Example: bachelor of science






Link to this page:

Please notify us if you found a problem with this document:

Other abuse


POLITICS, STRUCTURE, AND PUBLIC POLICY:THE CASE OF HIGHER EDUCATION Jill Nicholson-Crottyand Kenneth J. MeierDept. of Political ScienceandGeorge Bush School of GovernmentTexas A&M University4348 TAMUSCollege Station, TX Educational PolicyABSTRACTThis study examines whether governance structures facilitate or impede political forcesby testing two competing hypotheses concerning the ability of bureaucratic structures to insulatehigher education policies from politics. Centralized structures both create autonomy andfacilitate access by environmental forces. This study examines the structures of higher educationboards to gain a better understanding of how they interact with politics to affect higher educationpolicy. To the extent that variation in governance structures is correlated with bureaucraticautonomy, it should limit the ability of elected officials to influence education policies. Thetransaction costs of individuals seeking to influence overall agency policy are lowered, however,in more centralized organizations. Political actors can focus their attention on a singlegeographic site rather than multiple sites that are adapting to different sets of institutionalarrangements and different local environments. These hypotheses are tested in a 47-state, eight-year words: higher education, politics, structure1POLITICS, STRUCTURE, AND PUBLIC POLICY:THE CASE OF HIGHER EDUCATION The relationship between government structure and performance can be dated to both thedebates over the Constitution and the attempts by ancient philosophers such as Aristotle todesign the ideal polity. Unfortunately, until recently, systematic empirical attention to this issuein higher education has not been a high priority (McLendon forthcoming). Using formal workfrom the political control of bureaucracy literature (McCubbins, Noll and Weingast 1987; Moe1990) as its guide, political scientists (Lowry, 2001 and Knott and Payne, 2002) have started anempirical investigation of the role that structure plays in determining higher education policyoutputs. A second branch of empirical work by education scholars grows out of organizationtheory and a substantive interest in higher education (see Hearn and Griswold 1994; Hearn,Griswold, and Marine 1996; McLendon, Heller and Young 2002; Volkwein 1986; 1989; Zumeta1992, 1996).The structure question in higher education links to an important theoretical literature inpolitical science. Structural questions are an essential part of the political control literaturebecause some advocates believe that structures are little more than hardwired political biases thataffect who benefits from administrative decisions (McCubbins, Noll and Weingast 1987). Structures, however, have a second function. They set the rules of game and, thus, advantagesome interests rather than others without actually hardwiring biases into the system. As anillustration, providing for legislative oversight of administrative agencies advantages thepolitical interests who control the legislature but does not guarantee that the wishes of onefaction will forever triumph over another (Moe 1990). Lowry (2001), in his seminal study ofstructures and higher education, demonstrated the first aspect of higher education structures; they2are associated with the distribution of policy benefits (in a different tradition see similar studiesby Hearn and Griswold 1994; Hearn, Griswold, and Marine 1996; McLendon, Heller and Young2002; and Volkwein 1986; 1989). This paper examines the second dimension of highereducation structures; do some structures facilitate or impede the political forces regardless ofideology? Any assessment of higher education governance structures must recognize that theagencies responsible for implementing postsecondary education policy differ markedly acrossthe nation. Three distinct structures are used to govern higher education at the state level--consolidated governing boards, coordinating boards, and planning/service agencies. While thenumber of consolidated governing boards approximately equals the number of coordinatingboards, only two states have planning/service agencies. Because of differences in the scope oftheir activities and the control they have over important functions such as budgeting, theseboards differ both in autonomy and the degree of centralization. This paper uses structuraltheories of politics to derive and test hypotheses about individual governing boards and theirability to insulate policies from politics. In an important addition to previous studies, thisanalysis also controls for the means by which agencies receive their power. First, the theoretical relationship between structure and policy outcomes will be , a description of the structures and powers of the three types of governing boards ispresented. Third, a short review of the literature that examines the effects of politics onbureaucratic structure will provide a set of contrasting hypotheses. Finally, a preliminary test ofthese hypotheses linking structure and politics will involve a quantitative analysis of how highereducation costs are distributed among tuition, state appropriations, and need-based STRUCTURES AND THEORY 3A number of structural dimensions must be addressed in order to comprehend fully how aboard s structure affects the insulation of the costs of higher education from political forces. Asa first step this study will deal with two issues conceptually autonomy and Wewill do so using insights from both organizational theory and the literature on bureaucraticpolitics. Seidman (1970) argues