View in EDS HTML Full Text PDF Full Text

Partisanship, White Racial Resentment, and State Support for Higher Education

Saved in:
Bibliographic Details
Title: Partisanship, White Racial Resentment, and State Support for Higher Education
Language: English
Authors: Taylor, Barrett J., Cantwell, Brendan, Watts, Kimberly, Wood, Olivia
Source: Journal of Higher Education. 2020 91(6):858-887.
Availability: Routledge. Available from: Taylor & Francis, Ltd. 530 Walnut Street Suite 850, Philadelphia, PA 19106. Tel: 800-354-1420; Tel: 215-625-8900; Fax: 215-207-0050; Web site: http://www.tandf.co.uk/journals
Peer Reviewed: Y
Page Count: 30
Publication Date: 2020
Document Type: Journal Articles
Reports - Research
Education Level: Higher Education
Postsecondary Education
Descriptors: Politics of Education, Higher Education, Racial Bias, Racial Attitudes, Whites, Educational Policy, State Policy, Disproportionate Representation, State Government, Educational Finance, State Aid, Political Attitudes
DOI: 10.1080/00221546.2019.1706016
ISSN: 0022-1546
Abstract: Dominant explanations of state higher education policy tend to emphasize economic models that foreground the business cycle or political approaches that cast ideology as fairly fixed. We instead foreground changing social context to conceptualize state appropriations as predicted not only by these classic explanations, but also by the interplay of racial representation and political party control. Drawing on the racial backlash hypothesis and quantitative analyses, we show that party control of state government and racial representation in higher education jointly explain state appropriations. Unified Republican governments spent more than Democratic or divided governments when White students were overrepresented. Republicans spent less otherwise. These results suggest that partisan attitudes toward racial representation in higher education may shape state government support for colleges and universities.
Abstractor: As Provided
Entry Date: 2020
Accession Number: EJ1264420
Database: ERIC
Full text is not displayed to guests.
FullText Links:
  – Type: pdflink
    Url: https://content.ebscohost.com/cds/retrieve?content=AQICAHj0k_4E0hTGH8RJwT4gCJyBsGNe_WN95AvKlDbXJGqwxwHU3BEITz1VRQXzJP5BEZ8kAAAA4zCB4AYJKoZIhvcNAQcGoIHSMIHPAgEAMIHJBgkqhkiG9w0BBwEwHgYJYIZIAWUDBAEuMBEEDDdxHVmpHtLKmrLUlwIBEICBmzKlvT24J4a0cFvGEXw1WZhdLSlgi2ne7bHzKzlgvyaCEOQj2JRULnBHlegvf-Pn08cEYNNCJkZX0nl8e8CFGElDdJ8tW8_XQRGIp2ibhWFJhML2k_casnyLxexMdcNsUffPKrMRs7TASnpr7VpLgE484x8loOcTvdDX64RoBiaRMlSlCWUleKDUEMVaaTIVRPfmPoeEWqFrocmz
Text:
  Availability: 1
  Value: <anid>AN0145086186;jhe01sep.20;2020Aug14.07:46;v2.2.500</anid> <title id="AN0145086186-1">Partisanship, White Racial Resentment, and State Support for Higher Education </title> <p>Dominant explanations of state higher education policy tend to emphasize economic models that foreground the business cycle or political approaches that cast ideology as fairly fixed. We instead foreground changing social context to conceptualize state appropriations as predicted not only by these classic explanations, but also by the interplay of racial representation and political party control. Drawing on the racial backlash hypothesis and quantitative analyses, we show that party control of state government and racial representation in higher education jointly explain state appropriations. Unified Republican governments spent more than Democratic or divided governments when White students were overrepresented. Republicans spent less otherwise. These results suggest that partisan attitudes toward racial representation in higher education may shape state government support for colleges and universities.</p> <p>Keywords: Politics of higher education; political partisanship; racial representation in higher education; White racial resentment; state appropriations for higher education</p> <p>Public trust in higher education has dropped sharply (Gallup, [<reflink idref="bib20" id="ref1">20</reflink>]; Pew Research Center, [<reflink idref="bib61" id="ref2">61</reflink>]) and state appropriations have fallen apace (State Higher Education Executive Officers Association [SHEEO], [<reflink idref="bib73" id="ref3">73</reflink>]). At the same time, enrollments in public four-year colleges and universities are larger than ever before (National Center for Education Statistics [NCES], [<reflink idref="bib56" id="ref4">56</reflink>], Table 303.70). Members of the public appear both to distrust higher education and to want a higher education for themselves or their children. This study examines state funding for public higher education within this context of ebbing public trust and rising demand.</p> <p>Two broad social conditions undergird declining public trust. The first is hyper-partisanship and its influence on higher education policy (Mettler, [<reflink idref="bib51" id="ref5">51</reflink>]). As voters and policymakers become more deeply committed to their political party (Mason, [<reflink idref="bib46" id="ref6">46</reflink>])—a phenomenon that is especially acute among Republicans (Grossman & Hopkins, [<reflink idref="bib25" id="ref7">25</reflink>])—compromise becomes more difficult to forge. Political decisions that were once made locally (e.g., funding for higher education) are often cast as referenda on national political figures and parties (Hopkins, [<reflink idref="bib35" id="ref8">35</reflink>]). With an array of issues viewed through the lens of party affiliation (Mason, [<reflink idref="bib46" id="ref9">46</reflink>]), confidence in higher education has declined far more among Republicans or Republican-leaning voters than among other Americans (Gallup, [<reflink idref="bib20" id="ref10">20</reflink>]; Pew Research Center, [<reflink idref="bib61" id="ref11">61</reflink>]).</p> <p>Second, partisanship is sorted by identity rather than ideology (Grossman & Hopkins, [<reflink idref="bib25" id="ref12">25</reflink>]; Lupton, Myers, & Thornton, [<reflink idref="bib44" id="ref13">44</reflink>]). Whites identify as Republicans (or lean Republican) over Democrats by a margin of about eight percentage points, while the overwhelming majority of Asian, Black, and Latinx Americans identify (or lean) Democratic (Pew Research Center, [<reflink idref="bib63" id="ref14">63</reflink>]). The backlash against demographic and cultural changes seems to unify a Republican Party whose membership is overwhelmingly White (Abramowitz, [<reflink idref="bib1" id="ref15">1</reflink>]; Barber & Pope, [<reflink idref="bib8" id="ref16">8</reflink>]; Cramer, [<reflink idref="bib14" id="ref17">14</reflink>]; Mason, [<reflink idref="bib45" id="ref18">45</reflink>]; Mutz, [<reflink idref="bib55" id="ref19">55</reflink>]). As the political scientists Sides, Tesler, and Vavreck ([<reflink idref="bib71" id="ref20">71</reflink>]) put it, "Trump's positions on immigration, Confederate monuments, and national anthem protests have proved more popular with Republican voters than have the GOP tax bill or Republican alternatives to the Affordable Care Act" (p. 214).</p> <p>In the context of these two trends—deepened partisanship and political parties that are sorted by identity—confidence in higher education may vary across social settings. Republican officials may be more skeptical of higher education funding when the presumed beneficiaries of government spending are racially diverse, and more sympathetic to government expenditures when the presumed beneficiaries are White. Anecdotal evidence from states such as Kansas and Wyoming — which featured strong Republican leanings, majority White populations and relatively robust government funding for higher education through 2017—supports this account. Research on voting patterns (Brunner & Johnson, [<reflink idref="bib11" id="ref21">11</reflink>]) and the distribution of government funding to institutions (Hill & Jones, [<reflink idref="bib31" id="ref22">31</reflink>]) also resonates with our understanding that deepening partisan commitments and the growing role of White racial resentment among Republicans are jointly associated with state appropriations for higher education.