Structural Effects in Education 113

Structural Effects in Education 113
case of soldiers with higlier levels of education being more likely than those with lower levels
to say that they had volunteered or should not have been deferred. Kendall and Lazarsfeld
interpreted the finding by arguing that better educated soldiers accepted induction because
they came from civilian environments in which few men were deferred, and thus had no legitimate
expectation of deferment because of their education (p. 149). This interpretation was
based on soldiers' membership in a social category (i.e., an environment with infrequent deferments)
affecting their self-assessments and beliefs about the appropriateness of being in the
Army; based primarily on soldiers' thought processes, it may or may not have been correct.
Whatever the case, the purpose of the exercise was more to develop a logic of analysis than to
pursue the substantive details empirically.
To explore structural effects, Kendall and Lazarsfeld (1950) developed a formalization
of connections between descriptions of individuals (personal data) and of aggregates, interactions,
categories, and groups (unit data), creating a typology to represent how personal data
and unit data are layered to express relations among levels of aggregation. This effort concluded
with a discussion of whether personal data and unit data can be used interchangeably,
a question that comes down to us as ecological correlation (Robinson, 1950) and as the ecological
fallacy (Selvin, 1958, 1960). "Our main interest here is directed toward the logical
relationship between personal data and unit data" (Kendall & Lazarsfeld, p. 188; my italics),
toward developing general principles of data analysis applicable to surveys regardless of content.
This enterprise differed from Merton's.
Lazarsfeld and Menzel (1961), paralleling Merton's (1957b) classification of group characteristics,
developed "A Typology of Properties Describing 'Collectives' and 'Members'"
(p. 526). Selvin and Hagstrom (1963) later proposed a more systematic version of the Kendall-
Lazarsfeld-Menzel scheme. The former referred to properties intrinsic to any level of group—
individual, pair, primary and secondary group, total society—each of which could be aggregated
upward, so that an "integral characteristic becomes the basis for aggregative characteristics
of successively higher-level groups" (p. 404); moving toward smaller units, "an individual or
a group below the total society can be described 'contextually' by the characteristics of successively
larger groups to which it belongs (p. 404). Selvin (1960) also identified subtle and
insidious forms of the ecological fallacy. This interest in group properties, in levels of aggregation,
and in the connection between social structure and individual conduct characterized
Columbia sociology for more than two decades, but scholars located elsewhere also contributed:
Robinson (1950) on ecological correlation; Inkeles (1963) on the relation between psychological
and sociological levels of analysis; Davis (1966) and his colleagues on statistical
problems in analyzing individual and collective evidence (Davis, Spaeth, & Huson, 1961;
Tannenbaum & Bachman, 1964).
There are palpable differences between styles of work emphasizing general analytic methods
and those focused on descriptive analysis, though they should not be exaggerated. Related
to this distinction is a preference for the generalizing efforts of surveys and the particularizing
character of case studies, though neither one precludes the other. An early attempt to reconcile
the two can be found in Coleman's essay, "Relational Analysis: The Study of Social Organizations
with Survey Methods" (1958-59), a discussion of how to develop indices of social structure
using surveys. He drew his examples from Columbia studies exemplifying structural effects
and the use of sociometric methods to characterize networks. (The Adolescent Society
[Coleman, 1961] employed this strategy.) A major theme of his article, however, was to show
how advancements in computer technology could contribute to new survey methods designed
to investigate social organization (1958-59, p. 28) by adapting them to deal with relationships
as units of analysis and by avoiding the atomism of past survey methodology



114 Robert Dreeben
It would be misleading to suggest that Columbia sociology at the time bifurcated into
methodological and substantive camps. Yet the affinities between Merton and Blau, on the
one hand, and between Lazarsfeld and Coleman, on the other, are unmistakable. Blau's agenda
was to explore the nature of work and authority in organizations. He employed a Durkheiminspired
methodology to examine substantive propositions about organizations derived largely
from Weber (via Merton, 1940; Merton, Gray, Hockey, & Selvin, 1952). Coleman contributed
to the organizational analysis of the International Typographical Union (Lipset et al., 1956)
but later moved away from questions about organizational structure and toward the application
of survey methods to studying school organization and its effects. The subsequent history
of the sociology of education was powerfully influenced by this development in his thinking,
particularly by Equality of Educational Opportunity (Coleman et al., 1966), a study whose
reliance on survey methods minimized the attention it could pay to the particulars of school
organization. That history was also influenced by the fact that those who participated in
Columbia's tradition of organizational analysis (e.g., Blau, 1955; Chinoy, 1955; Gouldner,
1954; Lipset, 1950; Lipset et a l , 1956; Selznick, 1949; Sills, 1957) had little if any interest in
research on educational organization (save for Blau's [1973] and Lazarsfeld and Thielens's
[1958] studies of academia).
