ P4 ~ ~ ~ O ~ P~ ~ C1, P~ O p] t4 ~ ~ ~ O ~, t) ~ ~ O ' - ' - 3 > 3 D4 :~: ~U)

9 as they should be, and that nonlegitimized factors have an influence on sentences that they should not have. For example, while over 80 percent of the respondents thought being well-to-do should have no influence on sentencing, fewer than 30 percent thought this was actually the case. There is good evidence that a substantial part of the American public perceives its system of criminal justice to be unjust, at least to some degree. Table 1-2 makes the additional point that black Americans are particularly likely to perceive minority offenders as receiving tougher sentences than whites (see also Hagan and Albonetti, 1982). The Pearson's r for this relationship is .18. This may not be surprising, but it does help to focus the concerns of this review. The latter finding leads to the final concern of this section: Past research has confused discussions of discretion, disparity, and discrimination in sentencing. For our purposes we regard discretion as the latitude of decision provided by law to someone in imposing a sentence we regard discrimination as a pattern of sentencing regarded as unfair, disadvantaging, and prejudicial in origin; and we regard disparity as a form of unequal treatment that is often of unexplained cause and is at least incongruous, if not unfair and disadvantaging, in consequence. An illustration of the confusion that can occur in the use of these terms is Farrell and Swigert's conclusion (1978a:450) from an important study of the impact of prior offense record on sentencing that "the use of a prior record as meaningful information in the disposition of a criminal case TABLE 1-2 Perceived Influence of Minority Status on Sentencing by Race of Respondent Influence Much Little No Little Much Race Lighter Lighter Influence Tougher TougherTotal White 4.7 14.7 49.0 26.3 5.4 (76) (237) (791) (424) (87)1,615 Black 1.7 3.9 40.4 32.6 21.3 (3) (7) (72) (58) (38)178 NOTE: Ganuna = .45; Pearson's r = .18.

10 compounds the discretion of prior adjudications." This use of the term discretion renders the conclusion of this study unclear: Do the authors mean only that the latitude of decision available to authorities has been increased in a legally acceptable manner? The context suggests that what is really meant is that reliance on prior record as a factor in sentencing institutionalizes a form of discrimination. In any case, our purpose in defining these terms is to make our use of them as unambiguous as possible. METHODOLOGICAL PROBLEMS IN SENTENCING RESEARCH Ideally, social science research is a commutative enterprise: Research problems are refined in their definition, increasing amounts of data are brought to bear, findings accumulate, and knowledge increases. The reality of sentencing research falls far short of this ideal. In this section we consider some of the methodological sources of this shortfall. Alternative Measures of Sentence Severity One problem in the commutation of results from sentencing studies is that they operationalize the dependent variable--sentence--in a variety of ways. The only clear area of agreement on this issue seems to be an implicit consensus that sentences can be ordered in terms of severity; the type of ordering applied, however, varies considerably from study to study. These orderings range from a basic binary division between those offenders sentenced to prison versus those who are not (see, for example, Clarke and Koch, 1976) to Uhlman and walker's 93-point scale that attempts to differentiate in considerable detail ". . . between and among degrees of deprivation of individual freedom and the varying severity of nonprison sanctions" (1980:327 and Appendix). Somewhere between these two approaches is the position taken by Hagan et al. (1979:516), who operationalize the concept of sentence severity by using two different binary codings of the same set of sentence outcomes. The first coding separates prison sentences from all others, while the second coding separates the most lenient disposition in their data set, deferred sentences, from all others. The assumption, confirmed in

11 their subsequent analysis, is that if there is a single dimension of severity, then we should expect to find similar effects of opposite signs when the two coding s are used to examine the determinants of sentencing. our purpose is not to favor one or another of the preceding approaches, but to make clear the diversity of dependent measures that have been used in sentencing studies. In order to cumulate findings from these studies, it is necessary to adopt a common standard, or variety of standards, to be used in some meaningful way across studies. Reanalysis of the original data may be the only definitive way of doing this. Sampling Problems Another problem that complicates the cumulation of results from sentencing studies is the variety of court settings and stages in the criminal justice system at which this research is done. Sentencing studies have been done in federal, state, and municipal courts, drawing samples from locations in the system as early as prosecutors' offices and as late as corrections institutions. There are good arguments for drawing samples from all of these settings and stages. Hagan et al. (1980) argue that federal court samples are important because they include larger numbers of white-collar offenders than lower courts, while Feeley (1979) argues that studies of municipal courts are needed because 90-95 percent of all cases are handled in these lower courts, with the result that "Next to the police, the lower criminal courts play the most important role in forming citizen impressions of the American system of criminal justice" (p. xv). With similar conviction, arguments are made for collecting data relevant to sentencing decisions at various stages of the criminal justice system. On one hand it is suggested that focusing exclusively on offenders sentenced to prison allows consideration of a homogeneous set of dispositions that can be compared in a straightforward way. On the other hand it is suggested that considering a broader range of offenders, to whom various kinds of sentences and other kinds of sanctions are attached, may reduce problems of bias resulting from nonrandom selection processes that may characterize the criminal justice system, from the earliest stages of detection through the imposition of final dispositions.

12 This problem of sample selection bias is dealt with in detail by Klepper et al. (in this volume). We have reason to raise this issue at several points below. Measurement of Legitimized Influences on Criminal Sentencing What we have called legitimized influences on criminal sentencing--for example, type of offense and prior record--should be expected to be strongly related to sentence outcomes. In practice this expectation confronts at least two problems: (1) accurate measurement of these influences and (2) determining how strong a relationship constitutes compliance with public wishes. These problems may be related in that the more measurement error there is, the weaker the observed relationships are likely to be. The issue of measurement error is of further significance in that determining the net influence of illegitimate variables requires effectively holding constant legitimized influences. This cannot be done, of course, without appropriate measurement. We now turn to a discussion of some of the measurements applied. Type of offense has been measured in sentencing studies in a variety of ways, with an equivalent variety of results. The early tabular studies, like that of Green (1961), measured type of offense in terms of legal categorizations, such as burglary and robbery; these different crimes were then considered separately. This tradition has endured (see LaFree, 1980; Bernstein et al., 1977a), but with the increasing popularity of the linear assumptions of regression analysis, researchers have begun to use scales of offense seriousness. One approach takes the maximum sentence allowed by law for an offense as a measure of statutory seriousness (e.g., Hagan et al., 1980). Another approach applies the Sellin-Wolfgang (1964) seriousness scale, based on several aspects of the crime and its victim (Hagan, 1982). A third approach asks judges to indicate how serious they perceive a particular offense presented for sentencing to be (Hogarth, 1971). A fourth approach uses survey responses from the public as to the type and length of sentences they would apply to particular offense descriptions (Blumstein and Cohen, 1980). Although this is not a full enumeration of the types of offense measures that have been applied, these four

13 measures alone have produced quite different results, increasing in the size of their correlation with sentence outcomes in approximately the order they are presented. The findings of prior research may be in part a result of the types of offense measures applied, leaving no clear indication as to what the strength of this legitimized influence actually is. It also must be said that none of these measurement approaches is clearly right or wrong; each approach represents a different, defensible way of conceiving offense seriousness. The concept of offense seriousness is one that indeterminate sentencing laws leave ambiguous, since they allow both public and judicial views to have an effect through the discretion judges have in determining offense seriousness. There may be no definitively correct way of measuring this concept. Similar problems emerge in the measurement of prior record. A variety of alternative measures have been used: presence and absence of prior arrests and/or convictions, number of prior arrests and/or convictions, presence or number of prior felony convictions, most serious prior conviction charge, conviction on the same charge previously, and most serious prior disposition. Again, the type of measure used makes a difference. For example, Wolfgang et al. (1972:227) find in an analysis of juvenile court dispositions that the severity of the prior disposition exerts a substantial influence on outcome. This version of a prior record measure has not been used in adult sentencing studies, and the measures that have been used yield a complicated pattern of results (see Hagan, 1974a). It should also be noted that there are other conceivably legitimate influences on sentencing decisions that have not been included in previous studies. For example, criminal intent, particularly the degree of intent, may be a legitimate consideration at sentencing. Similarly, the type, quality, and quantity of evidence may play a role in sentencing that the public may judge legitimate. Finally, it may be that the sheer bloodiness of a crime may influence sentencing in ways that abstract measures of offense seriousness do not suggest. Again, these other plausibly legitimate influences may be particularly important insofar as they mediate, and thereby serve to justify, the influence of the non- legitimized variables we consider next. Probably in contrast to those who study sentencing professionally, the public seems to be equally concerned with the

9 as they should be, and that nonlegitimized factors have an influence on sentences that they should not have. For example, while over 80 percent of the respondents thought being well-to-do should have no influence on sentencing, fewer than 30 percent thought this was actually the case. There is good evidence that a substantial part of the American public perceives its system of criminal justice to be unjust, at least to some degree. Table 1-2 makes the additional point that black Americans are particularly likely to perceive minority offenders as receiving tougher sentences than whites (see also Hagan and Albonetti, 1982). The Pearson's r for this relationship is .18. This may not be surprising, but it does help to focus the concerns of this review. The latter finding leads to the final concern of this section: Past research has confused discussions of discretion, disparity, and discrimination in sentencing. For our purposes we regard discretion as the latitude of decision provided by law to someone in imposing a sentence; we regard discrimination as a pattern of sentencing regarded as unfair, disadvantaging, and prejudicial in origin; and we regard disparity as a form of unequal treatment that is often of unexplained cause and is at least incongruous, if not unfair and disadvantaging, in consequence. An illustration of the confusion that can occur in the use of these terms is Farrell and Swigert's conclusion (1978a:450) from an important study of the impact of prior offense record on sentencing that "the use of a prior record as meaningful information in the disposition of a criminal case TABLE 1-2 Perceived Influence of Minority Status on Sentencing by Race of Respondent Influence Much Little No Little Much Race Lighter Lighter Influence Tougher Tougher Total White 4.7 14.7 49.0 26.3 5.4 (76) (237) (791) (424) (87)1,615 Black 1.7 3.9 40.4 32.6 21.3 (3) (7) (72) (58) (38)178 NOTE: Gamma = .45; Pearson's r = .18.

15 One response is to argue that socioeconomic status actually is a discrete rather than a continuous variable (Hopkins, 1977); the great mass of incarcerated offenders then falls into a "lower class" and perhaps a few represent a "middle or upper class counterpart." Chiricos and Waldo (1975) demonstrate that measurements based on this continuum approach, manipulated in a variety of ways, yield neither consistent nor significant correlations with length of prison sentences. A common response to this kind of outcome in the stratification literature is to argue that the continuum measure itself, from which these divisions are drawn, is problematic in its conceptualization. The base of the problem, it is suggested, is that mobility and income determination are neither as open nor as fluid as a continuum conceptualization suggests. It therefore is argued that what is needed is a "relational" rather than a "gradational" measure of class position; that is, a measure based on "common positions within the social relations of production" (Wright, 1980:326). This approach distinguishes one class from another largely on the basis of two criteria: ownership of the means of production and purchase of the labor power of others. A second and related approach distinguishes classes on the basis of their relation to authority. This approach, based on the work of Dahrendorf, is concerned with whether authority is exercised over the work of others. Combining these two models, Robinson and Kelley (1979) have conceptualized and measured class positions as a collection of discrete categories. In an analysis of national survey data drawn from the United States and Great Britain, Robinson and Kelley are able to show that all three of the models we have discussed make an independent contribution to the explanation of men's income. The implication is that each of these models could contribute to the explanation of variations in criminal sentencing as well; however, these types of distinctions have not been drawn in the sentencing literature. Another type of nonlegitimized influence that has been considered sporadically in sentencing research is the relationship between the victim and the offender. Early studies concentrated on the race of the offender and the victim (e.g., Garfinkel, 1949). As we note later in this review, there is convincing historical evidence of discrimination in some offender-offense combinations; however, victim-offender relationships are considerably

7 and are not regarded as legitimate by a surveyed population. The top half of Table 1-1 presents data on public attitudes toward nine factors that may influence sentencing decisions. These data come from a national survey of American adults interviewed in 1977 to measure their perceptions of and experiences with local, state, and federal courts as well as their more general attitudes toward the administration of justice (see Inter-University Consortium for Political and Social Research, 1979). These data seem to indicate two rather different kinds of concerns. The first is that offenders with a prior record, offenders previously convicted of the same crime, and offenders convicted of a violent crime should receive tougher sentences than offenders who have done none of these things. Well over 80 percent of the respondents endorsed tougher sentencing of these kinds of offenders. The second concern is that whether an offender is well-to-do, poor, or of minority status should have no influence on sentencing; similar levels of support are apparent for this position. In terms of measured attitudes, it is clear that the American public regards prior record and type of offense as legitimate influences on sentencing and that they do not regard economic and ethnic characteristics as legitimate influences on sentencing. The bottom half of Table 1-1 summarizes the responses of elite members of the community--i.e., judges, lawyers, community leaders--who were surveyed separately. A similar pattern is apparent. In the remainder of this paper we speak of the variables in Table 1-1 as legitimized and nonlegitimized influences on sentencing decisions. Of course, as we noted above, these data cannot resolve in any ultimate moral sense the issue of whether the legitimacy of variables such as prior record is or is not deserved. In addition, all relevant influences are not considered in this survey. For example, these data do not inform us as to public attitudes toward the influence on sentencing of an offender's employment record. It is also important to note that the data presented in Table 1-1 indicate that the American public as well as some of its elite members believe that neither of its primary concerns is translated fully into the admin- istration of criminal justice in America. In general these data indicate that both groups believe that legitimized factors do not result in sentences as severe

