IIIl_ ‘ {leT‘I‘IILIJ‘C' 1'! \N “ 'I‘JP 3 :_‘I ~ -: '{NzlfI' LI-I‘RI' ,\‘\ , yam-3.5:. .3 I; :(HI “I: '9‘""“’" """C ' ,i‘ " II: I'-' f". I ' ' .“.‘.‘C.¢'ICI‘. },I'§¢.‘ "JI\I§I I {II I'xy." WWI that. I I ;. fr; )3 '1. - —-‘ I: I"? \"II'I'II. ' W". l\ ‘ I ' I ' I I . I .fiyfinIIEI'N {Zn-5". :‘W'hI' Inge!” \IIILL? ' .)".' ‘ , .3 ‘IA‘ I'I -- ‘I .IW‘II'S :U ‘Iv ‘4 I AI)” 1‘3“.)53‘9 L "‘~"I'-":“(II‘"‘JI.' RIII HI; 1'" 5" II. n... ¢ Ier‘cll .I ‘l nun-1“) in”) '( .Ik‘m .‘I, f w..» 81‘ Ann.» 3:17. :I’f‘II'IIWJ‘V" 'IIY.I:?‘tI.‘I"§{II“.L-"III2$ “55%" . C'3I'I""II"IL " ' IIrI, 1" “:Q‘II‘ ”w. I":.").'!“II.‘: Ir: (:‘IJJECI' n£Wva5“INW WJMM MAW .: I \Z." 1‘5"! I ”it. :“Vf'tlfigr' _ "II ' II :- 'V;V'U:F’"g. E{' :IIInIgin-Lflw‘nv “:13." “IRA! II . i . I \:.':‘f‘:’;:$:3'3 Fact“ “(‘1 ‘33,» 5. W54)”. #3.": {’1‘ C . . Ck 0' mi“ K-I'I I I 0551'..¢( r\‘) :‘L': “1 3A" . S$‘wrtx"¢1""1‘:‘2f{'” C)".- ' ’35“. .II )IIIL' “3‘," z. 5.1““ " _' .I-III'JIIIM ‘ ' ‘fr'I'u II ‘ (I “"‘L ME: {H.il. Tad"; I' I t.“ LII, ‘I:"7- lbw? MILK}? I‘l‘I'I ”IA “:2: ‘ " In I'II'J'IH II M- 'I' 1.).” '9” 'IC"‘I I“--‘:: ,.:' '_‘ . 7-, '.",I'I' :3; II ”Ins: .\‘.. milk"... “1' c‘c ‘ I .' j . I - Il-ti' \ , Liv." .1. 'I in]? ~~;>;‘: "I." I'.‘I ll Il‘ {VCR}. V 13.0 “‘I‘E‘V'I1QII‘IE:|D:'I‘ "in.” II '|"'V":'1:"':" . 6"- I‘! III' I . I :h. {KI/M I :Q‘IA'I 11'“ "."‘:'.‘71I:|: ‘I I I I " t" .':"|fl|"'I"’ "3 . i“ .‘ ‘C.‘x“':""‘).\\" | EN \' I '1'"?! 7 "'91?an l‘ .IK “ 51' ' , M “:":'¢-':'.I1:I:v ‘ ‘HFT - 1“,“: ‘1'.“ "I'.‘\'.-' I‘ia' “ ‘ #)';J‘:';V \v'; at?! '. _ ‘g‘fi’Q'vIII '13-" .I::'J:'I'l‘\'t' Whig ‘ ‘I .‘ ,‘I"-"‘I'.' 'C' '3‘ Wuhfi EU I.I\5.; \‘u’u 1;.“ . 5;..5: , I“I: : - v - 9maXur” I. w: 3 n ‘* I\-I‘¢"Ih\ I :‘I ah: $1: I‘xwvlz.‘ -.II;-. . 53.53%”? . "."Iwi. "”1"“ 'mnv' "MWI' III ‘I " I'.‘-, I'II'III' “Ml?“ Illl(" . 6?}. $1 ‘I‘II‘II ‘ID...£ . H “It” ”gym" ‘ :fl'i'u‘u' I ‘35:.le . I“'g(’\'uv:\'\‘:~'."i}.{b~= ’2 §LVWH I' II“ Iy'pl". 'I?“ . L‘." ‘I'\" ‘t‘V| IM"I "”XV‘IIWI I. " I VII.” JILIMW~ v I “l ’1 '33 “' "*’I"1‘[‘I§Izl"‘ja.:f I: A 'W ' 'nm iv “HIM? 'H.|...'\‘ “I. I:,:;' l .:om=oa acoucmp=_ama=m .a mo.m mm.om oo.o eoppazpb< xupzuaa .m a. .3 a... 3.222.. .wfimfiwm .. no~.mm aaa.ma Noo.o cowampauwcpaz coupauacu cocoa: .m uo~.e amo.m_ woo.o apex baoaoao .N m~.m~ mm.¢m om.ma pcmea>mwgu< “causum ._ com: gap: 304 m_ampgc> .me> Foozum omimmmp mew mcwgso mpopgum_o Poocum Npug commsuPz cw zuw>mmco4 ucmucmucpgmaam ecu cowupguu< zupzumm .mmepmcmeEmm avgmcmpogum “Pew: chowumz .cowumpzowgumz cowumusum Lunar: .oamm «sauces .ucmsm>mw;u< ucmuaum com cam: use ma:mzu-.¢um m4m< .m No.- .mmi co. Amosv a~am bowcbmwo .N No.- ac. mmq Panza can cowamzpm> umepmacm mumpm ._ m N p mpnmwgm> .Loggm ugmucmum mum mmewu mmggu can» meozaa "Cm 0" a" mom A_m mmv Pm mm mm. ~_ooooo. Foooooo.- .vop.- aaapam coautumw=_sa< mmaaa>< .m mm. «Po. Ne. ..mo. Amosv mNCm powepmso .N mo. mmoooooo. mmoooooo. aims. .aaza can :o_uaspm> umepmacm mpmpm .P L m\mm m comm a_aaaaa> .xngom Loumgumwcwsc< mmmgm>< use Amogv macaw co copumpucmgommwo m>wpmgumwcweu< mo cowpwmoqeoumo new cowmmmgmmm mpnaupzzuu.mim m4m umNPPmscm 102 upon the dependent variable in the metric of the respective independent and dependent variables; (3) the standard error of the nonstandardized regression coefficient, SE/B, which is used to determine the statistical significance of the relationship between the independent and the dependent variable; (4) the simple correlation, r, which is equivalent to the zero-order standardized regression coefficient without controls; (5) the squared coefficient of multiple correlation or coefficient of determination, R2, which indicates the amount of the variability of the dependent variable accounted for by the independent variables in the regression equation; (6) the number of cases, n, from which the statistics were computed; and (7) the magnitude of the regression coefficient relative to the magnitude of its standard error, * or **, which, as was indicated above, reveals the statistical significance of the relationship between the independent and dependent variable, other conditions equal. (A regression coefficient that is less than twice its standard error is considered to be insignificant.) These statis- tics were computed at the Michigan State University Computer Center using the REGRESSION subprogram of the "Statistical Package for the Social Sciences, Version 6.5" (Michigan State University, 1976). The lower portion of Table 3-5 decomposes the relationships observed in the upper portion of the table. The underlined values in the diagonal of the matrix are the standardized regression coefficients (Beta's) from the upper portion of the table. As was indicated above, these values represent the direct effect of the variable in the row, x, upon the dependent variable, y (administrative differentiation). 103 Values outside the diagonal represent the indirect connections between the variable in the row, x, and the dependent variable, y, produced by the variable in the column, i. These values are obtained by multiplying the simple correlation between variables in the respective row and column, r1.x (see Appendix B), by the standard- ized regression coefficient, Betayi. Except for rounding errors, the sum of the values in each row is equal to the simple correlation between x and y shown in the upper portion of the table. Assuming that the variables in the equation are ordered according to their causal sequence, values to the left of the diagonal represent the amount of the association between x and y which is spurious due to the effects of common antecedents. Values to the right of the diagonal represent the indirect effects of x on y which are mediated by the intervening variables listed in the columns. Decomposition coefficients of less than .10 are unimportant and receive no atten- tion in the analysis. Anticipating the analysis in Chapter IV, the regression equation summarized in Table 3-5 indicates that the extent of administrative differentiation in Michigan K-12 school districts is almost entirely a functioncfiithe size of the school district (row 2). The decomposition coefficient to the left of the diagonal in the second row of the lower portion of Table 3-5 (d.c. = .00) indicates that no portion of the simple correlation between district size (log) and administrative differentiation (r = .89) is spurious. In addi-' tion, the very small decomposition coefficient to the right of the diagonal in the second row of the lower portion of the table (d.c. = 104 -.O7) indicates that the influence of the independent upon the dependent variable is virtually unmediated. Thus the larger the school district, the greater the spread of district administrators across specialized assignment categories. More specifically, the nonstandardized regression coefficient (8) in the second row of the upper portion of Table 3-5 indicates that, on the average, a one unit (i.e., ten teacher) increase in the size of a school district is associated with a 42% increase in the extent of administrative differentiation. In addition, administrative differentiation is somewhat contingent upon the level of administrative salaries and the value of local residential, commercial and industrial property per pupil. Although the decomposition coefficient to the left of the diagonal in the third row of the lower portion of the table indicates that the simple correlation between average administrator salary and administrative differentiation (r = .56) is largely spurious due to the antecedent influence of district size (log) upon both variables (d.c. = .65), the small but statistically significant regression coefficient (Beta = -.10) indicates that this factor has a negative impact upon the dependent variable when the effect of district size is controlled. This finding makes sound intuitive sense: the greater the proportion of administrative assignment categories occupied in a school district, the greater the probability of incumbancy in one of the less highly remunerated administrative positions--e.g., assistant principal, school-community coordinator, etc. The nonstandardized regression coefficient (8) in the third row 105 of the upper portion of the table indicates the extent of this contingency in the metric of the variables under consideration: all else equal, a one unit increase in the extent of administrative differentiation is associated with a reduction of 6,757.38 units (dollars) in the salary paid to the typical district administrator (mean = $20,975). By the same token, the extent of administrative differentia- tion is also somewhat contingent upon the value of the taxable real estate and personal property in the district (row 1). It is not entirely clear why state equalized valuation per pupil (which is a measure of raw tax base) should achieve statistical significance when neither operating millage nor local revenue per pupil do so when entered into the same multiple regression equation (not presented). However, given the assumptions concerning the causal sequence of the variables in the equation, the first row of the lower portion of. Table 3-5 indicates that no portion of the simple correlation between state equalized valuation per pupil and administrative differentiation (r = .09) is spurious and that its influence upon the dependent variable is virtually unmediated. However, when the standardized regression coefficient (Beta) in the first row of the upper portion of the table is translated into the metric of the original measures (8), the importance of the relation is largely academic: it would require an increase of approximately $50,000 in state equalized valuation per pupil (mean = $26,037) to change the extent of administrative differentiation by one unit, ceteriS'paribus. 106 In summary, the larger the school district, the greater the differentiation of the administrative hierarchy into specialized administrative functions. Further, although large school districts tend to have higher average administrator salaries than small school districts (r = .70), the salary differential among administrative positions is such that extensive administrative differentiation has the effect of reducing the average salary of district administrators. Although the nexus between state equalized valuation per pupil and administrative differentiation under these conditions is statistically significant, the relationship has virtually no practical importance. This chapter has discussed the hypotheses and questions to be examined in this investigation; the variables utilized to measure the environmental, contextual, structural and performance characteristics of the organizations examined; the procedures employed in selecting the study population; the sources from which data were obtained; the methods used in collecting and recording the data; and the pro- cedures utilized in the analysis of the data.' Chapter IV and Chapter V discuss the findings obtained concerning the environmental and contextual conditions and the performance consequences of the structural characteristics of Michigan K-12 school districts. CHAPTER IV FINDINGS I: ENVIRONMENTAL CONDITIONS OF THE FORMAL STRUCTURE OF MICHIGAN K-12 SCHOOL DISTRICT ORGANIZATIONS Introduction This chapter presents the results of an empirical investiga- tion of the environmental conditions of three dimensions of the formal structure of Michigan K-12 school districts: division of labor, hierarchy of authority and administrative apparatus. The inquiry answers two fundamental questions: (1) What are the environmental conditions of the structural characteristics of Michigan K-12 school districts? and (2) To what extent is the formal structure of Michigan K-12 school districts homologous to the formal structure of other organizations despite differences in goals, functions and internal procedures? Each dimension of school district organization is presented in order. Following a description of the indicators utilized to measure each dimension, the discussion proceeds to identify those factors in the external and internal environments of Michigan school districts which have a significant influence upon each dimension. The fbllowing chapter is concerned with the performance consequences of these three dimensions of school district organization. 107 108 Division of Labor Formal organizations are characterized by an extensive division of labor whereby organizatiOnal functions are differentiated into more or less specialized occupational positions and distributed across a variety of differentiated organizational subunits and/or work locations. Since no single measure can capture the full com- plexity of this variable, this investigation employs four indicators of school district division of labor: faculty differentiation, faculty distribution, faculty dispersion and faculty per building. Faculty_differentiation refers to the number of teaching assignment categories (e.g., social science, mathematics, elementary education, vocational education, etc.) filled by the district staff. This information is collected routinely by the Michigan Department of Education for inclusion in the Register of Professional Personnel (a personnel record system maintained on magnetic tape by the Michi- gan Department of Education). The specific measure is the propor- tion of ninety-five teaching assignment categories actually occupied as first or second assignments by district faculty members during the 1975-76 school year. The higher the district faculty differ- entiation score, the greater the number of teaching assignment categories occupied by the district faculty. In addition, since several teaching assignment categories represent sub-specialties of a more general category (e.g., "sociology," "psychology" and "anthropology" are classified as sub-specialties of "social science"), faculty differentiation also measures the extent of faculty speciali- zation. Thus, the higher the district faculty differentiation score, 109 the greater the number of specialized teaching assignment categories occupied by the district faculty. Faculty members in the typical Michigan K-12 school district occupy 30.4 teaching assignment categories as first or second assignments. Faculty distribution refers to the spread of faculty across occupied teaching assignment categories. Data for this measure are derived from the same source as faculty differentiation but are manipulated to indicate the extent to which staff members are either concentrated within a relatively small number of teaching assignment categories or evenly distributed across a wide range of assignments. A high faculty distribution score indicates that the staff occupy a broad range of teaching assignment categories and that they are relatively evenly distributed within those categories. A low district faculty distribution score indicates that the faculty occupy a narrow range of teaching assignments with relatively high concentrations within one or a few categories. The specific measure for faculty distribution is the Gibbs and Martin (1962, 1966) formula for mea- suring division of labor: 1 _ sum x2 (sum x) where the unit of analysis (x) equals the number of district faculty members occupying a faculty assignment category as a first or second assignment. The typical Michigan K-12 school district has a faculty distribution score of .82, ranging from a low of .67 to a high of .90. 