Lazarsfeld and Merton (1954) addressed the difference between the two styles of social
science in their essay "Friendship as A Social Process" on the formation of and change in
friendship patterns among residents of housing projects to demonstrate both the complementarity
and distinctiveness of the two perspectives (p. 19). Merton's section of the essay dealt with
how friendships form, continue, and dissolve (p. 24); it conceptualized interaction and social
process and the structures emerging from them over time. Lazarsfeld's section formalized the
argument. The implication of this joint effort was that methodology should be directed as
much to understand the structure, the interaction patterns, and the mechanisms of social process
as to codify the analysis of empirical associations. The larger point was that the two
perspectives were applied together, guided by the formulation and analysis of a substantive
question. Lazarsfeld's section brought little to the essay that was not already available in
Kendall and Lazarsfeld (1950), except as it applied to friendship. It is hard to imagine
Lazarsfeld's section preceding Merton's, which is not to gainsay the former's contribution to
identifying gaps in the substantive argument and in generalizing it. The Lazarsfeld-Merton
essay brought together two strands of the Columbia tradition, one based on the logic of survey
analysis, the other on organizational studies and their concern with social form and process. It
is difficult to detect any residue it may have left in later studies of school effects.
COLEMAN'S CONTRIBUTION
Equality of Educational Opportunity (EEO; Coleman et al., 1966) manifested an implicit
connection between the idea of school climate and its intellectual antecedents in Durkheim,
mediated through Columbia sociology. A curiosity of this work is the inexplicitness of its
conceptual scheme and origins. That EEO does not treat its own origins, however, does not
mean it lacks them. The conventional view is that EEO employed the economic formulation of
the production function, a form of input-output analysis (Averch, Carroll, Donaldson, Kiesling
& Pincus, 1972; Hanushek, 1986; Lau, 1979), and in a partial sense this is so. It was also
propelled by an unexpressed argument originating in the same traditions that generated the
formulations of Wilson and Blau. The structure-conduct linkage in that argument, moreover,
is highly compatible with the input-output logic of production functions.

Hasil (Bahasa Indonesia) 1:

Struktural efek dalam pendidikan 113kasus tentara dengan higlier tingkat pendidikan yang lebih mungkin dibandingkan dengan tingkat yang lebih rendahuntuk mengatakan bahwa mereka telah secara sukarela atau harus tidak dirapatkan. Kendall dan Lazarsfeldmenafsirkan temuan dengan berargumen bahwa tentara yang diterima induksi berpendidikan lebih baik karenamereka datang dari lingkungan sipil di mana beberapa orang itu ditangguhkan, dan dengan demikian memiliki tidak sahharapan penangguhan karena mereka pendidikan (p. 149). Interpretasi ini adalahBerdasarkan para prajurit keanggotaan dalam kategori sosial (yaitu, lingkungan dengan deferments jarang)mempengaruhi penilaian diri mereka dan keyakinan tentang kelayakan menjadiTentara; terutama didasarkan pada proses berpikir para prajurit, mungkin atau tidak mungkin benar.Apapun masalahnya, tujuan dari latihan ini adalah lebih untuk mengembangkan logika dan analisis daripada kemengejar rincian substantif secara empiris.Untuk mengeksplorasi efek struktural, Kendall dan Lazarsfeld (1950) dikembangkan Formalisasihubungan antara Deskripsi individu (data pribadi) dan agregat, interaksi,Kategori, dan kelompok (unit data), menciptakan tipologi untuk mewakili data bagaimana pribadidan unit data berlapis Check hubungan antara tingkat agregasi. Upaya ini menyimpulkandengan diskusi tentang apakah data pribadi dan unit data dapat digunakan secara bergantian,pertanyaan yang datang kepada kita sebagai ekologi korelasi (Robinson, 1950) dan ekologifallacy (Selvin, 1958, 1960). "Our main interest here is directed toward the logicalrelationship between personal data and unit data" (Kendall & Lazarsfeld, p. 188; my italics),toward developing