</p> <hd id="AN0145086186-2">Higher education funding and racial backlash</hd> <p>Predicting direct state support for higher education is a complex undertaking (McLendon, Tandberg, & Hillman, [<reflink idref="bib49" id="ref23">49</reflink>]; Tandberg, [<reflink idref="bib74" id="ref24">74</reflink>], [<reflink idref="bib75" id="ref25">75</reflink>]). Many explanations adopt an applied neoclassical economic approach that relies on rational choice models and examines marginal changes created by policy interventions in static situations (e.g., Paulsen & Toutkoushian, [<reflink idref="bib60" id="ref26">60</reflink>]). Perhaps the most powerful explanation for state budgeting in this tradition is the "balance wheel," which emphasizes state economic conditions (Delaney & Doyle, [<reflink idref="bib15" id="ref27">15</reflink>]; Hovey, [<reflink idref="bib36" id="ref28">36</reflink>]). When the economy is down, states' budgets contract. The brunt of cuts have fallen on higher education because, unlike other state agencies, colleges and universities can generate revenue by raising tuition prices. Cuts to higher education funding under these circumstances do not indicate declining confidence in the enterprise. Rather, according to the balance wheel model, declining state appropriations reflect a calculated attempt to navigate fiscal realities.</p> <p>A second major approach to understanding higher education appropriations emphasizes state political conditions. Research by McLendon, Hearn, and their colleagues (e.g., McLendon, Deaton, & Hearn, [<reflink idref="bib47" id="ref29">47</reflink>]; McLendon, Hearn, & Mokher, [<reflink idref="bib48" id="ref30">48</reflink>]) considers political factors such as partisan control, legislative professionalism, and higher education governance structure. Assuming that policy choices are associated with underlying ideology on a liberal/Democratic to conservative/Republican spectrum, these models assert that politics matter, but tend to accept political parties as entities that are broadly stable across states and time periods.</p> <p>Tandberg ([<reflink idref="bib74" id="ref31">74</reflink>], [<reflink idref="bib75" id="ref32">75</reflink>]) synthesized insights from both of these accounts by proposing that states' social and demographic conditions, along with economic and political factors, influence policy choices. Gándara, Ness, and colleagues adapted Tandberg's insights to conceptualize the ways in which policies are socially conditioned by non-governmental policy actors (Gándara, [<reflink idref="bib21" id="ref33">21</reflink>], [<reflink idref="bib22" id="ref34">22</reflink>]; Gándara, Rippner, & Ness, [<reflink idref="bib23" id="ref35">23</reflink>]; Ness & Gándara, [<reflink idref="bib58" id="ref36">58</reflink>]; Ness, Tandberg, & McLendon, [<reflink idref="bib57" id="ref37">57</reflink>]). Foster and Fowles ([<reflink idref="bib19" id="ref38">19</reflink>]) explicitly explored the racial composition of a state's population, finding that declining White majorities within states predicted declining funding for higher education. This result implied that racism shapes higher education policy but did not attend to the complex dynamics of a highly partisan environment that was sorted on race.</p> <p>We build upon prior analyses by using critical political theories to conceptualize the relationship between partisan control, racial representation, and state support for higher education. We do not dismiss the explanatory power of rational choice economic and political models. Evidence supports these accounts, and we include variables that correspond to their tenets in our quantitative analyses. We seek to expand, refine, and complement these models by identifying social contexts in which different explanations hold.</p> <p>Racism and White privilege characterize American higher education (Harper, [<reflink idref="bib28" id="ref39">28</reflink>]; Ladson-Billings, [<reflink idref="bib41" id="ref40">41</reflink>]). These forces also shape policymaking, with the result that policies that do not directly benefit White voters may be met with skepticism from state lawmakers (Jardina, [<reflink idref="bib38" id="ref41">38</reflink>]). For example, following federal welfare reform in 1996, the share of recipients who were Black was a powerful predictor for the imposition of "tough" restrictions on eligibility for income assistance within the states (Soss, Schram, Vartanian, & O'Brien, [<reflink idref="bib72" id="ref42">72</reflink>]). Similar dynamics can be seen in higher education, where race-based affirmative action policies have encountered political resistance (Pusser, [<reflink idref="bib64" id="ref43">64</reflink>]).</p> <p>In order to conceptualize these conditions, we draw from Grogan and Park's ([<reflink idref="bib24" id="ref44">24</reflink>]) racial backlash model and Baker's ([<reflink idref="bib6" id="ref45">6</reflink>]) application of racial threat theory. The racial backlash model assumes that party ideology interacts with White resentment to produce policy outcomes. Grogan and Park built the model through a study of state Medicaid expansion. Following the passage of the Affordable Care Act, states had the option to expand Medicaid eligibility backed by federal financial support. The expansion was more likely when the state's population was predominantly White, and less likely when Black Americans accounted for a substantial share of state residents. Similar patterns characterize the passage of voter ID laws (Rocha & Matsubayashi, [<reflink idref="bib67" id="ref46">67</reflink>]).</p> <p>Baker ([<reflink idref="bib6" id="ref47">6</reflink>]) integrated racial threat theory with established models for analyzing policy diffusion to explain state-level affirmative action bans. A sense of racial threat activated by fear of competition for resources—in this case, access to desirable universities—could prompt retaliatory policy. In this way, a ban on affirmative action could be a means of ensuring that White students retain privileged access to selective public institutions. Results from Baker's study implied that state higher education policy might be associated with shifting underlying social conditions related to racial resentment.</p> <hd id="AN0145086186-3">Policymaking, partisanship, and White racial resentment</hd> <p>Recent evidence questions the universal applicability of rational choice explanations based on economic conditions and/or fixed ideologies. Baker's ([<reflink idref="bib6" id="ref48">6</reflink>]) study of affirmative action bans is one compelling example of research that complicates established theories of policy adoption. Another example is Li's ([<reflink idref="bib43" id="ref49">43</reflink>]) study of performance funding metrics, which reveal that, even when equity metrics are in place, policies that emphasize output accountability can heighten inequality of access and success (see also Hillman & Crespín-Trujillo, [<reflink idref="bib32" id="ref50">32</reflink>]). The adoption of these policies and others seems to reflect the interests of dominant social groups and powerful organizations, not just efforts to improve higher education performance (Baker, [<reflink idref="bib6" id="ref51">6</reflink>]; Gándara, [<reflink idref="bib21" id="ref52">21</reflink>], [<reflink idref="bib22" id="ref53">22</reflink>]; Gándara et al., [<reflink idref="bib23" id="ref54">23</reflink>]; Hertel-Fernandez, [<reflink idref="bib30" id="ref55">30</reflink>]; Miller & Morphew, [<reflink idref="bib52" id="ref56">52</reflink>]). It is therefore crucial to understand the role of power and context in shaping policy decisions. We highlight two dimensions of a changing social context—deepening political partisanship and White racial resentment—that are likely to shape direct state funding for higher education.