110 Whereas faculty differentiation and faculty distribution measure the functional division of labor in Michigan K-12 school districts, faculty dispersion and faculty per building_measure the spatial division of academic labor. Faculty dispersion refers to the spread of faculty members across a set of district buildings or work locations. Like faculty distribution, the measure for faculty dis- persibn is based upon the Gibbs and Martin (1962, 1966) division of labor formula except that the unit of analysis (x) is equal to the number of faculty members assigned to a district building or work location. Thus, a high faculty dispersion score indicates that the district faculty are spread across a large number of buildings with relatively even distributions within each building. The mean faculty dispersion score for Michigan K-12 school districts is .69 with a low of zero (indicating that all faculty members are assigned to a single building) and a high of .99 (indicating that faculty members are spread across a relatively large number of district buildings with relatively equal numbers in each building. Faculty per building is a straightforward measure of the average number of faculty members assigned to district buildings or work locations and is obtained by dividing the total number of faculty by the total number of occupied classroom buildings. The higher the faculty per building score, the larger the number of faculty members assigned to district buildings. On the average, Michigan K-12 school districts have 22.56 faculty members per district building, with a high of 54.00 and a low of 5.67. (Faculty 111 per building differs from principal span of control since some principals supervise more than one building or work location.) In summary, the division of academic labor in Michigan K-12 school districts is measured by fbur variables: faculty differentia- tion, faculty distribution, faculty dispersion and faculty per build- ing. The first two measure the spread of district faculty members across a finite set of teaching assignment areas; the last two mea- sure their spatial distribution within school district buildings. The remainder of this section documents the environmental conditions of each dimension of school district division of labor. Faculty Differentiation What environmental conditions influence the extent of faculty differentiation in Michigan K-12 school districts? Studies reviewed in an earlier section indicate that the key determinant of an organization's division of labor is the size of the organization: the larger the organization, the greater the spread of employees across occupational positions and work locations. Does this pattern apply to the formal structure of educational organizations as well? Do other environmental conditions have any effect upon this aspect of the formal structure of school district organizations? Specif- ically, what factors determine the proportion of ninety-five teaching assignment categories actually occupied as first or second assignments by district faculty members in Michigan K-12 school districts during the 1975-76 school year? 112 Table 4-1 summarizes the multiple regression equation that includes all the factors with a significant influence upon the extent of faculty differentiation in Michigan K-12 school districts. The beta weight (Beta) of school district size (109) represents over ninety-three percent of the simple correlation (r) between the independent and dependent variables. The decomposition coefficient TABLE 4-1.--Multip1e Regression and Decomposition of Faculty Dif- ferentiation on District Size (Log) and Faculty Qualifications. Variable Beta 8 SE/B r 1. District Size (Log) .85** .20 .0054 .91 2. Faculty Qualifications .09** .00063 .00017 .61 R2 = .83 (E = .83); n = 508. ** More than three times its standard error. Variable l 2 1. District Size (Log) ;§§_ .06 2. Faculty Qualifications .53 .09 in the first row of the lower portion of Table 4-1 indicates that no part of the influence of district size upon faculty differentiation .06) is is spurious and that only a very small portion (d.c. mediated by the other independent variable. Does this mean, as some have suggested (Hall, 1972), that this aspect of the division of labor is simply a surrogate for the size of the school district? Not 113 at all. Although a school district must have some minimum number of faculty members to fill ninety-five faculty assignment categories as first or second assignments, there is no logical necessity for doing so. On the contrary, even a very large school district might elect, for whatever reasons, to limit the scope of its teaching staff to some minimum number of basic subject matter areas and ignore those academic, aesthetic and vocational courses which require a more highly specialized and differentiated faculty.) Furthermore, although most of the influence of faculty qualifications (i.e., the percentage of district faculty members holding an advanced degree) upon faculty differentiation is rendered spurious by the antecedent influence of school district size (d.c. = .53) upon both variables, the significant beta weight in the second row of Table 4-1 indicates that a highly differentiated faculty is also contingent upon the academic qualifica- tions of the faculty. Translating these relationships into the metric of the original measures, the influence of school district size upon the dependent variable is such that, on the average, an increase of one faculty member to the staff of a Michigan K-12 school district would increase its rate of faculty differentiation by two teaching assign- ment categories. The same rate of faculty differentiation would require an increase of approximately thirty-two percent in the number of faculty members with advanced degrees, other conditions equal. In summary, the rate of faculty differentiation in Michigan K-12 school districts is highly contingent upon the number of faculty members in the district. However, the small but significant 114 influence of the academic credentials of the faculty upon this aspect of school district division of labor suggests that faculty dif- ferentiation has both qualitative and quantitative dimensions. Further evidence of this suggestion is presented in subsequent sections. Faculty Distribution Whereas faculty differentiation measures the total number of teaching assignment categories occupied by district faculty members, faculty distribution is concerned with the number of faculty members in each teaching assignment category across the total range of teaching assignment categories. For example, given two school districts with a faculty differentiation score of .47 (i.e., the district faculty occupy forty-five teaching assignment categories), the district having the greater number (or more even distribution) of faculty in each occupied teaching assignment category will have the higher faculty distribution score. In practical terms, this might mean that whereas both districts have twelve teachers assigned to the social sciences, one district might have nine teachers assigned to "social science" and one each to "psychology," "sociology" and "anthropology." The other district might have three teachers assigned to the more general category and three each to the more specialized disciplines. Thus, while both districts have the capacity for offering the same range of courses in the social sciences, the latter district is capable of teaching more students in a broader range of more specialized courses. In short, the measure for faculty distribution is designed to answer the following question: 115 To what extent are faculty members concentrated in one or a few teaching assignment categories and to what extent are they evenly distributed across a broad range of teaching assignment categories? The multiple regression equation summarized in Table 4-2 indicates that the distribution of district faculty members within teaching assignment categories is a function of one structural and two environmental influences: the extent of faculty differentiation, the financial resources of the school district and the number of non-public school students in the district jurisdiction. TABLE 4-2.--Multip1e Regression and Decomposition of Faculty Distribu- tion on Non-Public School Membership, Operating Expense per Pupil and Faculty Differentiation. Variable Beta 8 . SE/B r 1. Non-public School Membership .13** .00054 .00018 .26 2. Operating Expense per Pupil .21** .000041 .000009 .37 3. Faculty Differentiation .26** .099 .018 .39 R2 = .21 (fi2 = .20); n = 508. ** More than three times its standard error. Variable l 2 3 l. Non-public School Membership Ll§_ .06 .08 2. Operating Expense per Pupil .04 .21 .12 3. Faculty Differentiation .04 .10 .26 116 The influence of faculty differentiation upon faculty distrib- ution is ambiguous: The spread of faculty members across teaching assignment areas provides no information about their distribution within those teaching areas. However, this ambiguity is resolved when the size of the school district is considered. The analysis of faculty differentiation (Table 4-1) indicated that this aspect of school district division of labor is overwhelmingly influenced by school district size (total number of faculty). Since the sub- stantial simple correlation between district size (log) and faculty differentiation (r = .91) precludes their inclusion in the same regression equation as independent variables, it is impossible to determine their independent effects upon the dependent variable. However, when school district size (log) is entered into the regression equation summarized in Table 4-2 instead of faculty dif- ferentiation, its smaller beta weight (.13, greater than twice its standard error) and the larger beta weights of operating expense per pupil and non-public school membership (respectively .26 and .15, both greater than three times their standard errors) plus the smaller amount of variance accounted for by the variables in the equation (R2 = .17) indicates that although faculty distribution is more highly contingent upon the size of the school district, faculty differentiation adds an increment of influence which is not accounted for by school district size. Thus school districts in which the faculty are more evenly distributed across a broader range of teach- ing assignment categories are both larger and have a more highly dif- ferentiated teaching staff, ceteris paribus. 117 The financial resources of school districts also exert an independent influence upon the extent of faculty distribution as indicated by the regression coefficient in the second row of Table 4-2. Although no part of this influence is rendered spurious by antecedent conditions, almost one-third of its impact is mediated by faculty differentiation (d.c. .12). Since operating expense per pupil fails to achieve statistical significance when entered into the regression of faculty differentiation, it must be assumed that the combination of a highly differentiated and broadly distributed faculty requires more extensive financial resources. Although it is reasonable to assume that the higher costs associated with extensive faculty differentiation and faculty distribution are a function of either student-faculty ratio, faculty experience, faculty qualifica- tions or average faculty salary, none of these conditions achieve statistical significance when entered into the regression equation with either school district size (log) or operating expense per pupil or both. The influence of non-public school membership on faculty distribution is statistically straightforward--one-ha1f of the influence represented by its simple correlation indicating a direct influence, one-half mediated by the combined influences of large financial resources and a high degree of faculty differentiation. However, the meaning of these relationships is not entirely clear. What difference should the number of private and parochial students in the district jurisdiction make for the way the faculty is dis- tributed in the public schools? Initially, the best explanation 118 seems to be that since the parents of non-public school students pay local property taxes at the same rate as the parents of public school students, their taxes actually increase the pool of financial resources available for the education of each public school student. However, when operating millage and local revenue per pupil are entered into a regression equation with the variables in Table 4-2, their regression coefficients are small and insignificant while the influence of non-public school membership upon the criterion remains virtually unchanged. A more speculative interpretation of this finding derives from an examination of the distribution of non-public school students in Michigan. Although the mean percentage of non-public school students in Michigan district jurisdictions is relatively small (5.42), the distribution is positively skewed (2.79) indicating that the mean is actually depressed by a large number of districts which have few, if any, non-public school students. The rank ordering of Michigan school districts according to percentage of non-public school students confirms this finding and reveals that districts with higher percentages of non-public school students are generally larger (or are adjacent to larger districts) and tend to be con- centrated in the metropolitan counties of southeastern Michigan. Furthermore, a visual examination of the lists of non-public schools in these counties reveals an impressive array of prestigious private and parochial schools which are widely known for the excellence of their educational programs. Assuming a highly differentiated faculty which is broadly distributed across a large number of teaching 119 specialties represents a dimension of curricular breadth and depth, than the mere presence of one or more of these prestigious, privately endowed educational institutions may have an exemplary effect upon both the educational values and the expectations of community residents and public school curricular programs. The persistent influence of the measure for non-public school membership upon this and other quality-indicating variables (c.f., pp. 189-191) suggests the importance of future research concerning these specula- tions about the influence of non-public schools upon the public schools. In summary, the distribution of district faculty members among teaching assignment categories is contingent upon the extent of faculty differentiation (which is a function of school district size), the financial resources available to the district and the number of non-public school students in the district jurisdiction. Translating these relationships into the metric of their original measures, a one percent increase in the rate of faculty distribution is associated with an average increase of ten percent in the rate of faculty differentiation, an average increase of $243.90 in district operating expenses per pupil and an average increase of 18.52% in the number of non-public school students. Faculty Dispersion Whereas faculty differentiation and faculty distribution are concerned with the functional differentiation of district faculty members, faculty dispersion and faculty per building measure their 120 spatial differentiation. The former provides a summary measure of the distribution of district faculty members within district build- ings or work locations while the latter measures staff density. Table 4-3 indicates that the distribution of district faculty members within district buildings is primarily contingent upon district size (row 1). No part of the regression coefficient of TABLE 4-3.--Multiple Regression and Decomposition of Faculty Dispersion on District Size (Log) and Operating Expense per Pupil. Variable Beta 8 SE/B r 1. District Size (Log) .85** .38 .013 .81 2. Operating Expense per Pupil -.09** -.00009 .000028 .29 R2 = .66 (fiz = .66); n = 508. ** More than three times its standard error. Variable l 2 1. District Size (Log) _gg; -.O4 2. Operating Expense per Pupil .38 -.O9 district size (log) is spurious, and its influence on the dependent variable is virtually unmediated, indicating that the faculty in larger school districts are relatively more evenly distributed throughout a larger number of district buildings than the faculty in smaller school districts, ceteris paribus. 