that political structures determine power and that organizationalstructures are an instrument of politics, position, and power. He contends that there are criticaldifferences between institutional types in terms of the composition of the directing authority(single or multi-headed), qualifications for appointment, procedures for the appointment andremoval of principal officers, method of financing, budget and audit controls, personnelregulations, and advisory councils and committees (Seidman 1970, 242; Meier 1980). According to Seidman these provisions determine the degree of organizational and operatingautonomy as well as the relationship between an agency and its political environment (but seeVolkwein 1986).The tension between autonomy and political control is obvious. Scholars argue that for abureaucracy to implement policy effectively it must have a certain level of autonomy and thatautonomy allows for insulation from politics (Meier 2000; Selznick 1948; Wilson 1989; Ingram1990; West 1997). This autonomy and power then influences the ability of an agency to reshapelegislation in a way that more closely fits its mission. According to Rourke (1969, 43), agenciesthat are highly professional in their orientation are allowed a degree of independence andautonomy not afforded to all public agencies. The level of autonomy is important to this study,as is the level of discretion that an agency has to choose the way that it will carry out its mission(Meier 2000). Agency structure can have some role in determining the level of autonomy. 4A combination of these theories leads us to the proposition that bureaucracies structuredto be insulated from politics have a greater degree of autonomy and greater control over policyoutcomes. If structures allow for influence from political changes, the likelihood of autonomyfor the organization decreases. This study examines the structures of higher education boards togain a better understanding of how they affect policies pertaining to higher education. To theextent that variation in governance structures is correlated with bureaucratic autonomy, it shouldcorrelate with differences in the ability of elected officials to influence policies pertaining second structural view of bureaucracy deals with the notion of centralization anddecentralization. To the degree that organizations are centralized, the transaction costs are lowerfor individuals seeking to influence overall agency policy. Political actors focus their attentionon a single geographic site rather than multiple sites that might be adapting to different sets ofinstitutional arrangements and different local environments. Despite the conclusions of Seidmanthat decentralized bureaucracy exists to match the structure of Congress (read the legislature),any political institution seeking to control a bureaucracy has lower transactions costs in acentralized bureaucracy, if all other things are equal. The perceived centralization benefit, infact, has specifically guided the design of higher education structures (see McGuinness 1999;McLendon forthcoming and the citations therein).HIGHER EDUCATION GOVERNANCE STRUCTURESMuch of the literature concerning postsecondary education policy debates how much thestructure of higher education systems affects higher education policies. Studies focus on thepower of the governor, the character of the governing system, and the effects of the et al. (1999) discuss the importance of political actors in the governance of higher5education as well as the increased interest of these parties in existing systems. With a sevenstate comparative study, they examine the structures of state education boards and how stategovernments influence education policies. Jones et al. (1998) provide a generalized overview ofhigher education governance and attempt to set a research agenda for further examination in thisarea. Zumeta (1992) provides a broad study of the effects of changes in state tuition and boardstructures on scholarships for private universities. Hearn, Griswold and Marine (1996) show thathigher education structures are associated with changes in tuition costs and financial aid, but therelationships are not always consistent across types of institution ( , four year versus twoyear). Hearn and Griswold (1994) find these structures affect innovation rates among highereducation systems. McLendon, Heller and Young (2002) examine six policy innovations dealingwith finances and accountability; they find modest influences of structure. Volkwein (1986;1989), in contrast, found no relationship between structural autonomy and the quality ofeducation offered. The present study examines 47 states that have either a consolidated governing board or acoordinating board. Because Michigan and Delaware do not have either type of board, they areexcluded from this study. Nebraska is excluded because its legislature is nonpartisan andlegislative partisanship is a key variable in our analysis. The starting point of our analysis is thework of Lowry (2001), Hearn, Griswold and Marine (1996), and McLendon, Heller and Young2002). Lowry is attempting to synthesize some of the literature and at the same time studyhigher education structures in all fifty states. His work examines the effects of institutions onpolicy implementation. Although he finds that universities whose trustees are selected bynonacademic stakeholders charge prices that are significantly lower than states withdecentralized systems where trustees chosen by academic stakeholders (see also Hearn,6Griswold and Marine 1996), he also suggests that structures can enhance political control (seealso Knott and Payne 2002, 6). Lowry does not actually