17 those characteristics by which aggregative units differ in their composition and are known to be correlated with and antecedent to the dependent variable. In a manner that Hauser (1970:14) skeptically calls the "method of residues," residual covariation of the contextual variable with the dependent variable is then identified as the effect of the context. Hauser points out that gross aggregate differences in a dependent variable are usually small when compared with the total variability of that variable, and that when other independent variables that must be controlled are held constant, aggregate differences in levels of the dependent variable are diminished further. This point should be kept in mind when considering the variation explained "contextually n in aggregate studies of jurisdictional differences in sentencing. There is also the problem of correctly interpreting these contextual effects. As Hauser points out, the choice of a variable to represent a context is arbitrary in the sense that most contextual variables are intercorrelated and thus can generate similar residual differences among groups. The problem is that the particular contextual variable chosen affects the interpretation of residual group differences. Furthermore, this interpretation will usually rest on equally arbitrary assumptions about the effects of unmeasured variables. For example, in a national sentencing study one can readily imagine the use of region as a contextual variable, with residual differences associated with this variable being interpreted as the product of differences in regional culture. A problem with this interpretation, however, is that regional culture is unlikely to be measured directly. The interpretation of a contextual effect, then, can be no more secure than the measurements on which it is based, and these measurements are often indirect in character. Finally, it should be noted that time as well as place can be a source of contextual variation. As we will see below, Thomson and Zingraff (1981) and Hagan and Bernstein (1979) have demonstrated that the practice of aggregating sentencing data over time to construct larger data sets can mask important shifts in sentencing patterns. Similarly, Greenberg (1977:175) notes that sentencing practices may be sensitive to long-term social trends, such as the growing number of blacks in the judiciary and increased public concern about racial

5 environment correlated with sentencing outcomes is a cause of those differences in outcome unless the variables considered in an individual-processual approach are taken into account. Similarly, it is impossible to know the generalizability of individual-processual variables apart from structural and contextual consider- ations. Thus the limitations on combining these approaches are clearly not conceptual but rather have to do with the availability of comparable kinds of data across contexts. We return to this problem several times in the course of this paper, for it is central to the advancement of this area of work. Meanwhile, we proceed to a discussion of a variety of more specific problems that plague the various kinds of sentencing research we have introduced. DEFINITIONAL PROBLEMS IN SENTENCING RESEARCH Confusion of central concepts has often made the collation of findings from sentencing studies difficult. The most important of these problems has involved the attempt to draw distinctions between legal and extralegal factors in sentencing decisions. Much of the sentencing research of the 1960 s and early 1970s was premised on such a distinction (see Green, 1961). The distinction frequently drawn was that offense seriousness (as indicated by the maximum sentence allowed by law) and prior conviction record (often written into the law as a basis for more severe sentences) were "legal" variables and that race, sex, age, and other characteristics not included in the law were "extralegal." Difficulties with this distinction cut in at least two directions. On one hand it has been noted that what are called legal variables vary from jurisdiction to jurisdiction (i.e., rankings of offense seriousness vary among states), that what is legal at one stage of decision making may not be at another (e.g., community ties may be considered relevant for bail decisions and irrelevant at sentencing), and that what is legal at the sentencing stage (e.g., prior record) may be the product of discrimination at earlier stages (e.g., by the police) (see Bernstein et al., 1977b). On the other hand, it can also be noted that what are called extralegal variables are directly or indirectly built into some parts of the criminal law. For example, probation statutes often include consideration of the offender's age, and there

19 (3) marital status, drug/alcohol dependence, prior criminal record. Case-Processing Attributes: Bail status, attorney type, plea, presentence report, recommendations of prosecutors and probation officers. (4) Contextual Attributes: Court characteristics, case load, identity of judge, community characteristics, social change over time. Unfortunately most of these variables are not consistently measured and considered in the sentencing literature. And many of these studies do not present their data in a form that can be readily cumulated with other studies for the purposes of reanalysis. For example, some studies report only mean scores that cannot be accumulated across studies (e.g., Tiffany et al., 1975; Perry, 1977; Levin, 1972), other studies using correlation and regression techniques frequently do not report zero-order correlations (e.g., Nardulli, 1979; Chiricos and Waldo, 1975; Kelly, 1976; Eisenstein and Jacob, 1977), and studies based on multivariate categorical techniques often do not report comparable statistics or a full cross-tabulation of their variables (e.g., Unnever et al., 1980; Burke and Turk, 1975; Uhlman and Walker, 1979, 1980; Thomson and Zingraff, 1981). This point is not made as a criticism of these studies; they were not done for the purpose of facilitating our review. The cumulation of these data sets in a central data bank would allow those interested in reanalysis to overcome some of these problems. Lacking immediate access to these data sets, we must work with what is provided in the form of published results. The possibilities that published results provide vary according to the attribute considered. For example, the best prospects for cumulation involve consideration of the offender's race: 51 studies reach conclusions about the relationship between the offender's race and sentence (these studies are described in greater detail in Appendix A). No other attribute benefits from such a large pool from which to draw conclusions. For this as well as for other attributes, however, there remains the issue of how cumulative conclusions can best be drawn. The most common approach to this problem takes a narrative form, in which the reviewer uses his or her own judgment to weigh the findings against one another. We too rely on judgment, but we first organize our

3 work on sentencing was to argue that when "legal" variables like offense type and prior record are taken into account, the relationship between race and sentence disappears. Implicit in this argument is the assumption that prior record and current offense mediate (in a causal and sequential sense) the race-sentence relation- ship. Later arguments have focused on whether race linked patterns of offense type and prior record should be taken as reflecting differences in criminal behavior or as reflecting earlier experiences of differential treatment by legal authorities (fair -~;~mnl~ Am 1i':~rrml 1 and Swiaerr. 1 C~7~;q~ ~ ~ ~ ~ ...= _ _, ~ - _ ~ . ~ , ~ ,. Both positions could be correct; what is important for our immediate purposes is that in either case it is assumed that offense type and prior record play a causally intervening role in the process by which judges reach sentencing decisions. Two different types of processes are involved, but each is an example of an individual-processual approach to the analysis and understanding of sentencing data. Much of the sentencing research of the 1970 s involved variations and elaborations of individual-processual models of the sentencing process. Most significant in the development of this approach was the introduction of a number of important "case-processing variables" into these models and the application of more sophisticated multivariate techniques in the effort to test the fit of these models with actual case data. Among the new variables considered were pretrial bail decisions (e.g., Bernstein et al., 1977a), plea and charging decisions (e.g. , Hagan, 1975c), and the presentence reconunendations of probation officers and prosecutors (e.g., Hagan, 1975b; Hagan et al., 1979). These studies made increasingly explicit the premise that sentencing is an end result of a decision-making process that involves offenders moving through a series of potentially important stages in a complex criminal justice system. Farrell and Swigert (1978a:442) make this point well: "The highly structured nature of the judicial system lends itself to a systematic analysis of legal processing. The discrete ordering of events--the social characteristics of the defendants prior to their entry into the system, their accumulated criminal histories, the type of legal representation, pretrial release, the mode of adjudication, and final disposition--constitutes a series of stages that allows the researcher to assert the causal sequence of relationships." Structural

21 TABLE 1-3 Cross-Classification of Control for Offense and Prior Record by Finding of Racial Discrimination and Time Period To 1968 1969 on No Discrimi- Discrimi- No Discrimi- Discrimi nation nation Total nation nation Total No Control 21.4% 78.6% 56.0% 33.4% 66.6% 23.1% for Offense (3)a tll)b (14) (2)c (4)d (6) and Record Control 72.7% 27.3% 44.0% 50.0% 50.0% 76.9% for Offense (8)e (3)f (11) (lo)g (1o)h (20) and Record Total 44.0% 56.0% 100.0% 46.2% 53.8% 100.0% (11) (14) (25) (12) (14) (26) aBedau, 1965~; Bensing and Schroeder, 1960; Conklin, 1972. bBedau, 1964; Bowers, 1974; Bullock, 1961; Garfinkel, 1949; Gerard and Terry, 1970; Johnson, 1957; Martin, 1934; Partington, 1965; Wolf, 1964; Wolfgang et al., 1962; Wolfgang and Reidel, 1973. CAtkinson and Newman, 1970; Perry, 1977. argan and Coates, 1974; Uhlman, 1979; zalman et al., 1979; Zimring et al., 1976. eBaab and Furgeson, 1968; Burke and Turk, 1975; Farrell and Swigert, 1978b; Green, 1961, 1964; Judson et al., 1969; Levin, 1972; Mileski, 1971. fLemert and Rosberg, 1948; Nagel, 1969; Tiffany et al., 1975. gBernstein et al., 1977a; Chiricos and Waldo, 1975; Clarke and Koch, 1977; Eisenstein and Jacob, 1977; Feeley, 1979; Hagan et al., 1979; Hagan et al., 1980; McCarthy, 1979; Pope, 1975a; Shane-Dubow, 1979. hClarke and Koch, 1977; Gibson, 1978; Hagan and Bernstein, 1979; Kelly, 1976; LaFree, 1980; Lizotte, 1978; Myers, 1979; Pope, 1975b; Thomson and Zingraff, 1981; Unnever et al., 1980. second time period that control for offense and record are more likely than those from the first period (with similar controls) to find discr iminatic~n ~ Sn n "rc1-nE compared with 27.3 percent). Although studies from the second period are still evenlY divided between those that do and do not find discrimination, the fact that an increased proportion of these studies conclude discri- mination deserves an explanation. One possible explanation we elaborate below is that, as researchers have increased their use of multivariate techniques facilitating the control of legitimized variables like offense and prior record, they also have focused more selectively on those structural and contextual conditions that are most likely to result in racial discrimination. We make this point below by individually considering the 10 studies based on data from 1969 on that find evidence of racial discrimination with offense and record controlled. It is important before turning to this discussion to repeat that the

1 Making Sense of Sentencing: A Review and Critique of Sentencing Research John Hagan and Kristin Bumiller One of the few certain things about criminal sentencing is that it is an increasingly common subject of empirical research. The bibliography of this paper lists more than 40 studies of sentencing published in the past decade. These studies are notable not only for number but also for their diversity of methods and results. Although early studies of sentencing relied heavily on the use of contingency tables, a striking feature of the past decade has been the widespread application of multivariate techniques, including the development of structural equation models and log linear analyses of the sentencing process. The results have often been provocative, touching most sensitively on issues of racial discrimination in sentencing. The results of recent studies are provocative not only because they raise important issues of equality before the law but also because they frequently appear to contradict one another. To cite only one recent example, while Eisenstein and Jacob (1977:v) conclude from a study of sentencing in Baltimore, Chicago, and Detroit that "blacks are not treated worse than whites . . . ," Lizotte (1978:577) uses some of the same data from Chicago to calculate that ". . . the 'cost' of being a black laborer is an additional 8.06 months of prison sentence. . . ." The purpose of this paper is to critically review these and other findings of sentencing research and to outline directions this research literature might usefully take in the future. 1

1 23 particularly likely to occur. Ethnic and racial hostility has long been thought to be a factor in American drug policy (MUSto, 1973; Bonnie and Whitebread; 1974), and the manifestation of this hostility in sentencing patterns might therefore be expected. Gibson (1978) This study focuses on 11 judges who sentenced 1,219 felony cases in the superior court of Fulton County (Atlanta, Georgia) from March 1968 to October 1970. A unique feature of this study is its attention to the sentencing behavior of individual judges. The index of discrimination in sentencing used attempts to capture the differential severity of sentences imposed by individual judges to black and white offenders net of legitimized criteria. The results reveal differences among judges that are masked in the aggregate. Gibson concludes that " . blacks are the victims of discrimination by some judges but the beneficiaries of discrimination by others" (p. 470). Gibson uses interview responses to show that the antiblack judges are tied strongly to traditional southern culture, concerned about crime, prejudiced against blacks, and relatively punitive in their sentencing philosophies; in addition they tend to rely more heavily on the defendant's attitude and prior record in making their sentencing decisions. Gibson is cautious in framing his conclusions, noting the limitations that a sample of 11 judges, only three of whom were clearly discriminatory, imposes. Nonetheless, it is interesting to note a parallel between this and a Canadian study by Hagan (1975a): namely, that the finding of discrimination is specific to a subset of judges considered. Only in this context is a culturally based finding of discrimination revealed. · - Hagan and Bernstein (1979) Using data from the 14-year period 1963-1976, this study analyzes the sentences imposed on 238 persons for selective service violations in one of America's largest cities. Unique features of this study are its identi- fication of two different social and political contexts in which these cases were sentenced and an examination of the influence of race within these contexts independent

/ Research on Sentencing: The Search for Reform

25 Oklahoma, black offenders receive longer sentences than whites for burglary. Blacks do not receive significantly longer sentences for homicide than whites, but Mexican- American and Indian offenders do receive shorter sentences than whites convicted of this offense. Although these findings are context-specific, as Kelly (p. 248) notes, The racial and ethnic differences by crime type are difficult to explain." LaFree (1980) The sample for this study consists of 881 suspects charged with "forcible sex offenses" in a large midwestern city between January 1970 and December 1975. Two unique features of this study are (1) its focus on the racial composition of the victim-defendant dyed rather than on attributes of the defendant or victim taken separately and (2) its consideration of a series of official processing decisions in these cases, from initial police reports to final dispositions. A variety of dependent measures are used. "To the extent that relationships between men and women in America are still defined by race-specific rules of sexual access, H LaFree (p. 843) argues, "an implicit ordering of official reactions to sexual assault by race of the victim and offender is suggested." Results of a stepwise multiple regression analysis confirm this expectation: compared with other defendants, black men who assaulted white women received (1) more serious charges and (2) longer sentences and were more likely to (3) have their cases filed as felonies, (4) receive executed sentences, and (5) be incarcerated in the state penitentiary. Overall, the inclusion of racial composition substantially improves the prediction of outcomes, the greatest increases occurring for later sentencing outcomes. These findings are certainly not unique to this study: Even stronger relationships between racial composition of the victim-defendant dyed and sentencing are apparent in studies focusing on rape cases in earlier eras (see Wolfgang and Reidel, 1973; Partington, 1965). The long history of black-white sexual segregation in the United States makes sexual assault cases one of the most likely structural contexts in which racial discrimination will be found.