121 The extent of faculty dispersion may be contingent to some extent upon the degree of faculty differentiation. However, when this variable is entered into the multiple regression equation of faculty dispersion instead of district size (log), the magnitude of its beta weight (.73) and the smaller amount of variance accounted fbr by the variables in the equation (R2 = .53) probably indicates the dependence of both conditions upon school district size. Assuming that this interpretation is correct, a one unit (ten teacher) increase in school district size is associated with a thirty-eight percent increase in the rate of faculty dispersion, other conditions equal. Operating expense per pupil exerts a small but significant negative effect upon the rate of faculty dispersion in Michigan K-12 school districts (row 2). Although most of this influence is rendered spurious by the overwhelming impact of school district size (d.c. - .38) upon both variables, a one percent increase in the criterion is associated with an average decrease of $109.89 in operating expense per pupil when district size is controlled. In summary, faculty members in larger school districts are relatively more evenly distributed throughout a larger number of district buildings than faculty members in smaller districts. This finding should cause no great surprise given the population densities associated with larger school districts. However, the finding that larger concentrations of teachers in a larger number of buildings requires relatively fewer financial resources per pupil is somewhat surprising. It is probably due to the additional costs required for 122 operating the one—, two- and three-room school buildings which characterize some of the smaller school districts in Michigan. Faculty Per Building_ As might be expected, the evidence with respect to the average number of faculty assigned to each district building or work location is similar to that of faculty dispersion. The large and significant beta weight in the second row of Table 4-4 indicates that, other things equal, the number of faculty per building is a function of school district size (log). In fact, under the condi- tions represented in Table 4-4, a one unit (ten teacher) increase in the number of district faculty members is associated with an average increase of 7.23 faculty members per building. TABLE 4-4.--Multiple Regression and Decomposition of Faculty per Building on Average Family Income and District Size (Log). Variable Beta 8 SE/B r 1. Average Family Income .l7** .00045 .00012 .41 2. District Size (Log) .41** 7.23 .82 .51 R2 = .28 (8 = .27); n = 508. ** More than three times its standard error. Variable l 2 1. Average Family Income .17 .23 2. District Size (Log) .10 .41 123 Why the affluence of district residents should exert a major influence upon the average number of faculty per building (row 1) remains unclear. Perhaps average family income is a surrogate for other indicators of school district affluence. However, none of these (i.e., state equalized valuation per pupil, operating millage, local revenue per pupil or operating expense per pupil) or any other environmental condition achieves statistical significance when entered into the multiple regression summarized in Table 4-4. The fact that over half of the influence of average family income is mediated by school district size (d.c. = .23) may indicate that the relationship is purely fortuitous: both high average family income and a large number of faculty per building may be a function of the population density characteristic of large metropolitan school districts. In summary, the evidence with respect to the functional and spatial division of labor in Michigan K-l2 school districts demon- strates substantial agreement with the evidence of other studies of formal organizations: the primary determinant of organizational division of labor is the size of the organization. The faculty in larger school districts are assigned to more (and more specialized) teaching assignment areas than the faculty members in smaller dis- tricts. This condition is somewhat contingent upon the advanced academic credentials of the faculty, but the faculty in larger districts tend to have higher academic credentials in any case. The larger the school district, the greater the probability that the faculty are evenly distributed across a broad range of specialized 124 teaching areas. This condition is contingent to some extent upon the financial resources of the district and the number of non-public school students in the district. While the former contingency makes sound intuitive sense, the latter does not, and the available evidence is insufficient to permit more than tentative speculations. Faculty in larger school districts tend to be more evenly distributed throughout a larger number of buildings and work locations than the faculty in smaller districts. Such distributions tend to be somewhat less expensive than the more uneven distributions that characterize smaller school districts. Finally, larger school districts tend to have more faculty per building than smaller districts. This undoubtedly results from the location of these districts in more densely populated communities. The finding that the average income of community residents also affects this condition probably derives from the location of more affluent people in more densely populated communities. Hierarchy of Authority Formal organizations are characterized by a hierarchy of circumscribed authority which mediates the distribution of organiza- tional power and status in such a way as to assure adequate super- vision and direction of organizational activities. As indicated in an earlier section, the authority structure of formal organizations is exceedingly complex, and for the purposes of empirical analysis may (and probably must) be differentiated into two dimensions: (1) the formal structure of authority relations (i.e., that relatively 125 stable pattern of super- and subordination revealed in the typical organization chart); and (2) the informal structure of authority relations (i.e., those unofficial and frequently shifting patterns of influence and favor which derive from thepersonal , social and political characteristics of organizational members, but which often determine both what and how "things get done" in organizations). Since the latter, however ubiquitous, is contingent upon the internal dynamics of particular organizations, it is not entirely amenable to the comparative analysis of a large number of organizations. Thus, the present investigation only analyzes three aspects of the formal structure of authority relations in Michigan K-12 school districts: the extent of administrative differentiation, the number of major divisions and the number of hierarchical levels. Three closely related variables--administrative ratio, supervisory ratio and administrative staff ratio-~are analyzed in the discussion of school district administrative apparatus. Three other variables that bear upon the distribution of authority in school district organizations-- superintendent span of control, supervisory span of control and principal span of control--are included in the analysis of both hierarchy of authority and administrative apparatus, but only as independent variables. Administrative differentiation refers to the functional division of managerial and administrative labor in school district organizations, and thus provides a context for the consideration of both hierarchy of authority and administrative apparatus. It is measured by the proportion of twenty-five administrative assignment 126 categories occupied as first or second assignments of district administrators during the 1975-76 school year. The administrative differentiation scores of Michigan K-12 school districts range from .04 to 1.00, with a mean of .29, indicating that the administrators in the typical Michigan school district are spread across approxi- mately seven separate administrative assignment categories. Major divisions refers to the horizontal differentiation of the school district hierarchy of authority, and is measured by the number of district subunits headed by an administrator who reports directly to the superintendent and who supervises two or more administrators other than building principals. This measure differs from the measure for superintendent span of control in that the latter includes all administrators reporting directly to the superintendent regardless of their supervisory responsibilities. The typical Michigan K-12 school district has 1.34 major divisions, ranging from a low of one to a high of eight. Hierarchical levels refers to the vertical differentiation of the school district authority structure and is measured by the number of supervisory strata between the superintendent and the faculty, with the superintendent and faculty counted as extreme strata. Michigan K-12 school districts have from two to seven hierarchical levels, with a mean of 3.32. Superintendent span of control refers to the total number of non-clerical personnel reporting directly to the district superin- tendent. This measure differs from major divisions in that it includes isolated specialists and consultants who report directly to 127 the superintendent but do not head major subunits. The spans of con- trol of Michigan K-12 superintendents range from one to eighty-six with a mean of 9.26. Supervisory span of control refers to the average number of non-clerical personnel reporting directly to administrators between (and excluding) the district superintendent and building principals. In the typical Michigan K-12 school district the average supervisory span of control is 5.84, ranging from a low of zero to a high of eighty. Principal span of control refers to the average number of district professional personnel (faculty, assistant principals and other administrators) reporting directly to a principal or building supervisor. This measure differs from the measure for faculty per building in that it includes assistant principals and other admin- istrators, and because some principals and building supervisors are responsible for more than one building. "Principal teachers" and "teachers-in-charge" are not counted as principals unless they are designated as administrators in the Register of Professional Personnel. The mean principal span of control for Michigan K-12 school districts is 24.87, ranging from a low of 4.25 to a high of fifty-nine. Administrative Differentiation As indicated above, administrative differentiation measures the proportion of twenty-five administrative assignment categories occupied by district administrators during the 1975-76 school year. 128 A high administrative differentiation score indicates that district administrators, regardless of their absolute number, have designated responsibilities for a relatively large proportion of twenty-five specialized administrative functions in the district hierarchy of authority. Administrators in the typical Michigan school district occupy 7.25 separate administrative assignment categories. What are the environmental conditions of the administrative division of labor in Michigan K-12 school districts? The regression equation summarized in Table 4-5 indicates that the extent of admin- istrative differentiation in Michigan K-12 school districts is almost entirely a function of the size of the school district. No portion of the regression coefficient in the second row of Table 4-5 is spurious and its influence upon the dependent variable is virtually unmediated. The larger the school district, the greater the spread of district administrators across specialized assignment areas. On the average, a one unit (ten teacher) increase in the size of a school district is associated with a 42% increase in the extent of administrative differentiation. Perhaps the distribution of administrative labor is also influenced by the distribution of faculty labor in Michigan K-12 school districts. Administrative differentiation is highly correlated with both faculty differentiation (r = .84) and faculty dispersion (r = .68). However, the high correlations between these aspects of school district division of labor (r = .73) and between each and district size (log) (r = .91 and .81, respectively) precludes their simultaneous inclusion in the same regression equation. When faculty 129 op.n mo. Po. zgmpmm Loumgampcw28< mmmgm>< .m No.- mm. co. Amosv «Nam ao_tmeo .N No.- as. .mmq _aa=a Lea copuas_a> um~w_a=am spasm .F m N P m_aaaaa> .Loggm ngmuzmum my? mmswu boggy cusp ago: we mom u : “Rpm. u my Fm. u m N mm. upooooo. pmooooo.1 «rop.u zgmpmm Lopmgamwcwsu< woosm>< .m mm. Npo. Ne. aaom. AmOAV mem puwgpmwo .N mo. mmoooooo. . mwoooooo. same. Fwasm Lma cowamspm> chPpmzam mumgm .P .xgapmm Loumgammcweu< mmmgm>< new Amodv onm pomsumwo .pwaza can cowuma—m> umNPPmacm macaw co :owumvucmgmme_o o>wpmgumwcwsv< co cowuwmanouma new cowmmmcmmm wpawupzz--.m-¢ m4m< .m mo.- mm. w_.- awn. mp.- _aa=a sea aneaaxm aeaeaeeac .N oo. oo.- oo. P~.- oeeam sepaees-peae=em ._ m a m N mpnewce> .soggm vceucepm m»? was?» mmccu cusp use: rs mom 6 e “Rem. n WV am. n mm mm.. .oo.- armm.- Paeeeau ea eeam Peaeee_ea .m mm. mmo. sake. aaaeaeeeesaeeeg asaeaeemaeeee< .a ~_.- mmoooooo. mmooooo.- s.oa.- asepem aaeeeamaePEe< amaea>< .m cm. Peooooo. mmoooo. seem. _aa=a sea eneaaxm meaeeseao .N _c. was. rsNN. cream sapseea-e=aeaum ._ a asem eFaeaee> .Focpcou mo seam Fenwucwsa ucm cow“ -mppcmemwmpo m>wumspmpcwsu< .xsepem sanctumvcwee< muesm>< .Ppnza can mmcmaxm mcwpesmac .owumm zupauemuucmcapm :o opuem m>¢uespmwcpsu< po cowuwmoaeoumo use cowmmmemmm mpawupzz--.mia m4m

ea seesaesases: .e mo.- ms. .mmn ea.- as. so.- as.. eaasesssa sane: .m __.- ac. N_. .mmqu as. oo.- s~.- asasem saeaseesessea eaasesa .a 40.- ea. as. _F.- .qu es.- es.- Psaaa sea enaeaxe assesseao .m co. co. so.- co. eN.- .Hmn Na.- asaea assaeas-saeeaem .N Na.. es. as. a..- me. .o. .mth amass ease sessemsa .s a e m a m N s e_aasse> .sosse useuceum m»? was?“ mess» cog» use: «a see u a mace. n my mm. 6 me am.. Paco. eooo.- P~.- _aseaae ea seam seaseassa .a as. Nae. soc. same. e_e>ea seesaesasesz .e as. sea. sea. area. measesssa some: .m 2.- 888a. 888.- .58.- :28 ssasemsaseea eaesesa .e as. eooaoo. mecca. same. ssaaa sea eeaeaxe massaseao .m .a. ease. moo. ease. aseaa »s_aeas eaeeaem .N mo.- ace. Na.- ream.- amass ensm sesseesa ._ s asem e aeea esaesse> .pospcou mo ceam Feawucwsa can mpe>e4 Peuwgusesmw: .mcowmw>wn some: .ase_em souesum_=wse< mmmsm>< .Pwaza sea emceaxm mcwaeseao .owaea aupzueanuceuapm .Amosv e~wm pu_sumwo :o ownea m>mpesme:PEu< we cowuwmoasouoo use :o_mmmsmem mPQPHP=Zuu.m1¢ m4m