incorporate political factors in hismodel, however, or interact those political forces with structures. McLendon, Heller, and Young(2002) without using an interaction term find direct effects of political variables on the adoptionof education policy innovations; political factors, in fact, far outweigh structural ones in terms ofinfluence. Both Lowry (2001) and Knott and Payne (2002) note the complex relationship betweenstructure and organizational outputs. Lowry (2001, 859) concludes that different structurescould lead to similar ends as demonstrated by his empirical results. Knott and Payne (2002)show structures including some political aspects (number of gubernatorial appointments ongoverning boards) affect decisions made by universities. Both of Hearn s studies show thatstructures are associated with different outcomes in different policy areas (but see McLendon,Heller and Young 2002). In short, these studies find that different structures can produce similarresults and vice STRUCTURE OF POSTSECONDARY EDUCATION GOVERNANCE BOARDSAlthough the structural governance of higher education varies from state to state,postsecondary education structures generally consist of three types of boards that can be arrayedon a continuum from more to less autonomy as well as on a continuum from more to lesscentralized (Lowry 2001; Knott and Payne 2002, but see Volkwein 1986). Of the three boards,consolidated governing boards appear to have the highest level of autonomy and are the mostcentralized. Consolidated governing boards are assigned most, if not all, authority forcoordinating postsecondary education. The primary responsibility of these boards is to governthe institutions under their jurisdiction. This responsibility, however, might be split between two7boards, one for four-year institutions and one for community colleges or technical of consolidated governing boards include the governance of a singlecorporate entity. The state of North Carolina, for example, has a consolidated governing boardsystem. It is responsible for all decisions concerning that system, and individual institutions donot have separate governing entities. Within this structure the boards have all the rights andresponsibilities of that corporation as defined by state law. Individual institutions under theboard s jurisdiction do not have separate corporate status. These boards also coordinatefunctions, including planning, setting a public agenda, policy analysis, and problem resolution. Governing boards are responsible for academic program review and approval, budgetdevelopment, and maintaining information and accountability. These boards participate indeveloping and implementing policy as well as advocating the needs of the institutions withinthe board s jurisdiction to the legislature and governor. Other responsibilities includeestablishing faculty personnel policies such as awarding tenure and serving as the final point ofappeal on faculty grievances. These boards also allocate and reallocate resources between theinstitutions in the jurisdiction and establish policies for, at times even setting, tuition and fees(ECS 1997; McLendon forthcoming). To perform these functions, governing boards haveprofessional staff, which contributes to greater autonomy (Lowry 2001, 848).Coordinating boards, in contrast, merely provide an interface between the stategovernment and the governing boards of the state s systems and individual colleges anduniversities. Coordinating boards are less likely to have the power and autonomy necessary toresist political influence. Some coordinating boards are responsible for approving academicprograms. Other advisory coordinating boards only have the power to review and makerecommendations to institutional governing boards regarding academic programs. Some carry8out only limited coordinating functions and primarily administer student financial aid orlicensure responsibilities (ECS 1997). Coordinating boards differ from consolidated governing boards in that they do not governinstitutions, they do not appoint institutional chief executives or set faculty personnel appoint, set compensations for, and evaluate only the board s chief executive officer andstaff. The governor is sometimes actually responsible for appointing the agency executive butusually on the recommendation of the coordinating board. Coordinating boards do not havecorporate status independent from the state government (ECS 1997). They focus on state andsystem needs and priorities rather than advocating interests. They may or may not review andmake recommendations on budgets for the state s system. The limited scope of these boardsmeans that they have less coordinating agencies recommend consolidated budgets for the whole systemwhile others make recommendations to the governor or state legislature on individual orsegmental budgets. Most coordinating boards, however, have responsibility for implementingbudget policy only for funds appropriated specifically to the agency (as opposed toappropriations to the individual campuses). Review or approval of academic programs andauthority to require institutions to review existing programs varies. In terms of faculty personnelpolicy they are only responsible for carrying out legislative mandates for studies of issues suchas faculty workload and productivity or tenure policy (ECS 1997). Overall coordinating boardsclearly have less power and autonomy than governing boards; Lowry (2001, 847), in fact, treatscoordinating boards as merely extensions of the legislature or similar political principals. The difference between consolidated governing boards and coordinating boards in termsof autonomy is offset by their