general principles of data analysis applicable to surveys regardless of content.This enterprise differed from Merton's.Lazarsfeld and Menzel (1961), paralleling Merton's (1957b) classification of group characteristics,developed "A Typology of Properties Describing 'Collectives' and 'Members'"(p. 526). Selvin and Hagstrom (1963) later proposed a more systematic version of the Kendall-Lazarsfeld-Menzel scheme. The former referred to properties intrinsic to any level of group—individual, pair, primary and secondary group, total society—each of which could be aggregatedupward, so that an "integral characteristic becomes the basis for aggregative characteristicsof successively higher-level groups" (p. 404); moving toward smaller units, "an individual ora group below the total society can be described 'contextually' by the characteristics of successivelylarger groups to which it belongs (p. 404). Selvin (1960) also identified subtle andinsidious forms of the ecological fallacy. This interest in group properties, in levels of aggregation,and in the connection between social structure and individual conduct characterizedColumbia sociology for more than two decades, but scholars located elsewhere also contributed:Robinson (1950) on ecological correlation; Inkeles (1963) on the relation between psychologicaland sociological levels of analysis; Davis (1966) and his colleagues on statisticalproblems in analyzing individual and collective evidence (Davis, Spaeth, & Huson, 1961;Tannenbaum & Bachman, 1964).There are palpable differences between styles of work emphasizing general analytic methodsand those focused on descriptive analysis, though they should not be exaggerated. Relatedto this distinction is a preference for the generalizing efforts of surveys and the particularizingcharacter of case studies, though neither one precludes the other. An early attempt to reconcilethe two can be found in Coleman's essay, "Relational Analysis: The Study of Social Organizationswith Survey Methods" (1958-59), a discussion of how to develop indices of social structureusing surveys. He drew his examples from Columbia studies exemplifying structural effectsand the use of sociometric methods to characterize networks. (The Adolescent Society[Coleman, 1961] employed this strategy.) A major theme of his article, however, was to showhow advancements in computer technology could contribute to new survey methods designedto investigate social organization (1958-59, p. 28) by adapting them to deal with relationshipsas units of analysis and by avoiding the atomism of past survey methodology114 Robert DreebenIt would be misleading to suggest that Columbia sociology at the time bifurcated intomethodological and substantive camps. Yet the affinities between Merton and Blau, on theone hand, and between Lazarsfeld and Coleman, on the other, are unmistakable. Blau's agendawas to explore the nature of work and authority in organizations. He employed a Durkheiminspiredmethodology to examine substantive propositions about organizations derived largelyfrom Weber (via Merton, 1940; Merton, Gray, Hockey, & Selvin, 1952). Coleman contributedto the organizational analysis of the International Typographical Union (Lipset et al., 1956)but later moved away from questions about organizational structure and toward the applicationof survey methods to studying school organization and its effects. The subsequent historyof the sociology of education was powerfully influenced by this development in his thinking,particularly by Equality of Educational Opportunity (Coleman et al., 1966), a study whosereliance on survey methods minimized the attention it could pay to the particulars of schoolorganization. That history was also influenced by the fact that those who participated inColumbia's tradition of organizational analysis (e.g., Blau, 1955; Chinoy, 1955; Gouldner,1954; Lipset, 1950; Lipset et a l , 1956; Selznick, 1949; Sills, 1957) had little if any interest inresearch on educational organization (save for Blau's [1973] and Lazarsfeld and Thielens's[1958] studies of academia).Lazarsfeld and Merton (1954) addressed the difference between the two styles of socialscience in their essay "Friendship as A Social Process" on the formation of and change infriendship patterns among residents of housing projects to demonstrate both the complementarityand distinctiveness of the two perspectives (p. 19). Merton's section of the essay dealt withhow friendships form, continue, and dissolve (p. 24); it conceptualized interaction and socialprocess and the structures emerging from them over time. Lazarsfeld's section formalized theargument. The implication of this joint effort was that methodology should be directed asmuch to understand the structure, the interaction patterns, and the mechanisms of social processas to codify the analysis of empirical associations. The larger point was that the twoperspectives were applied together, guided by the formulation and analysis of a substantivequestion. Lazarsfeld's section brought little to the essay that was not already available inKendall and Lazarsfeld (1950), except as it applied to friendship. It is hard to imagineLazarsfeld's section preceding Merton's, which is not to gainsay the former's contribution toidentifying gaps in the substantive argument and in generalizing it. The Lazarsfeld-Mertonessay brought together two strands of the Columbia tradition, one based on the logic of surveyanalysis, the other on organizational studies and their concern with social form and process. Itis difficult to detect any residue it may have left in later studies of school effects.COLEMAN'S CONTRIBUTIONEquality of Educational Opportunity (EEO; Coleman et al., 1966) manifested an implicitconnection between the idea of school climate and its intellectual antecedents in Durkheim,mediated through Columbia sociology. A curiosity of this work is the inexplicitness of itsconceptual scheme and origins. That EEO does not treat its own origins, however, does notmean it lacks them. The conventional view is that EEO employed the economic formulation ofthe production function, a form of input-output analysis (Averch, Carroll, Donaldson, Kiesling& Pincus, 1972; Hanushek, 1986; Lau, 1979), and in a partial sense this is so. It was alsopropelled by an unexpressed argument originating in the same traditions that generated theformulations of Wilson and Blau. The structure-conduct linkage in that argument, moreover,is highly compatible with the input-output logic of production functions.

Hasil (Bahasa Indonesia) 2:

Efek struktural dalam Pendidikan 113
kasus tentara dengan tingkat higlier pendidikan menjadi lebih mungkin dibandingkan dengan tingkat yang lebih rendah
untuk mengatakan bahwa mereka telah secara sukarela atau tidak seharusnya ditangguhkan. Kendall dan Lazarsfeld
ditafsirkan temuan dengan menyatakan bahwa tentara yang lebih terdidik menerima induksi karena
mereka berasal dari lingkungan sipil di mana beberapa orang yang ditangguhkan, dan dengan demikian tidak sah
harapan penundaan karena pendidikan mereka (hal. 149). Interpretasi ini
didasarkan pada keanggotaan tentara dalam kategori sosial (yaitu, lingkungan dengan deferments jarang)
mempengaruhi penilaian diri dan keyakinan tentang kelayakan berada di mereka
Darat; terutama didasarkan pada proses berpikir tentara, hal itu mungkin atau mungkin belum benar.
Apapun masalahnya, tujuan dari latihan lebih untuk mengembangkan logika analisis daripada
mengejar rincian substantif secara empiris.
Untuk mengeksplorasi efek struktural, Kendall dan Lazarsfeld (1950) mengembangkan formalisasi
hubungan antara deskripsi individu (data pribadi) dan agregat, interaksi,
kategori, dan kelompok (data satuan), menciptakan tipologi untuk mewakili bagaimana data pribadi
dan data satuan yang berlapis untuk mengekspresikan hubungan antara tingkat agregasi. Upaya ini menyimpulkan
dengan diskusi tentang apakah data dan satuan data pribadi dapat digunakan secara bergantian,
pertanyaan yang datang ke kita sebagai korelasi ekologis (Robinson, 1950) dan sebagai ekologi
kekeliruan (Selvin, 1958, 1960). "Perhatian utama kami di sini diarahkan logis
hubungan antara data pribadi dan data unit "(Kendall & Lazarsfeld, hal 188;. miring saya),
. untuk mengembangkan prinsip-prinsip umum analisis data yang berlaku untuk survei terlepas dari konten
perusahaan ini berbeda dari Merton .