.Losgm usmucuum m“; mas—.5 $923 can» 95: is. .Lossm cgmvcmum my? mots» can» «so: as mom u a Has. 1 as me. u Na oo.- mosses. soo.- ssam.- sososoo so seam seasoassa .e N_.- mmoooo. assoc. sacs. sososoo so seam ssoossseaom .N _N.- Nsoo. moo. ssoN. aseseo seessososess .o N_.- woos. Nmoo. samN. eaosossss some: .m _o.- omoooo. msooo.- «os.- oaosooossssooo sosooes .o m..- ooooooo. «Noose. .sNN. ssaaa sea eoaeaxe assesseao .m No.- Nmooo. soc. sst. oseos aosooos-oaeooom .N oo.- eNoo. emo.- rsoe.- sooav eNsm eossoosa ._ s assm N sees e_oosso> .Posusou so seam soososssa use Fospsou so seam asoms>sossm .mpo>os poossososos: .msosms>so some: .msosaoussspooo xupoooa .Psooa soo omsooxm mssuosooo .osuom appoooaiusououm .Aoosv oNsm possumso so osaom.>soms>seoom so sososmoosooos use sosmmesmem oposupzz--.osio mom<~ 155 school districts. The greater the span of control of the district principals, the lower the number of principals and thus the lower the overall supervisory ratio. Most of this influence is rendered spurious by the antecedent influence of the size of the school district (d.c. = -.46) upon both principal span of control and supervisory ratio, but its negative regression coefficient indicates a strong independent influence when district size (109) is controlled. Other conditions equal, the addition of ten teachers to the typical span of control of building principals is associated with a one percent decrease in the district supervisory ratio. Since each additional unit of vertical or horizontal dif- ferentiation adds at least one supervisor to the administrative staff, and since both dimensions are largely contingent upon the size of the school district, the findings in rows 5 and 6 of Table 4-10 cause no surprise. The antecedent influence of district size (109) upon both major divisions (d.c. = -.53) and hierarchical levels (d.c. = -.54) reduces their impact upon the dependent variable, but their significant regression coefficients indicate an independent influence when the number of faculty is controlled. Translating to the metric of their original measures, either an additional 1.9 major divisions or an additional 1.25 hierarchical levels would increase the supervisory ratio of the typical Michigan K-12 school district by one percent, ceteris paribus. The significant beta weight of operating expense per pupil (row 3) indicates that when district size (log) and its correlates are controlled, higher supervisory ratios require greater financial 156 resources. Under these conditions, a one percent increase in supervisory ratio is contingent upon an additional $434.78 in operating expenses per pupil. The student-faculty ratios of Michigan K-12 school districts have a marginal but significant impact upon the size of their supervisory ratios. Although most of this influence is mediated by operating expense per pupil (d.c. = -.13), a one percent increase in supervisory ratio is associated with an average increase of ten students per teacher. The meaning of this finding is not entirely clear. Although purely speculative, this may be a result of declining enrollments in Michigan schools. That is, under conditions of declin- ing enrollment, the administration must adjust the size of the teach- ing staff. Since further declines may be anticipated, superintendents may decide to reduce the number of classrooms and increase the number of students in each. This would have the effect of increasing the student-faculty ratio, reducing the principal span of control and increasing the district supervisory ratio--exact1y the pattern observed in Table 4-10. At first glance the impact of supervisory span of control upon supervisory ratio (row 7) appears to contradict the more reasonable expectation suggested by their simple correlation (r = -.12). Most of the influence of supervisory span of control upon the dependent variable is rendered spurious by the antecedent influence of district size (log) (d.c. = -.22) upon both conditions. However, the regression coefficient of supervisory span of control indicates a small but independent influence upon supervisory ratio 157 when other conditions are controlled. The key to this situation is the decomposition coefficient for principal span of control (d.c. = -.O7). A low principal span of control tends to increase the relative number of principals and thus the span of control of their immediate supervisors. The magnitude of the regression coefficient in row seven makes the issue largely academic, however. Other conditions equal, a one percent increase in the average super- visory ratio is contingent upon an average increase of 52.63 subordinates per supervisor (mean = 5.84). The regression coefficient in the fourth row of Table 4-10 suggests an interesting line of specualtion. It appears that school districts in which a high proportion of faculty members have advanced degrees require fewer supervisors than other school districts, other conditions equal. Although most of the influence of faculty qualifications upon supervisory ratio is rendered spurious by the antecedent effects of school district size (d.c. = -.50) upon both variables and part of its impact is mediated by principal span of control (d.c. = -.13), its small but statistically significant regression coefficient indicates an independent influence when other conditions are controlled. However, under these conditions a 1.5% decrease in the district supervisory ratio would require a 100% increase in the number of faculty members with at least one degree 31.34%). beyond the baccalaureate level (mean In summary, the supervisory ratio of Michigan K-12 school districts is largely a function of the size of the school district. 'The larger the district, the lower the average supervisory ratio. 158 Secondly, high principal span of control tends to depress district supervisory ratios. Thirdly, as with the overall administrative ratio, it is not so much the size of a school district as the com- plexity of its administrative hierarchy--i.e., the extent of vertical differentiation--which increases the supervisory ratio. Under these conditions, school district supervisory ratios also tend to have some independent influence upon the magnitude of school district supervisory ratios, but their respective influences are probably too small to make any practical difference. Administrative Staff Ratio Administrative staff ratio refers to the relative magnitude of that component of the administrative apparatus of Michigan K-12 school districts which is comprised of administratOrs who exercise few if any supervisory responsibilities and whose primary contribu- tions to the district derive from their specialized knowledge and technical expertise. It is measured by the ratio of the total number of district administrators (excluding clerical and supportive personnel and teachers in quasi-administrative posts) who do not supervise two or more non-clerical personnel to the total number of district faculty. The typical Michigan K-12 school district has one administrative staff person for every fifty teachers. The major prerequisite for a large staff of consultants and specialists in the administrative apparatus of Michigan K-12 school districts is a highly differentiated central administration, as is indicated by the large and significant regression coefficient in the 159 fourth row of Table 4-11. The greater the extent of administrative differentiation, the higher the administrative staff ratio. None of this influence is rendered spurious by the antecedent influence of other variables in the equation, but part of its impact is mediated by a relatively small number of hierarchical levels (d.c. = - .10). In light of the high simple correlation between administra- tive differentiation and the logarithmic transformation of district size (r = .90) and faculty differentiation (r = .84), these vari- ables were examined in separate analyses otherwise containing the same independent variables. Their regression coefficients were .48 and .43 (each greater than three times its standard error), respectively, and the variables in the respective regression equa- tions accounted for forty-five and forty-three percent of the variance of the dependent variable. These findings suggest that although the degree of administrative differentiation is the best predictor of the magnitude of the administrative staff ratios of Michigan K-12 school districts, this condition (like the extent of administrative dif- ferentiation) is also contingent upon the size of the school district and the extent to which its faculty members are differentiated into specialized functions. In addition, when district size (109) is sub- stituted for administrative differentiation in the regression equation summarized in Table 4-11, and when major divisions is entered to approximate the full extent of administrative differentiation, the regression coefficient of the latter is .18 (three times its standard error), despite the antecedent influence of district size (log) upon both major divisions (d.c. = .32) and administrative staff 160 Ns. so.. am. No. oo.- oo.- soseaoo so seam seasoassa .e as. .wflqu am. as. mo.- so. aseses seesaososesz .e ea. os.- .mmq me. so. so.. aosoosoaesesssa essoosoosessos .o No. oo.- oN. .flfln eo.- No.- aoseaasseess sesooos .e es. eo.- es. mo. .mequ so.- ssosom soooseosessos eoese>< .N oo.- oo.- .o.- No. es. .iflq oseos sesaoos-oseeoom .s e m o m N _ esoesso> .sosso useuseum mos moss» moss» sesu esoz «a eom u a ”see. 6 ms em. n Ns so. seoooo. eNooo. sst. .oseaoo so seam seasoessa .e mo. ssoo. eeoo.- seas.- osesee seesaososesz .e as. moss. oso. saNe. eosooseeesessss esseeseosasses .o eN. oso. see. sass. aosoaassoesa eosooos .e em. NNoooooo. esoooooo.- aaNs.- asosem soeesoosssses eoosess .N No. NNooo. eases. sass. osoos aesooos-oseooom .s s msem e ooee esoosso> .pososou so seam seasosssa use mpo>oo Feossosesosz .sosp resusosossss o>suesomssseu< .soseoosspmss aapooes .xsesem sooesumssssu< emeso>< .osuea aossoesiosouopm so asses sseom o>suesemssssu< so sospsmoosoooo use sosmmosoos opaspssz--.~sus usmos sesouos .osues hopsoesiusousam .eos< so osoem ssepm o>sesoooem so sospsmooeoooo use sosmmosmom oposppozuu.m_us msmmm=o4 ucmucmucwgmaam copupsuu< zupaumm upmepmeweesmm apgmeepoeum page: Pacemuez coppmpaovgpmz covenant“ gmgmv: mama uzoaoga o o o o c—NMQ'LDCDN Fo.- _o.- oo.- Po mo - m¢.- .hpw copppmoae°u Peeuam sumczssou . Fe. cc. me. me. No. Po.- 5,. «souefi a__saa . m o m e m N _ mpnmeea> .socem vgmvcmum my? mos?» mossy coca one: «« Cot—gm vgmtcsm mt. mots“ :9.“ 932 a. mom u e ”Ame. n «V cm. u mm m_. mmo. PF. sac. abysmmeos bemucmuewemaam .N mp.- mmo. o_.- «top.- eowpwepp< awesome .m mm. mm. mm. «ace. mpmw.aewcwsmm awsmcepogum page: Paeoweaz .m NN. ope. moo. e*m_. eo.6m_=o.epez gawpeuaum Logos: .e we.. «mo. em.- ase~.- mama “sonata .m cm.. mmo. mm.- «eme.- cowavmoasou Parana saveaseoo .N Pm. «mecca. emcee. «INF. 6506:“ AP.Eea .F g m\um m mama mpampgm> .auw>mmcog ucmucmucwemasm can cowuwguu< znpaumd .mumWPmcmmpsmm nwgmgmpogum awed: Pacowumz .corumpauwgumz compmuaum cosmm: .wumm uaoqoga .cowupmoasou pmwumm zupcas -Eoo .msou=H APPEmu co acmsm>mwgu< acmuzum we cowawmoQEoumn can cowmmwgmmz mpgwupazuu.pum m4mmcgu< bemu=»m_ e «9.- “mm. mo.- ao.- Amoov mNcm “accents .m mo.- op. .Hfiqu eo.- acgmemaEmz Foogum ur_n=a-=oz .N m_.- NN. mo.- .mflqu asoueH spcsea .P e m N P mpaawea> .coccm ccmucmum.mpw was?“ omsgp cog» «no: «« .Losgm vngcmum mu: mutfi. =93 msoz .4. mom a e "Aom. u my Fm. u mm 04.- “Po. NF.- as_e.- cememsmcgu< “emuapm .4 mN. mm. mN.N *«mm. Amosv mNFm Soceumco .m PF.- __o. mo.- «PF.- acgmemasmz coagum ocpnaa-=oz .N mo.- «cocoa. m_ooo.- *sm_.- aeouec spread .P e m\mm m epmm m_aa.ee> umpnsaucoz .pcmsm>mw;u< acmuzum can Amodv wumm auvcummo .amgmcmasoz Foocum .mEoocH xpmsmm :o mama uzoqoco mo cowuwmoqeoomo use cowmmmcmmx «pawupszu-.~-m m4m<~ 188 effect of family income (d.c. = .09), its large beta weight indicates an independent influence when the effect of family income is con- trolled. Thus, the larger the school district, the higher the per- centage of students who leave school prior to graduation. More specifically, a one unit (ten teachers) increase in the size of the school district is associated with a 2.28% increase in the district dropout rate, ceteris paribus. But why should larger school districts have a higher dropout rate than smaller school districts? Could it be that some students, particularly those from low income families or who have low achievement scores, simply get lost in the depersonalized machinery of large school districts? As tempting as this suggestion may be, none of the variables typically associated with depersonalized bigness or overcrowding (e.g.,student-faculty ratio, faculty per building, faculty dispersion, principal span of control) attain statistical significance when entered into the regression of high school dropout rate. 0n the other hand, since large school districts tend to be located in or near large commercial or industrial centers, it may not be the "push" of alienating school conditions but the "pull" or more attractive nonschool opportunities in the surrounding environment which is responsible for this phen- omenon. In addition, average family income (row l) has a small but significant effect upon the high school dropout rates of Michigan K-l2 school districts. The higher the average family income, the lower the dropout rate, other conditions equal. Although none of this influence can be attributed to antecedent conditions, some 0f 189 it is mediated through the contradictory effects of school district size (d.c. = .22) and student achievement (d.c. = -.l3). When these influences are controlled, however, a $7692.3l increase in the average family income of Michigan K-l2 school districts is associated with a one percent decrease in the high school dropout rate. It is interesting to note in this respect that, when other conditions are controlled, community and student body racial composi- tion are not significantly related to the high school dropout rates of Michigan K-lZ school districts. When the measures for community racial composition and student body racial composition are entered into the regression equation sunmarized in Table 5-2, their beta weights fail to attain statistical significance (.06 and -.04, respectively) and contribute nothing to the coefficient of determina- tion (R2). Further, the decomposition of the simple correlations between these variables and the dependent variable (.37 and -.l8, respectively) indicates that most of their influence is mediated through school district size and student achievement. In short, if race is a factor in determining high school dropout rates, it appears to be so only because minority students tend to be con- centrated in larger school districts and to have somewhat lower achievement scores. Finally, the regression coefficient in the second row of Table 5-2 presents an enigma. Why should the rate of non-public school membership have any influence upon the dropout rate of the public schools? Although this relationship is very small (it would take a 32.26% increase in non-public school membership to effect a 190 one percent increase in the public school dropout rate), and most of its influence on the dependent variable is mediated by the effects of school district size (d.c. = .l0), its persistence under controls invites specualation. The suggestion that availability of the non- public school alternative encourages transfers from the public schools is precluded by the definition of the variable. By the same token, the suggestion that potential high school dropouts are attracted to or channeled into the non-public schools because of their need for stricter discipline is probably too flattering both to anyone's ability to identify potential dropouts and to the allegedly superior discipline of the non-public schools. Perhaps the solution is less direct but more consistent with the organizational unit of analysis. Although the students in most public school districts have access to one or more non-public schools, most of the private and parochial schools in Michigan are located in the larger commercial- industrial centers of the state, particularly in the populous counties of southeastern Michigan. Further, a visual scan of the list of non- public schools in these counties indicates that although most are parochial schools, many are private preparatory schools, widely known for their academic standards, that attract students from many other regions. As suggested earlier, part of the influence of school district size upon dropout rate might derive from the "pull" of the non-school opportunities presented in the surrounding commercial- ‘industrial environment. Could it be that a similar--though contradictory--dynamic operates with respect to the non-public school? Could it be that the mere presence of these schools, and the attitudes 191 and values they represent, have an exemplary influence upon the educational norms operating in their host communities, norms which tend to increase interest in scholarship and school achievement and thus reduce the high school dropout rate? This suggestion would be much more convincing if the data indicated that non-public school membership has a similar influence upon the student achievement scores of the public school districts, but this influence, if present, is not revealed under the conditions included in the analy- sis. Consequently, the suggestion raised here must remain in the realm of speculation pending further investigation of the impact of non-public schools upon various dimensions of public school performance. In summary, the dropout rates of Michigan K-l2 school districts are influenced primarily by district student achievement scores. Low achievement from an early age-~and the kinds of reinforcement associated with low performance--probably create a climate which increases the likelihood of dropping out of school. Secondly, perhaps the location of large school districts in the major commercial-industrial centers presents students--particularly those with low achievement--with attractive non-school alternatives, thus increasing district dropout rates. Thirdly, average family income, associated with both district size and student achievement and unediated through both, has an expected negative influence upon dropout rate. Finally, the very small but significant influence of non-public school membership upon dropout rate may reveal an exemplary 192 influence deriving from the mere presence of non-public schools in the public school jurisdiction. Higher Education Matriculation What are the structural and environmental conditions that influence the rates of higher education matriculation in Michigan K-l2 school districts? The variables listed in Table 5-3 include one structural characteristic, eight contextual characteristics and, with one exception, no surprises. The number of district national merit scholarship semi- finalists (row 8) exerts a significant influence upon the extent to which students continue their education in institutions of higher learning. In fact, other conditions equal, a one percent change in the number of national merit scholarship semifinalists is associated with an average change of 4.42% in the number of students enrolled in two- or four-year colleges and universities. Although the direction of influence between these two variables is open to question, the interpretation suggested by the present arrangement is that just as the presence of one or more semifinalists indicates high achieve- ment and academic excellence on the part of those students as individuals, so their presence in the organization contributes to an atmosphere which encourages a continuation of educational pursuits. An alternative interpretation--i.e., that a longstanding tradition of college attendance increases interest in scholarship competition--may be equally true. 193 mo.