differences in centralization. Governing boards with their9autonomy and capacity are centralized, making the logistics of interacting with them relativelyeasy. Coordinating boards do not centralize power but rather leave the actual power of theinstitutions decentralized in the various institutions. All things being equal, the lowertransactions costs of the centralized governing boards means that political influence should beeasier (Lowry 2001, 848). Quite clearly all things are not equal since the centralized governingboards also have greater expertise and autonomy. This combination sets up an interestingempirical contrast between centralized structures with autonomy versus decentralized structureswith less autonomy. POLITICAL INFLUENCE ON HIGHER EDUCATION GOVERNANCEIn addition to governing boards, other political institutions are also involved in thegovernance of higher education within individual states. The governor and lieutenant governorhave some authority over higher education. Some governors and lieutenant governors exertinfluence through formal powers such as appointment of board members or executives, whileothers exert power through influence over the legislature. In addition to governors andlieutenant governors, certain members of state legislatures also affect higher education positions in the legislature are allowed to appoint members to the boards, some powersdelineated from the constitution are given to the legislature, and those boards whose powerresides in statute can experience loss or gain depending on the character of the state legislature. In addition, a variety of committees in state legislatures exert power over policies pertaining tohigher education (ECS 1997). Based on the education policy and bureaucratic literature, this study examines how wellgoverning boards are able to insulate higher education policies in individual states from study differs from previous studies because it not only examines the structure but the means10by which these agencies receive the powers that they have and therefore what process wouldhave to be under-taken by political actors to influence these boards. In addition, we willexamine the importance of how the board executive came to power, how powers are delineated,and the ability of the higher education boards to insulate the governance of institutions of highereducation from policies driven by political 1: Consolidated governing boards will provide more insulation forpolicies from politics because of the structure and autonomy that they have andthat will result in less political influence on education 2: Consolidated governing boards will generate lower transactionscosts to political actors owing to their centralization and this will result in greaterpolitical influence on education two hypotheses are essentially contradictory predictions. The reason for this is thatthe design of governing boards is influenced by two different organizational principles. Thisinherent theoretical conflict, in fact, might be why McGuinness (1999) concludes that theevidence is inconclusive on the benefits and costs of various structures. It also could explain thevaried findings of Lowry (2001), Hearn and Griswold (1994), Hearn, Griswold, and Marine(1996) and Volkwein (1986; 1989).METHODSOur distinct theoretical contribution to the literature is to operationalize Lowry shypothesis as an interaction between structure and politics. No prior test of this interactionexists although most of the literature in the area has not precisely specified that an interactionexists (see McLendon forthcoming). To test the hypotheses, we will code our structural variable(S) as equal to 1 if the state has a coordinating board (rather than a consolidated governing11board). We will then define a vector of political variables (P) as well as a vector of controlvariables (C). Whether or not structure facilitates or restricts the influence of political factorsthen can be tested by interacting structure with politics and using a joint f-test for whether theslopes of the interacted variables are different from zero (and thus different in states withconsolidated governing boards from those in states with coordinating boards). More formally, O = $1S + $2P + $3SP +$4CWhere O is some measure of output, and the key test is whether $3 is equal to zero. DEPENDENT VARIABLESOur concern is the costs of higher education and who incurs them. Four dependentvariables are used. The first is simply the total dollar cost per student of public higher educationin the state. Education, in an ideal world, has redistributive consequences; and to the extent thatthe costs of education are low, these redistributive aspects can materialize. The other threevariables measure the distribution of costs; they are tuition per student, need-based scholarshipsand financial aid per student, and state/local appropriations per student. The basic issue is whopays for the costs of education. States that emphasize tuition place the burden on the individualstudent; this burden is significantly higher if states do not provide much need-based aid. Incontrast states that fund more of higher education via appropriations take the burden of paymentoff of students and use the general tax All data were taken from the Digest ofEducational Statistics (National Center for Education Statistics 1998). Because the distributionof each of these variables is skewed, they will be subjected to a log transformation. INDEPENDENT VARIABLES POLITICAL INFLUENCEThree clusters of political forces will be examined in this research partisanship,ideology, and legislative characteristics. Politics in many cases means political parties. All12models will include