structural Effects in Education 127
CLOSING OBSERVATIONS: THE STATICS OF STRUCTURAL EFFECTS
The structural effects argument developed out of Durkheim's claims for the significance of
social facts in accounting for stable rates of individual conduct. The connection between
Durkheim and recent investigations of school effects, mediated through Columbia contributions
to general sociology, looks linear from a distance but convoluted up close. Durkheim's
(and others') ideas about the relation between social structure and individual conduct underwent
elaboration at Columbia in distinct yet related directions: first, in studies focused on
organizational structure and events, interactions among people at work, and short-term historical
change; second, in methodological investigations designed to formulate general principles
of survey analysis.
The significance of Columbia sociology for how the subsequent study of school effects
took shape rested in both the presence and the absence of its influence. The first direction
stressed organizational case studies rather than organizational case studies. Their contribution
regarding structure and conduct came more from the substantive analysis of organizations
than from case study methodology. As to structural effects, Blau was a key figure in elaborating
on that idea as an outgrowth of his substantive agenda (as shown by his extraction of
structural effects reasoning from The Dynamics of Bureaucracy [1955]), not mainly as a mode
of data analysis. Dynamics was emblematic of the organizational side of the Columbia tradition.
Although instrumental in developing structural effects argumentation in general, his work
(and that of others from Columbia) exerted little if any direct substantive impact on later
developments in school effects research. No one working in the sociology of education in the
middle to late 1950s, moreover, showed much interest in school organization. Not until 1965
did an important contribution to that area of study appear (Bidwell, 1965), and a long time
subsequently passed before others gained interest.
A different picture emerges from the second direction. Structural effects was one of several
themes developing out of Lazarsfeld's and his students' contributions to principles of
survey analysis. Their later application to education in general and to school effects in particular
can be attributed primarily to Coleman. Of the participants in the Columbia milieu of the
1950s, he was almost alone in later devoting a substantial proportion of his intellectual efforts
to educational topics. (An exception is Trow, who, however, wrote largely about higher education.)
Most noteworthy were The Adolescent Society (1967) and EEO (1966), as well as his
influence on the design of large, nationwide surveys of education from the mid-1960s onward.
Those contributions were as important for what they did not take from Columbia as for what
they did. Coleman did not apply the substantive agenda of the Columbia organizational studies
to the analysis of schools, nor did he continue the research style of the case studies. There
was certainly an organizational side to his approach to school effects, but it was conceptually
implicit and fashioned out of the empirical materials that large surveys could yield; namely,
global variables and aggregations of individual-level data resembling the nonsubstantive
schemes of aggregation developed by Lazarsfeld, Kendall, Menzel, Selvin, and Hagstrom.
The later acronymic surveys were not substantively designed around conceptions of school
organization and how it operated. My speculation is that his views about policy analysis assigned
greater persuasiveness to evidence gathered from nationwide surveys than from intensive
organizational analyses.
Studies of school effects followed Wilson's formulation, schematically similar to Blau's
but substantively unconnected to organizational analysis. His position contained a rather narrow
(in hindsight) view of structure as the aggregation of individual characteristics and of
normative climate as the dominant mechanism influencing individual outcomes. The norma
128 Robert Dreeben
tive climate perspective endured as a thread running through structural effects studies for
decades to come. Although prominent, it was and is not the only mechanism employed in
investigations based on structural effects arguments. The alleged influences of school composition
based on race, on sex, or on teacher qualifications, for example, did not necessarily
represent normative versions of such arguments but are nevertheless examples of the general
scheme.
Traditions other than Columbia's could have led to the emergence of a sociology of
education that paid significant attention to the substance of school organization, its inner workings,
and its effects. Abbott's (1997) analysis of the Chicago School, developing around Thomas,
Park, and Burgess (like Merton's [1995] of Columbia), for example, maintained that
. . . [T]he Chicago School thought—and thinks—that one cannot understand social life without
understanding the arrangements of particular social actors in particular social times and
places . . . [N]o social fact makes any sense abstracted from its context in social (and often geographic)
space and social time. (Abbott, 1997, p. 1152)
His portrayal of the Chicago School fastened on the "interactional fields" (p. 1156) of ethnic
groups, of neighborhoods and other segments of cities, of occupational groups, and the like.
The common ground between this conception of what is sociologically important and that
prevailing earlier at Columbia is plain. Despite that similarity, neither a sociology of education
nor a concern with school effects developed in Chicago (at least not until after the mid-
1960s, and then not based on the patron saints of the Chicago School). Chicago sociology
showed little concern with organizational structure or with education and schooling (Becker,
1952,1953, was an exception), focusing more on other units of society: neighborhoods, occupations,
urban areas, and racial ethnic groups. The Chicago version of Durkheim focused on
area rates of crime, delinquency, and forms of social disorganization, excluding analogous
patterns of educational phenomena.
Hindsight reveals the legacy that Durkheim's (1897/1951) Suicide left for the study of
school effects. Along with the Rules (1895/1938), it laid the groundwork for a sociology
based on social facts. It also employed methods that relied on evidence drawn from public
records, an analogue of survey evidence collected for later secondary analysis prompted by
questions different from those generating the original surveys. Although we remember social
facts as external and constraining (and Durkheim's political reasons for a superindividual
definition of them), they also represent static indices of social conditions: the proportion of
Catholics and Protestants in given geographic areas defined religious society; the configuration
of sex, marital status, and parenthood defined domestic society; and so forth. The substantive
core of his enterprise was to identify a mechanism—social integration—to explain
why certain social facts account for stable rates of suicide. That effort was hampered by limitations
of the evidence, which provided no direct information on how church polities operated
and family life transpired or of social integration. We continue to admire the ingenious research
design of his secondary analyses and his digging into public records to find intervening
and contingent conditions to argue the explanatory case for the mechanism of integration. He
looked a lot like a late 20th century student of school effects rummaging through a body of
available survey data short on evidence about the mechanisms of schooling. His interpretations
were plausible but speculative. Halbwachs (1978), Henry and Short (1954), Hyman
(1955), and Inkeles (1963), for example, identified different mechanisms from Durkheim in
linking social states and rates of suicide.
In my view, the conceptual vulnerability of structural effects argumentation, as it has
been applied to explaining school effects, is that it rigidifies social structure and social process.
This is primarily a substantive problem; secondarily a methodological one occasioned
Sedang diterjemahkan, harap tunggu..