- .bbw- oo.- No. No.- No.- Fo.- Fo.- _o.- coquepa< sppauac .m co. Fm. Po. mo. Po. No. .o. No.- no. mpmwpaccccsmm amgmea_ogum “cam: Panacea: .m co.- no. mo.- .- No. Po. oo.- so. No.- mung “sonata .e Po. mo. so. _P. .ppw me. _o. eo.- no. SemEm>mpgu< bemuspm .8 Po. mo. Fo.- co. N_. .mpw Fo.- No. mo. ocpam ccapm ascaeepmcecee< .m .o. no. Po.- No. mo. op. mp. co. Pp. mcocpauvccpezc »6_=uad .e _o.- Po. oo.- Po. Po. _o.- m..- Po.- No.- ocpaa sepsuaa-u=mc=pm .m oo.- «0. co. me. _o.- oo.- oo.- o~.- .wpw cocppmoasou pacuam seam bemuzpm .N co. co. Po. mo. mo. mo. _o. Po.i ow. esouec chsmd .P m m A o m e m N _ mpaacee> .LOLLm Ugmtcmum mu: mm—Eb 09:3 can» 952 «:1 .Logsm Ugmucmum mu: $2.5 :05» «so: a, mom u e “Rum. u my mm. a me up.. cc. om.- «mo.- coc»_cpp< supzuac .m mm. mm. N¢.¢ «._~. mumcpaepcwsmm e.gmeaPogum “Pew: _eeocpaz .m mp.- mm. m¢.- «mo.- mama “sonata .N NN. op. em. app. Semem>mpgu< acmcapm .0 mm. mo.¢m a~.ma .~_. ocuem cceum m>wuaepmcemsu< .m cm. mac. Nae. top. m=o_ceuccc-=o appzuac .q _N.- NN. mm.- ««m_.- ovum“ su_=uac-pemu=pm .m a..- Spa. eeo.- ««o~.- cecpcmoasou Pmcuam seam gemuapm .N Fe. mmooo. mmooo. seem. «soocc spPEac .P e m\mm m comm apnaPLa> .cocpceu64 seesaw; new mpmcpaccccemm acgm -cmpogum awed: Pecowumz .mpmm paoaoeo .ucwsm>mw;o< acmuaum .owumm ccmum m>wumcumwcrsu< .ucowpoopmmezc xp_:umu .ompmm xupaommnucmuzum .cowupmoasou meomm xuom acmuaum .meoocH zpwsmd co comme=chpmz copumoaom gmgmw: co :ovuvmoaeoumo use copmmmcmmm mFaPuPazuu.mim m4m<fi 194 That school districts with more affluent families send more students on to institutions of higher education is indicated by the regression coefficient in the first row of Table 5-3. No portion of this relationship is spurious and its influence is not mediated by other variables in the equation. Its interpretation is equally straightforward. In addition to representing an expense which the more affluent are better able to afford, college attendance also represents a value commitment which is highly correlated with socio- economic status. The statistics in the second row of Table 5-3 indicate that the next most potent influence upon college attendance rates in Michigan K-lZ school districts is the racial composition of the stu- dent body. The higher the percentage of caucasian students in the district, the lower the higher education matriculation rate. More specifically, to change this rate by one percent would require a l3.5l% increase in the number of minority students in the district, other conditions equal. Although none of the community type variables attained statistical significance when entered into the regression equation presented in Table 5-3, it seems reasonable to suggest that school districts with larger concentrations of minority students send more graduates on to higher education because they are located in more densely populated areas which have greater access and exposure to both two- and four-year colleges and universities. Student faculty ratio (row 3), administrative staff ratio (row 5) and faculty qualifications (row 4) each make an independent, positive contribution to the rate of higher education matriculation 195 in Michigan K—l2 school districts. The influence of student- faculty ratio is direct and unmediated. A reduction of l.l4 students per teacher is associated with a one percent increase in college attendance, other conditions equal. The influence of administrative staff ratio is partially spurious because of the antecedent influence of average family income and faculty qualifica- tions (d.c. = .08 and .05, respectively), but the statistically significant regression coefficient which remains after controls indicates an independent influence. In terms of the original metric of both variables, one additional administrative staff specialist is associated with a one percent increase in the number of high school graduates who continue their education in two- or four-year colleges or universities. Faculty qualifications-~i.e., the percentage of the district faculty which have masters, specialist or doctoral degrees--makes a significant contribution to higher education matriculation rate despite the antecedent influence of average family income on both variables (d.c. = .ll). In terms of the metric of their original measures, however, a one percent increase in the rate of higher education matriculation would require a l0.31% increase in faculty qualificatidns. Quite apart from the substantive contributions of these three variables to the dependent variable, their presence in the regression equation itself is significant because they represent the only so-called "school effects" which have any statistically significant influence upon any of the performance criteria examined in this study. 196 Student achievement and high school dropout rate each have a small influence upon the higher education matriculation rates of Michigan K-12 school districts. The beta weights of these variables (row 7 and row 9, respectively) are substantially lower than their simple correlations with the dependent variable because of the influence of antecedent and mediating conditions, but their signife icant beta weights indicate an independent influence when these conditions are controlled. Their influence is such that, other conditions equal, the attainment of either an additioanl 4.17 reading and math objectives or a 2.22% decrease in the high school dropout rate would produce a one percent increase in the school district higher education matriculation rate. Finally, faculty attrition makes a very small but significant contribution to the percentage of school district graduates who enroll in some form of advanced education. Almost half of the simple correlation of this variable with the dependent variable is rendered spurious by the presence of antecedent conditions in the equation, but when these conditions are controlled, a 3.33% decrease in the rate of faculty attrition is associated with a one percent increase in the college matriculation rate. In summary, the higher education matriculation rate of Michigan K-12 school districts is a function of community, school district and student environmental influences. School districts with somewhat higher average family incomes and somewhat larger proportions of minority students send more graduates on to higher education. This is especially true if those school districts also have lower 197 student-faculty ratios and higher adminsitrative staff ratios and if their faculty members have relatively more advanced degrees. Finally, much of a school district's higher education matriculation rate depends upon the student environment. More students tend to enroll in colleges and universities when the presence of one or more national merit scholarship semifinalists is present in the system, when district student achievement is high and the high school drop- out rate is low and when the rate of faculty turnover is relatively stable. None of the structural characteristics of Michigan K-12 school districts examined in this study has any influence upon the higher education matriculation rates of those districts. National Merit Scholarship Semifinalists How do the structural characteristics of Michigan K-lZ school districts influence the number of national merit scholarship semi- finalists in those districts? The multiple regression equation summarized in Table 5-4 indicates that none of the structural characteristics examined in this investigation have any impact upon the percentage of district graduates who attain this honor. More- over, the exceedingly small amount of variability of the dependent variable explained by those conditions which do have a statistically significant impact upon the criterion (R2 = .17) indicates that this perfbrmance characteristic is virtually unpredictable (at least in terms of the organizational characteristics examined in this investi- gation). This may be an artifact of the extremely small number of national merit scholarship semifinalists in the state and the even 198 TABLE 5-4.--Multip1e Regression and Decomposition of National Merit Scholarship Semifinalists on Average Family Income, Student Body Racial Composition, Student Achievement and Higher Education Matriculation. Variable Beta 8 SE/B r 1. Average Family Income .13* .000029 .00001 .28 2. Student Body Racial Composition .12* .0021 .00076 .11 3. Student Achievement .13* .014 .0047 .26 4. Higher Education Matriculation .25** .012 .0022 .32 R2 = .17 (fi = .16); n = 508 * More than twice its standard error. ** More than three times its standard error. Variable l 2 3 4 1. Average Family Income :13. .01 .04 .10 2° 3333§2§tififly RaCiaI ~01 .212 .03 -.04 3. Student Achievement .04 .02 .13 .07 4' 3328.883?” .05 ~02 .04 . .._2_s_ smaller number of school districts which have even a single national merit scholarship semifinalist. Or, more probably, this phenomenon may be explained by the fact that the characteristics which determine 'the performance represented by this achievement are so fundamentally ‘individual that they are unaffected by either structural or environ- niental influences. 199 The full impact of these considerations becomes evident when the statistically significant beta weights in Table 5-4 are trans- lated into their nonstandardized values. A one percent increase in the number of district national merit scholarship semifinalists would require a 83.33% increase in the percentage of district graduates matriculating in institutions of higher education, an additional 71.43 reading and math objectives in district student achievement scores, a $34,482.76 increase in the average family income of school district families and a 476.76% increase in the average number of district students classified as caucasion, other conditions equal. Thus, whereas these findings contribute to an understanding of the statistical predictability of the number of district students who become national merit scholarship semifinalists in a given dis- trict, they offer precious little guidance for the establishnent of policies which could significantly increase the number of such scholars. Faculty Attrition The performance characteristics of Michigan K-12 school districts considered in previous sections have been based upon various aspects of student behavior--the reading and math achievement of district fourth and seventh graders, the district high school dropout rate, the percentage of district graduates matriculating in institutions of higher education and the percentage of district graduates named as national merit scholarship semifinalists. The two remaining school district performance characteristics--faculty 200 attrition and superintendent longevity--measure the employment stability of the district faculty and superintendent. The measure for faculty attrition is the average number of district faculty removed from the Register of Professional Personnel for any reason during the 1974-75 and l975-76 school years. Two weaknesses of this measure--i.e., the limitation of a two-year purview and the inability to measure intra- or interdistrict mobility--were discussed in an earlier section (p. 88). A.third weakness is revealed in the coefficient of determination reported for Table 5-5 (R2 = .13). This statistic indicates that the vari- ables included in the regression equation account for only 13% of the variability associated with the dependent variable. Thus, although the statistical significance of the regression coefficients reported in Table 5-5 permit a high degree of confidence in the predictive value of the variables examined, the magnitude of the coefficient of determination indicates that the primary determinants of this measure of faculty attrition are simply not included in the present investigation. Thus, other things equal, the factor that has the largest influence upon faculty attrition is faculty experience--i.e., the average teaching experience (years) of the district faculty prior to the l975-76 school year. The longer the average experience of the district faculty, the lower the district attrition rate. Translating to the original metric of both variables, two additional years in the average teaching experience of the district faculty is associated with a one unit decrease in faculty attrition, ceteris paribus. 201 No. NF.- mP.- mo. NtaFam Napsaaa amaaa>< .a so. .mmqu ao.- No. aueaceaaxm accsuaa .N mo. so.. .quu ao. Amaav aNem pacacm_a .N Po. so.-. mo.- .mflq eacpcaaasau Fawaam Nucaassau .P a m N _ apaawsa> .Loccm uceucmum mew mos?“ omega can» ego: we mom u a ”ANN. u mv NP. a Na FN.- apooo. apooo. No. Neapam supsaaa amaaa>< .a om.- NNo. om. - aaom.- aaaaaaaaxm sapsaaa .m .N.- mm. NN._- aapN.- Amaav aNcm Nacspmpo .N No. FNo. ”me. asap. cacacaaasau Facaam Na_==esau .P a N\mm N aaam apaacaa> .zcmme mapsumu mmmcw>< one mucmwcoaxm appaumu mmwcm>< .AmoAV m~wm uomcpmwo .comu -wmonEou pmwuma xuwcsseoo co cowuwcap< xupaomm co comuwmoasouuo new covmmmcmmm mpawupazii.mim m4m .LOLLw usmucwum mu: mmE: wmszu :23. use: .3. .coccm ocmvcmpm mu? mowzp can» mcoz « mom u a mANN. u mv NN. a Na mp. No. FNO. aao. pcaeasacga< paaaaam .N am. Pmc. Na. aapm. asscae Neaaeaaepeaaam .N NN. moooo. Fpooo. aaN_. stapam pcaacaueasaasm ._ e N\Nm m apam apaa_sa> pcmucmucvcoasm co apw>mmco4 .u:mEm>m?;u« ucmuzam use mcacmh acmvcmvcvcmasm .zcmpmm acmucmucpcmazm mo copummoaeoomo use cormmwcmmm mpnpupazin.oim w4m