Lazarsfeld dan Menzel (1961), paralel (1957b) klasifikasi Merton karakteristik kelompok,
mengembangkan "Tipologi Properti Menggambarkan 'Kolektif' dan 'Anggota'"
(hal. 526). Selvin dan Hagstrom (1963) kemudian mengusulkan versi yang lebih sistematis Kendall-
skema Lazarsfeld-Menzel. Mantan disebut sifat intrinsik untuk setiap tingkat kelompok-
individu, pasangan, kelompok primer dan sekunder, jumlah masyarakat-yang masing-masing dapat dikumpulkan
ke atas, sehingga sebuah "karakteristik yang tidak terpisahkan menjadi dasar karakteristik agregatif
dari berturut-turut kelompok-tingkat yang lebih tinggi "(hal 404.); bergerak ke arah unit yang lebih kecil, "seorang individu atau
kelompok bawah masyarakat keseluruhan dapat digambarkan 'kontekstual' oleh karakteristik berturut-turut
kelompok yang lebih besar mana ia berasal (hlm. 404). Selvin (1960) juga mengidentifikasi halus dan
bentuk berbahaya dari . kekeliruan ekologi bunga ini dalam sifat kelompok, tingkat agregasi,
dan hubungan antara struktur sosial dan perilaku individu ditandai
Columbia sosiologi selama lebih dari dua dekade, tapi ulama yang terletak di tempat lain juga memberikan kontribusi:
Robinson (1950) korelasi ekologis; Inkeles ( 1963) pada hubungan antara psikologis
dan sosiologis tingkat analisis, Davis (1966) dan rekan-rekannya di statistik
masalah dalam menganalisis bukti individu dan kolektif (Davis, Spaeth, & Huson, 1961;
. Tannenbaum & Bachman, 1964)
Ada perbedaan teraba antara gaya kerja menekankan metode analitik umum
dan orang-orang terfokus pada analisis deskriptif, meskipun mereka tidak perlu dibesar-besarkan. Terkait
dengan perbedaan ini adalah preferensi untuk upaya generalisasi survei dan particularizing
karakter studi kasus, meskipun tidak satu pun menghalangi yang lain. Sebuah usaha awal untuk mendamaikan
kedua dapat ditemukan dalam esai Coleman, "Analisis Relasional: Studi Organisasi Sosial
dengan Metode Survey "(1958-1959), diskusi tentang bagaimana mengembangkan indeks struktur sosial
menggunakan survei. Dia menarik contoh-contohnya dari studi Columbia mencontohkan efek struktural
dan penggunaan metode sociometric untuk mengkarakterisasi jaringan. (The Remaja Masyarakat
[Coleman, 1961] dipekerjakan strategi ini.) Sebuah tema utama artikelnya, bagaimanapun, adalah untuk menunjukkan
bagaimana kemajuan teknologi komputer dapat memberikan kontribusi untuk metode survei baru yang dirancang
untuk menyelidiki organisasi sosial (1958-1959, hal. 28 ) dengan mengadaptasi mereka untuk berurusan dengan hubungan
sebagai unit analisis dan dengan menghindari atomisme metodologi survei terakhir 114 Robert Dreeben Akan menyesatkan untuk menyarankan bahwa Columbia sosiologi pada saat itu dipisahkan antara kamp metodologis dan substantif. Namun kedekatan antara Merton dan Blau, di satu sisi, dan antara Lazarsfeld dan Coleman, di sisi lain, yang jelas. Agenda Blau adalah untuk mengeksplorasi sifat pekerjaan dan wewenang dalam organisasi. Dia mempekerjakan Durkheiminspired metodologi untuk memeriksa proposisi substantif tentang organisasi sebagian besar berasal dari Weber (via Merton, 1940; Merton, Gray, Hoki, & Selvin, 1952). Coleman berkontribusi terhadap analisis organisasi dari tipografis Union (Lipset et al., 1956) , tetapi