</p> <hd id="AN0145086186-4">Deepening partisanship</hd> <p>We use the term <emph>partisanship</emph> rather than <emph>polarization</emph>. Partisanship indicates the sorting of political actors into "teams" and intense devotion to victory. Polarization, by contrast, indicates ideological sorting based on idealized models of how government ought to work. Where partisanship is a social process that often involves other forms of identity (e.g., race, gender, sexual orientation, religion), polarization is a cognitive process reflecting ideological cohesion. The two major US political parties are ideologically heterogeneous, including a wide range of attitudes about government spending and policy priorities. Especially for Republicans, however, partisan identification and demographic homogeneity are high (Barber & Pope, [<reflink idref="bib7" id="ref57">7</reflink>]; Lupton et al., [<reflink idref="bib44" id="ref58">44</reflink>]; Mason, [<reflink idref="bib45" id="ref59">45</reflink>], [<reflink idref="bib46" id="ref60">46</reflink>]; Pew Research Center, [<reflink idref="bib63" id="ref61">63</reflink>]).</p> <p>Grossman and Hopkins ([<reflink idref="bib25" id="ref62">25</reflink>]) show that the Republican Party coheres around the ideas of solidarity and victory while the Democratic Party is a coalition tied together by a commitment to delivering policies in support of diverse interests. The core of the Republican Party seems to consist of the Party itself, not of a particular agenda for governing. In such a context, negative partisanship—an opponent against whom to define who belongs to the party and who is excluded—becomes particularly important (Abramowitz, [<reflink idref="bib1" id="ref63">1</reflink>]; Abramowitz & Webster, [<reflink idref="bib3" id="ref64">3</reflink>]).</p> <p>Higher education is a likely target for negative partisan sentiment. Contemporary Republicans have low levels of trust in higher education (Johnson & Peifer, [<reflink idref="bib39" id="ref65">39</reflink>]) and often view it as removed from the realities of everyday life (Cramer, [<reflink idref="bib14" id="ref66">14</reflink>]). Large numbers of Americans define their identity in terms of party affiliation (Mason, [<reflink idref="bib45" id="ref67">45</reflink>], [<reflink idref="bib46" id="ref68">46</reflink>]), including a growing number of "party loyalists" whose identification with the GOP is virtually unshakable (Barber & Pope, [<reflink idref="bib7" id="ref69">7</reflink>]). Various Republican factions emphasize different issues (e.g., abortion, immigration, tax rates), but all—with the possible exception of intellectually conservative ideologues (Barber & Pope, [<reflink idref="bib8" id="ref70">8</reflink>])—tend to remain loyal to the party itself (Lupton et al., [<reflink idref="bib44" id="ref71">44</reflink>]). In such a context, it is difficult to imagine that motivation to support higher education would outweigh partisan commitment.</p> <p>Negative partisanship makes most political contests seem national rather than local or regional (Abramowitz & Webster, [<reflink idref="bib2" id="ref72">2</reflink>]; Hopkins, [<reflink idref="bib35" id="ref73">35</reflink>]). It is therefore difficult to imagine that state-level Republicans will abandon the suspicion of higher education that is widespread in their party (Johnson & Peifer, [<reflink idref="bib39" id="ref74">39</reflink>]). To use Miller and Morphew's ([<reflink idref="bib52" id="ref75">52</reflink>]) terminology, partisanship is a powerful narrative "frame" that guides policymaking. For Republicans, that frame appears to include declining trust in higher education and a willingness to use nonpartisan institutions such as higher education to marshal negative partisan sentiments.</p> <hd id="AN0145086186-5">White racial resentment</hd> <p>Republicans are more likely to identify as White, male and Christian than is the country as a whole (Pew Research Center, [<reflink idref="bib61" id="ref76">61</reflink>]). These distinctive demographics are reinforced by a tendency to socialize with other Republicans (Mason, [<reflink idref="bib46" id="ref77">46</reflink>]) and to consume partisan media (Bail et al., [<reflink idref="bib5" id="ref78">5</reflink>]; Benkler, Faris, & Roberts, [<reflink idref="bib9" id="ref79">9</reflink>]). The resulting coalition foregrounded race, immigration and other cultural political issues as the core of the Republican Party (Hooghe & Dassonneville, [<reflink idref="bib34" id="ref80">34</reflink>]). Although media accounts often emphasized the role of "economic anxiety" in the 2016 Presidential election, social scientific analyses (e.g., Mutz, [<reflink idref="bib55" id="ref81">55</reflink>]; Schaffner, MacWilliams, & Nteta, [<reflink idref="bib69" id="ref82">69</reflink>]; Sides, [<reflink idref="bib70" id="ref83">70</reflink>]) instead emphasized the role of racial resentment among White voters.[<reflink idref="bib1" id="ref84">1</reflink>]</p> <p>The alignment of White voters with the Republican Party was not new in 2016 (Sides, [<reflink idref="bib70" id="ref85">70</reflink>]). Since the 1970s there has been not only "a growing gap between the two electoral coalitions but [also] a dramatic increase in racial resentment among white Republican voters" (Abramowitz, [<reflink idref="bib1" id="ref86">1</reflink>], p. 126). Cramer ([<reflink idref="bib14" id="ref87">14</reflink>]) demonstrated that White resentment often extended to anger at higher education. Unsurprisingly, higher education policies and practices that explicitly aim to redress racial inequality—from race-based affirmative action (Baker, [<reflink idref="bib6" id="ref88">6</reflink>]; Douglass, [<reflink idref="bib16" id="ref89">16</reflink>]; Pusser, [<reflink idref="bib64" id="ref90">64</reflink>]) to Black studies programs (Rojas, [<reflink idref="bib68" id="ref91">68</reflink>])—routinely generated backlash from a political Right that found higher education to be "too liberal/political" (Gallup, [<reflink idref="bib20" id="ref92">20</reflink>]) and inundated with (Leftist) politics (Pew Research Center, [<reflink idref="bib62" id="ref93">62</reflink>]).</p> <hd id="AN0145086186-6">Synthesizing partisanship, White racial resentment, and the racial backlash model</hd> <p>Our arguments are about the overlap of partisanship, White racial resentment, and higher education funding in certain contexts, not about the fixed character of higher education or the two major political parties. We do <emph>not</emph> claim that higher education is free of racism. Higher education in America was established in a context of slave labor and colonial expansion (Wilder, [<reflink idref="bib78" id="ref94">78</reflink>]), and Black students continue to be profoundly underserved in virtually every state in the US (Harper & Simmons, [<reflink idref="bib29" id="ref95">29</reflink>]). We also do not suggest that Democrats and independents are anti-racist (Ostfield, [<reflink idref="bib59" id="ref96">59</reflink>]).</p> <p>We argue that higher education is a potential target for negative partisanship by Republicans because any step toward greater racial equality stirs White racial resentment. Just as the racial backlash model explained which states adopted Medicaid expansion (Grogan & Park, [<reflink idref="bib24" id="ref97">24</reflink>]), we posit that the politics of White racial resentment may explain variation in state appropriations for higher education. Just as affirmative action bans became more likely as admission to a flagship university became more competitive (Baker, [<reflink idref="bib6" id="ref98">6</reflink>]), we expect this relationship to be more intense in certain social contexts. We expect Republican mistrust of higher education to be blunted when the beneficiaries of higher education spending are more likely to be White, and to be intensified when enrollments become more racially diverse.</p> <p>While elections are categorical, social change is continuous. Because we are focused on contexts rather than on covering laws, meaning time- and context-invariant social processes, we expect a change in White overrepresentation and state funding for higher education to occur simultaneously as conditions on the ground change. As Mettler ([<reflink idref="bib50" id="ref99">50</reflink>]) documented, the Tea Party revolt that followed the election of President Barack Obama only appeared to be sudden. Its roots stretched over years of White voters' frustration. Accordingly, by the time a unified Republican government has been seated, we expect that a state's context already will have changed dramatically. This conceptualization is consistent with Hertel-Fernandez's ([<reflink idref="bib30" id="ref100">30</reflink>]) recent work on state policymaking, which employs a "problem-driven" methodology to address pressing real-world concerns that cannot easily be addressed with formal social scientific modeling. White overrepresentation cannot be randomly assigned but is an urgent policy problem that demands examination. This state of affairs, along with methodological reasons laid out below, explains why we eschew any causal claims. Instead, we suggest that the social conditions that underlie declining trust and funding for higher education simultaneously yielded unified Republican control of many state governments.