Preface major administrative and secretarial support throughout the work of the panel; and her dedication was notable. Jane Beltz provided comparable support at Carnegie-Mellon University. We would also like to express our appreciation to the National Institute of Justice. Robert Burkhart and Cheryl Martorana of the institute attended most of the meetings of the panel and were most helpful in providing advice and information on the institute's program on sentencing research. ALFRED BLUMSTEIN, Chair Panel on Sentencing Research xiii

27 Myers' research are connected, in that although the racial composition of the victim-offender dyed did not notably affect sentencing directly, it did have a significant effect indirectly, through probation _ _ ~ ~ officers' recommendations. The ultimate effect was greater leniency for blacks convicted of victimizing blacks compared with white-white and black-white dyads, net of other legitimized factors. The decision to imprison was the dependent measure for this analysis. As Myers (p. 530) notes, the subset of cases considered in this analysis represents a small minority of all cases sentenced: Most cases do not go to trial, and the restriction to cases with specified individual victims further narrows the focus. Nonetheless, the indicated role of probation officers' recommendations and their liability to bias has been noted before (see Hag an, 1975b; Unnever et al., 1980, below), and we have discussed the influence of racial composition above (e.g., LaFree, 1980). Both of these variables may identify structural contexts in which the likelihood of racial discrimination is increased. Pope (1975) This study is based on 32,694 offenders sentenced in 12 counties in California between 1969 and 1971. Unique features of this study are (1) its separate consideration of sentencing decisions in urban and rural counties and (2) its separate use of type (e.g., probation, jail, prison) and length (i.e., length of confinement or probation) of sentence as dependent variables. The technique used to take account of legitimized influences in this analysis is test factor standardization. When type of sentence is the dependent measure, the results of this analysis indicate that rural courts sentence blacks more severely than whites. Blacks sentenced by rural courts, for example, were substantially more likely to be confined and less likely to obtain a probation disposi- tion. Bivariate differences by race in urban areas disappeared when legitimized influences were taken into account. When length of sentence was used as the dependent measure, no racial differences were evident for either urban or rural areas. The latter finding may be a consequence of sample selection bias. In any case, the former finding of discrimination in rural but not urban settings has a parallel in at least one other, Canadian

Preface The Panel on Sentencing Research is an outgrowth of the ferment that significantly affected sentencing practice in the 1970s. That ferment is reflected in a variety of sentencing "reforms," many of which had their roots in research, much of which involved technical questions of some complexity. The Panel on Sentencing Research was established in September 1980 to review that research on sentencing and its impact. The panel was created in response to a request from the National Institute of Justice to the National Academy of Sciences, as a panel of the Committee on Research on Law Enforcement and the Administration of Justice of the Commission on Behavioral and Social Sciences and Education of the National Research Council. The panel's task was to assess the quality of the available re- search, to indicate how the application of research techniques could be improved, and to suggest directions for future research, especially that supported by the National Institute of Justice. To address this range of issues, the panel was composed of specialists representing a variety of academic disciplines, methodological approaches, and operational exper . . . . . terse In t he criminal Justice system. The issue of sentencing is very broad, and so the panel very early had to limit the scope of its work. Much of the public concern over sentencing relates to its effects on crime, but those effects were explicitly excluded from the panel's efforts because two other panels of the Committee on Research on Law Enforcement and the Administration of Justice- the Panel on Research on Rehabilitative Techniques and the Panel on Research xi

29 ratio corresponding to the chances of probation compared with prison for whites is predicted to be 2.3 times that for blacks. Unnever et al. also find that when a control is introduced for the inclusion of probation officer's recommendations in the final disposition equation, the effect of race is reduced substantially. The implication, as noted above in Myers' study (1979; see also Hagan, 1975b), is that the incorporation of probation officers in the sentencing process can result in discrimination by race. We have made the point elsewhere (Hagan et al., 1979) that the historical process by which probation officers were included in the sentencing process represents a structural change in the American courts with significant symbolic consequences. This study illustrates that the implications of this structural change can be instrumental as well, in this case with racial consequences. We have talked in some detail about recent studies that have found evidence of racial discrimination and about the structural and contextual conditions that may give rise to these findings. _ ~. ~ . It is important to note as we. ~ snag one or those studies that does not find evidence of racial discrimination since 1969 does report class-linked disparities in sentencing. This study (Clarke and Koch, 1976) analyzes the experiences of 798 burglary and larceny defendants in the criminal courts of Mecklenburg County, North Carolina, in 1971. Clarke and Koch report that both income and race have a substantial first-order relationship to whether a defendant goes to prison, but while income continues to show at least a small effect on this outcome when other variables are taken into account, race does not. It is also shown that most of the effect of income on sentence is mediated by two variables: bail status and defense counsel. In the county under study, the only alternative to pretrial detention for most defendants was bail bond, which meant depositing in cash the full amount of the bond set for the offense charged or obtaining a professional bondsman as surety in exchange for a nonreturnable fee. An assigned counsel system was used in the county to provide private representation for defendants who could not retain private counsel. Clarke and Koch (pp. 83-84) found that in the county considered ". . . most of the influence of income on the likelihood of imprisonment among the defendants studied is explained by poorer opportunity of the low-income defendant for bail and his

Contents Volume PREFACE SUMMARY Introduction Determinants of Sentences Structuring Sentencing Decisions Assessment of the Effects of New Sentencing Policies Sentencing Policies and Prison Populations Research Agenda 1 INTRODUCTION: SENTENCING PRACTICES AND THE SENTENCING REFORM MOVEMENT The Processes That Constitute Sentencing The Goals of Criminal Sanctions American Sentencing in Comparative and Historical Perspective Scope of This Report 2 DETERMINANTS OF SENTENCES Issues Findings Conclusion ix

31 Interestingly, the increased tendency to control for legitimized variables in sentencing studies has not resulted in fewer findings of racial discrimination. In fact, the more recent studies that include these controls are more likely to report discrimination than the older studies with similar controls; the recent studies are equally divided between those that do and do not conclude discrimination. The challenge is to explain why some studies find discrimination while others do not, and why among those studies including controls for legitimized variables the proportion finding discrimination has shown s igns of increasing. Our explanation is that with increasing sensitivity, those researchers who find evidence of discrimination have specified for study structural contexts in which discrimination by race is most likely to occur. This type of work marks the emergence of what we have called a structural-contextua 1 approach in sentencing research. It also can be noted that this approach has roots in an earlier period. In the earlier part of this century a large number of studies both with and without controls for offense and prior record found evidence of racial discrimination in the use of the death penalty, particularly in the South (Bedau, 1964; E. Johnson, 1957; wolf, 1964; Wolfgang et al., 1962; Wolfgang and Reidel, 1973). Notwithstanding their incomplete control of legitimized variables, many of these studies found such large relationships between race and sentence outcome that it is difficult to imagine that further control would have eliminated evidence of racial discrimination (e.g., Wolfgang and Reidel, 1973; Hagan, 1974a). The use of the death penalty early in this century in the South represents one structural context in which racial discrimination seems clearly to have occurred; however, the declining use of the dearth penalty in this century has diminished the importance of this context for our i~Tunediate concerns. Thus we have focused most of our attention in this review on mar e recent studies. S tudies with data sets drawn over the last decade have identified a number of structural contexts in which racial discrimination seems to persist. A number of these studies reveal racial discrimination, for example: in rural but not in urban settings (Pope, 1975a; Hagan, 1977); among judges with culturally linked prejudicial attitudes (Gibson, 1978; Hagan, 1975a); for crimes like rape (LaFree, 1980) and robbery (Thomson and Zingraff, 1981) that are interracial (see also Myers, 1979); among

Contents PRE FACE 1 MAKING SENSE OF SENTENCING: A REVIEW AND CRITIQUE OF SENTENCING RESEARCH John Hagan, Department of Sociology and Faculty of Law, University of Toronto Kristin Bumiller, Department of Sociology, University of Wisconsin DISCRIMINATION IN THE CRIMINAL JUSTICE SYSTE M: A CRITICAL APPRAISAL OF THE LITERATURE Steven Klepper, Department of Statistics, Carnegie-Me/lon University Daniel Nagin, Pennsylvania Department of Revenue, Harrisburg Luke-Jon Tierney, Department of Statistics, Carnegie-Mellon University ·e V11 X1 55

33 weak. Indeed, the relationships for recent studie s g enerally are weaker than the relationship noted early in this paper between the race of surveyed respondents and their tendency to believe that the courts discriminate by race. The weakness of the race-sentence relationship is not necessarily surprising. An important feature of the individual-processual approach is its conceptualization of race as an exogenous variable exercising its influence through an extended causal chain that includes such intervening variables as offense type, prior record, bail status, and recommendations by various control agents. S ince we ordinarily expect that the largest correlations will occur between adjacent variables in such a chain, and since all of these correlations are assumed to be less than perfect, we should therefore expect that the smallest correlations will occur between those variables that are furthest removed from each other in the causal chain (Blalock, 1964): i. e., race and sentence. It is also important to note that the influence that race may have on earlier decisions such as pretrial detention may be even more punishing that the final sentence imposed (see Feeley, 1979). Meanwhile, the increasing attention that has been- given to this causal chain has confirmed that other variables do have a regular and importan t impact on sentence. In fact, there is considerable evidence to confirm that the closer one gets to sentence in such causal chains, the stronger the observed correlations become. Thus it has been noted that presentence recommendations by probation officers and prosecutors exhibit a substantial relationship to sentence (Hagan, 1975b, 1977; Hag an et al., 1979; Myers, 1979; Unnever et al., 1980) and that judges' perceptions of offense seriousness and offender culpability are very strongly related to sentence (Hogarth, 1971). Indeed, in the latter case the relationships are so strong that one may reasonable begin to question the conceptual and methodological separateness of the independent and dependent variables. Nonetheless, the pattern of relationships is consistent with the type of causal chains implied by the individual-processual approach. This approach to the issue of race and sentencing is also important because it can alert us to the ways in which mediating variables like employment can be incorporated in a disadvantaging way into the sentencing process. We have noted that employment has a legally mandated role to play in bail decisions and that

COMMITTEE ON RESEARCH ON LAW ENFORCEMENT AND THE ADMINISTRATION OF JUSTICE ALFRED BLUMSTEIN (Chair), School of Urban and Public ~Affairs, Carnegie-Mellon University LEE P. BROWN, Chief of Police, Houston, Texas JOSEPH B. KADANE, Department of Statistics, Carnegie-Mellon Uni versity SAMUEL KRISLOV, Department of Political Science and Law School, University of Minnesota RICHARD LEMPERT, Law School, University of Michigan NORVAL MORRIS, Law School, University of Chicago RICHARD D. SCHWARTZ, College of Law, Syracuse University LEE B. SECHREST, Center for Research in the Utilization of Social Knowledge, Institute for Social Research, University of Michigan JUNE STARR, Department of Anthropology, State University of New York, Stony Brook JACK B. WEINSTEIN, U.S. District Court, Brooklyn, New York JAMES Q. WILSON, Department of Government, Harvard University ANN WITIE, Department of Economics, University of North Carolina MARVIN E. WOLFGANG, Department of Criminology, University of Pennsylvania v

35 sample selection are, one has to admit that among them are the variable characteristics of the settings in which the selection processes occur. Indeed, variation in the selection process across variable structural contexts is an intriguing and important topic for study. Ignoring this kind of structural variation would simply constitute a macrolevel source of sample selection bias. The important research on race and sentencing of the future will involve individual-processual analyses of the sentencing process that are also able to take the types of structural and contextual variation we have discussed into account. This research will require data that allow consideration not only of the characteristics of individuals and their cases as they move through the criminal justice system, but also of the structural and contextual conditions in which they are processed; in other words, data collected on individual offenders across stages and settings. This kind of data is costly to collect but necessary for full consideration of the types of theoretical issues we have raised. Sentencing is an outcome of the contextualized processes in which it occurs.