both the political party of the governor (coded 1 if a Democrat and 0otherwise) and the proportion of the legislature that are Democrats (Statistical Abstract, variousyears). A priori one would expect that Democratic control would be more likely to decrease thecosts of education to students by increasing aid or by holding tuition one can argue that the Democratic Party is always more liberal than theRepublican Party in a given state, party is only a rough surrogate for ideology. To more directlycapture the influence of ideology, we include Berry et al. s (1996) measures of both governmentand citizen ideology. Government ideology attempts to measure the political ideology ofgovernment officials and thus fills in the gaps in the party measure. Citizen ideology is ameasure of mass preferences. Although the general public is unlikely to affect higher educationdecisions directly, they could influence government officials to act on their behalf, that is to actas delegates rather than trustees. Both measures are coded on a scale of 0 to 100 with 100 beingmost liberal and 0 being most conservative. The two remaining political variables are the professionalism of the state legislature anda variable for conflict in the state legislature. The conditions that these variables represent arelikely to affect the policies and implementation in any state. Scholars have found that in times ofhigh conflict in a legislature, the interest and influence of members of Congress in policies alsoincreases (Bond and Fleisher 2000). More legislation is introduced, and more attempts tooversee the implementation are made in an effort to claim responsibility for actions that might bebeneficial to one party or another. We expect that increased activity by a legislature could resultin the inability of boards to insulate the policy outputs from political influence. At the same timegreater legislative conflict could keep the legislature focused on internal battles and leave it littletime for overseeing the higher education system. Our indicator of legislative conflict is13measured by how closely the legislature is to a perfect fifty-fifty party split. Legislative professionalism is an important variable because it provides the capacity tointeract with the bureaucracy on a more equal basis. Full time legislatures who are paid moreare able to commit more time and resources to actual legislation and conduct more oversight intothe actions of higher education governing boards. The greater resources that moreprofessionalized legislatures have at their discretion allows them to overcome problems ofinformation asymmetry. According to the literature, legislatures who spend more timeinfluencing policy are more likely to be concerned with equity, and this should result indecisions that affect the distribution of costs for higher education (Lowry 2001). Althoughmulti-indicator measures of legislative professionalism exist, they do not exist on an annualbasis. Studies have found, however, that legislative salaries are good surrogates for legislativeprofessionalism and that measure will be used here (Fiorina 1994; 1999; source: Council of StateGovernments various years).INDEPENDENT VARIABLES CONTROLSBesides political variables, the model includes additional structural variables that mightaffect the implementation of policies. Several of these pertain to how a board receives its power. First, we code whether the legal basis for board actions comes directly from the state constitutionor if the powers are statutory (coded 1 if statutory, a historical justification for this variable canbe found in McLendon forthcoming). Second, a dummy variable is included that measureswhether or not the chief executive of the board is a member of the governor s cabinet; one mighthypothesize that a cabinet post would allow the governor to have greater influence on policy. The final structural variable included is whether or not a board is elected or appointed. This is adummy variable coded 1 for the four states with elected boards and 0 otherwise. Appointments14can be done by a state legislature, the governor, or another high-ranking state governmentalofficial. States that are coded elected are popularly elected by the general population. Allstructural variables are from the Educational Commission of the States (1997).In addition to controlling for structural variables, the model also controls for twoeconomic/demographic variables as well. Per capita personal income should influence the needthat a state might have for either more need-based scholarships and grants or for a reduction intuition as well as the state s ability to do so based on its income (Statistical Abstract, variousyears). This ability might also affect the overall distribution of higher education costs. The totalfall undergraduate enrollment is included in all models in case economies of scale change thecost factors (ECS 1997). All variables other than the dummy variables were logged and thus canbe interpreted as analysis is a pooled time series of 47 states from 1989 to 1996. A few data pointsare missing for a few states and thus the pool does not total exactly 376 cases. To control forany time dominance in the pool, dummy variables for individual years are included in the initial exploratory nature of this research and the two contradictory hypothesesabout how structure might interact with political forces, the discussion of the findings will focuson general patterns. The use of models with several interaction terms such as these are oftenaffected by collinearity and thus the individual regression coefficients might be unreliable. Tables 1 through 4 present the results for the total higher