</p> <hd id="AN0145086186-7">Methods</hd> <p></p> <hd id="AN0145086186-8">Data and sample</hd> <p>The sample for this study is US states over the period 2006 to 2015. We compiled data from several different sources to facilitate our analysis. Data on state higher education systems were aggregated from the Delta Cost Project. These data are unsuitable for institution-level analysis of public higher education because of "parent-child" reporting (Jaquette & Parra, [<reflink idref="bib37" id="ref101">37</reflink>]). When used carefully, they are appropriate for state-level analysis. Additional data were accessed from the American Community Survey (ACS) of the US Census Bureau, the Bureau of Labor Statistics (BLS) and the National Conference of State Legislatures (NCSL).</p> <p>We observed states from 2006–2015 because ACS data were not available prior to 2006. We controlled all finance figures for inflation using the BLS' Consumer Price Index. Finally, as is common for state-level analyses that focus on the politics of higher education, we omitted Nebraska due to its nonpartisan legislature (e.g., McLendon et al., [<reflink idref="bib49" id="ref102">49</reflink>]). This meant that our final sample included 49 states.</p> <hd id="AN0145086186-9">Analytic strategy</hd> <p>Consistent with recent work in political science (e.g., Hertel-Fernandez, [<reflink idref="bib30" id="ref103">30</reflink>]), we sought to understand the social process that cannot be modeled experimentally. Our goal was not to isolate a cause and its effects, but to use a robust technique to highlight the conditions under which independent variables and the dependent variable were related to one another. We sought to do this with as much rigor as possible. "Difference-in-differences" (DID) techniques have proven powerful tools for researchers interested in specifying social processes (Murnane & Willett, [<reflink idref="bib54" id="ref104">54</reflink>]). However, our analysis does not feature a "treatment" that would allow us to calculate the first difference as state behavior before and after policy adoption. Further, our study highlighted continuous and interrelated processes rather than discontinuities. Leveraging the arithmetic insights of DID techniques nonetheless allowed us to strengthen inferences about the social contexts in which higher education policymaking occurs by reducing unobserved heterogeneity in data. We therefore focused on changes in variables rather than observed values, and centered these changes on the average change across all sampled states in a given year.</p> <p>We calculated a first difference by comparing a state to itself from the year prior. This ensured that our analysis focused on changes. Rather than the variable <emph>x<subs>it</subs></emph>, then, we analyzed the difference between <emph>x<subs>it</subs></emph> and <emph>x<subs>i(t-1)</subs></emph>. We then calculated a second difference by subtracting the sample-wide average of the first difference for all states in a given year <emph>t</emph> (i.e., we centered all differences for year <emph>t</emph> on zero). Unless otherwise noted, all variables were twice differenced in this way. This approach aligned with recent empirical work (Li, [<reflink idref="bib42" id="ref105">42</reflink>]) showing that volatility in state funding is patterned rather than randomly distributed across states.</p> <p>Our dependent variable was the annual change in state appropriations per full-time equivalent (FTE) student, relative to the average annual change for all states in that year. Our conceptual model, drawing guidance from Grogan and Park ([<reflink idref="bib24" id="ref106">24</reflink>]) and Baker ([<reflink idref="bib6" id="ref107">6</reflink>]), explicitly linked policy decisions to the individuals who enrolled in higher education. Rewarding some individuals while punishing others is likely in a social environment characterized by heightened and racialized partisanship. In other words, if policymakers used appropriations to signal resentment, this behavior would be most evident on a per-student basis. We consider alternative dependent variables under <emph>assumption checks</emph>.</p> <p>Our primary independent variable was the change in White overrepresentation in higher education, relative to the average change for all states in that year. Our secondary independent variable was Republican control of state government. Given Li's ([<reflink idref="bib42" id="ref108">42</reflink>]) findings that unified governments (i.e., those in which a single party controls both the state legislature and the governorship) demonstrated the most volatility, we adopted a narrow definition of Republican government as unified control of both branches. We did not differentiate this variable because we assumed that current control of government interacted with changing demographics (i.e., that political control was contingent upon social change).</p> <p>In keeping with our conceptual model, we estimated a second regression in which we interacted Republican control with White overrepresentation (twice differenced, as outlined above). This interaction term assessed the non-independence of political control and racial representation. We then tested for the joint significance of these three variables to assess the explanatory power of our theoretical model (Brambor, Clark, & Golder, [<reflink idref="bib10" id="ref109">10</reflink>]). In much the same way that interacting a discontinuous "treatment" with years since the treatment occurred can be interpreted to indicate the effects of treatment (e.g., Hillman, Fryar, & Crespín-Trujillo, [<reflink idref="bib33" id="ref110">33</reflink>]), we interpret this interaction term as the estimated effect of unified Republican response to White resentment of higher education. To ensure that we interpreted the interaction term appropriately (Hainmueller, Mummolo, & Xu, [<reflink idref="bib27" id="ref111">27</reflink>]), we provide extensive robustness checks to scope and temper our claims.</p> <p>In addition to our three regressors of interest, we included several control characteristics. Total population, per capita income, and unemployment rate accounted for balance wheel-type factors that might predict the ability to support higher education. Total FTE enrollment in a state and the percentage of public institutions that were research universities controlled for variation in a state's system composition over time. States with growing enrollments might have lower appropriations due to economies of scale. By contrast, those with a growing percentage of research universities might have faster increases in appropriations per students because these institutions are expensive to operate (College Board, [<reflink idref="bib13" id="ref112">13</reflink>]). The model also controlled for change in the number of public four-year institutions in the state. This variable was included for two reasons. First, states with a growing number of institutions may be less likely to reduce higher education appropriations because colleges and universities, like other interest groups, may pressure state legislatures (Gándara, [<reflink idref="bib22" id="ref113">22</reflink>]). Second, if the number of institutions changed over the study period, we would expect appropriations per FTE to vary due to the fixed costs inherent in operating multiple campuses. These control variables were also twice differenced to account for heterogeneity across states and over time.