PANEL ON SENTENCING RESEARCH ALFRED BLUMSTEIN (Chair), School of Urban and Public Affairs, Carnegie-Mellon University SYLVIA BACON, Superior Court of the District of Columbia RICHARD A. BERK, Department of Sociology, University of California, Santa Barbara JONATHAN D. CASPER, Department of Political Science, University of Illinois, Urbana JOHN C. COFFEE, JR., School of Law, Columbia University SHARI S. DIAMOND, Department of Psychology, University of Illinois, Chicago Circle FRANKLIN M. FISHER, Department of Economics, Massachusetts In stitute of Technology DON M. GOTTFREDSON, School of Criminal Justice, Rutgers Univer- sity JOSEPH B. KADANE, Department of Statistics, Carnegie-Mellon Uni- versity NORVAL MORRIS, Law School, University of Chicago DAVID J. ROTHMAN, Department of History, Columbia University RUTH L. RUSHEN, Department of Corrections, Sacramento, California JAMES Q. WILSON, Department of Government, Harvard University SUSAN E. MARTIN, Study Director DIANE L. GOLDMAN, Administrative Secretary JACQUELINE COHEN, Consultant, School of Urban and Public Affairs, Carnegie-Mellon University MICHAEL H. TONRY, Consultant, School of Law, University of Mary- land ·~e

37 l U} U) tG 1 ~ ~ o o, _1 ~ 4 aJ ~ ~ 4 a' r _~ 4 0 ~ ~ q~ ·- ~ O 4J ~ C Q' ~ O 4 ~ ·r1 ~ U] 4 ~ ~, >t ~ ~ C .,' O ~ ~ ·,4 >~ 4 S O O ~ 4d a, Q) C 0) ~ 4 . - a) 4 O O a) ~ c: tU ~ 4 . - 4 Ol ~ ~ U] - Q. ~ _ L. C ~U] ' ~ O QJ ~C C ~ C ~ C 4 ~ ·0 - ~- - ·,1 · - 4 ~ 4 a U] ~ V 4 0 4 ~ 0) ·- ~ ~ 0.~ al U) ·- a ~a, ~ Q n~ 4 U1 3 ·- Q~ ~ ~> ~ E 4 ~ ~ ~ ~ 4 O >, ~ >' ~ 4 C S O ~ ~ 4 S ~U1 V C [g tT' o Q~ 0 4 (0 C Ul O C dJ U) ·- (g ~ 4 4-J 0 4 ~ C 4 a) J~J to OJ > 0 U) 1 >1 q~ 1 a) O ~ 1 a) ~ lq _/.- ~ 4 C Y Q. 4 0' O O Q' 4J ~= m~- ~ O >~ u' 3 s ta .,..,. ~ s JJ ~ ~ s · - 4 m. - > ~ ~ o Q (U · - u' u' s ~a 4J ~ O QJ U) U] ' C C U) · - Q} ' 4~ 4J IJ > 4J ~ C ~ o ~ C 01 al ~ a ta ~ ~ ~ 0 4 rn a, ~ q~ 4 C O C O (L~ QJ 3 Q~ a~ Q. 4 Q. 4 4 U] ~ >' 4 ~ ~ > S C O ~ O C ~ ~ C ~J 4 4 O C O aJ o Q) ~ U) =1 y 4 C U ~ ~ ~ '13 ~ ~ Q) _~ Q~ N q~ N Q ~ · - O · - C C QJ 4 ,~ ~ C O S 4~ -' · - ~ ~ n5 4 4 aJ s a, ~ O ~ C C Q ~ >' 0, 4 0 U] a, 4J Q' c Y U) ~ ~ O C C C .,. ·,1 ·rl 0 -~ ~0 ~ . - ~ · - 1 JJ ·~1 ~ U] U. ~ 4 ~ O u~ .- O ~ {O 4 U) ~ U} -~ tt5 · - 4 ·- O > ~ O er ~D 1 ~1 1 1 co ~r O O 11 11 11 1 1 4 ~1 t.~) l C 4 · - ' ~ O 4~ ~ >t~ N C ~ ~ U] ~ ~W ~C ~ · - 0 0 ·O ~ 4 ~ _Io nW C · - ·~/ ~ u' O lU· - 4 ~ C 4 C 4 U] >, Co~ Q O ~w wl ~ ~w ~ 4 ~W 4 · ~1 4 ~ ~ > ~ ~ ~4 ~ ~ ~1 4 tO O C · - O 4 .,' U) ° X w O ~1 ns 4~ad 1 1 w ~wnw OO (1505 44 w U] C w IL~ w C n l ~w O V ~ q~ Q o U] ~. CO c m U, 0 4 w, · - · - ~w X ~ 4~ ~ ~W w Q. O ~ w N ·,' S o V ~ ~S C C o ~ ~W . ~ U) ~ ~ Wl >, O IJ C · - 4~ ~W C > ~ ~W ~W O U] U] U] 11 u~ C I ~C ' ~ o Enw o >' c -I O ~· - 4 ~ ,C ~4 U] ~ U) ~- 3 ~ ~1 ~ ~W U~ U' ~W ~ ~W O O ~ ~ ~W ~ ~W · - ~- ~w ~ 4 ~ 4 · - ' C ' X ' ~ (d 3 ~ · - O ~ O ~ w ~ ~q 4 ~ O S ' u, 1- 1 · - · - ~D o C 3 ~ ' oi 4 ai ~Iw ~> >' ~ O ~ X ~ U) ~ 1C 4J 1 >d 1 W' U1 ~U] r~,0 C 0 ~r C ~ ~ i~ wl 0D ~ >' ~ · - ~- 3 ~D ~ O ~ ~ ~J ~in X :, O S a, w ~c° _I ~ C JJ ~a a ~ wl O ~ z ~ ~a ~ ~ ~ ~ ,0 ~ ~ Wl ~ _~ >~ 4 C ~ w 4 {O V ~ · - ._~ ~ ~ ~W U) · - ~ Y · - ~ ~ O U] 4 ~ ·-' 0 U) C _I a~ ~w · - Q ~w m 0, 4 · - ~ >~1 - ~W ~ ~ ~ O S r~ JJ ~ C C ~ ~ C -- - ~ O · - O ~ · - O C ~ C U) ~ ~W ~ · - ~W W 4 dJ ilw C J~ 4J ~w C O ~ C · - > w! rd ~ ~w S O U) 4 U) U) 3 4 C O O a ~ · - ~ C 4 ~J w 0 ~ ~a O ·~/ 4 4 ~ ~ O 4 ~ ~ V ~a W o ~ o O ~ C ~a O ul . - w U} C ~w nw (Li a, >~ 0 4J V 4' S C O O ~ 4 · - O S 4 C 4~1 Q4 w o 4 w >~ JJ Aw ·~' 4 o >1 O ~ C · - nW 4 ~w JJ w C > ~ A I ~w O U) ~ U] 1 u' C C o o U ·V O ~ ' 1 - > C O V > ~E 4 C~' 4 ~ ~ I_ ~W O O a, E ~ o o O >. 1 ~ v ' U] V C O 1- C ~ · - ~w a, 0 0 c ~ ~: C) O e ~ a i.d c a ~c . - ~C {o a, ~ _ y ~ ~ ~ _ ~ 0 ~ _ o - - C ~ _ ·- o 0 U] ~, ~ t- 4 ~O- ~0 ~U ~da ~ ~ U1 L~ tD C, - 1- a~ 1- ~ ~D y y I_ O1 ~ 1- C ~ ~ - = ~3 ~- = ~ 4 ~ o ~W ~ ~ ~O ~ ~ ~ ~ ~ O m cn- m x ~- m- m- m ~- oc)-

VOLUME II Research on Sentencing: The Search for Reform Alfred Blumstein, Jacqueline Cohen, Susan E. Martin and Michael H. Tonry, Editors Pane} on Sentencing Research Committee on Research on Law Enforcement and the Administration of Justice Commission on Behavioral and Social Sciences and Education National Research Council NATIONAL ACADEMY PRESS Washington, D.C. 1983

39 u7 · - - . - C la _4 ·- U1 CI, V ~ X ~ U a~ ~ ~ C C O > S .,, V C ~ . - C ~ u7 ~V C ~ O · - V ~ U :' U O C Q ·-, 4~ ·- V ~ C V ° a, v o~e u, ~ l ~5 c u) c ~ · - .,, ~ ~ v 0 ~ .^ ° XQ' u v )J 0 ~ c 0 o v q~ 0 c ~ u .G v ~ c .,~ u~ .- v c c ~ ~ q~ u, ~ U4 0 ·- 0 l o . 11 1 1 1 1 1 ) ~1 u7 1 ut ~ v~4 ~ ~ u c ~ c ° e 0 ~ c V.,' ~a ~v~ 8 Ll · - ~ · - O · - a, V a,, ~ ~ v )J V V D ~ o, a, ~ .,4 . - C 54 V ~ - > ~ ~ 3 U O O C ~5 01 c ~ ·r, q5 C~ C3. a ~·-l u1 u7 ·,'v ~i ~Vq VV ~ ~ O ~n O· - C ~v Q 05 ' ~ ~ ~ > ~ >q ·a a, ~ O ~ C · - U. l l I U C Ll q.4 V ~ U a' c Q. Q o V U h C . - 01 ~ a) ~1 1 a) 1 ra u ~C ' ~ ~ ~ ~ O C ~ O ·^ v )~ a, ~ ~ ~ ~ ~c ,c ~ c .~ C ~, U) C ~' ·- ~ ~ V ~ · - ·- V ~a O ~ ~ 0 v )J ~V ~ U ~ V 3 · - ~ ~ ~rJ . - OJ ~ O V rn U r ~C Q ~ la ~ V ~ · - . - C ~ ~ ~ ~ ~ 0 ~O ·^ V ~ U V Q~ CL O X ~ ~ ~ ~U) V ~rl > ~ > ·- C ~' ~ Ql V ~ ~ O :D ~a ~ ~ a) ~ a O q~ U ~ V C C Q u~ ~ ~ ~ ~ O O ~ o' ~ C ~ O ·~- - ~ ~C ~ C > tq q~ V JJ ~ C O U Ul U, 4d ·~ C =1 U) C U Ul U3 V U) ~ QJ ~ ~ ~ ~ ~ ~ U3 C ~ o ~ ~ ~ ~ C ~ ~ V ·- ~ ·- ld a, ~ ·^ v 0 ~ 0 v ~ ~ 0 ~ ~· - Q · - V- - v ~ ~a ~ ~ · - c c · - ~ ~ c u U C ~ · - O V U ~ V ~p V ,,, ~ i, ¢1 ~ 3 ~ `0 C U 1 C, U3 U3 ~: ~ E ~ O ~ ~ O ~ C ~ ~ · - C L~ Ct. ~ O Ul ~J V G~ c U3 ~ ·- ~ ~ ~ ~ a~ a~ ~ a, C _~ 0 ~u ~ ~ ~ ~ > c ~ ~ v ·- 3 ~ ~ U3 0 fa V ~ C ~ 0~ (U fa X V _! ·- ~ V _4 C O C O · - · - · - C >1 0 ~·'t ~ S S (1) C o N · - C 3 -I ·,4 U ~ V Q ·~1 V U 4J 1:1 Q1 V U U) . - > ·- ~ .r~ U] _ 1- ~ ~r tD o ~r r~ o 1 11 1 1 C ~ut ~ O ~ :~ >, V ~ U3 V L. V O ~ 10 U3 . - aJ C _' I· - o~ C ~, Q IQ X U, U V . - ·~ > ~ C U O > ~ U ~ i., U) la ~_ ' O C 3 U) OJ .- O Q a, 01 ~ "1 U i' ~ D. V ~a ~ C 0) o .,' ~ v 01 .,1 v~ ~ u, O ~ Li o, ~ ~ O ·,, O 1 E U3 ~nl a, C O '01 ~ · - U3 ' t~ ~ U3 ~ t q~ 0 0 C Q U JJ 1 U3 >~ C ~ ~ Ll ~ O C rrJ Q U ~ ~-- O C ~ ~ > ·r4 V U] O ~ ~ ~ ~ · - Q U ~L a' ~ 0 0 0 C ~ :' ~ ~ O ~ > .'- - · - U3 C ~.e · ~ ~ a~ O a (U V U] U3 S ~ C . - · - · - ~ S, ~ U] .^ U 3 ~ U] S U] U3 ·,' >, a, ~ ~J e ~- 0- ~ <~ <, O ~ O V ~ C s ~ ~ ~ ~ ~ ~ c V al ~ ~ v Ld ~ O C S Ql C 0 ~ · - ~ V ,, ~ a, V U) O V O U) 11 C' 1 Q' - a, I Q > C 1 _. C.,, QJ V U U~ .,, O > O C U U] (d 54 q~ O c: >1 o v 4, a · - U Q) u3 .,' ~C ~ > > C ~q~ ~ o C q.d u3 U U O - q.i O q~ U VC U U · - U VC O Cu U C ~n5 v C ~C ~V ~C V g ~o~ C V C ~V g ~D. ~ VC V D~ oD °J Uv ~ u3 U u>] a, al U>, C ~u~ =` vu3 u,V u3 ~E C 1 ~1 :,0 1 0 ~o ~ ~o) rJ ~ u3 L.l OS C C o u,v ~ ~· - U O O ~ C ~ a)a' u C E U U V u3~ u3 U 'v co ~ ~3 ~aa, ~ u ~ ~r~ ~ u3o ~5 ~ U C :3~ c a ~ ~u ~ ~ ~0~ . O q' ~C4 ~ ~ ~oo U U~ U r~l > v 4' 1 ~ O ~ ~O~ v ~3o ~ r~ ~ ~ u3 V Q, ~ ~4 u) ~J u3 ~ E o, ~ ~ c u3 <~ a ~I ~ ~ a I s ~ ~ c I ~ 0 v V v c~ V U ~ v v~-, a ~0 V ~d C~ u~ C erl ~ ~4 .,. ~o ~ `, C ~ > ~ ~ C ~ u3 ~ °\ ~ ~°` ~ S C ~ Z ~ V~ ~ O ' - ~ ~ O a, c C o, 4~ 0 4t ~C V ' Y - u3 ~ _ ~ ~ ~ _~ _ C C ~ ~ ~ ~ ~ a' q~ o ~Ll >q O u~ ~ ·- ~ ~ ~Ll o ~)J ~ a 0 ·.' C ~a ~ ~ ~ ~ ~ W ~ - ~ U] -~ - {.7 - ~ E~ -