education costs per student, tuition perstudent, state appropriations per student, and need-based aid per student respectively. [Tables 1-4 About Here]15Before progressing to the main set of hypotheses, the general impact of coordinatingboards versus consolidating governing boards should be noted. The first regression coefficientin each table indicates how much impact a coordinating board has relative to a consolidatedgoverning board. Because the variable is a dummy variable, it can be converted to a percentagechange by using a binomial expansion (see Tufte 1974). All other things being equal, states withcoordinating boards have 40% lower total costs for higher education and their tuition costs perstudent are 52% lower (both appropriations and scholarships appear unrelated to the distinctionbetween coordinating boards and governing boards). These are substantively large differencesthat indicate coordinating boards are associated with providing relatively inexpensive education. Whether this might be a preference for low cost methods of delivery of education ( ,community colleges or four year schools versus comprehensive universities) is the subject offuture two key hypotheses are whether coordinating boards facilitate the ability of politicalfactors to influence higher education policy or not. Unfortunately, the pattern of coefficientsdoes not reveal a clear and consistent set of relationships. An optimal pattern might, forexample, show significant relationships for the coordinating board interactions and norelationship for the noninteracted relationships (or vice versa). Instead we get a general patternwhere some of the coordinating board relationships are significant and some of the governingboard relationships are significant. What is clear from the tables is that higher educationstructures do significantly affect the ability of political forces to influence higher education. Inall four cases the joint f-test shows that the sets of coefficients are significantly different fromeach other. This concisely shows that how politics affect higher education in states withconsolidated governing boards is different from how politics affects higher education in states16with coordinating boards. To illustrate the differences, the impact of legislative professionalism on tuition costswill be used (see Table 2). In states with consolidated governing boards, a one percent increasein legislative professionalism is associated with a .0437 percent decrease in tuition per student. In states with coordinating boards, however, this relationship changes dramatically; and a onepercent increase in legislative professionalism is associated with a .058 percent increase intuition costs per student (the impact in coordinating board states is the sum of the twocoefficients). Both coefficients are significantly different from zero and significantly differentfrom each widely varying pattern of coefficients as politics interacts with structure suggeststhat the relationships are highly complex. Providing an explanation for the patterns and howthose patterns should appear will require additional theoretical work. One possibility is that therelationships are even more complex than the current regressions reveal them to be. Forexample, the direction of impact of legislative professionalism might be a function of both thestructure of higher education and the ideology or partisanship of the legislature. This notionsuggests a three way or perhaps even a four way interaction of these terms. The other structural factors that are not part of the interactive hypothesis merit somediscussion. All other things being equal, elected boards are associated with 14 percent lowertuition per student and 11 percent lower state appropriations. In other words, states with electedboards favor lower educational costs. Cabinet rank also appears to matter. All other thingsbeing equal, states that provide cabinet rank for the chief education officer are associated with a3% higher total cost per student, 6% higher tuition charges, and 4% lower state appropriations. In short, such states appear to impose more of the costs of higher education on the student. 17Finally, states that provide the legal authority for their higher education board via statute areassociated with 8% greater total costs per student, 14% higher tuition, 6% lower stateappropriations, but compliment that with a 58% percent greater allocation of financial aid. CONCLUSIONStructural questions are an essential part of the political control literature. Building fromthe work of Lowry, Hearn, McLendon and colleagues, who first demonstrated that highereducation structures are associated with the distribution of policy benefits, this paper examinedwhether structures facilitate or impede political influences. Unfortunately, the theoreticalliterature does not suggest a clear directional hypothesis regarding the impact of structure. Acombination of the theories of political structures led us to two competing hypotheses regardingstructure and political influence. First, we posited that bureaucracies structured to be insulatedfrom politics might have a greater degree of autonomy and greater control over policy outcomes. Alternatively, any political institution seeking to control a bureaucracy has lower transactioncosts in a centralized bureaucracy. Our hypotheses concerning the ability of particular structures to insulate or facilitatehigher education from politics produced mixed findings. The pattern of coefficients does notreveal a clear and consistent set of relationships in terms of direction and signs. What is clearfrom this analysis is that higher education structures do significantly affect