kemudian pindah jauh dari pertanyaan tentang struktur organisasi dan arah penerapan metode survei untuk mempelajari organisasi sekolah dan dampaknya. Sejarah selanjutnya dari sosiologi pendidikan sangat dipengaruhi oleh perkembangan ini dalam pemikirannya, terutama oleh Kesetaraan Peluang Pendidikan (Coleman et al., 1966), sebuah studi yang ketergantungan pada metode survei diminimalkan perhatian yang bisa membayar untuk ihwal sekolah organisasi. Sejarah itu juga dipengaruhi oleh fakta bahwa mereka yang berpartisipasi dalam tradisi Columbia analisis organisasi (misalnya, Blau, 1955; Chinoy, 1955; Gouldner, 1954; Lipset, 1950; Lipset et al, 1956; Selznick, 1949; Sills, 1957 ) memiliki sedikit jika ada minat dalam penelitian tentang organisasi pendidikan (kecuali Blau [1973] dan Lazarsfeld dan Thielens itu [1958] studi akademisi). Lazarsfeld dan Merton (1954) membahas perbedaan antara dua gaya sosial ilmu dalam esai mereka " Persahabatan sebagai Proses Sosial "pada pembentukan dan perubahan pola persahabatan antara penduduk proyek perumahan untuk menunjukkan kedua saling melengkapi dan kekhasan dari dua perspektif (hal. 19). Bagian Merton esai berurusan dengan bagaimana persahabatan bentuk, terus, dan melarutkan (p 24.); itu dikonsep interaksi sosial dan proses dan struktur yang muncul dari mereka dari waktu ke waktu. Bagian Lazarsfeld yang diformalkan argumen. Implikasi dari upaya bersama ini adalah bahwa metodologi harus diarahkan sebagai banyak untuk memahami struktur, pola interaksi, dan mekanisme proses sosial untuk menyusun analisis asosiasi empiris. Titik yang lebih besar adalah bahwa dua perspektif yang diterapkan bersama-sama, dipandu oleh perumusan dan analisis substantif pertanyaan. Bagian Lazarsfeld yang membawa sedikit esai yang belum tersedia di Kendall dan Lazarsfeld (1950), kecuali karena diterapkan pada persahabatan. Sulit untuk membayangkan bagian Lazarsfeld yang sebelumnya Merton, yang bukan untuk membantah para mantan kontribusi mengidentifikasi kesenjangan dalam argumen substantif dan generalisasi itu. The Lazarsfeld-Merton esai dibawa bersama dua helai tradisi Columbia, yang didasarkan pada logika survei analisis, yang lain pada studi organisasi dan keprihatinan mereka dengan bentuk sosial dan proses. Hal ini sulit untuk mendeteksi residu mungkin telah meninggalkan dalam studi kemudian efek sekolah. KONTRIBUSI COLEMAN'S Kesetaraan Peluang Pendidikan (EEO;. Coleman et al, 1966) diwujudkan implisit hubungan antara ide iklim sekolah dan anteseden intelektualnya di Durkheim , dimediasi melalui Columbia sosiologi. Sebuah rasa ingin tahu dari pekerjaan ini adalah inexplicitness nya skema dan asal konseptual. Itu EEO tidak memperlakukan asal sendiri, bagaimanapun, tidak berarti itu tidak memiliki mereka. Pandangan konvensional adalah bahwa EEO digunakan rumusan ekonomi fungsi produksi, suatu bentuk analisis input-output (Averch, Carroll, Donaldson, Kiesling & Pincus, 1972; Hanushek, 1986; Lau, 1979), dan dalam arti parsial ini begitu. Hal itu juga didorong oleh argumen terpendam yang berasal dari tradisi yang sama yang dihasilkan formulasi Wilson dan Blau. Struktur-perilaku hubungan dalam argumen itu, apalagi, sangat kompatibel dengan logika input-output dari fungsi produksi.