40 1 0 1 ::' O QJ q~ C JJ C . - . - 4 ~o ~ ~ c ~ ~ · - .~t e m~ ~ ~ . ~r1 i4 a U ~ ~ a~ U) JJ · - x ~ ~ c 0 ~ ~ U ~,., al U) c U) ~ ~ ~ ~ C ~ ~ ~ 0 C O ~ ~ m ~ ·~ Ql ~ C .,..} > U 0 ~mc C e ~ a~s ~ 0 ~ U 11 ~a . - > ~ a, ·,~ CI~ C Ld C ~ 0 .,- .m, .Q 1 q~ O · - U] C Q) U ·C Q) ~ Ll 0 Q) ~ ~ ~ c > .,, q~ S ~a O Q a ~u u ~ ~ a) C O O U . - ~ C > 4~ ~ C C U O a, ~ ~ ~ , t71 U] i4 4J =1 1tO 01 ~ u~a) ~ /1) > t0 c ~ ~ a, ~ ~ U ~ > O S ~ ~ ' - V O 4~) U h O c a' 0 > a, ~ > u .,' ~ C ~ S ~r1 U) O U' ~C · - ~ C ~ U) U) Q. u a) ~ ~ 4~ ' - ~ C S Q > 3 (1) aJ o Ll a) a) ~ u 1 ~1 ~ C 1 D4 1 ~ a) ~mE ~C C U ~ '. - O U7 O ~ O q~ ~ ~ ·- C QU' a) QU, O ·.~, C Q~ O mU i,, ~ U V ~O ·> ~ U) U O a' ~ O a) JJ > ·- O X-- U ~ ~ C C ~,0 C a) Q. O Q~ i. ~u~ ~4 ~ ~ Q ~ U U] C a) u~ U] ~ C O a) . - Q) O U} O >,~C ~ C .~1 a) ~ 4) 3 4~ ~ > 0 C > 0 ~ ~ ~ ~ aJ u] 00 (0 Q o - ~o SmU X Q' eq Q) ~C q°- U) . - .= ~ c Q\ o-' V . - U' c ~a ~ q~ y .,, ~ n' O U ~Q' C ~ ~ >~ ~1 QJ ~ Q ~, a, ~ ~s U a, ~ a, C ~ U1 ~ Q) > an .r4 .,4 ~J h 3 ~ · - 43 U) ~r 03 a' C I q ~C ~ . - O ~ ~O U U) q) 0 c a' u U) ~ ~ C JJ U) X Q ~ ·^ JJ C 4 CO' ~ C ·,4 U] ~n U O Ld ~ Q ~ ..' U) ~ C) <: ID Q C ·,1 a, ~ V q~ O ~: U) Q) 4U aJ ~ Ul Q C .,' ~J ,4 q~ ~ O QJ U C Q) ~ U) C ~ to £ a, N . - U) ~ E~ _ ~ C U] ~ C O U, 4) U ~0 >1 U) ~: E~ 1 1 U ·,4 q~ ~ .-C U, >, C ' C O U] C O ~ ~n ~ v C ~ U a' Q · - C O C .,' U, a' >' ~ · - U ~ O ~ C · - O Q) Q} c > ~a Q) ~ U ~ ~n ~ u, U U] ~ - ~ I C q~ UD a) a~ C c ra O a~ u C~ ~ 1 · - Q ' k4 O tn Q) ~ · - ~, ,~i Q Q X ~ U] O ~ ~ C 13 tU O ~ _'~ O s q~ C ' ~ ~ ~ O ·a ~ <L, '' X C O ~a 1)·u ~ a U]C ~ ~ ~ ~ c ~ u C > C v a' a) ~ ~ ~ u O s ~i Q >~ U~ ~a ~ ~ ~ ~ u a~ ~ O a, c 0) 4J JJ J~ h U U ~ U) U ,~s a) s ~ a, ,~ `4 3 s ~ O 11 l C 1 o ~ C ~.,~ ~ o a~ O ~a U) 4~ ~I J~ Q - la p~ ~ o ~ ~ U ~ QJ C Q) Q, U a , , q~ ~ ~ ~ c ~ o Q ~ ~ O U ~ O U U U ~ ~ C h ~4 la Ql ~ u, 10 c c ~ aJ QJ qJ DJ q~ ~ O ~O q~ 0 a' C U S C 43 aJ al C C >' 4~ ~ u, C au JJ a., C fL, u' ~ Y O 3 ~ C o ~· - U ' 4 Ll ~ Ll O X ~ IJ ~ ' - C U] S . - \,0 ~ ~ ~ > a' ~J ~ 3 U, · - C ~ ~ ' O ~ ~ ~ O ~ U q-4 ~ · - ~ C c q ~ a) ~ 0 0 a O ~q ~ U ~ - S O 4J ~ S 04 . - U] 3 · - U ~ · - U U~ q~ ~ q~ Q' 4~ JJ U U} ~ a, L' O ~a ~ u S ~ ~ C C . - ~a ~ JJ ^ C U, 0` C i4 01 a~ - .,. ~ 3n ~ 4~ ~ _ C ~ ~: ~ ' - c5, 3 ra ° ~ Q, C ~ := 31 ~a -

41 ·r1 C s · - · - ~ X 1 ~ u' ~4 ~ ~ c o 0 1 ~ P4 0 u7 c o ~ s · - v ~ ~ ~ ~ eq o · - q~ · - ~ · - . - C O S U) r4 0 Q. 3 ~ ~ a, ·a U] C ~ C ._' ~ . - ~ ~ Q. 'C ~ O ~ E C tr C ~ · - 0 · - a' > C ~ ~ o · - ~ C ~ ~ ~ a, a) ~ ~ s o ~ lq 1 ~ ~ o ~ ~ ~ ·a O ~ ~ m~ ~ ~ ~- - ~ ~ O ~ C ~.a ~ ~ O ~ td o o . ~ 1 1 11 11 a' ~a > > a' .,, ~ U) N .IJ 4d 4,~-_d o o U .- Ll a, 4 U~ 4J ~ ·-1 ~ U) C O Q) ~a c S O O · - ~ O a) ~ a' Ll ' L4 '= O X C o u' E O ~ ~ O O ~-- tU E a ~ E O ~ C O on ~a ~ u' Q. U] U) C ~ ~ JJ a, ~ 0 ,4 U] 1 ~ 54 JJ JJ ~ U] a, ° · - O 'C ° o, 1 14 ~1 L. C ~ C~ :, ~JJ O . - 0\ 0~ Q °' ''- - ~I ~4 s ., ~u~ 0 >1 a, u~ 1 -4 ~ 430 ~u, ~ ~ C .Q ~>' c ~ u, ~ (a 43 U7 ~ c ~ ~ ~ cu' C ~ Y G) S N QJ ~ ~ ~ O ~ ~ C ~ ~ y -· - O · - >' JJ~ S C O ~ ~a ~u' ~Y ).4 0 ~ C S JJ ~C 4~ ~ ~ ~-,1 a, ~O E 0, tU u~ a~ ~ ~a~ ~ 0a~ q~ ~ C n ~ ~ > C o 43~- E Y 3 q ~·- Q) C >1 ~ 4~ 0 n C 3 ~ ~q~ C O ~ O ~a, ' ~ o U' 1 0 0 a, ~ C c 0 a, ~ o, C c ~ ~,- ~ ~ ~ ~ ^ ~ ~E ~ C C ~ ~ > ~ o ,, C u' ~ s 3 ~ ~ ta . - O .~ q ~ o, O e) Q) a, O 0 4~ O ~ 4' E O - ~y i 0 E ~ ~ ~ ~ O ~ 0 0 ~ Q a, a, mO c ~ 4, ~ ~ ~a · - 0 4~ 0 0 0 q ~0 ~ ~ 0 c ~0 1 >' ~ o E ~ ~ ·~ c Q ~EE > 0 ~Y ~1 ~ a, ~ ~:, . - ~ 0 0 ~ 1 a) ~.- .,4 C ~, 0 ~0 0 ~ i4 ~ 0 ~ 0 ~O ~ ~ · - ~ · - , ~a ~ E a, ~ ~ ~ · - > 0 0 0 E c O w ~ 0 ~ ~ O ~ a) ~ x ~ > ~ 0 S ~c ~ · - ~a ~ O ~ G, ~ ~' ~ Z~4~ 0 ~Q a, 0 0n 3 3 ~, ~0 ~> ~s ~"a ~> 0 o~ 0 ~D 0 ~. 1 11 11 , 1 1 1 1 ~1 1 >1 O Q, i 0 ~ ~ L, J 0 :^ :3 c al c0 0 ~ ~ O :>, >.. - q~ ~ c ~ v 0 ~a ~ O C .,, ~ o0 4~ 0) 0 c~.Q~, c 0 D E c ~ :~ a, ~ 0 ~ ~ 0 04 0 ~ ~ O ~ I 0 0 a, ' ' QJ c u E 43 E 0 C 0 O L' .,. ~ JJ > O ~ ~ O X ~ Ql C a C I O 3 c O 0 i, O ~c 54 0 ~ a) c ~ ~0 ~ O O C ~ Ln . - a) Q) > ~ ~ O ~ ~a kD .- U, ta ~ c a, C) o, h O 1 S 1 ~4 a, ~ ~ a3 ,1 O ~ C u~ _I O ~ ~ ~a Z C 0 C S ~ O ~ U, a' · ~ _ O O ~J-, 0 ~a Q Q' o Ql 0 ~·,. O >1 0 C k4 ~ ~ O ~ le u' O · - JJ c ~ O ~ Q (5\ c 1 ~ a ~D ~ JJ tD 0 `o o' ~ :~ 0 c .~ 4 ~ 4 ~C ~ ' ~ _ ~ ~ _ c ~ c 0 ~ u' O m~ a' -4 E · - Z ~: - u~ 1 c · _ O ~ a' U) ~ o, 43 ~ a, - l 1 1 q~ o a) Q4 ~U >, 0 C C )~1 O V V 1 a 1 O V ~ O .,1 ~ . - ).4 ~J E ~ ~v c O ~ ~ 0 0 o, V .r4 Q, dIS 0 4J Li 1 ~ 0 c O . - E E O L. o o c ~ O 0 · - QJ ., Q, · - ~ ~ >1 0 ~ C o o r~ o~ Y ~ O