how political forcesinfluence higher education. In all four cases the joint f-test shows that the sets of coefficientsare significantly different from each other. This clearly demonstrates that political forces affecthigher education differently in states with coordinating boards than in states with consolidatedgoverning boards, even though the research does not provide an answer concerning exactly howstructure affects political findings such as these are obviously an invitation for further research, but theyalso provide insights that should be incorporated into future studies of higher education. Ourfindings suggest that structures affect how politics matters. Models of higher education policyshould include both structural variables and political factors. This insight should not besurprising to students of public organizations, where findings consistently demonstrate thatbureaucratic structures influence outputs. It is nonetheless an element that can be explored inmore depth in future studies of state-run higher education systems. Even more work remains tobe done theoretically to develop a set of more precise hypotheses. We need to know what kindsof political forces are mediated by the various structures, and how in combination these affecthigher education policy. Models will also need to reflect the possible complex patterns ofinteractions among the various structures and the myriad political forces. 19Table 1: Determinates of Logged Total Cost Per Student of Higher Education in the American States, 1989-1996Independent Variables Slope t-score Consolidating Governing Board * Political VariablesCitizen Ideology .1373 *Government Ideology *Legislative Professionalism * Democratic Governor .0394 *Democratic Legislature Percent .1142 *Legislative Conflict .1277 *Political Variables Interacted With StructureCitizen Ideology .1023 Government Ideology .0199 .49Legislative Professionalism .0895 * Democratic Governor *Democratic Legislature Percent .34Legislative Conflict .0621 .56 Elected Board .0195 Cabinet Rank .0327 *Statutory Authority .0806 *Per capita Income .4592 *Student Enrollment (millions) .14_____________________________________ __________________________________R-Squa red = .67 Standard Error = .0654 N = 372 F = f-test for political variables (6, 346) = p = .0000Coefficients for dummy variables not reported*significant p < .05 20Table 2: Determinates of Logged Total Tuition per Student of Higher Education in the American States, 1989-1996Independent Variables Slope t-score Consolidating Governing Board * Political VariablesCitizen Ideology .0208 .24 Government Ideology .23 Legislative Professionalism * Democratic Governor .0061 .34 Democratic Legislature Percent .0549 .68Legislative Conflict .1893 *Political Variables Interacted With StructureCitizen Ideology .2490 *Government Ideology .70Legislative Professionalism .1018 * Democratic Governor Democratic Legislature Percent .1259 .91Legislative Conflict .0832 .52 Elected Board *Cabinet Rank .0575 *Statutory Authority .1332 *Per capita Income .5304 *Student Enrollment (millions) .90_____________________________________ _________________________________R-Squar ed = .65 Standard Error = .0953 N = 375 F = f-test for political variables (6, 349) = p = .0000Coefficients for dummy variables not reported*significant p < .05 21Table 3: Determinates of Logged Total Appropriations Per Student of Higher Education in the American States, 1989-1996Independent Variables Slope t-score Consolidating Governing Board .2012 .94 Political VariablesCitizen Ideology *Government Ideology .1142 *Legislative Professionalism .0903 * Democratic Governor .81 Democratic Legislature Percent .1399 Conflict .48 Political Variables Interacted With StructureCitizen Ideology .2861 *Government Ideology *Legislative Professionalism * Democratic Governor .0354 Democratic Legislature Percent .24Legislative Conflict .36 Elected Board *Cabinet Rank *Statutory Authority *Per capita Income .0986 .92 Student Enrollment = .33 Standard Error = .0950 N = 376 F = f-test for political variables (6, 350) = p = .0000Coefficients for dummy variables not reported*significant p < .05 22Table 4: Determinates of Logged Scholarships/Aid Per Student of Higher Education in the American States, 1989-1996Independent Variables Slope t-score Consolidating Governing Board .66 Political VariablesCitizen Ideology *Government Ideology .5044 *Legislative Professionalism .0089 .17 Democratic Governor Democratic Legislature Percent Conflict .5052 Political Variables Interacted With StructureCitizen Ideology .0680 .17 Government Ideology .45Legislative Professionalism .2839 * Democratic Governor .0496 .51 Democratic Legislature Percent .7125 Conflict .8384 Elected Board Cabinet Rank .1085 Statutory Authority .4585 *Per capita Income .1913 .48 Student Enrollment .4178 *_______________________________________ ________________________________R-Square d = .64 Standard Error = .3565 N = 376 F = f-test for political variables (6, 350) = p = .0088Coefficients for dummy variables not reported*significant p < .05 23REFERENCESBernstein, M. (1955). Regulating business by independent commission. Princeton, :Princeton University , and R. Fleisher. (2000). Polarized politics: Congress and the president in apartisan era. Washington, : CQ of State Governments. (various years). The book of the states. Lexington: , A. (1967). Inside bureaucracy. Boston: Little , (1994). Divided government in the United States: A byproduct oflegislative professionalism? American Political Science Review 88, , (1999). Further evidence of the partisan consequences of legislativeprofessionalism. American Journal of Political Science 43, Commission of the States. (1997). 