the relationships between degree of organizational complexity and levels or kinds of adminis- trative skills would determine the veracity of this suggestion and, as was suggested above, could have important implications for the recruitment, training, placement and evaluation of administrators in these and other organizations. Although highly serendipitous and not related to the central thrust of this investigation, two observations concerning school district performance deserve further study. First, it was found that the percentage of non—public school students in-a public school district jurisdiction has a small but statistically significant influence upon the functional division of labor and the dropout rates of the public schools. The interpretation of these findings was highly speculative and the relationships observed may be the fortuitous result of unknown antecedent conditions. However, the persistence of the relationships under controls may suggest a structural effect which merits further exploration. Secondly, it was found that community racial composition has a greater influence upon school district student achievement scores than student body racial composition. This difference may derive from the different dates of data collection (1970 versus 1975) or from differences in the operational definitions of the respective independent variables 228 (percent black versus percent caucasion). However, the magnitude of the differences suggests the importance of further exploration of these relationships. APPENDICES 229 APPENDIX A VARIABLE DEFINITIONS, BASIC STATISTICS AND SOURCES This appendix furnishes the operational definitions of the 52 variables analyzed in the tables. It provides the mean and standard deviation for each variable and the number of cases on which these data are based. The parenthesized, capitalized letters imnediately following the variable name are used to designate each variable in the correlation matrix on Appendix B. The parenthesized Roman numerals following each operational definition refer to the source(s) from which the data for each variable were obtained. A key identifying each source is at the end of this appendix. 230 231 mom com. ewe. mom 5mm. mmp. mom mmw. 0mm. mom mum. mmo. mom mop. one. 2 cowuww>wo cum: aaaaaapm MHV .Apocam ma umwwwmmmpu use zucsou vexez :P mmpumcaseou 6: page N muxm .mmcwcu cane: sous: m>onm umcwmmu cmuwewgoa on» mcwmuzo mow, by new .apwu mgou caupponocumz co emcee; cans: .czoe .xpmo m we muse; cam cu m_ mmmcuum uuwcpmvu on» up to .oom.m cusp mmwp mo cowumpsaoa a we; ..m.wv Pmczm we we: »a vmwmpmmmpu agpcaseou a cw m? Noycumwu map page acmumuvccp apnorem> xseaa A4Pm cwgumz mm xupcassou ecu Amv so “have ogou cauvpoaocumz a mo mmpwe cm» argue: we Xuvcaseou may Amy ”muses 9mm a m? aw mmmpcs zuwu ocou :muwpoaocumz a my auwcassou on» we mmmcuum mcvams mg» APV “mmcmuwcu mcvzoppow on» mo mco umamp an mpmoe ..m.wv emcee; cone: mm was an vmwmwmmupu Appeaseou m cm me uuwcumwc mg» was» mcvumuwucw mpnmvcm> mezzo Azp_u mgou copppoaocumz m we umpmvmmmpu we: we can egos so ooo.o— mo coruepsaoa a we; ..m.mv xuwo N me me: »n umwwpmmmpu mavczssou a a? ma Nacspawa as» pas» mecpaucucc apaacsas Nessa A>efiuv Nave AHV .Azuvu mcou cmawpogoguwz a mu uwvmpmmm—u apmsow>mcn mm; Amy co mmpoco an my Amv mfima cam: aaaaaaam A” .mna ac op mama oa coaamscommcmga cascaacmmoa AuNHmwoav op mNam AHH>V .aaas .aaaam aN-mNm_ aaa newest aaaeamaa as» an aasa_aea Naaaaac ca amass: _aaaa aaa ANNHNV aNam AHNV .a_;msaaeas we 0— mama oa coaamsgommcmca oasgaagmmoa Atmzwoav amoav aasmcmasmz AHHV .mNoa amN amasaaaamv has sagas meazaaaoa savage cacao; mga co Foogum mo mmoau mga am acacamau msa ca weapogcm s__amaa magmasaa Na-¥ ca amass: aaaoa sea Ammmzmzc aagmaaasaz Aa>v .aaa» .aagaa oN-mNma aaa mamasa coaauaumaezn auagamau mga casaaz weave acmzcm amaoa mg» AHV .ouimnma ca maoogum uaanznicoc ca ooaaocco xepmmmp zaacaseoo mga ca macmuaam Naix mo omaacmucma as» Asmazv magmcmnEmz Foogom oapnanicoz Aav .ouma mo mzmcmo .m.= msa ca ocmmz we cmwwammapo macwuammc xaacss iEou mo mmmacmucma mg» Amum as» Amzouzav meooca aaasmu “Hay .gasmcmnsmz an vmva>ac .mmimuma ca acacamau oga ca Acoammaesoo xm» mamam mga xn umNaamaom zaamcam mm mzam> cmmmmmmmv Aacwaoca amcomcma new mamamm Fame ac maam> manmxwa on» A>mmv :oaawapm> umNaamauw macaw .mp .Fp .op 233 mom mm.~wp n¢.mmp~ mom pw.mm m¢.mm mom NN.-~ m~.pm¢ mom mo.mpm mm.o¢~ mom mo.m m~.mm mom ov. po.p mom mp.m mo.FN z coaama>mo cmmz ataaaaam AHHV .magmcmasmz an umua>av .mmua>cmm acmuaam ucm mmua>cmm xaacassou .amaaao amaaamu acavspuxm aza acmam we mean -cmacams vcm coaamcmmo vcm mmmcmgm cmxaa .coaamacoamcmca .sapmm; .mocmucmaam .coaamcamacasum .coaauacamca meauaachaUacamau Foogmm mga we mmmcmnxm mammn msa Aaxmov Fan=m\mm:mnxm mcaamcmnc AHHV .aagmamasmz an aaaasaa .aaa» Possum mN-mNm_ mca Lea mamam mga an cmamooaammc mm co apaumcau cmsaam acmscgm>ow amcmnmu mca seem um>amumc macaw ap< Ammuv paa:m\m::m>mm pmcmumu AHHV .aagmcmasmz an umua>ac .mucam mamam mo azo mamam mna an umamagaosaam mcmz seas: Lama Foogum whimmmr mga acacsu mamam mga seem um>amumc muczw aa< Ammmv pan=m\m=cm>mm mamam AHHV .aagasaasmz zn umua>au .umuauocm acaosm mga ca amacamac mca ca manuaam>m ucm maoogum mga acaamcmmo Lea Lama Foogom whimmma mga acacam auacaman msa cazaaz vmuauocm m=:m>mm Ammav pam=m\m=:m>mm ammo; AHHV .cmm» aoosum enumnmp mca mcagsu mpoosom mo coaamcmao mca so» macm>mc muauoca oa auacamau mga ac coaamaam> cmnaamzam mamam aga ca umaaaaa aaaas ca maaa xaa asa Amm .HHV .maaas aaagam mN-mNma aca NN-_Naa mga amazaaa aagataaeaz aoacamau ca mummcumu\mmmmcu=a mamcoaagomocm mgh Azhzommv gazosw a__> .aac .aNam Na aaaasaa aagaaaasaz Aamo cam: eaaecaam Aae>v .cem» aeegem omimuma msa meagee mceaecamacasee amasamae we aceaem aesaeecacee mmmcm>m mca Am< AHH>V .emma aeecem ouimuma mga measee meemacmexm meea>mce e: saw: smegma zapeema maemgeepmeeee Lem» amcaa e we Aseamm ameaeecaeee mmegm>e ms» AmV .cem» aeecem enumuma mga meacee aaaeeem aeacamae mga me ageaem ameauecaeee mmmgm>e mew Am< AH~>V .cema aeesem enumnma mga measee mmcmme amaaeaemem Le mceaeeo .mcmamez e mewewe; maaeeem auagamae ae mmeaemecme mga Am<=ouV .aaas Possum eN-mNea age aa saaaa Naasame auaaaaaa mam ca meem mecmacm xm eageema m esm>e mna mxmoV .cema aeenem emimuma mga meacse seameeeee we emamammeae mceaecamacaeee aeacamae ae mmeaemecme mze AmuV .cem» aeegem chimnma mga meacee eeameeeee me ewwmammmau xapeeem aeacamae me mmeaemecme mzp Amumo cam: aeaaaaam AHHH> .HH>V .taa» _oa;aa eN-mNm_ aga eawese maaaaaaaacasea aewcamwe an emweeeee mmwcemmame acme:mwmme m>waecamwcw5ee Ammv m>ww -Naaaza we caaasaaoaa mew Aaaeezev .cem» weegem owimwaa mga acacee mcweawse aewgamwe seem ea em: amme aawzemw we emcee: mmecm>e mew Aoaamu .mcweawee aewcamwe seem ea ememwmme zaaeumw we cme22: mga n x Meme» maxmga we gem msav we emew>we x mca we 53m mga meewa mce ..m.w iigem» weegem owimwmw mga mcwcee mmewewweev meewaeeep ace: cacawz saweeew aewcamwe we cewaeewcamwe mgw AmmaouV .mmwgeamaee acmsemwmme waaeoew Aamv meenaasmwm we seem cw aaaeeew we amass: msa u x meme: Ax mga we see mcav we emew>we Nx mea we see mca meews mce ..m.wiimema aeesum owimwma mga mcwcee mmwgemmaee acmE=mwmme aaazeew emweauee cwsaw: xawzuew aewcamwe we cewaaewcamwe mew AwmaeuV .aaaa Possum eN-mNma mga mewsaa xaweeew aewcamwe mza he emweeeee mmwcemmame acmsemwmmm xaaeeew Apmv acciaasmwm we cowawoaecn msh Aummaue .aemeemacwcmeem emme we; acmeemaeacmezm Axum» weenem owimwma mga we mmv aemccee mga mgemz we amass: mew Azmweamv mcecmw acmecmacwcmesm AHH>V .saa» Possum eN-mNma map neweae aaaeaaaaasaaaa aeaeaaae mga we acmaem wesaeecaeee mcw Aamo :emz eaeeeeam Aaaa> .Ha>v .aaa» aaaeea eN-mNm_ msa meagee zaaeeew aewcamwe we cmesec weaea msa ea flameeemcme m>waceeeem ece aeuwcm—e meweewexmv mceaegamwewsee auwcamwm ca tease: peace as» we capes mew Aa .HH>V .cemw>eme=m mewe—wee Le weewucwce e ea mewaceemc xaaeeew aewsamwe we LmeE== mmecm>e mg» “umszev Fecaeeu we :eem weewecwee AHHH>V .cem» Feesem mwimwmw mca meweee maeeweewge mcwewwee mga use acmecmaewgmeam aewgamwe mga Amcwezpexm eeev cmmzame mgeaegamwewEee ea awaemcwe mcwageemc amecemcme weewemru ice: we amass: mmecm>e mew Aummamv aegaeeu we :eem agemw>cmq=m AHHH>V .aam» Possum eN-mNm_ age mewsze aaaeemaea igmeem aewcamwe mga ea xaaemcwe mewageemg wmeeemgme Peuwcmre ice: we cmesee weaea mew Aumwemv wegaeeo we :eem acmeemaewcmaam Aaaa>v .eaegam msmcaxm me emaeeeu za—eeew ece acmecmacwcmeem mza saw: .Lem» weecem owimwma mca acwcee zaaeeew aewcamwe mca ece acmeemaewgmeem aewgamme mca emmzame eaegam agemw>cmeem we Lmesee mew Amau>mav mam>mm weewsucecmwz AH~H>V A.:ewaw:wwme he .cewmw>wo gene: mce m>eg acmeemacwcmeem mca ea awaemgwe aceemg mcemw>gmeem mcww amcww cmgae ece maeeweewge meweawee gewgz cw maewcamwov .gem» weegum owumwma mca mcwcee .mceaecamwewsee mcee ce eza mmmw>gmeam es: eee acmecmacacmeem mza ea xaaemcwe maceemc es: Aaeeweewce mcwewwae e ceza cmgaev weaecamweweve :e an emeem; maweeesm aewwamwe we amass: mew A>Hov mcewmw>wo gene: .wv .oc .mm .mm 237 mom mm.m ow.¢ mom ow.m wm.mw mom ma. aw. mom co. mo. mom No. No. mom No. mo. 2 :ewaew>mn :em: eeeeeeem Axe .meeeecmaae mcwaemwwe acenew we mmmewaw .gaeme .maewwamwe weecem wmsae ea mwmwmcewa eeca emcae cememc Ace sew enumwma cw maaew ewgmwmesme weecem mga Eecw em>eemw nmrimv maemeeam xweeceumm aewwamwe we mmeaemewme mew Awaoeomov maeeeewo aeecem new: Axav .mwimwma ea ammw acmEmmmmm< :emwguwz mca ce memeewm sacm>mm ece gaweew auwwamwe we aaaemgceuicmwmzmce mm>waemnee saes eee meweemw we wmesec mmewm>e mew Amwee<.aemeeam wee> .NV .eeee aeeeee ew-mwew msa mewwee zaweeew aewwamwe we emesec weaea mga ea wweam mueecma ucwes eee wweam Feweeamee .wweam :ewaeaweemceca .wweam mew>cmm eeew .mmewe caaemg .mmewe weewwmae .mmewe zgeceww .mmmwe wmgeema we Leesee weeea eee we eweee eew waeeaeemv ewaee wweem esweeeeeem AHH> .HV .Lem» poo:0m chimwmp mga mcwcze aaweeew aewwamwe we smesec weaea mga ea mmewe weewwmre eeweemwe we weeEee weeee eee we eweee eea Aa .ee>v .eeee aeeeee ewimwew eea eeweee saweeew eewweewe we weasee waaee eea ea ameeemwme weeawmweiee: mwee we eza mmw>wme=m aee en es: wamccemwme m>waweeeem ece weewwmwe meweewexmv mweaecamwewsee auwwamwe we weesee aeaee eea we eweee eea waeemev ewaee wweam esweeeaewewse< wewee .Heev .weee weeeee mwimwma mca mcwwee xaaeeew aewwamwe we emessc mga ea ameeemwme weewcmweicec mwee we eza mmw>wmeem es: mweaegamwewsee aewwamwe we wmesee weaea mca we ewaew mew awgmesm .ov .mv 238 mom mm.e m~.m mom mm.m mo.m mom om. em. mom ww.mp om.mm z :ewaew>ma eemz eeaeeeem Aaaw>v .maemecmaewcmeem aewwamwe mewememwe eza eee Amwimwma we mev acmeweu mg» we acmeemacw -wmeem we mweema mmewm>e mew Aozomwemvxawzwmmem acmecmacwwmeem weex .eeexv .eeee» aeeeee ew-mwew eee ew-ewew eea eeweee cememw ace Lew amazemcme wecewmmmwewe we wmamwmmm mga Eeww em>eemw eeaeeew eeweemwe we cease: eeewese eea Aweewawmm aemEmmmmm< eee eewaeewe>m .cewemmmm meewaeuzem we acmaneemo cemwgewz .waeeeewwaw .ew-mwew .aewweewe weeeem eee weeeee we meeeeewe weeeem ewweee eeeweewz= .mwea .eemwcewz .mcwmcem .mmew>wmm acmEmmmmm< ece cewaeeae>m .eewemmmm meewaeeeem we acmsaweeme cemwsewz .Ameew ewameaezv =Eewmewe aemsmmmmm< cewaeueem ceawsewze .wwmw .m— sewez ece mwmw .ow wmeEmemo :mmzame aw—w u :v weaemwamm>ew msa we emwmamwcwsee we Awwe u cw emwmamwcwEeeiwwmm we: gown: mwweecewammae < =.xm>w:m cowaewemwweeu m>waewamwewse<= .mwm— .eewaemm mmew>wmm aemseeam>mo weeewmmmwewe eee :ewaeweemwe wmgeemw .cewaeeeeu we acmaneemn cemw;Uwz .Ameew ewamemezv ewmccemwme weeewmmmwewe we wmamwmmm owmwimwmws .mwmw .eemwgewz .aewmcem .cewaemm aceww weegem .mmew>wmm acme immecez weegem .cewaeeeem we aemsaweeme cemwgewz .Aaeeaewwev =eaea aewwamwo «aux mwmaimwmpe .Nwew .eeeweewz .eeweeee .eewaeeeew we weaseweeee eeeweewz .wNwew eweewweev =eaee weweeeeww eeaeewem we maewwamwe weeeem eewz ewweee eeeweewz we eewweema .mwmw .cemwgewz .mewmeea .cewaeezem we aemsaweemo cemwgewz .Aaeeaewwev =aweemm awessem ewgmwmeEmz ewwezeucez peeee<= .owmw .cemwcewz .aewmeem .wawceaweeeo :ewaeeeem weeem we mewwwo .cewaeueem we acmsaweemn eeuwgewz .Aaeeacwwev ememcmo ewegamuwewuem weegeme .owmw .cemwgewz .mewmcem .eewaeeeem we aemsaweema cemwgewz .Amaoa ewaewweev =eaee weweeeeww eeeeewem we eeewwaewe weeeem sew: ewweee eeeweewz we eewxeee= .mwmw .cemwcuw: .mcwmcea .wo:wm>ow mza we muwwwo .Ameew owamcmezv eeaeo aowwamwo Fw<= mmumaom .xH .HHH> .HH> .H> .>~ .HHH .HH 240 .ewma .cemwzewz .mcwmcem .cewaeewwwawmo wmgeemw we ceamp>ae .cewaeeeem we aemanee eeeweewz .waeeaewwev =ew-ewew weew weeeem wew eewewese aez eweeeeea mw-ewew we aweeemm .mwma .eemwgewz .mewmeem .cewaeewwwawmo wmzeemw we :ewmw>ao .eeaaeeeem we acmaneem eeeweewz .weeeaewwew =ew-ewew weew weeeem wew eeweweEm eez eweeeeaw ew-mwew we eweeeem . .ewmw .eewaeweeweu ewcmwe e e awwm Fecewaez e :ewawameeeu ewemwewecem awwmz wecewaez weeee< eeeemmiwaemzw mza cw mamwwnemwwsmmu .26.. .95“. .mmugwmm mucmumwmmdw meucmcE ucm 3“ :5 no: we aeeeeweeee eeeweewz .waeeeewwew .meaeeeewe weeeem eewz eeeweewz ewew enemee wWWEEemw .>Hx .HHHx .HHx .Hx APPENDIX B MATRIX OF SIMPLE CORRELATIONS This appendix furnishes the simple correlations between each of the variables examined in this investigation. The variable num- bers heading each row and column correspond to the variable numbers in Appendix A. The capital letters heading each row and column represent a shorthand version of the variable name.* These also are included in the variable definitions in Appendix A for ease of cross reference. 241 242 mme.eu o.e.e eew.ei aee.e ewa.e "Ne.ei me.e eNe.e- wee.e ~ve.e eee.e- eea.e eme.e- aaN.e- an.e mea.e «ne.e mme.ei escapee «m nee.e- eme.ei wNe.e mee.e eeN.e- eee.e- mew.e- Nee.e- eme.e- me..ei Nee.e Nee.e- one.ei oea.e .eve.e- nee.e- nee.ei mee.eu anger: an ema.e qu.e ene.e- aee.ea em~.e eve.e ee~.e eNe.e «me.ei Nee.e Nee.ei NeN.e eve.e ema.e- ove.e eve.e em~.e awe.e mg: on ee~.e nee.e maa.ei meN.e- aan.e maa.e neN.e wN~.e ame.e- Nee.e eee.e eev.e ee_.e man.ei wee.e ee~.e e_~.e vma.e sum ev eme.ei ooe.e Nee.ei eNe.e amN.e vNN.e NmN.e vaN.e gee.e- oe~.ei wen.e cme.ei «we.ei ee_.e- mne.ei nee.e- ea~.e mc~.e peeaeae me Nma.e ema.e vee.e wee.e- eve.e me.ei one.e vn~.eu nae.e ava.e mnm.e- «an.e mee.e mm~.ei nea.e ~e~.e eae.e ee~.ei «m we eve.e- ena.e- on.e eeN.e em..e- Hae.ei Nna.e- mee.e ~e~.e me~.e- Nna.e eNN.e- nee.e mmN.e mme.ei meN.e- eve.e- eee.e hexemem o. Naa.e ve~.e Nea.ei eee.e Ne~.e ee~.e neN.e eme.e mee.e- Nee.e eee.eu ae~.e ene.e veN.ei Noe.e oma.e eNe.ei ame.e eezemau no mNN.e eon.e nee.ei ewe.e mne.e emn.e ene.e Nen.e ne~.ei MMN.e emN.e men.e wee.e eev.ei ee~.e vnN.e mea.e nee.e a¢¢m< vc mee.ei mmN.e- ame.e qu.e- mee.e- eea.ei Nee.e- eee.ei aee.e we..ei wee.eu nee.ei eee.e oav.e eva.e- oc~.ei eN~.ei see.ei eeuwem n. ama.e Nee.e mee.ei vee.e eve.e- ae~.e woe.e- ema.e nee.ei mee.e eN~.e vve.ei mma.e e~e.e ame.ei ave.e- wee.ei «ma.e a¢¢xe< No eee.e emN.e eme.e. Nee.ei Nem.e veN.e ecm.e neN.e vm_.ei eee.e eva.e ace.e Nee.ei ven.ei pee.e emN.e nNN.e vea.e umzaee av eNe.e NNa.e mee.ei ane.e meN.e Nee.e meN.e wee.e vee.ei eme.e eNe.e ee~.e nae.ei ee~.e- mna.e eme.e mne.e wee.e umeam ev Nee.ei «ma.ei wee.e eme.e Nea.ei wee.ei ene.eu see.e -N.e mme.e- mNe.e- una.ei eve.e- wea.e oee.ei eee.ei oee.ei nee.e enema on Nea.e eae.e eo~.ei eee.e vee.e evm.e eee.e eev.e mea.ei mea.e nnN.e man.e voe.e ave.ei aee.e eeN.e man.e men.e maneua on eea.e eon.e mme.eu eNe.eu eee.e Nee.e eve.e 0mm.e e-.ei eea.e em~.e meN.e Nee.e own.ea ewe.eu aua.e ama.e oeo.e >~e we emN.e oev.e mea.eu wae.ei mee.e aNm.e vee.e eve.e eaN.e- evN.e eeN.e emv.e eoe.e mmm.ei NNe.e enn.e wee.e evo.e wwaexe< on maa.e wen.e nae.e nee.e- eem.e mNN.e Nem.e eoa.e maN.ei Noe.e va~.e mev.e nee.e- «an.ei e~e.e amu.e Nee.e oee.e eeaeucw an eaa.e aav.e Naa.ei mee.e eee.e men.e Nam.e eeN.e oeN.ei moa.e oea.e mev.e e~e.e eem.ei mna.e ove.e na~.e «NN.e mmaeozw en mnN.e men.e Nea.ei waN.ei emN.e eca.e aeN.e e-.e v-.eu mmN.e ave.ei veN.e eN~.e em~.ei vNe.e meu.e mNa.e mee.e emaeuzw en mnN.e Nam.e vo~.en Nee.ei mem.e nev.e nem.e vee.e enN.ei me~.e eeN.e eNm.e wee.e e~e.en eve.e ven.e ven.e Nee.e. wwaeo00 00 000.0 000.0 000.0 000.0 000.0 000.0: «00.0 000.0 000.0 000.0 000.0 000.0 000.0: 000.0: 000.0: «00.0 «00.0 000.0: 0000204 00 000.0 000.0 000.0 000.0 000.0: 000.0 000.0 «00.0 000.0 «00.0 000.0 000.0: 000.0: 000.0: 000.0 000.0: 000.0 0000000 00 000.0 000.0 000.0 000.0 000.0 000.0 000.0 000.0 000.0 000.0 000.0: 000.0: 000.0: 000.0 000.0: 000.0: 0000000 00 000.0 000.0 000.0: 000.0 000.0 000.0 000.0 000.0 000.0 000.0: 000.0 000.0 000.0 «00.0: 000.0: 9000000 00 000.0 000.0: 000.0 000.0 000.0 000.0 000.0 «00.0 000.0: 000.0: 000.0: 000.0 000.0 000.0: 0000000 «0 000.0 000.0 000.0 000.0 000.0: 000.0: 000.0: 000.0 000.0 000.0 000.0: 000.0: 000.0 200000 00 000.0 000.0 000.0 000.0 000.0 000.0 000.0: 000.0: 000.0: 000.0 000.0: 000.0: 010000 00 000.0 000.0 000.0 000.0 000.0 000.0: 000.0: 000.0 000.0 000.0: 000.0: 010100 00 000.0 000.0 «00.0 000.0 000.0: 000.0: 000.0: 000.0 000.0: 000.0: 00000. 