</p> <p>We included several adjustments to our model to address the panel nature of our data. We included state-level fixed effects to account for unobserved variation between states. We also included fixed year effects to account for historical time. Finally, we clustered standard errors by state to adjust for serial correlation.</p> <hd id="AN0145086186-10">Findings</hd> <p></p> <hd id="AN0145086186-11">Descriptive analysis</hd> <p>Table 1 reports descriptive statistics on observed (i.e., non-differenced) values of variables of interest. These figures are useful for interpreting the practical significance of relationships discovered in regressions. Table 1 also suggests trends over time. Average state appropriations per FTE were much lower in the Recessionary environment of 2011 and at the end of the sample in 2015 than they had been in 2007. Despite drops in state funding, total FTE enrollments grew and became more racially diverse. White overrepresentation declined from −3.24% to −6.44% in the average state. The number of unified Republican governments more than doubled over the same time period. These descriptive characteristics indicate that the conditions we seek to understand were present in our sample. On average, states funded higher education at a lower level as enrollments became more racially diverse and Republicans controlled more governments.</p> <p>Table 1. Sample description.</p> <p> <ephtml> <table><thead><tr><td>Variables</td><td>Grand mean (standard deviation)</td><td>2007 Mean (2007 standard deviation)</td><td>2011 Mean (2011 standard deviation)</td><td>2015 Mean (2015 standard deviation)</td></tr></thead><tbody><tr><td>State appropriations per FTE</td><td>$8,318.11</td><td>$9,502.19</td><td>$7,857.37</td><td>$7,687.19</td></tr><tr><td /><td>(3,553.02)</td><td>(3,383.87)</td><td>(3,337.11)</td><td>(3,756.3)</td></tr><tr><td>White overrepresentation</td><td>−4.46%</td><td>−3.24%</td><td>−4.36</td><td>−6.44%</td></tr><tr><td /><td>(4.6)</td><td>(4.59)</td><td>(4.43)</td><td>(4.54)</td></tr><tr><td>Percentage of states with unified Republican control of state government</td><td>29.2%</td><td>20%</td><td>18%</td><td>46%</td></tr><tr><td /><td>(45.5)</td><td>(40.4)</td><td>(38.8)</td><td>(50.3)</td></tr><tr><td>State population</td><td>6,278,893</td><td>6,099,456</td><td>6,310,216</td><td>6,496.581</td></tr><tr><td /><td>(6,881,186)</td><td>(6,697,604)</td><td>(6,975,440)</td><td>(7,253.984)</td></tr><tr><td>Per capita income</td><td>$44,340.17</td><td>$44,541.21</td><td>$44,224.63</td><td>$47,001.33</td></tr><tr><td /><td>(7,114.74)</td><td>(6,897.2)</td><td>(7,178.97)</td><td>(7,443.15)</td></tr><tr><td>Unemployment rate</td><td>6.5%</td><td>4.37%</td><td>8.2%</td><td>5.05%</td></tr><tr><td /><td>(2.2)</td><td>(1.0)</td><td>(1.87)</td><td>(1.03)</td></tr><tr><td>State's four-year FTE enrollment</td><td>126,082.9</td><td>117,434.1</td><td>129,879.0</td><td>133,415.2</td></tr><tr><td /><td>(117,222.2)</td><td>(108,914.6)</td><td>(119,616.0)</td><td>(128,495.1)</td></tr><tr><td>Count of public institutions</td><td>10.3</td><td>10.3</td><td>10.3</td><td>10.2</td></tr><tr><td /><td>(7.5)</td><td>(7.6)</td><td>(7.6)</td><td>(7.45)</td></tr><tr><td>Pct of public institutions that are research universities</td><td>39.3%</td><td>39.3%</td><td>39.3%</td><td>38.5%</td></tr><tr><td /><td>(24.1)</td><td>(24.5)</td><td>(24.5)</td><td>(22.9)</td></tr><tr><td>Observations</td><td>490</td><td>49</td><td>49</td><td>49</td></tr><tr><td>Number of states</td><td>49</td><td>49</td><td>49</td><td>49</td></tr></tbody></table> </ephtml> </p> <p>Another way to consider trends over time is to consider average changes in a given year, which indicate periods in which the higher education environment was particularly volatile. Accordingly, Figure 1 reports <emph>changes</emph> in state appropriations per FTE (in constant dollars) and White overrepresentation. These two measures tended to move in tandem over time. There were sizable drops in state appropriations during the Great Recession. There were small reductions or even increases in White overrepresentation during this period, suggesting the racial inequities of growing tuition reliance among public institutions (Taylor & Cantwell, [<reflink idref="bib76" id="ref114">76</reflink>]). As appropriations stabilized from 2012–2015, the average state saw negative changes (reductions) in White overrepresentation.</p> <p>PHOTO (COLOR): Figure 1. Changes over time in White overrepresentation and state appropriations per FTE.</p> <p>Although appropriations recovered—as indicated by positive changes in Figure 1 after 2012—they never returned to pre-Recessionary levels (Table 1). This was expected given that appropriations often function as the "balance wheel" of state budgets (Doyle & Delaney, [<reflink idref="bib17" id="ref115">17</reflink>]). The pattern of state funding over time is illustrated in Figure 2, which disaggregates state appropriations per student by partisan control of state government. The solid line shows average appropriations under unified Republican control and the dashed line reports average appropriations for all other forms of government. These two lines generally move in tandem, with each—as expected—declining around the Recession. However, there were important nominal differences in the two trends. From 2007–2013, states with unified Republican governments provided lower average appropriations than did all other states. This was expected given the historical preference of the GOP for lower levels of government spending. In the last two years of our sample, however, this pattern was reversed. States with unified Republican governments funded higher education at nominally higher levels than did other states.</p> <p>PHOTO (COLOR): Figure 2. Change in state appropriations per student (by unified Republican control) and count of states with unified Republican control.</p> <p>As indicated by the dotted line of Figure 2 (indexed to the right-hand axis), far more states had unified Republican governments in 2015 than had been the case even a few years before. As more states entered the unified Republican group, average levels of funding in that category changed. States with unified Republican governments also differed demographically from other states. As implied in Table 1, White students were overrepresented in fairly few states—only 12 states in 2007, 7 in 2011, and 2 in 2015. However, White students were far more likely to be represented at close to population-proportional levels in states with unified Republican control. The average state under unified Republican control had a level of White overrepresentation (−2.78) almost half as negative as that of all other states (−5.11). This finding supported our conceptual model, in which White overrepresentation and political control were social processes that occurred simultaneously rather than transactions that occurred sequentially.</p> <hd id="AN0145086186-12">Regression analyses</hd> <p>Regression results are reported in Table 2. The first column reports the results of "main effects" regression without the interaction term. In this analysis, the White overrepresentation variable (the change in White overrepresentation relative to the change in the average state) does not predict the dependent variable. By contrast, unified Republican control of state government did predict variation in the dependent variable. This result supports the salience of theories that identify enduring preferences among political parties. Net of other factors, Republican governments tended to reduce appropriations for higher education by about $220 more per FTE than did other state governments (i.e., the cut was $220 per FTE deeper than the average change). This is a substantively meaningful relationship that is approximately equal (in absolute value) to the largest average annual change observed in Figure 1.</p> <p>Table 2. FE regression results predicting state appropriations per FTE.