1997 state postsecondary educationstructures sourcebook: State coordinating and governing boards. Denver, CO: EducationCommission of the , M. (2000). Higher education Politics, access, and economic Behavioral Scientist 43, , and Griswold, (1994). State-level centralization and policy innovationin postsecondary education. Educational Evaluation and Policy Analysis 16, , , Griswold, , and Marine, (1996). Region, resources, and reason: Acontextual analysis of state tuition and student aid policies. Research in Higher Education 37, , H. 1990. Implementation: A review and suggested framework. In Public24administration: The state of the discipline, N. Lynn and A. Wildavsky, eds. Chatham, :Chatham House Publishers, , D., Ewell, P., and McGuinness, (1998). The challenges and opportunitiesfacing higher education: An agenda for policy research. Washington: The National Center forPublic Policy and Higher Education working , H. (1969). Administrative decentralization and political power. PublicAdministration Review 29, , and Payne, (2002). The impact of state governance structures on highereducation resources and research activities. Paper presented at the Texas A&M conference onpublic management, , (2001). Governmental structure, trustee selection, and public universityprices and spending: Multiple means to similar ends. American Journal of Political Science 45, , , Noll, , and Weingast, (1989). Structure and process, politicsand policy: Administrative arrangements and the political control of agencies. Virginia LawReview 75, , (1999). The states and higher education. In American highereducation in the twenty-first century, P. Altback, ed. American Society of Higher , (forthcoming). State governance reform of higher education: Patterns,trends, and theories of the policy process. In J. Smart, Higher education: Handbook of theoryand research. Vol. XVIII. New York: Agathon , , Heller, , and Young, (2002). State postsecondary policyinnovation: Politics, competition, and the interstate migration of policy ideas. Paper presented at25the Midwest Political Science Association meetings, , (1980). The impact of regulatory organization structure: IRCs or Review of Public Administration 3, 427-443Meier, (2000). Politics and the bureaucracy: Policymaking in the fourth branch ofgovernment. Fort Worth: Harcourt College , (1985). Control and feedback in economic regulation: The case of the Political Science Review 79, , (1990). The politics of bureaucratic structure. In Can government govern? Chubb and Peterson, eds. Washington, : The Brookings Center for Education Statistics. (1998). Digest of education , : Government Printing Office. Richardson, , Bracco, , Callan, , and Finney, (1999). Designing statehigher education systems for a new century. Phoenix: American Council on Education and , (1969). Bureaucracy, politics, and public policy. Boston: Little, Brown , H. (1980). Politics, position, and power: The dynamics of federal York: Oxford University , P. (1948). Foundation of the theory of organization. American SociologicalReview 13, 25-35. Statistical abstract of the United States. (various years). Washington: Bureau of , (1967). Organizations in action. New York: McGraw-Hill. 26Tufte, (1974). Data analysis for politics and policy. Englewood Cliffs, NJ:Prentice , (1986). Campus autonomy and its relationship to measures of universityquality. Journal of Higher Education 57, , (1989). Changes in quality among public universities. Journal of HigherEducation 60, , (1997). Searching for a theory of bureaucratic structure. Journal of PublicAdministration Research and Theory 4, , (1989). Bureaucracy: What government agencies do and why they do York: Basic , and Waterman, (1991). The dynamics of political control of thebureaucracy. American Political Science Review 85, , J. (1965). Industrial organization. London: Oxford University , W. (1996). Meeting the demand for higher education without breaking thebank. Journal of Higher Education 67, , W. (1992). State policies and higher education: Policies, correlates, andlinkages. Journal of Higher Education 63, StatementsKenneth J. Meier is the Charles Puryear Professor of Liberal Arts in the Department of PoliticalScience and holds the Sara Lindsey Chair in the Bush School of Government at Texas A&MUniversity. His work on education policy generally combines the themes of politics, equity, andrepresentation. His research agenda in education includes a new national study of race, ethnicityand education in 1800 school districts and an on-going project on how school systems can bemanaged Nicholson-Crotty is a PhD candidate in political science at Texas A&M University andresearch director of the Texas Educational Excellence Project. Her research interests are publicadministration and public policy with a special emphasis on gender, representation, andmanagement. She is currently working on projects involving gender and the management ofschool districts, police and the enforcement of sexual assault laws, and the politics of publichealth are a wide variety of other structural issues such as whether the board hasstatutory or constitutional powers, whether it is part of the governor's cabinet, and thedegree of control vested in the governor versus other elected officials. On the last issuesee Knott and Payne (2002). Whether or not structure actually translates into autonomyis an open question. A series of studies by Volkwein (1986; 1989, and referencestherein) seek to measure autonomy and find it is not closely linked to structural other major sources of funds for higher education are research grants and privatecontributions. ENDNOTES

Related search results