42 U. .,, ~C ·,1 c c ~ ~4 C · - o U U] 0 c a' ~ tn 0 u ~ c oa ·,i ~a c a tZ Id ~ U) ~ 4 4d ~ x n a, · - C JJ O JJ U ~: c .~. V) U) ~ U) u a, O JJ ~ :1 P4 1 ·~1 U1 JJ U ~ 0 C ·^ q~ 0 ~ U. U) ~ C · - h 4) O ~ - o C) - l ~: E~ O C Ll a, U) c ·a U) a, N .,' u~ a E~ ~ ~C U] ~ C C O ·_' ~n v U O U) ~Q O >, ~ U? U)I U. ~ to a~ ~ U1 q~ a, c ~l Y >~ a, ~ 0 c 1 0 ~- Ll · C aJ aC O ~ U ~ ~ ~ 4J .^ ~a~ s ~ o.- ~ ~ U) o.^ ~S 4J U ~ C ~.rl O ~ U U) O ~ ~ ~ ·~ (U 0~ ~-Q J ~U) ~ ·~14) S ~ >~ 01 C ~.a 44 ·r~ e .- a) c >' c n ~ 3 0) -' C ~ ~O ~ ·- C ·- u~ X )~ O U ~O C ~U) _I (U .,~ ~1 i4 U) C U] V U1 O s C ~ a, ~ ~ ·- v ·- O ~d O O O ~ Y ~ C O ~ ~ ~C ~ ~ 3 0) t~ ~: ~ ~ .3C C ~ ~ s ~ ~s o ~ ~ ~ ~ ~ o ~ ~ O > o c 36 - .~ ~ ~ ~u ~ ~ ~ ~ 0 ~ ~ E _ ~ 4 ~ ~ s 1 ~ ~ ~ ~ C > C ~ ~ ~ ~ ~ · - £ ~ 4 ~ a~ O O c ~ s~ ~ 1 ~c ~ U~ L~ o .r4 ~ ~ ~ ·- ~ u ~ ~ ~ ~ ~ ~ w ~ ~ ~ ~ > ~ ~ ~ ~ ~ C · - ~ o ~ U ~ U o o ~0 U V 4~ L ~ro ·- ~ u) ~ u E v al c 0 Q ~ 4 ~u, u ~ ~ ~a ~ 0 ~ > u: a~ ~ u~ ~ c E ~ > u~ O ~a~ ~ 4~ .- ~ a~ ~ · - :^ u~ c u~ ~ 0 ~ ·a ~ Q, a' ·~ u' U ~ 43 ~ ~V 0 ~ ~ 0 ~ ~ 0 ~ u ~n c u > E ~ 0 ~c c ~ v ~d ~.^ ~ ~ L ~1n 4J t ~ 3 u Y a, ~ ~ ~ · - ~ u E ~ ·- 4, ~s ~ ~Y c ~ ~ c al u v V a~ 0 al ·a v v a~ ~, c 3 ~ 4~ 4~ o, u 4~ c s .,, a~ u~ ~ ·~ ~a C ~ ~~ 0 ~ 4, 4, u c ~ c ,<' c 3 o, ~ c la ~ 3 v u s ~ ~u c > ~ ~ ~ S ~ ·- ~ O C ~ ~0 0 ·- n~ 0 ~ ~ 0 ~a V ~ tw a, u, ~ JJ ~ ~ v v ~ > u, E · - :; ~c ~ ~ E ~.o > :>. 3 ~ O ~ O u, ~(D 0 CD X ~r · . . I' 11 11 I )J ~I ,~ v [, ~ 1 0 ~I I I ,i u 1 1 1 v 0 E ~ ~ ~ ° 2 o o ~ o E u' O ~. ~a u u ~ ~ ~ ~ u °' a, c I ~ ~ a, I ~ ~ I Q, n U1 ~ E 1 C -,4 O v .,' u' c ~ v Q. ~ O ~ O~ o ~ ~ :~ JJ O ~ ·- ·~4 ~ uu ·~ x c u ul U ~ E a~ ·- O ~a~ E~ ~a >, ~ ~ ~ ~ ~ ~ v ~ ~ ~ 0 (,, u u' ~u 0 ~u u ~ v ~O ~ ~ ~ ,., Q,·~ ~, ~L' ~ > ~Q~ ~ E s u' u, a' QJ U)V 00 U) ~Ll ~ C ~U) c Q, °~ ~ ° a,c tu ~ ~ ~0 ~ ~ ~ 04 q~ :^ 1 ~ a~4 ~ a~ q~d >1 0 4J 1 0 u,a4 > Q, Oo V a' O V ~ C ~O a, cu~ v c u ° c u c c c ~C ~a ~ ~ V ~ 'v c v ~ vc ~U O ,[' ~VC C ~C Q. ~ ~ vc u, va. Q' ~ O u u, ~ v v ~ u, U1 U) ~ ~ Ul 1C ~(U .15 1 _'~ u I · - u ~ u ~o u~t-, ~ ~ ~v v ~4 u ~ ~ ~ ~0 ~ er ·-U~ ~ U] > iD ~t_ ·- ~ U7 c u 1 ~a 0~ ~ ~ v ~ O ~ 0 ~ u u, c ~ ~ ~s ~ >, u~,,~ ~ i~ ~ ~ ~ O`0 0 iD ~U~ U) ~ U) ~n ~o~ ~ a~ ]) ° ~ ~-01 a, ~ a~ v1 a~ I u ~ ~0 ~ v ul 1 oo 41~ ~ c O v ~ O O c ~ :,tn C vo V [~ .^ ~ ~ X r. 3 0\ C Oo, . - o' .^ ~ "C Oi O Q1 ~ /u -~ ~ U_4 :£ _~ : ~ U -~ U E ~ Z o U V ~' _ ~ ~C: ~V _' ~ o,

43 C o o o ~ U) a ~ C ' - - o . - a, ~ ~ o ~ ~ - q~ C ~ ·,1 ~1 al 0, h ° e~= C 1 0 1 k4 .~' ' ~4 0 ~ ~o ~ 0 ~ · - ~ c. - ~ O > Ll a, 4~ ~., .- C ~ ~ O ~ 43 0 ~ q~.- C U) ~ 0 0 > o.^ l o .~. 0 0 0 Ll ~ Ul U1 c U) C 0 ~ . - q~ q~ 0 a o C L V C ~C ~ a, ~a, >' U] ~U) t05 U) 1 c1 k4 - - m a,c ~ ~ q~ ~ · - > C ~ >- - O · - ~ ~ q~ oo C O C q~ a' O > ~ O ~D a' 1, C U1 1 ~ · - · - ~ O cn C ~ ~ ~a C C ~U C 1 U] k4 ~ ~ ~ U) 0\ 0 ~ O O ~ C N i.,, 05 "- - ~4d oN U) ~ > ~U) (D ~ U1 ~ ~ 0\ · - ~ · - O ~ O ~ ~ ~ O ~ ~ O O ~ ~ ~ ~ · - U' O Q) >1 Ll C ~ U) Q} (U ~ ~ ~ S c ~ O > ·r, 4J ~ ~O s a, O q~ 3 u~ ~C ~ u, q~ ~ C 0 1" ~ ~ ~ o, O O u, ~ 3 C ·^ .' U ~ O O · - 4~ U1 ~ O '~ O ~ ~ C 4J ~ ~ c ~ a, a' a, ~ la ·,1 ~ :1 ~, C ~ U) ~ O q~ ~a O ~ ·^ c · - U1 ~ ~, · - h tt U 10 aJ ~ 4 a~,~ c · - > C ~ c ns > ~ ~ ~ a, ~ c ~a O JJ U) ~ (Q £1 la 4) l ~C o o a~ (p o~ ~n O ~ L~ r, a, U) C ~ O a, ·rd ~ ~ ~ ' 04 C O ~ C >' (U ~ ~ . - JJ ~ C ' U] · - O ~ ·~ .,1 ,0 C . - QJ O ~ ~ · - .~ O O ~ O O JJ U) U) a 4J O U) C ·- O OOc ,a ~ 0 a, o 0 0 Q) ~ 0 00 ~ CC O q~ q~ ~ 0 0 0 c ~a 0 ~a c a, a 4J Q, ~ C · - ~ >~ 0n ~ I 0 ~a ~a.- c -4 C ~·^ ~ 0 ~a ~, ., 0 ~ 0 ~ ~ ~ ~ >~ ,a ~ ~ ~ C 0 0? a~ 3 U) u~ ~a ~ 0 0 ~a u ~ O Q. 0 o, ~ 0 · - a, ~ c) .-, _ 0` o, ~ 0` ~a ~la ~ Z - Z _ v ~ C c 0 ~a 1 0 o ° ~ ., ~a, a, q~ la 0 .- ~r~ , ~ ~ ~ r~ C 0 ·- C ~Y 0 m.- C} 0 · - ~ F; C 0 C1, Q' > C C 0 ~ 0 e .' ~a O e a) e .- ~ " 0 ~0 ~ ' C ~ ~ ~ ~a ·- c 0 ~a O L' ~q~ > 0- - 'J c IJ C .a · - 4~ 0 rn ~ ~ O · - 0 o. - U ~ ~ 0 a, ~ > "' ~ O O ° ~ 0 0 c ~, ~a O c ~' l o. ~· - ~ : 0 ~ 0 a C C ~a 0 C C ~ ~ ~· - ~ 4.1 ~ o, 0 0 >1 0 C ~ ~a 0. ~a 1 C o Y · - U ~a ~ 0 ,, ~a c n 4., C q~ C ~ 0 ~a ~ 0 ~ ~ q~ W 0 .~. · - q~ c ~a G, C . - ~ c 0 .rl IU 0 C q~ X C 3 r~ U~ I I 1 1 r~ 11 1 1 ~1 1 q.4 0 0 o C ·,' aJ ~ a C 0 ~ a) ~ ~1 >~ 0 ~ r, 1 0 C O ~ .,' · - o ~a ~ · - (U 0 F ~a ~ e ~ ta ~a ~° o -l ~ Q! ~ a, L. ~1 ~ ~ JJ Q' 1 ~, ~a 54 .a 0 e ~ ' °o ·.~, ~ 0 q~ 0 ~ ~ 4J 0 · - ~ 0 · - > 0 .~' c aJ ·- > 3 (L, O ~ C U ~a C ~ 0 ~a ~a ~ 0 0 ~ o a, U) 0 C ,0a a) I ~ q~ 1 U O _ ~ U) . - ~ IU · - h 0 ~a u c ·~' u] U? ~ a' Q) ~ a, ~ U ~/ ·r. O~ .m U 0 . - c ~a ~0 I · - ~ C V 00 a, ~ 0 ~ u O ~ O 0\ ~ ~ ~ ~ ~ U) U JJ · - C O ~ a, cr ~o, C ^ .,~ tn ~· - ~r 4~ kD 4J ~ ~ a, ~ ~ ~ a' ~ ~a c ~ ~ P4 - p' ,0 ~ _ 0 C ~ C C · - Q) > ~ ~ q~ c E o