</p> <p> <ephtml> <table><thead><tr><td /><td>(1)</td><td>(2)</td></tr><tr><td>VARIABLES</td><td>Main effects</td><td>Interaction model</td></tr></thead><tbody><tr><td>White overrepresentation</td><td>−61.62</td><td>−66.02</td></tr><tr><td /><td>(31.73)</td><td>(39.63)</td></tr><tr><td>Republican control of legislature and governorship</td><td>−219.7*</td><td>−220.7*</td></tr><tr><td /><td>(83.56)</td><td>(83.46)</td></tr><tr><td>White overrepresentation X Republican control</td><td>-</td><td>20.03</td></tr><tr><td /><td>-</td><td>(59.67)</td></tr><tr><td>State population</td><td>0.00381**</td><td>0.00381**</td></tr><tr><td /><td>(0.00130)</td><td>(0.00131)</td></tr><tr><td>Per capita income</td><td>−0.0755</td><td>−0.0744</td></tr><tr><td /><td>(0.0715)</td><td>(0.0729)</td></tr><tr><td>Unemployment rate</td><td>−174.7*</td><td>−174.5*</td></tr><tr><td /><td>(73.40)</td><td>(73.28)</td></tr><tr><td>State's four-year FTE enrollment</td><td>−0.0293*</td><td>−0.0294**</td></tr><tr><td /><td>(0.0109)</td><td>(0.0108)</td></tr><tr><td>Count of public institutions</td><td>28.62</td><td>29.10</td></tr><tr><td /><td>(148.5)</td><td>(148.0)</td></tr><tr><td>Pct of public institutions that are research universities</td><td>−10.00</td><td>−10.17</td></tr><tr><td /><td>(5.470)</td><td>(5.538)</td></tr><tr><td>Constant</td><td>−3,306*</td><td>−3,325**</td></tr><tr><td /><td>(1,251)</td><td>(1,236)</td></tr><tr><td>Observations</td><td>441</td><td>441</td></tr><tr><td>R-squared</td><td>0.068</td><td>0.068</td></tr><tr><td>Number of states</td><td>49</td><td>49</td></tr></tbody></table> </ephtml> </p> <p>1 Robust standard errors in parentheses</p> <p>2 **p <.01, *p <.05.</p> <p>Column two reports regression results that include the interaction term. A skeptical observer might note that the coefficient of determination (R<sups>2</sups>) is little changed between the main effects and interaction model (Table 2). Here it is important to note that R<sups>2</sups> is a measure describing the amount of variance explained by model parameters, not a goodness-of-fit statistic indexing model fit to a known distribution. This means that R<sups>2</sups> might change little even if a model improved upon prior estimates if, for example, a social process was difficult to determine (Rabe-Hesketh & Skrondal, [<reflink idref="bib66" id="ref116">66</reflink>]). Model fit is therefore better judged by statistical tests.[<reflink idref="bib2" id="ref117">2</reflink>] The White overrepresentation variable, unified Republican control, and the interaction of these two terms are jointly significant (<emph>p</emph> <.01). Given this result, along with robustness tests described below, we are confident that our preferred model fits the data well. However, we also acknowledge that alterative interpretations are possible, and consider these possibilities in our discussion.</p> <p>As in the first column of Table 2, unified Republican control of government predicted a negative change of a little more than $221 in the dependent variable (relative to the change in the average state). Increases in the White overrepresentation variable (the change in White overrepresentation in that state relative to the change in the average state) also predicted declines in the dependent variable. These two results were conditioned by the interaction term (Brambor et al., [<reflink idref="bib10" id="ref118">10</reflink>]).</p> <p>We present the interaction of White overrepresentation and unified Republican control visually in Figure 3, which identifies the specific conditions under which certain relationships held.[<reflink idref="bib3" id="ref119">3</reflink>] The gap between the two lines—the estimated return to the interaction term—is the phenomenon of interest. Unified Republican governments were expected to enact smaller cuts to appropriations per FTE (relative to the average change in appropriations per FTE) when White students became more overrepresented (relative to the change in the average state). This is indicated by the gap between the two lines of the positive side of zero. The opposite was true when White students were not overrepresented. States with unified Republican governments were expected to cut spending especially deeply (relative to the average state) when White student overrepresentation declined more rapidly than it did in the average state.</p> <p>PHOTO (COLOR): Figure 3. Predicted state appropriations per FTE under unified Republican or other form of government as White overrepresentation varies.</p> <p>A few control characteristics merit comment. Increases in a state's unemployment rate (relative to the change in the average state's employment rate) were associated with the dependent variable, confirming the ongoing importance of the balance wheel account. Increases in state support for higher education relative to the average state also were associated with changes in the state population relative to the average state's change. This may have reflected economies of scale resulting from population increase; the nth citizen is less necessary to support baseline operations than was the (n-1)st citizen, allowing greater support for higher education.</p> <hd id="AN0145086186-13">Assumption checks and limitations</hd> <p>Although we have used contemporary theory and robust quantitative techniques, our analysis—like all social scientific work—rests on a set of assumptions. We must test these assumptions to ensure that our account is not model-dependent. Our first assumption was that we should focus on changes in social context rather than the context itself. It is possible, however, that differencing introduced rather than removed "noise." Appendix 1 presents the results of two regressions that reproduce the analyses from Table 2 using observed (non-differenced) values. The "signs and significance" as well as a test for joint significance in the interaction model match the results presented above. We therefore concluded that our analytic strategy would not be improved by using observed rather than twice-differenced values. This supports our decision to focus on changes, in keeping with recent research on patterned volatility across states (Li, [<reflink idref="bib42" id="ref120">42</reflink>]).</p> <p>The second set of assumptions to test is our operationalization of concepts as variables. While we used a series of dummy variables for each year in our sample, historical time could instead be modeled as a trend. A log-linear trend would give time a fairly gentle slope. A linear trend would grow steadily, while a quadratic or cubic trend would account for possible variations in the economic cycle that might correspond to higher education's "balance wheel" role in state budgets. Our dependent variable also could be operationalized differently. We could consider total appropriations rather than appropriations per FTE, thereby ensuring that the dependent variable and FTE (a regressor in the model) were not functionally related to one another.</p> <p>Because there were plausibly two different versions of the dependent variable and five different models of time, we tested 10 regression models. These 10 models appear in Appendices 2 (appropriations per FTE) and 3 (total appropriations in millions). We also supply results of a test of joint significance for the interacted variables. In 8 of 10 models, the "stars and significance" of these three variables of interest matched one another. This concordance is strong evidence that our preferred models are robust to alternative specifications. At the same time, because the results are concordant without being unanimous, we present all results in the interest of transparency.</p> <p>Our theoretical model emphasizes the role of White racial resentment in the contemporary Republican Party. White Democrats often manifest racial resentment as well (Ostfield, [<reflink idref="bib59" id="ref121">59</reflink>]), making it crucial to test the possibility that White overrepresentation also conditions the behavior of states under unified Democratic governments. Appendix 4 reports results that reproduced the analyses presented in Table 2 but replaced unified Republican control with unified Democratic control. The first column reports the main effects. As in the first column of Table 2, the relationship between the change in White overrepresentation (relative to the change in the average state) and the dependent variable is negative. Changes in White overrepresentation matter whether we focus on unified Republican or unified Democratic governments. Crucially, however, the role of unified Democratic control does not appear to be conditioned by changes in White overrepresentation. The two regressors of interest and their interaction are not jointly significant. This null finding indicates that our conceptual model, which links the Republican Party and White racial resentment, rests on a reasonable assumption.