00 000.0 000.0 000.0 «00.0: 000.0: «00.0: 000.0 000.0: «00.0: 000000 00 000.0 000.0 000.0: 000.0: 000.0: 000.0 000.0: 000.0: 0:000<0 00 000.0 000.0: 000.0: 000.0: 000.0 000.0 000.0: humus» 00 e00.0 0e0.0 000.0 000.0: 000.0: 000.0: 0000:00 00 000.0 000.0 000.0: 000.0: 000.0: nUldUlfi 00 000.0 000.0: 000.0: 000.0: nugam 00 000.0 000.0 000.0: 0300 00 000.0 000.0 000 00 000.0 «:0 00 and 00 000000: 00 spruce 00 03000 00 0000000 00 0000 00 202000 00 800:0: 00 00:0 00 :00: 0 nu¢¢o 0 010020 e >00 0 04:0: 0 34.00 0 2:00 0 0000 0 0:00: 0 00 00 00 00 00 an 00 00 mm em 00 we 00 00 Ne 00 00 00 wwHexe< ueemu.0 5n 0:.0xn: on 00.00:: a. 00.00:: 0. 00.00:: an 00.00:: 0. gunman .. 0:000» 00 .0m:0: a. 4:000: 00 .4mu:u 50 04000:: 00 0000:: an 00¢:xn: 0. nu::u:~ n. 00::0 «« 0x00 .0 0:: on gun a. 0:. a. 00:00.: 5. :rxaau 0. 0:0.» 0. nu.muo. 0. an.» n. :n:00. a. 000:»: .. :u:: 0. :ma: 0 nugxu 0 n:oo:. 5 >0» 0 0:10: 0 0:0:0 0 3:0. n .r.u a 0:00: . ~m .m 0. 0v 0» 50 on .0 v. .0 N0 .0 00 0. 0. 5. ozoqeam a.¢pa< ms: :0: vacuoaa .0 APPENDIX C ADMINISTRATIVE CONFIGURATION SURVEY INSTRUMENTS 245 246 MICHIGAN STATE UNIVERSITY COLLEGE OF EDUCATION EAST LANSING ' MICHIGAN ' 48824 DEPARTMENT OF ADMINISTRATION AND HIGHER EDUCATION December 13' 1976 ERICKSON HALL Dear Superintendent: School district organizations are frequently described as top-heavy bureaucracies which squander scarce resources on administrative ”frills" to the detriment of the processes of teaching and learning. We believe that this characterization is patently false and are engaged in an extensive study which will document the conditions and consequences of key administrative characteristics of school districts in Michigan. Although some of the information we need for our study is available in published documents and records, our success will depend upon information which only you provide. we have designed the enclosed survey instrument to gather the necessary informa- tion at minimal costs to yourself. In essence, we need to know: 1. the dates of service of the current (as of the l975-1976 school year) and two preceding superintendents of your district: 2. the title of the immediate supervisor of each administrator in Your district: 3. the number of non-clerical personnel under the immediate supervision of each administrator in your district: and 4. the titles of the administrators responsible for each of the buildings in your district. All information requested refers to the 1975-1976 school year. Please read the accompanying instructions carefully and return your completed survey instrument in the stamped. self-addressed envelope at your earliest convenience. No individual or school district will be identified in reporting the results of this study and your responses will be treated with the strictest standards of professional confidentiality. If you have any questions or comments about our study, please do not hesitate to call either of us (collect, of course) at the numbers listed below. Thank you in advance for your assistance in this important study. sincerely, Frederick R. Ignatovich Stanley E. Hacker Robert H. Richardson Department of Administration Department of Administration Survey Director and Higher Education and higher Education (517) 372-1369 (517) 353-5342 (517) 355—4595 Enclosures 247 Code ADMINISTRATIVE CONFIGURATION SURVEY Please indicate the name and telephone number of the person completing this survey instrument in case we have questions requiring further clarification. Name Telephone SECTION I Instructions: We need to know the dates of service of the current (as of the 1975- 1976 school year) and two preceding superintendents of your district. On line one please enter the name of the current (as of the 1975-1976 school year) superintendent of your district and the date on which he/she began his/her super- intendency. On lines two and three. please enter the names of the two preceding superintendents of your district and the dates of their respective superintendencies. Name From 23 1. 6/30/76 2. 3. Thank You PLEASE CONTINUE TO THE NEXT SECTION 248 ADMINISTRATIVE CONFIGURATION SURVEY Section 11 Instructions: We need to know (1) the title of the immediate supervisor of each administrator in your district during the 1975-1976 school year and (2) the number of individuals supervised by each administrator in your diStrict during the 1975-1976 school year. The table on the following pages lists the 25 administrative classifications utilized by the Michigan Department of Education in the “Register of Professional Personnel.” Column 1 indicates the title of each position. Column 2 indicates our code for each position. (NOTE: Since some districts have more than one person in a given position. and since we are prevented from using the names of individuals, this is our way of differentiating between individuals. For example, AH-l and AH-Z refer to two different Secondary Principals). An ”X" in Column 3 indicates that you had an individual in a given position during the 1975-1976 school year. Please enter the appropriate codes and numbers in Columns 4 and S as follows: Column 4 - Immediate Sppgrvisor: Please enter the Position Code (from Column 2) of the immediate supervisor and each administrative position indicated by ”X“ under column 3. Position Occupied 1975-1976. Column 5 - Number Supgrvised: Please enter the total number of non-clerical personnel under the immediate supervision of each administrator indicated by "x" under Column 3, Position Occupied 1975-1976. E X ATIfl P L E (1) (2) (3) (4) (5) sunnvzson. SECONDARY AQ-l x ”6.3 3 SUPERVISOR. sscouoasy ap-z ’ snczaz. rouca'rxou DIRECTOR AR-l x ””4 /6 This ficticious school district had one suns/1503. SECONDARY tag-1) who was supervised by a. ascend of two ASST. SDPIIINTBNDENT. GENERAL (Ab-2) and who was responsible for the supervision of three non- clerical personnel during the 1975-1976 school year. The district also had one SPECIAL ED. DIRECTOR (AR-l) who reported directly to SUPERINTENDERT, GENERAL (AA-l) and who supervised 16 non-clerical personnel during the 1975-1976 school year. in the following table. please enter the appropriate Position Code and Number Supgrvised for each administrative position indicated by 'x" in Column 3. Position Occupied l975-1976. 249 Code ADMINISTRATIVE CONFIGURATION SURVEY Section 21 (l) (2) (3) (4) (S) POSltlon Title Position Position Immediate Supervisor , Number Supgrvised 2295 92532159 (Enter Position Code of the (Enter the total number of 197S~1976 . —— . —— —- ----- immediate supervisor of each non-clerical personnel under administrator indicated by the immediate supervision of “x" under Position Occupied each administrator indicated 1975-1976.) by ”X" under Position Occupied 1975-1976.) SUPERINTENDENT. GENERAL AA-l SUPERINTENDENT. GENERAL AA’Z ASST. SUPERINTENDENT. GENERAL AB-l ASST. SUPERINTENDENT. GENERAL A3-2 ADMIN. Of FINANCE OR BUSINESS AC-l ADMIN. OF FINANCE OR BUSINESS AC-Z ADMIN. OF INSTRUCTION AD-2 ADMIN. OF INSTRUCTION AD-2 ADMIN. OF PLANT 5 FACILITIES AE-l ADMIN. OF PLANT S FACILITIES . AE-2 ADMIN. OP DIPLOYED PERSONNEL AF-l ADMIN. OP DJPLOYED PERSONNEL Alf-2 AMIN. 0! RESEARCH AG-l ADMIN. or assume: AG-2 PRINCIPAL. SECONDARY All-l 4* It will not PRINCIPAL. SECONDARY Afl-Z parwcxpar. ssoouoanr All-3 b. mess-Irv PRIKIPAL. SECONDARY Ali-4 ' to report PRINCIPAL. SECONDARY Afl-S amount. swam-m AI-l m PRINCIPAL. WAR? AI-2 spanned PRIICIPAL. WAR! AI-3 seminar. was! AI-4 9°! PRIICIPAL. WAR! AI-S ASST. PRIEIPAL. 8W AJ-l Principals ASST. PRINCIPAL. snowman! as-z 0* AIST. PRIKIPAL. M! AJ-J ' Indicates second assign-ant PLEASE cow-mm TO NEXT use 250 Code ADMINISTRATIVE CONFIGURATION SURVEY Section II (1) (2) (3) (d) (5) Position Title Position Position I-ediate Supgrvisor mar Sumrvised LN. $335336 (Bite: W of the (hater the gels; _n__umbu of i-ediate supervisor of each non-clerical personnel under (administrator indicated by the immediate supervision of '1' under Position chpied each administrator indicated 1975-1976.) by "X" under Position Occupied 19754976.} ASST. PRINCIPAL. ELDIEN'I'ARY AK-l Assistant ASST. PRINCIPAL. ELEMENTARY Alt-2 ‘ Principals ASST. PRINCIPAL. ”HINT“? Alt-3 CONSULTANT. SUBJECT m AL-l CONSULTANT. SUBJECT AREA AL-2 CQISULTANT. ELDAENTARY Ali-1 consuum. ELEMENTARY ADI-2 CONSULTANT. SECONDARY AN-l CONSULTANT. SECONDARY AN-2 COORDINATOR. SUDJECT AREA AO—l COORDINATOR. SUBJECT AREA AO-Z SUPERVISOR. ELDIENTAR‘! AP-l SUPERVISOR. ELEMENTARY AP-2 SUPERVISOR. SECONDARY AQ-l SUPERVISOR. WY AQ-Z SPECIAL NATION DIRECTOR AR-l SPECIAL EUJCATION DIRECTOR AR-Z mason. STATE. 5 PM. PROGS AS-l (INSULIN. STATE 5 no. MS AS-Z WIT! SCIKDL DIRECTOR AT-l WRIT)! SCHCDL DIRECTOR AT-2 DIRECTOR. VOQTIGIAL sums-10w AD-l DIRECTOR. VOCATIQIAL ERICATIGI All-2 DIRECTOR. DATA PIOCESSING AV-l DIRECTOR. DATA PROCESSING av-z DIRECTOR. TRANSPORTATIGI Alt-l DIRECTOR. TRAISPORTATICI Nl-Z ~ DIRECTOR. noun-m. SD. AX-l DIRECTOR. ADDLFGJNT. ED. AX-S SUPERVISOR. SPECIAL NATION AY-l SWIM. SPECIAL mead-ion AY-Z 0 Indicates second assignment I m1: 3G1 P'OR WM SEC?!“ II 251 Code ADMINISTRATIVE CONFIGURATION SURVEY Section III Instructions: us need to know the titles of the administrators responsible for each of the buildings in your district during the 1975-1976 school year. The following table lists the buildings utilised in your district during the 1975-1976 school year. Column l indicates the building Code and Column 2 indicates the Utilization Code of each building. In Column 3. you are requested to enter the Position Code (from Section II of this survey instrument) of the administrator responsible for each building utilised in your district during the 1975-1976 school year. EXAMPLE 402 .. 1. JJN’ Iii/.3 . 5678 fill" 7990 u - I In this fictitious school district. building 402--an Elesentary building-~was supervised by a PRINCIPAL. ELEMENTARY (AI-2): Building I243o-a Riddle School--was supervised by a PRINCIPAL. ELEMENTARY (AI-l): building 5678--a Senior nigh School-~was supervised by a PRINCIPAL. SECONDARY (AN-3)) and building 7890-- a Vocational Center-~was supervised by a DIRECTOR, VOCATIONAL EDUCATION (AU-l). (PLEASE NOTICE that the Superintendent kindly corrected two errors in our table). In the following table. please enter the Position Code (from Section II of this survey instrument) of the building Supervisor of each building utilised in your district during the 1975-1976 school year. (I) (2) (3) puilding Code Utilisation Code buildipg Smrvisor gpilisation code Roy A Sleaentary School P or R-S or S I Special Education Can tar b Junior high School 7-b or 7-9 I Administrative building 3 ::=::r High School 9-12 or lO-lZ L Library a school I vocational. Reini s E Junior-Senior sigh School 1-12 0 Other ng Apprentice Center P Elementary through High School 252 Code ADMINISTRATIVE CONFIGURATION SURVEY Section Ill (1) (2) (3) building Code Utilization Code building Suggrvisor QUAIR YOU FOR COH?LITING SICTIGN III 253 MICHIGAN STATE UNIVERSITY COLLAGt OF EDUCATION EAST LANSING ' MICHIGAN ' S8824 DEPARTMENT OF ADMINISTRATION AND HIGHER EDUCATION mcxsox HALL January 21, 1977 Dear Superintendent: Last month we wrote to you to request your assistance in the Administrative Configuration Survey--a research project designed to document the conditions and consequences of key administrative characteristics of school districts in Michigan. We are now in the process of analyzing the survey instruments which have been returned--a1most 500 of the 530 districts surveyed--and we regret that we have not received a response from your district. Since we are anxious to include every Michigan x-12 district in our research, the absence of information from your district diminishes the impact of the study. We are, therefore, sending you another set of survey instruments with the request that you complete and return them at your earliest convenience. All information requested refers to the 1975-1976 school year. No individual or school district will be identified in reporting the results of the study and your responses will be treated with the strictest standards of professional confidentiality. If you have any questions or comments about the study, please do not hesitate to call one of us at the numbers listed below. If you mailed the original survey instruments after January 21. please ignore this request. Thank you for your cooperation and assistance in this important study. Sincerely. Frederick R. Ignatovich Stanley E. Becker Robert H. Richardson Department of Administration Department of Administration Survey Director and Higher Education and higher Education (517) 372-1369 (517) 353-5342 (517) 355-4595 ' Enclosures clh BIBLIOGRAPHY 254 Aiken, Michae1, and Hage, Jera1d. 1968 "OrganizationaI Interdependence and Intraorganizationa1 Structure. " American SocioIogica1 Review 33 (December 1968):912-930. Anderson, James G. ' 1968 Bureaucracy in Education. Baltimore: The John Hopkins Press, 1968. Anderson, Theodore R., and Warkov, Semour. 1961 "OrganizationaI Size and Functiona1 Comp1exity: A Study of Administration in Hos ita1s." American SocioIogicaI Review 26 (February 1961 :23-28. Argyris, Chris. 1960a Understanding,0rganizationa1 Behavior. Homewood, 111.: Dorsey Press, 1960. Argyris, Chris. 1960b "Individua1 Actua1ization in Comp1ex Organizations. " MentaI Hygiene 44 (1960): 226- 237. Argyris, Chris. 1964 Integrating the Individua1 and the Organization. New York: John NiIey and Sons, 1964. Argyris, Chris. 1972 The App1icabi1ity of 0rganizationa1 Socio1ogy. New York: Cambridge University Press, 1972. Baker, A1ton R., and Davis, Ra1ph C. 1954 Ratios of Staff to Line Emp1oyees and Stages of Dif- ferentiation 0f Staff Functions. Co1umbus: Bureau of Businessr Research, The Ohio State University, 1954. Barnard, Chester I. 1938 The Functions of the Executive. Cambridge, Mass.: Harvard University Press, 1938. 8e11, GeraId D. 1967 "Determinants of Span of Contro1. " American JournaI of Socio1ogy 73 (Ju1y 1967):100-109. Bendix, Reinhard. 1956 Work and Authority in Industry. New York: Harper and Row, 1956. 255 256 B1au, Peter M. 1963 The Dynamics of Bureaucracy(rev. ed.). Chicago: The University of Chicago Press, 1963. B1au, Peter M. 1965 "The Comparative Study of Organizations." Industria1 and Labor Re1ations Review 18,3 (Apri1 1965):323-338. B1au, Peter M. 1967 Exchange and Power in Socia1 Life.- New York: John Ni1ey and Sons, 1967. B1au, Peter M. 1968a "The Study of Forma1 Organizations." In Peter M. B1au, 0n the Nature ofOrganizations, pp. 27-36. New York: John Hi1ey and Sons, 1974. B1au, Peter M. 1968b "The Hierarchy of Authority in Organizations." American Journa1 of Socio1ogy 73 (January 1968):453-467. B1au, Peter M. 1970 "DecentraIization in Bureaucracies." In Mayer N. ZaId, ed., Power in OrganizatiOns. Nashvi11e, Tenn.: Vanderbi1t University Press, 1970. B1au, Peter M. 1973 The Organization of Academic Hark. New York: John Hi1ey and Sons, 1973. B1au, Peter M., and Scott, H. Richard. 1962 Forma1 0r anizatiOns. San Francisco: Chandier Press, 1962. BIau, Peter M.; Heydebrand, Wo1f V.; and Stauffer, Robert. 1966 "The Structure of Sma11 Bureaucracies." American> SocioIogjca1 ReView 31 (Apri1 1966):179-191. B1au, Peter M., and Orum, Amy N. 1968 ' "Conditions of Bureaucratic Forma1ization." In A1bert J. Reiss, Jr., ed.,'C001ey'and Socio1ogica1‘Ana1ysis. Ann Arbor: University of Michigan Press, 1968, pp. 68-86. B1au, Peter M., and Meyer, Marsha11 w. 1971 Bureaucracy and Madern‘Society. 2nd ed. New York: Random House, 1971. B1au, Peter M., and Schoenherr, Richard S. 1971 ’ The StrUcture ofggrganizatIOns. New York: Basic Books, 1971. 257 B1auner, Robert. 