44 U] o~ c .,' c ·r/ c c 1 1 ~ q ~C · - o a' ~ U U] QJ o C Q. o U) o o ~ C U) · - ~ C 4~ U) td :, X ~ a' · - C o C) C~ ·,' U) U] a' u' U QJ O JJ P. ~ 1 · - U] J.] ~ 4. C) ~ U) a) 4J Q C · - Q) ~ q~ JJ q~ O ~ U) u, n c · - (U h 43 O ~ . o C) - E~ Q) O C ~ (U 3 U) C U] Q) N .,' U] E~ C~, e ~a ~ c U] ~ C C 0 .-, ~n ~ h ld U ~ O >4 ~ >1 U] u' ~ 1 1 1 n5 ~ a' ~ ~ 1 1 c ~ s c, > 0~ ~ 0 C ~e ~ ~0 · - ~ 4J Ll O ~ O a) > ~ ~ ~O ~ ~ O . - ~ C e u~ =.- a, ~ ~ ~ C ~ a' O c .- c >~ ~ 0 0 · - v, u' 3 c ~ c ~0 ~ u) ~a).,' ~a, ~a ~, Q' ~ Q. as 0 3 0 0~ ~ a' Q) Q) ~ c ~ ~L' c ~ u' 0 ~ e ~ tn O 00 ~ ~ JJ.- 0 ~ Q c ~ ~ ~ ~ `~ ~ e ~ ~ ~ >' `~ ~ a' ~ O JJ ~ O ~ ~C ~ u ~Q) C o, >1 JJ a, ~ ~ U) 4~ 0 C a, c ~ ~ e C ~ ~ 3 > a~s~ =- c cmq ~Q m~ c ~a 0 a' . - 0 QJ ~ ~v Y ~C 4) ~ ~ · - ~ ~ C c ~C~ q~ 3 c~ o ~ q~ 00 c c ·- 3 n ~ ~0 ~ ·- c a' ~ c ~ c~ 0 ~ ~ ~ · - u' ~ ~ ~ u, ~ ~ c ~ - - · - ~ ~ O c i4 ~ x u~ ~ q~ i4 4~ ~ ~ ~c ~ ~ > ~4~ c O ~ ~ u~ O q~ Q) c> ~ ~ ~ ~ a' ~ ~ ~ Q) ~ ~ u' ~c~ ~ 4~ 0 0 C u>~ .~- ~ ~ > ~ ~ O c O e ~ ~ ~ 0, ~ 0) 13 R 4 0 ~ >' 3 a' u) ~ a, q ~ O e 4 >' u~ C ~, )J ~ 4 ~ >, O · - 0 0 0 ~ O ~· - ~ O O ~ 0) ~· - ~ a, ~ O (,q ~ ~ C ~ C~ q~ C ~ ~ ~ ~ U] . - JJ ~ 4J . - ~ O ~ C 4 ~ C ~ a' ~a ·^ ~ a, 4 ~a, u, 0 ~ Q, ~ ~d 4 0, 4 u, 4 a O ~ Q) u~ 4~ ·- ~ c~ ~ S 4 4 C 0 4J ~ O ~ aJ C 4 ~ ~ 0 C 0 o, C 4 C > ~ 3 ~, O G, ~ C U1 C 0 > 0 `0 O > C 4 4J X ~ Q, q5 ~ ·- ~ ~ ~ O 0 5 ·- e c ~.- ~ ~ c ~: ~ c c~ U~ ~ o tn 4 C ~4 u~ u' n O v ~ :, ~ ~ e 4 ·- u, 3 >~- c u~ ~o, tD oo ~D 1 1 1 -1 1 1 1 0 . ~ 1 11 11 1 1 1 1 1 1 4 4 u~>, ~C O O - ~ >' O 1 ~ 1 · - ~ 1 ~4 '. - ~ ~-' 4 ~ 4 4 4 ~/ 4 4 O u~ ~ a a, ~ Q, Q) Q) ~ ~ ~JJ ~. - U? ~ ~O > 4 > ~, > 4 > Q ~ 0 Q C ~ 0 ~ ~ ~:1 : ~o 4 ~ ~ · - >1 'C C 4 4 u' ~C 4 4 u' ~Q' ~ ~ O ~ 4 ~ u, ~ Q' 4 ~ ~ 4 u~ O · - O 4 ~C Q' ~ Q u' 3 ~n ~n U) 3 m :' 0 ~q ~ ~ ~n ~e 4 4 ~ O ~ 0 4 ~ 0 ~ 0 4 O > q~ O ·- O ~1 u' ~ o, O ~ u~ ~ °0 O ~ 4 ta O O ~ ft Q) .,l u ~a n 0 0 a) u' C q~ ~ ~ O O C u' ~4 4 ~ ~a, 0 ~ ~n c e n e c u~ 4 ~ 4 1 ~Q ~ O Q~ L O C c~ ~ ~ ~ ~4 O c~ ~ · - · - ra ~ ~ ra Q) 0) ~ 0 4 JJ 4 ~' 4 Q, 4 C X 0 QJ C' 4 X C ~O O ~ O a~ a~ ~° ~ ~° a~ ° JJ 4 ~ O ~ u~ 4 ~15 4 rx5 4~ ~u~ ~ C ~ ra .- v C 4 C C 4 4~ q' ~4 4~ ~ ~ 4 e O ~o ` ~ ~ e O ~ u O O tO 4 4 ~ 4 4 Q ~`~5 ~ ¢) o ra 4 Q~ 0 u, Ch 0 c~ ~4 aJ ~ 4 4 0 Q' O a) a) a, ~ u ~C E cc c ~ 0 · - a~ a' a) a~ Q, 4U ~, c~' ~ 0 1 q~ >~ a) ~1 o JJO V 1 0 JJ 3 >1 ~q~w a,~ 0 ~a) >1 ~ O Q)O ~ OC 0 0 ~C UO C~ S C C · - `O S CS C ~v a ~ 4 4~ Q 4~ ala~ ~ ~4~ a) Q) ~ ~cn 4 C R~ Q' C C > ~ C~ C U] ~ ~ ~U' U1 ~ ~ U,~ U) u) ~ u ~I i (p >,C 4 ~ o u~ 4 4~ O O ~ c~ 4 ·a ~r ° 0 ~r ~ u~ ~ o'cn 4 4 ~r~ v, u, ~ Q) a~ 4 t~3 0 ~ <~ >t - u~ u~3 ~=~- -~ e 4 :~ ~ u~ ~ O ~ O ~ C ~ C 4 a O~ ~ u, 0 1 ~ O u~ 1 1 4 ~ C -I 0 1 3 U] ~ C 1 - ~ 4J 1- = ~1- ~4 C U~ o a ~1 - ·- a~ [~ u) J~ a, . - ~a, ~ ~ 0 C cn ~ ~ ~a O ~ ~ O ~ C ~ C ~ Ql 1 nl 0 1 05 0 '~ · - 1 0 ~ _I S u' a' c ~u ~ c ~U]O· O ~ 0 ~ ~ ~ ~ ~ ~ ~C- u' ~r~ ~D ~ ~ C a, ~ · - ·,l a, ~ . - · - a, · 4 C ~ . - ~s, C 0 4 C ~ ~I C ~ ~ ~ ·- ~ ~ ~ U~ a, C _ - C q~ ~a ~u, _ I _ 0 ~ - u ~un ~ ~D ~ 3 ~u~ 4 ~ ~- ~- -- c o ~e t~= ~ 0\ ~ a, 0 o, ~ ~ 01 0 c a, 0 ~0 ~s ~S ~ ~S ~ ~ , - ~x - V) ~ - E~ ~ -

45 U) V~ ~ U) 1 a, C ~ C Q. C Ul ~ U) ~ ~ > ~ o ~ C q~ . - s o o ~, a) ~ 4J ~ U) )J o o, a, C) v ~- - C -' a Ll ~ Ll ~ u, O q~ O a., ~ ~ o s .` ~ C o C' C C O ~ o' u ~1 Q) ~u ~>~ ~ U) C m- - U ~C ~ > 0 u, ° ~ 0 ,= ~ ~"a o ~ s ~ ~ ~ 4J ~ 0 ~ ~ ~n ~ ~QJ S ~ .-1 0 ~ ~ ' - .^ ~ ~ ~ ~ ~C o Q) ~ ~ ~ s ~ ~ ~ ~ ~ > 3 0, V tO ~ ~ ~ ~ O ~ ~ ' - ~ ~ . - .c: U~ U] .- Q~ ,Q U) o O O n5 S > .- ~ ~ a U) ~ ~ ~ ~ ~ ~ ~ ~ 3 ~ ~ ~ C ~ ~ ~ ~ ~ ~ ~ ' - '- ~ ~ ~ ~ ~ O ~ C ~ ~ ~ ~O c ~ 0) (d 0~ ~ Q) ~ l > ~ ~U ~ 3 Q. .33 ~.o ~ u ~ > > ~ 0 ~·- s v C O i4 ta ~. c ~s ~ ~ .,~ c ~Q) 4J O a) ~ ~ s ~ ~ c ~ aJ '- 05 ~ (d ~4J s h {0 Ll (0 C U ~4 ~> U} O C 54 .C c ~l ~a ~ Q) od -l ~ U] (U J~J C O C u~ J~ ·, ~ C ~·, ~ /15 t~ (d L~ ~O V C ·~ O ~ ~Ll C 0~ ~1 a~ Y ·,4 3 S .-i > 0 u) >, C 4J ~1 43 · - _I ~1 a~J ~ _I U7 ·' JU ~` ~ O ~ O ~ O ~ ~ 3 0 ~ 3 ~ 4~ ~ ~ ~ ~ ~ C ' - ~ ~Y ~ ~ ~ Y Y ~ ~ ~ 0) ~ c ~ ~ ~ ~ ~ ~ C ~ · - ~ ~ c ~ Q C ~ ~ F ~- ~ s ,0 tu ns V ° 1~ 0 ~ ~ C ~ Q) ~ C Q5 ~ ~ ~ ~ C ~ n5 JJ 4J ~J C ~ld ~ i4 ~ ~ o ~ ~ > 0 a) C s ~ tr 0 ~a 0 U) a} Q5 U ~Q~ 3 V Q U, U > Q V ~U ~O V Q Q ~ U ~L 01 11~1 0 u ~tD ~C ~ o 1 ~1 1 1 O u ~kD ~ ·~\ o C S o dJ U] (U V C ~ ~ .- :3 J~ U) O ~ ~ 0 a' u, U~ ' ~ ~ C ·,~ ,4 ~ (15 0 ~4 .Q Q U O ~ O a' ~ u' Ul · - L' C Q~ Q' ,~ ~ q~ > E ta ~: 0 tn C U U v U ~ C · - C · - ~ ~ a' Q) V ~ ~ C > ~C a' ~ u aJ U) ~ U] U) 11 ~I U] I C U] ~ ·- ~ C O 4~ L4 · - ' O JJ C U ~, Q ~J · - 1~1 0 ~ ~ O ~ ~ U O Q~ U ·~ U · - ~ ~ a, qJ Ll C W 1 ' 0 4J nl 0 1 X U] ~3 a ,0 ~ V 4~ ' O ~ C U) C U Q Q) ~a ~ ~ · - ~a 4U O ~ ~ ~ ·- U O · - . V ~ ,J C ~ > a O ~ U 4J Q) C ' - > ~ a al ~ 4J > ~ C > ~ ~ U ~ ~._1 ~ U7 U) U ~,- ~ . ' ~ ~''1 V O c ~ a' - c _ ~a s ~, u ~ ~ tU U 1- F 1- q~ .- ~ 0\ ~ 01 ,1 > C -I S-I E-4 94: ~ - ~ _ u, 1 ~ ~ QJ U) ~ U ~ ~ C r ~aJ ~ O C aJ U) a, 4J (~1 ~ CN 3 ~ ' ~ h ~I V Q) ~ _ > ' - ~ O a, N L oo C ~ ~ C C'\ C ~ C ~ ~ ~a :e _ ~: O C Q) U] ~ C U) U1 1 ~J U ql C ~ . - >, S Q 4~ :^ v a, QJ -l C ~ ~J ~ . - Q1 ~a U U ~ C '- U) Q) S -i V Q ~1 C (0 q~ U) O O ~ C2 o o C _4 ~ al aJ c U] 11 11 C' a) Ql U a' a' u~ Q) ~ u, c ~ ~ U (n C ' O O C ~ U a, a) (1, ~ a Q~ ~, ~ ,0 >1 a) ~ ~ s ~ O ~ 3 > 0 U '- 1 1 L, C Q, > ~ S ~ UD IJ ~ ~ ~a s u a' ~ c 'C a k0 ~ U ·n kD C a~ ~ U] 1 ~ a 4 ~r ~ ~ ax O JJ ~ U) U) GJ ~ U) ~ C C ~ O ~ O O u ~ ~a Q) C U) O ~ C U1 F 3 Q5 ~ O ~ O U 4 U U) 1 1U v 1 1 F ~C O . - U) ~r~ V ' - ' U O O ~ u ~a' u - ~ .- ~ Q, > ~a5 ra 0 ~ ~ ~ ~ V Ll ~ ~ C a' ~ c ~ ~ 0 O a) ,C ~ > .,, U ~ U ~ · - ~ o (U Q L O ~ ~ U Ql ' - q~ ~ o o _ ~ a 4J ~ V ~ 3 U a' Ll ~ x a) O ~ > U O aJ C U ,0 S C 4J a a) ~ JJ C C >~ a' a JJ ~ U) V} O ~ o .,, C ~ O C ' _C C~,0 (d voF ~ O V ~a v, Q~ -C~ ~ ' r~ -

46 ~7 c . c . ~ Q) c ~ ~C ·- o 0 c a' ~ o U tC C U) · - ~ C <: ~ ~ m ~ u~ m a, ~ 43 v X ~ a' · - 4 C O C) ~: c .,. u) u~ O ~ 1 · - u' 41 m u) 4J Ll Q. C · - QJ L~ ~ IJ q~ O <: u~ a, u~ ~ C · - a~ ~4 q~ q~ 4J O ~S . o - ~: a: c v u C m u, N ·_' U) ~ >~ E ~ m c u, ~ C c ~ O . - v' ~ 4 m O >' v v, 1 a, > O Q `' m `, O ~ C ~ ~ c ~ a, ~ m 3 ~ c u, u~ ~ ~ ~ u' s m ~ ~ q~ ·,~ ."l ~ ~ 3 C > m a~ a, 0 u q~ ~ ~4 C · - ~ O ~ ~ Y 3 ~ C ·- m ~ q~ -4 ~ 0 · - 3 ~ · - > ~D o 11 l >, ~4 ·r-, Q) a m a) ~ C ·,4 ~ C c .,' u' 1 u) Q} _ m 0 ~ ~ ~ · - · - a' ~ ~ PH cL e o, ~ C ' ~ _ ·rl C _~ \0 e ~£ ~ .,. . - , ~ b4 ~ O ~ ~n C o u, o CR s .,1 aJ - C 4~ C u' C ta U o, - q~ o C .,1 4d · - U O u~ o u, - ~ a, ~ ~ Y s ~ · - q4 a) O :~: C ~ ~ (a h S:; ~ ~ C U) ~ m N e e QS D O O m U1 U) .,4 ~a > a, .,' ~ a m u, u~ ~a Q U

47 REFERENCES Atkinson, David N., and Dale A. Newman 1970 Judicial attitudes and defendant attributes: some consequences for municipal court decision-making. Journal of Public Law 19:68-87 Baab, G.A., and W.R. Furgeson, Jr. 1968 Texas sentencing practices: a statistical study. Texas Law Review 45:471-503. Balbus, Isaac D . 1973 The Dialectics of Legal Regression: Black Rebels Before the American Criminal Courts. New York: Russell Sage Foundation. Bedau, Hugo Adam 1964 Death sentences in New Jersey. Review 19:1-2. 1965 Capital punishment in Oregon 1903-1964. Oregon Law Review 45:1-39. Bensing, Robert C., and Oliver J. Schroeder 1960 Homicide in an Urban Community. Springfield, Ill.: Charles Thomas. Bernstein, Ilene Nagel, William R. Kelly, and Patricia A. Doyle 1977a Societal reaction to deviants: the case of criminal defendants. American Sociological Review 42:743-755. Bernstein, Ilene Nagel, Edward Kick, Jan T. Lehng, and Barbara Schultz 1977b Charge reduction: an intermediary stage in the process of labelling. Social Force 56(2):362-384. Black, Donald 1976 The Behavior of Law. New York: Academic Press. Blalock, Hubert 1964 Causal Inferences in Nonexperimental Research. New York: Norton. Blau, Peter, and Otis D. Duncan 1967 The American Occupational Structure. New York: Wiley. Blumstein, Alfred, and Jacqueline Cohen 1980 Sentencing of convicted offenders: an analysis of the public's view. Law & Society Review 14(2):223-261. Bonnie, Richard, and Charles whitebread 1974 The Marihuana Conviction: A History of Marihuana Prohihiti an in Who rTni. ^~ C! Rutgers Law Charlottesville: . University Press of Virginia.