</p> <p>Finally, we tested the assumptions of our interaction term using Hainmueller et al.'s ([<reflink idref="bib27" id="ref122">27</reflink>]) techniques. Figure 4 shows a disaggregation of observed values of the dependent variable and independent variable of interest. The left-hand plot shows data when a state does not have unified Republican control, while those with unified GOP control are depicted on the right. We imposed both a locally weighted regression (LOESS) fit line and a linear fit line on observed data. Hainmueller and colleagues' guidelines indicate that the linear interaction term may be used when the two lines: (<reflink idref="bib1" id="ref123">1</reflink>) have similar slopes in both of the panels, and (<reflink idref="bib2" id="ref124">2</reflink>) seem approximately linear rather than curvilinear.</p> <p>PHOTO (COLOR): Figure 4. Testing assumptions of linear X dichotomous interaction term using linear and LOESS fit, by unified Republican control.</p> <p>Comparison of these two lines offers qualified support for our estimation strategy. In both figures, the lines slope in the same direction, indicating that the data do not behave in decidedly different ways across the two sub-samples. Notably, however, the LOESS fit line is slightly curved in the right-hand figure. Quality of fit is strongest when White overrepresentation is between −3 and 3. This can be seen in Figure 5, which shows linear and LOESS fit for only this region of the data. The curve observed in Figure 4 reflects the influence of high-leverage outliers on the LOESS fit line. To ensure the robustness of our claims, we re-estimated the models presented in Table 2 but included only observations in which the White overrepresentation variable ranged from −3 to 3. We found that the "signs and significance" of coefficients remained the same (see Appendix 5), suggesting that results presented in Table 2 are not model dependent.</p> <p>PHOTO (COLOR): Figure 5. Comparison of linear and LOESS fit under unified Republican control as White overrepresentation varies from 3 to −3 percentage points.</p> <p>Given the results of these robustness checks, we interpret our results with some caution. We do not generalize beyond the area (±3 percentage points in the White overrepresentation variable) in which most observations are clustered. For this reason, Figure 2 is limited to the range where support for inferences is strongest. We explicitly declaim implications for states in which White students are highly underrepresented or overrepresented. For example, in 2009 Maine's population was 94% White. Because its higher education system is relatively small, the demographics of Maine's enrollments can vary considerably from 1 year to another. Enrollments were notably more racially diverse in 2009 than they would be the next year, with the result that the change in White student overrepresentation was −9.4. This made Maine 2009 the outlier seen in the far left of the "not unified Republican control" panel of Figure 3. We do not think our model explains much about higher education funding in this or other outlying cases. For the remaining cases (more than 89% of observations that fell in the ±3 percentage points range), our results are robust to alternative specifications but are not causal.</p> <hd id="AN0145086186-14">Discussion</hd> <p>In this study, we posit that political parties and policies should be situated in specific contexts. Given the large share of the Republican Party that identifies as White and unifies around racial resentment (Abramowitz, [<reflink idref="bib1" id="ref125">1</reflink>]), we tested the hypothesis that changes in White overrepresentation (relative to the average state) and unified Republican control jointly predicted changes in a state's higher education appropriations relative to the national trend. Our results generally supported our conceptual model. In states controlled by unified Republican governments, rising White overrepresentation (relative to the average state) predicted increases in appropriations (relative to the change in the average state). By contrast, shrinking White overrepresentation (relative to the average state) and unified GOP control predicted especially deep cuts to higher education, even compared to a national environment in which state appropriations were generally declining.</p> <p>Our results illuminate the complex world in which policymakers allocate resources. Rational choice is certainly a factor. The balance wheel continues to explain some funding decisions; our findings indicate that the change in the unemployment rate is negatively associated with changes in higher education appropriations, both relative to the change in the average state. Just so, traditional political approaches that take party preferences as relatively fixed remain salient. Unified Republican governments tend to fund four-year institutions at lower levels than do other states (Table 2, column 1). Readers who are skeptical of our analysis due to the small nominal changes in R<sups>2</sups> (see Table 2) may find that our confirmation of longstanding accounts adds credibility to our model.</p> <p>At the same time, results from this study suggest that funding is not solely geared to address social problems within the constraints of available resources and fixed political preferences. Our study supports the centrality of changing social contexts to the processes that shape state higher education funding. Under Republican control, changing enrollment demographics may yield particularly rapid divestment from higher education regardless of available resources. Conversely, conditions that continue to favor White student enrollments may result in smaller than expected cuts despite the GOP's fixed political preferences. These results demonstrate how critical models that consider social context—in our study, power and racism—can advance our understanding of higher education policy. By allowing the relationship between party control and state appropriations to fluctuate based on changing social conditions, we can supplement rational actor models with accounts that foreground contextual factors of power and inequality.</p> <p>Our analysis is limited in important ways. We may not account adequately for the wide range of interest groups that shape higher education policymaking. Building on work by Gándara ([<reflink idref="bib22" id="ref126">22</reflink>]), our model includes higher education institutions themselves, but does not account for advocacy groups, philanthropies, and other organizations that also shape state policy (e.g., Haddad & Reckhow, [<reflink idref="bib26" id="ref127">26</reflink>]; Hertel-Fernandez, [<reflink idref="bib30" id="ref128">30</reflink>]; Miller & Morphew, [<reflink idref="bib52" id="ref129">52</reflink>]). Greater attention to these bodies will yield an even richer understanding of the policymaking process.</p> <p>Additionally, some within-state policy changes are not accounted for by our model. Several states have de-regulated tuition-setting authority. Recent studies demonstrate that institutional policy response (Kramer, Ortagus, & Lacy, [<reflink idref="bib40" id="ref130">40</reflink>]) and access patterns (Flores & Shepherd, [<reflink idref="bib18" id="ref131">18</reflink>]) vary based on differences in tuition-setting authority. Changes in appropriations resulting from changes in tuition-setting authority are not captured in this study because we assumed unspecified state effects to be fixed. Future research should test and refine this assumption.