1964 A1ienation and Freedom. Chicago: University of Chicago Press, 1964. Bobbitt, Frank1in. 1913 "The Supervision of City Schoo1s: Some Genera1 Prin- cip1es of Management AppIied to the Prob1ems of City Schoo1 Systems." Twe1fth Yearbook of the National Sggiety1fgg the Study of Education, Part I. BIoomington, .: 9 . Bo1and, Na1ter R. 1973 "Size, ExternaI Re1ations and the Distribution of Power: A Study of Co11eges and Universities." In No1f V. Heydebrand, ed., Comparative Organizations: The Resu1ts of Empirica1 Research. ‘EngIewood C1iffs, N.J.: Prentice-Ha11, Inc., 1973, pp. 428-440. Borger, R. and Cioffi, F., eds. 1970 Exp1anation in the Behaviora1 Sciences. C.U.P., 1970, pp. 313-343. Borgotta, Edgar F. 1968 "My Student the Purist." Socio1ogica1 QuarterLy 9 (1968):29-34. ' Cap1ow, Theodore. 1957 "Organization Size." Administrative Science Quarterly 1 (March 1957):484-505. Cap1ow, Theodore. 1964 PrincipIes of Organization. New York: Harcourt, Brace and Nor1d, 1964. Cap1ow, Theodore, and McGee, Reece J. The Academic Market P1ace. 1958 New York: Basic Books, 1958. Car1son, Richard O. 1961 "Succession and Performance Among Schoo1 Superinten- dents." 'Administrative Science Quarter1y 6 (September 1961):210-227. Chapin, F. Stuart. 1951 "The Growth of Bureaucracy: An Hypothesis." American Socio1ogica1 Review 16 (December 1951):835-856. Co1eman, James S. 1975 "Racia1 Segregation in the Schoo1s: New Research with New Po1icy Im Iications."' Phi De1ta Kappan 57,2 (October 1975 :75-78. 258 CoIeman, James 5.; Campbe11, E. 0.; Hobson, C. J.; McPartIand, J.; 1966 Mood, A. M.; Heinfe1d, F. 0.; and York, R. L. Equa1ity of Educationa1 Opportunity. Washington, D.C.: U.S. Department of HeaIth Education and We1fare, Office of Education, U.S. Government Printing Office, 1966. Co1eman, James 5.; Ke11y, Sara 0.; and Moore, John. 1975 Trends in Schoo1 Segregation, 1968-1973. Hashington, D.C.: Urban Institute, 1975. Corwin, Rona1d G. 1965 "Professiona1 Persons in Pub1ic Or anizations." ‘Educa- tionaI Administration Quarter1y 1 (1965):1-22. Crozier, MichaeI. 1964 The Bureaucratic Phenomenon. Chicago: University of Chicago Press, 1964. Da1ton, Me1vi11e. 1950 "Coanicts Between Staff and Line Manageria1 Officia1s." American Socio1ogica1 Review 15 (1950):342-351. Dewey, John. 1972 Experience and Education. New York: CoIIier Books, 1972. Durkheim, EmiIe. 1964 The Division of Labor in Spoiety. Trans1ated by George Simpson. New York: The Free Press, 1964. Etzioni, Amitai 1964 Modern Organizations. Eng1ewood C1iffs, N.J.: Prentice- Ha11, 1964. Etzioni, Amitai. 1964 The Active Society. New York: The Free Press, 1968. Fayo1, Henri 1949 Genera1 and Industria1 Management. London: Sir Isaac Pitman, 1949. Francis, Roy 6.; and Stone, Robert C. 1956 Service and Procedure in Bureaucracy. Minneapo1is: University of Minnesota Press, 1956. Georgopou1os, Basi1 S., and Mann, F10yd C. 1962 The Community Genera1 HospitaI. New York: Macmi11an, 1962. 259 Gibbs, Jack P., and Martin, Halter T. 1962 "Urbanization, Technology and the Division of Labor: International Patterns." American Sociological Review 27 (October 1962):667-677. Gibbs, Jack P., and Browning, Harley L. » 1966 "The Division of Labor, Technology and the Organization of Production in Twelve Countries." American Sociologi- cal Review 31 (February 1966):81-92. Goodman, Paul. 1962 The Community of Scholars. New York: Random House, 1962. Gou1dner, A1vin W. 1954 Patterns of Industrial Bureaucracy. Glencoe: Free Press, 1954. Gou1dner, A1vin H. 1957-58 "Cosmopolitans and Locals." Administrative Science Quarterly 2 (l957-58):281-306, 444-480. Grusky, Oscar. 1961 "Corporate Size, Bureaucratization and Manageria1 Succession." American Journal of Sociology_67,3 (November 1962):26l-269. Gulick, Luthen,and Urwick, Lyndall. 1937 Papers on the Science of Administration. New York: Institute of Public Administration, Columbia University, 1937. Haas, Eugene; Hall, Richard H.; and Johnson, Norman J. 1963 "The Size of the Supportive Component in Organizations: A Multi-organizational Analysis." Social Forces 43 (October l963):9-17. Hage, Jerald. 1965 "An Axiomatic Theory of Organizations." Administrative Science Quarterly 10 (December 1965):289-320. Hage, Jerald, and Aiken, Michael. 1967a "Program Change and Organizational Properties." American Jaurnal 0f SOCIOIOgy 72,5 (March 1967):503-509. Hage, Jerald, and Aiken, Michael. 1967b "Relationship of Centralization to Other Structural Properties."' Administrative Science QUarterjy 12,1 (June l967):72-92. 260 Hage, Jerald, and Aiken, Michael. 1969 "Routine Technology, Social Structure and Organizational Goals." AdministratiVe Science Quarterly 14,3 (September 1969):366-376. Haire, Mason. 1959 "Biological Models and Empirical Histories of the Growth of Organizations." In Mason Haire, ed., Modern Or aniZatiOn Theor . New York: John Wiley and Sons, Inc., I959, pp. 287-293. Hall, Richard H. 1962 "Intra-organizational Structure Variation." Administra- tive Science Quarterly 7,3 (December 1962):295-308. Hall, Richard H. 1963 "The Concept of Bureaucracy: An Empirical Assessment." American Journal of Sociology 69,1 (July 1963):32-40. Hall, Richard H. 1968 "Professionalization and Bureaucratization." American Sociological Review 33 (February 1968):92-104. Hall, Richard H. 1972 Organizations: Structure and PrOCess. Englewood Cliffs, N.J.: PrenticelHall, Inc., 1972. Hall, Richard H.; Haas, J. Eugene; and Johnson, Norman J. 1967 "Organizational Size, Complexity and Formalization." American SociolOgjcal Review 32,6 (December 1967): 903-912. Hall, Richard H., and Tittle, Charles R. 1966 "Bureaucracy and Its Correlates." 'American Journal of Sociology 72,3 (November 1966):267-72. Hawley, Amos H.; Boland, Walter; and Boland, Margaret. 1965 "Population Size and Administration in Institutions of Higher Education." ‘American Sociological ReView 30 (April l965):252-255. Herzberg, Frederick; Mausner, Bernard; and Snyderman, Barbara. 1959 The Motivation to Work. New York: John Wiley and Sons, 1959. Heydebrand, Wolf V. 1972 "The Logic of the Gini Index." In Wolf V. Heydebrand, Hospital Bureaucracy: ‘A camparative Study Of'Organiza- tiOns. New York: Dunellen-University Press of Cambridge, 1972, chap. 10. 261 Heydebrand, Wolf V. 1973 "Autonomy, Complexity and Non-bureaucratic Coordination in Professional Organizations." In Wolf V. Heydebrand, ed., Comparative Or anizations: The Results of Empirical Research. Englewood C iffs, N.J.: *Prentice—Hall, 1973, pp. 158-188. Heydebrand, Wolf V., and Noell, James J. 1973 "Task Structure and Innovation in Professional Organiza- tions." In Wolf V. Heydebrand, ed., Comparative Organizations: The Resultsgof Empirical Research. Egglewood C1iffs, N.J.: Prentice-Hall, 1973, pp. 294- 3 . Hickson, D. J.; Pugh, D. 5.; and Phesey, Diana C. 1969 "Operations Technology and Organizational Structure: An Empirical Reappraisal." Administrative Science Quarterly 14,3 (September 1969):378-397. Hughes, Everett C. 1958 Men and Their Work. Glencoe: The Free Press, 1958. Indik, Bernard P. 1964 "Organization Size and Supervision Ratio." Administra- tive Science Quarterly 9 (December l964):301-312. Kerlinger, Fred N., and Pedhazur, Elazar J. 1973 Multiple Regression in Behaviora1 Research. New York: Halt, Rinehart and Winston, Inc., 1973. Kimberly, John R. 1976 "Organizational Size and the Structuralist Perspective: A Review, Critique, and Proposal." Administrative Science Quarterly 21,4 (l976):571-597. Kornhouser, William. 1963 'Scientists in Industry. Berkeley: University of California Press,21963. Lawrence, Paul R., and Lorsch, Jay W. 1967 Qrganizatidn and Enyironment. Cambridge: Harvard Graduate School of Business Administration, 1967. Lazarsfeld, Paul F., and Menzel, Herbert. 1961 "On the Relation Between Individual and Collective Properties." In Amitai Etzioni, Com lex Or anizations. New York: Holt, Rinehard and Winston, 1961, pp. 426- 35. 262 Lewin, Kurt. 1943 "Forces Behind Food Habits and Methods of Change." Bulletin of the National Research COuncil 108 (1943): 35-65. Likert, Rensis. 1961 New Patterns of Management. New York: McGraw-Hill Book Co., 1961. Lindenfeld, Frank. 1961 "Does Administrative Staff Grow as Fast as Organization?" School Life 43,8 (May 1961):20-24. MacKay, D. A. 1966 "Using Professional Talent in a School Or anization." Cagadian Education and Research Digest 6 1966):342- 35 . Mann, Floyd C., and Williams, Lawrence K. 1960 "Observations in the Dynamics of a Change to Electric Data Processing Equipment." Administrative Science Quarterly 5 (1960):221 ff. March, James G., and Simon, Herbert A. 1958 Or anizatiOns. New York: John Wiley and Sons, Inc., 958. Maslow, Abraham H. 1956 Eupsychian Mana ement. Homewood, I11.: Richard D. Irwin, Inc., 19 6. Mayo, Elton. 1933 The Human Problems of Industrial Civilization. New York: The Macmillan Co., 1933. McGregor, Douglas. 1960 The Human Side of Enterprise. New York: McGraw-Hill Book Co., 1960. Melman, Semour. 1951 "The Rise of Administrative Overhead in Manufacturing Industries of the U.S., 1899-1947." ‘Oxford Economic ‘Papers 3 (1951). Merton, Robert K. 1940 "Bureaucratic Structure and Personality."' Social Farces 18 (1940):560-568. Merton, Robert K. 1968 Social Theory and Social Structure, 1968. Enlarged ed. New York: The Free Press,41968. 263 Metzner, Helen, and Mann, Floyd C. 1953 "Employee Attitudes and Absences." Personnel Psychology, 6 (Winter 1953):467-485. Meyer, Marshall W. 1968 "Two Authority Structures of Bureaucratic Organizations.” Administrative Science Quarterly 13,2 (September 1968): 211-228. Meyer, Marsha11 W. 1972 BureauCratic Structure and Authority: Coordination and Control in 254 GovernmentyAgencies. New York: Harper and Row, 1972. Michigan State University. 1976 "Statistical Package for the Social Sciences, Version 6.5." Michigan State University, 1976. Miller, George A. 1967 "Professionals in Bureaucracy: Alienation Among Industria1 Scientists and Engineers:" American Sociological Review 32,5 (October 1967):755-768. Moeller, Gerald H., and Charters, W. W. 1966 "Relation of Bureaucratization to Sense of Power Amon Teachers." Administrative Science Quarterly 10 (1966?: 444-465. Montagna, Paul D. 1968 "Professionalization and Bureaucratization in Large Professional Organizations." American Journal of Sociology 74 (September l968):138-145. Mooney, James 0., and Reilly, Allan C. 1931 Onward Industry. New York: Harper and Row, 1931. National Merit Scholarship Corporation. . 1976 "Semifinalists in the Twenty-second Annual Merit Scholarship Competition." National Merit Scholarship Corporation, 1976. Nie, Norman H.; Hull, C. Hadlai; Jenkins, Jean 6.; Steinbrenner, 1975 Karin; and Bent, Dale H. Statistical ngkgge for theg§ocia1 Sciences. 2nd ed. New York: McGraw-Hill Book Co., 1975. Parkinson, C. Northcote. 1957. ParkinSOn's'Law. New York: Ballantine Books, 1957. 264 Parsons, Talcott. 1960 Structure and Process in Modern Societies. New York: The Free Press, 1960. Parsons, Talcott. 1959 "The School as a Social System: Some of Its Functions in American Society." Harvard Educationa1 Review XXIX (Fall 1959):297-318. Parsons, Talcott. and Platt, Gerald M. ~ 1973 The American University. Cambridge, Mass.: Harvard University Press, 1973. Pelz, Donald C. , 1956 "Some Socia1 Factors Related to PerfOrmance inea Research Organization." ’Administrative Science Quarterly 1 (1956):310-325. Perrow, Charles. 1970 Organizational Analysis. Belmont, California: Brooks- Cole Publishing Co., 1970. Perrow, Charles. 1972 Complex Organizations. Glenview, I11.: Scott, Foresman andPCo., 1972. Pondy, Louis R. 1969 "Effects of Size, Complexity and Ownership on Administra- tive Intensity." 'Administrative Science Quarterly (March 1969):47-61. Porter, Lyman W., and Lawler, Edward III. 1968 'Managerial Attitudes and Performance. Homewood, 111.: Irwin, 1968. Presthus, Robert. 1962 ’ The Organizational satiety. .New York: Vintage Books, 1962. Price, James L. 1968 Organizational EffectiVeness: An InVentory‘Of'Pro- positiOns.PHomewood, 111.: Richard D. Irwin, Inc., 1968. Price, James L. 1972 Handbook 0f Orggnizational MeaSUrement. Lexington, Mass.: 0. C. Heath and Co., 1972. Pugh, D. S.; Hickson, D. J.; Hinings, C. R.; and Turner, C. u .n .1 ~ 1968 "Dimensions of Organizational Structure." ‘Administrative Science Quarterly (June l968):65-106. 265 Roethlisberger, Fritz J., and Dickson, William J. 1939 Management and the Worker. Cambridge: Harvard University Press, 1939. Raphael, Edna. 1967 "The Anderson-Warkov Hypothesis in Local Unions: A Comparative Study." American Sociological Review 32,5 (October 1967):770-776. Rushing, William A. 1966 "Organizational Size and Administration.“ Pacific Sociological Review 9 (September 1966):100-108. Rushing, William A. 1967 "The Effects of Industry Size and Division of Labor on Administration." Administrative Science Quarterly (September l967):267-295. Scott, Richard. 1965 "Reactions to Supervision in a Heteronomous Professional Organization." Administrative Science Quarter1y_10 (June 1965):65-81. Seashore, Stanley E., and Yuchtman, Ephraim. 1967 "Factorial Analysis of Organizational Performance." Administrative Science Quarterly 12,3 (December 1967): 377-395. Selznick, Philip. 1948 "Foundations of the Theory of Organizations." American Sociological Review 13 (1948):25-35. Selznick, Philip. 1949 T.V.A. and the Grass Roots. Berkley: University of California Press, 1949. Sergiovani, Thomas J. 1967 “Factors which Affect Satisfaction and Dissatisfaction of Teachers." The Journal of Educational Administration 5 (1967):66-82. Simon, Herbert A. 1957 Administrative Behavior. 2nd ed. New York: The Free Press, 1957. Simon, Herbert A. 1964 "On the Concept of Organizational Goal." Administrative Science Quarterly 9,1 (June 1964):1-22. 266 Simpson, Richard L., and Gulley, William H. 1962 "Goals, Environmental Pressures and Organizational Characteristics." American Sociological Review 27 (June 1962):344—51. Spaulding, Frank. 1955 School Superintendent in Action in Five Cities. Rindge, New Hampshire: 1955. Starbuck, William H. 1965 "Organizational Growth and Development." In James G. March, ed., Handbook of Organizations. Chicago: Rand McNally and Co., 1965, pp. 451-533. Stinchcombe, Arthur L. 1959 "Bureaucratic and Craft Administration of Production: A Comparative Study." Administrative Science Quarterly (September 1959):168-187. Taylor, Frederick W. 1911 Principles of Scientific Management. New York: Harper and Row, 1911 . Terrien, Frederick W., and Mills, Donald L. - 1955 "The Effect of Changing Size Upon the Internal Structure of Organizations." American Sociological Review 20 (February 1955):11-14Z Terrien, Frederick W. 1963 The Effect of Changing Size Upon Organizegions. San Francisco: Institute for Social Science Research, San Francisco State College, 1963. Thompson, Victor A. 1950 The Regulatory Process in OPA Rationing. New York: KingTs Crown Press, 1950. Thompson, Victor A. 1961 Modern Organizations. New York: Alfred A. Knoph, Inc., 1961. Tsouderos, John E. ~ 1955 "Organizational Change in Terms of a Series of Selected Variables." ‘American Sociological Review 20 (April 1955):206-210. Udy, Stanley H. 1959 "The Structure of Authority in Non-Industria1 Production Organizations." “American Journal of SociOlOgy 64 (May 1959):582-584. 267 Whyte, William H., Jr. 1956 The Organization Man. New York: Simon and Schuster, Inc., 1956. Weber, Max. 1 1946 From Max Weber: Essays in Sociology. Translated by H. H. Gerth and C. Wright Mills, eds. New York: Oxford University Press, 1946. Weber, Max. 1947 The Theory of Social and EconOmic Organization. Trans- lated by A. M. Henderson and T. Parsons, T. Parsons, ed. New York: The Free Press, 1947. Wilensky, Harold L. 1964 "The Professionalization of Everyone?" American Journal of Sociology 70 (September 1964):137-158. Woodward, Joan. 1962 Industrial Organizations. London: Oxford University Press, 1962. Yuchtman, Ephraim, and Seashore, Stanley E. 1967 "A System Resource Approach to Organizational Effective- ness." American Sociological Review 32,6 (December 1967):89T-903.