48 Bowers, William J. 1974 Executions in America. Lexington, Mass.: D.C. Heath. Box, George E.P., and Gwilym M. Jenkins 1970 Time Series Analysis: Forecasting and Control San Francisco: Holden-Day. Bullock, H.A. 1961 Significance of the racial factor in the length of prison sentences. Journal of Criminal Law, Criminology and Police Science 52:411-417. Burke, Peter, and Austin T. Turk 1975 Factors affecting postarrest dispositions: a model for analysis. Social Problems 22:313-332. Cargan, Leonard, and Mary A. Coates 1974 Indeterminate sentence and judicial bias. & Delinquency 20:144-56. Chiricos, Theodore G., and Gordon P. Waldo 1975 Socioeconomic status and criminal sentencing: an empirical assessment of a conflict Indeterminate sentence and indicial bias. Crime proposition. American Sociological Review 40:753-772. Clarke, Stevens H., and Gary G. Koch 1976 The influence of income and other factors on whether criminal defendants go to prison. Law & Society Review 11:57-92. 1977 Alaska Felony Sentencing Patterns: A Multivariate Statistical Analysis. Alaska Judicial Council, Anchorage. Conklin, John E. 1972 Robbery and the Criminal Justice System. Philadelphia: Lippincott. Cook, Beverly B. 1979 Public opinion and federal judicial policy. American Journal of Political Science 21:567-600. Duncan, Otis D. 1961 A socioeconomic index for all occupations. Pp. 109-138 in A.J. Reiss, Jr., O.D. Duncan, P.K. Hatt, and C.C. North, eds., Occupations and Social Status. Glencoe, Ill.: Free Press. Eisenstein, James, and Herbert Jacob 1977 Felony Justice. Boston: Little, Brown. Farrell, Ronald A., and Victoria Lynn Swigert 1978a Prior offense as a self-fulfilling prophecy. Law & Society Review 12:437-453. 1978b Legal disposition of inter-group and intra-group homicides. Sociological Quarterly 19:565-576.

49 Featherman, David 1980 Social stratification and mobility: two decades of cumulative social science. American Behavioral Scientist 24(3):364-385. Feeley, Malcolm M. 1979 The Process is the Punishment: in a Lower Criminal Court. Sage Foundation. Handling Cases New York: Russell Garfinkel, Harold 1949 Research note on inter- and intra-racial homicides. Social Forces 27:369-381. Gerard, Jules, and T.R. Terry 1970 Discrimination against Negroes in the administration of criminal law in Missouri. Washington University Law Quarterly 1970:415-437. Gibson, James L. 1978 Race as a determinant of criminal sentences: a methodological critique and a case study. Law & Society Review 12:455-478. Green, Edward 1961 Judicial Attitudes in Sentencing: A Study of the Factors Underlying the Sentencing Practices of the Criminal Court of Philadelphia. Vol. 15, Cambridge Studies in Criminology. London: Macmillan. 1964 Inter- and intra-racial crime relative to sentencing. Journal of Criminal Law, Criminology and Police Science 55:348-358. Greenberg, David F. 1977 Socioeconomic status and criminal sentences: is there an association? American Sociological Review 42:174-176. Hag an, John 1974a Extra-legal attributes and criminal sentencing: an assessment of a sociological viewpoint. Law & Society Review 8(spring):357-383. 1974b Criminal justice and native people: a study of incarceration in a Canadian province. Canadian Review of Sociology and Anthropology. Special issue published in conjunction with the International Sociological Association Meetings (August):220-238. 1975a Law, order and sentencing: a study of attitude in action. Sociometry 38(2):374-384. 1975b The social and legal construction of criminal justice: a study of the pre-sentencing process. Social Problems 22(5):620-637.

50 1975c Parameters of criminal prosecution: an application of path analysis to a problem of criminal justice. Journal of Criminal Law, Criminology and Police Science 65:536-544. 1977 Criminal justice in rural and urban communities: a study of the bureaucratization of justice. Social Forces 5S(3):597-612. 1982 The corporate advantage: a study of the involvement of corporate and individual victims in the criminal justice system. Social Forces 60 (4):993-1022. Hagan, John, and Celesta Albonetti 1982 Race, class and the perception of crimina 1 injustice in America. American Journal of Sociology 88(2) :329-355. Hagan, John, and Ilene Bernstein 1979 Conflict in context: the sanctioning of draft resisters, 1963-76. Social Problems 27 :109-12 2. Hagan, John, John Hewitt, and Duane Alwin 1979 Ceremonial justice: crime and punishment in a loosely coupled system. Social Forces 58(2):506-527. Hagan, John, Ilene Nagel, and Celesta Albonetti 1980 The differential sentencing of white collar offenders in ten federal district courts. American Sociological Review 45(0ctober):802-820. Hall, E.L., and A.A. Simkus 1975 Inequality in the types of sentences received by native Americans and whites. Criminology 13:199-222. Hauser, Robert 1970 Context and consex: a cautionary tale. American Journal of Sociology 75:645-664. John Sentencing as a Human Process. Toronto: University of Toronto Press. Andrew Hogarth, 1971 Hopkins, 1977 Is there a class bias in criminal sanctioning? American Sociological Review 42:176-177. Inter-University Consortium for Political and Social Research 1979 Public Image of the Courts, 1977: General Publics Data. Ann Arbor, Mich.: Inter-University Consortium for Political and Social Research.

51 Johnson, Elmer H. 1957 Selective forces in capital punishment. Social Forces 36:165-169. Judson, Charles J., James J. Pandell, Jack B. Owens, James L. McIntosh, and Dale L. Matschullat 1969 A study of the California penalty jury in first degree murder cases. Stanford Law Review 21:1297-1431. Kalleberg, Arne, and Aage B. Sorensen 1979 The sociology of labor markets. Of Sociology 5:351-379. Annual Review Kelly, Henry E. 1976 Comparison of defense strategy and race as influences in differential sentencing. Criminology 14:241-249. LaFree, Gary 1980 The effect of sexual stratification by race on official reactions to rape. American Sociological Review 45:842-854. Lemert, Edwin M., and Judy Rosberg 1948 The Administration of Justice to Minority Groups in L.A. County. of California Press. Levin, Martin A. Berkeley, Calif.: University 1972 Urban politics and judicial behavior. Journal of Legal Studies 1:220-221. 1977 Urban Politics and the Criminal Courts. University of Chicago Press. Chicago: Lizotte, Alan J. 1978 Extra-legal factors in Chicago's criminal courts: testing the conflict model of crimina l justice. Social Problems 25:564-580. Lotz, Roy, and John D. Hewitt 1977 The influence of legally irrelevant factors on felony sentencing. Sociological Inquiry 47:39-48. Martin, Roscoe 1934 The Defendant and Criminal Justice. Bulletin No. 34-37. Bureau of Research in the Social Sciences, University of Texas. McCarthy, J.P., N. Sheflin, and J.J. Barraco 1979 Report on the Sentencing Guidelines Project to the Administrative Director of the Courts: On the Relationship Between Race and Sentencing Sentencing Guidelines Project, State of New Jersey Administrative Office of the Courts, Trenton.

52 Mileski, Maureen 1971 Courtroom encounters: an observation of a lower criminal court. Law & Society Review 5:473-538. Musto, David 1973 The American Disease. University Press. New Haven, Conn.: Yale Myers, Martha A. 1979 Offended parties and official reactions: victims and the sentencing of criminal defendants. Sociological Quarterly 20:529-540. Nagel, Stuart 1969 The Legal Process From a Behavioral Perspective. Homewood, Ill.: Dorsey Press. Nam, C.B., and M. Powers 1968 Changes in the relative status level of workers in the United States, 1950-1960. Social Forces 47:158-170. Nardulli, Peter 1979 The Courtroom Elite. New York: Ballinger. Partington, Donald 1965 The incidence of the death penalty for rape in Virginia. Washington and Lee Law Review 22:43-75. Partridge, Anthony, and William B. Eldridge 1974 Second Circuit Sentencing Study: A Report to the Judges of the Second Circuit. FJC No. 74-4. Washington, D.C.: The Federal Judicial Center. Perry, R.W. 1977 Justice system and sentencing: the importance of race in the military. Criminology 15:225-234. Pope, Carl E. 1975a Sentencing of California Felony Offenders. National Criminal Justice Information and Statistics Service. Washington, D.C.: U.S. Department of Justice. 1975b The Judicial Processing of Assault and Burglary Offenders in Selected California Counties. National Criminal Justice Information Statistics Service. Washington, D.C.: U.S. Department of Justice. Rhodes, William M. 1976 The economics of criminal courts: a theoretical and empirical investigation. Journal of Legal Studies 5:311-340.

53 Robinson, Robert V., and Jonathan Kelley 1979 Class as conceived by Marx and Dahrendorf: effects on income inequality and politics in the United States and Great Britain. American Sociological Review 44:38-58. Sellin, Thorsten, and Marvin E. Wolfgang 1964 The Measurement of Delinquency. New York: Wiley. Shan - Dubow, Sandra 1979 Felony Sentencing in Wisconsin. Wisconsin Center for Public Policy (Mimeo). Randall J., and Matthew T. Zingraff Detecting sentence disparity: some problems and evidence. American Journal of Sociology 86:869-880. Tiffany, Lawrence, Yakov Avichai, and Geoffrey W. Peters 1975 A statistical analysis of sentencing in federal courts: defendants convicted after trial, 1967-1968. The Journal of Legal Studies 4:369-390. Treiman, D. 1977 Occupational Prestige in Comparative . Perspective. Thomson, 1981 New York. Academic Press. Uhlman, Thomas M. 1979 Racial Justice: Black Judges and Defendants in an Urban Trial Court. Lexington, Mass.: Lexington Books. Uhlman, Thomas M., and N. Darlene Walker 1979 A plea is no bargain: the impact of case disposition on sentencing. Social Science Quarterly 60:218-324. 1980 "He takes some of my time; I take some of his: an analysis of judicial sentencing patterns in jury cases. Law & Society Review 14:323-341. U.S. Bureau of the Census 1963 Methodology and Scores of Socioeconomic Status. Working Paper No. 15. Washington, D.C.: U.S. Department of Commerce. Unnever, James D., Charles Frazier, and John C. Henretta 1980 Race differences in criminal sentencing. Sociological Quarterly 21:197-207. Wolf, Edwin D. 1964 Abstract of analysis of jury sentencing in capital cases. Rutgers Law Review 19:56-64.

54 Wolfgang, Marvin E., and Marc Reidel 1973 Race, judicial discretion, and the death penalty. The Annals of the American Academy of Political and Social Science 407.119-133. Wolfgang, Marvin E., Robert Figlio, and Thorsten Sellin 1972 Delinquency in a Birth Cohort. Chicago: University of Chicago Press. Wolfgang, Marvin E., Arlene Kelly, and Hans C. Nolde 1962 Comparison of the executed and commuted among admissions to death row. Journal of Criminal Law, Criminology and Police Science 53:301-311. Wright, Erik Olin 1980 Varieties of Marxist conceptions of class structure. Politics and Society 9~3):299-322. Zalman, Marvin, Charles W. Ostrom, Jr., Phillip Guilliams, and Garret Peaslee 1979 Sentencing in Michigan: Report on the Michigan Felony Sentencing Project. Michigan State Court Administrative Office, Lansing. Zimring, Franklin E., Joel Eigen, and Sheila O'Malley 1976 Punishing homicide in Philadelphia: perspectives on the death penalty. University of Chicago Law Review 43:227-252.

Next: 2 DISCRIMINATION IN THE CRIMINAL JUSTICE SYSTEM: A CRITICAL APPRAISAL OF THE LITERATURE »
Research on Sentencing: The Search for Reform, Volume II Get This Book
×
Buy Paperback | $125.00
MyNAP members save 10% online.
Login or Register to save!
Download Free PDF
  1. ×

    Welcome to OpenBook!

    You're looking at OpenBook, NAP.edu's online reading room since 1999. Based on feedback from you, our users, we've made some improvements that make it easier than ever to read thousands of publications on our website.

    Do you want to take a quick tour of the OpenBook's features?

    No Thanks Take a Tour »
  2. ×

    Show this book's table of contents, where you can jump to any chapter by name.

    « Back Next »
  3. ×

    ...or use these buttons to go back to the previous chapter or skip to the next one.

    « Back Next »
  4. ×

    Jump up to the previous page or down to the next one. Also, you can type in a page number and press Enter to go directly to that page in the book.

    « Back Next »
  5. ×

    To search the entire text of this book, type in your search term here and press Enter.

    « Back Next »
  6. ×

    Share a link to this book page on your preferred social network or via email.

    « Back Next »
  7. ×

    View our suggested citation for this chapter.

    « Back Next »
  8. ×

    Ready to take your reading offline? Click here to buy this book in print or download it as a free PDF, if available.

    « Back Next »
Stay Connected!