</p> <p>Similarly, we acknowledge that direct state funding is not the only policy means by which state governments can influence access and affordably. Complex tax incentives (Mettler, [<reflink idref="bib51" id="ref132">51</reflink>]) and portable student financial aid (Toutkoushian & Paulsen, [<reflink idref="bib77" id="ref133">77</reflink>]) are other means by which state governments support higher education. It is possible that unified Republican governments merely prefer to fund higher education indirectly rather than through direct appropriations. Even if some funding is provided by other means, divestment from direct support signals a fundamental restructuring of the social compact that binds states and citizens together (Pusser, [<reflink idref="bib65" id="ref134">65</reflink>]). Direct support for higher education indicates investment in broad educational opportunity and "participatory readiness" for citizenship (Allen, [<reflink idref="bib4" id="ref135">4</reflink>]). Portable incentives only benefit those who attend, and so do not support higher education's public purposes in the same way. Future research on the social context of tax incentives and portable student financial aid is likely to illuminate different, likely complementary, dimensions of power and policymaking.</p> <p>Acknowledging the limitations and possible alternative explanations, our results cohere with a vibrant body of research and are robust to many, though not quite all, alternate model specifications. Our findings are limited, and are not conclusive, but stand up to robustness tests and are congruent with Baker's ([<reflink idref="bib6" id="ref136">6</reflink>]) work and with recent research in political science. Our analyses therefore entail important consequences for research on higher education policymaking.</p> <p>Orthodox policy thought in higher education assumes good faith actors. In other words, researchers tend to accept that policymakers have a genuine interest in reforming higher education. Under this set of assumptions, more information, diffusion of best practices, and more sophisticated tools for accountability—the fruits of the research—will result in better policy. Findings from our study question these assumptions. Some political actors might take positions on higher education in an effort to win electoral support by stirring negative partisanship and White racial resentment. The consequences for higher education might be irrelevant when policy positions are intended to win votes. In this context, researchers should be wary of assuming good faith.</p> <p>It also may be imprudent to assume that voters act rationally when expressing their preferences for higher education. The case of White voters who identify as Republicans is illustrative. These voters are the most likely to express distrust for higher education and disapproval about the rising price of tuition (Pew Research Center, [<reflink idref="bib61" id="ref137">61</reflink>]). Yet college participation is nearly universal among middle- and upper-income White people (Cantwell, [<reflink idref="bib12" id="ref138">12</reflink>]). High demand for higher education implies that many White voters want their children to go to college, and yet support politicians who promise (and deliver) reduced state support for colleges and universities. Simultaneous demand and support for disinvestment is a version of the paradox with which our paper opened. By focusing on underlying social conditions, our study helps to explain why this apparent paradox is no paradox at all. Why do growing numbers of White Americans both distrust higher education and demand access to it? Why do some voters want a college education for their children, but do not want to pay taxes or tuition? These voter preferences are not about applied rationality or enduring ideological commitments, but about winning and exercising power in order to punish the opposition even at the expense of self-interest. This sort of politics defies rationality but seems to characterize the policy process under certain conditions.</p> <p>In the broadest sense, a conclusion from our study is that researchers ought to understand higher education policymaking not only as technical processes that aim to improve system functioning within the constraints of differing resources and ideologies, but also as part of a wider struggle to define the state through empowering some actors and disempowering others (Pusser, [<reflink idref="bib65" id="ref139">65</reflink>]). Determining the level at which to fund public higher education is not just about calibrating the operations of a complex social system. It is also about exercising power, determining winners and losers, and marshaling support in future elections. This process plays out differently in different social contexts. Researchers who fail to attend to these dynamics are likely to produce partial accounts that leave fundamental paradoxes unaddressed.</p> <hd id="AN0145086186-15">Disclosure statement</hd> <p>No potential conflict of interest was reported by the authors.</p> <hd id="AN0145086186-16">Appendix 1. Alternative model of observed values rather than changes (relative to the average...</hd> <p></p> <p> <ephtml> <table><thead><tr><td /><td>(1)</td><td>(2)</td></tr><tr><td>VARIABLES</td><td>Main effects</td><td>Interaction model</td></tr></thead><tbody><tr><td>White overrepresentation</td><td>−47.48**</td><td>−48.01**</td></tr><tr><td /><td>(14.26)</td><td>(15.49)</td></tr><tr><td>Republican control of legislature and governorship</td><td>−240.0**</td><td>−232.1*</td></tr><tr><td /><td>(83.61)</td><td>(93.93)</td></tr><tr><td>White overrepresentation X Republican control</td><td>-</td><td>1.665</td></tr><tr><td /><td>-</td><td>(11.19)</td></tr><tr><td>Are the interacted terms jointly significant?</td><td>-</td><td>Yes</td></tr><tr><td /><td>-</td><td>(p ≈.0002)</td></tr><tr><td>State population</td><td>0.00376**</td><td>0.00378**</td></tr><tr><td /><td>(0.00132)</td><td>(0.00132)</td></tr><tr><td>Per capita income</td><td>−0.0750</td><td>−0.0750</td></tr><tr><td /><td>(0.0716)</td><td>(0.0717)</td></tr><tr><td>Unemployment rate</td><td>−189.5*</td><td>−189.3*</td></tr><tr><td /><td>(76.64)</td><td>(76.53)</td></tr><tr><td>State's four-year FTE enrollment</td><td>−0.0300**</td><td>−0.0301**</td></tr><tr><td /><td>(0.0109)</td><td>(0.0110)</td></tr><tr><td>Count of public institutions</td><td>67.12</td><td>65.45</td></tr><tr><td /><td>(139.5)</td><td>(140.8)</td></tr><tr><td>Pct of public institutions that are research universities</td><td>−7.174</td><td>−7.204</td></tr><tr><td /><td>(5.047)</td><td>(5.046)</td></tr><tr><td>Constant</td><td>−3,546**</td><td>−3,558**</td></tr><tr><td /><td>(1,254)</td><td>(1,257)</td></tr><tr><td>Observations</td><td>441</td><td>441</td></tr><tr><td>R-squared</td><td>0.069</td><td>0.069</td></tr><tr><td>Number of states</td><td>49</td><td>49</td></tr></tbody></table> </ephtml> </p> <ulist> <item>3 Robust standard errors in parentheses.</item> <item>4 **p <.01, *p <.05.</item> </ulist> <hd id="AN0145086186-17">Appendix 2. Alternative model specifications with dependent variable of appropriations per FT...</hd> <p></p> <p> <ephtml> <table><thead><tr><td /><td>(1)</td><td>(2)</td><td>(3)</td><td>(4)</td><td>(5)</td></tr><tr><td>VARIABLES</td><td>Dummy variables</td><td>Log-linear time trend</td><td>Linear time trend</td><td>Quadratic time trend</td><td>Cubic time trend</td></tr></thead><tbody><tr><td>White overrepresentation</td><td>−66.02</td><td>−65.15</td><td>−65.42</td><td>−65.35</td><td>−65.84</td></tr><tr><td /><td>(39.63)</td><td>(38.99)</td><td>(39.46)</td><td>(39.10)</td><td>(39.59)</td></tr><tr><td>Republican control of legislature and governorship</td><td>−220.7*</td><td>−219.0**</td><td>−201.3*</td><td>−219.2*</td><td>−211.2*</td></tr><tr><td>(83.46)</td><td>(79.78)</td><td>(78.80)</td><td>(82.40)</td><td>(80.75)</td></tr><tr><td>White overrepresentation X Republican control</td><td>20.03</td><td>16.14</td><td>16.15</td><td>17.31</td><td>19.40</td></tr><tr><td>(59.67)</td><td>(59.57)</td><td>(59.29)</td><td>(59.55)</td><td>(60.20)</td></tr><tr><td>Are the interacted terms jointly significant?</td><td>Yes</td><td>Yes</td><td>Yes</td><td>Yes</td><td>Yes</td></tr><tr><td>(p ≈.0086)</td><td>(p ≈.0109)</td><td>(p ≈.0161)</td><td>(p ≈.0127)</td><td>(p ≈.0147)</td></tr><tr><td>State population</td><td>0.00381**</td><td>0.00385**</td><td>0.00389**</td><td>0.00384**</td><td>0.00382**</td></tr><tr><td /><td>(0.00131)</td><td>(0.00127)</td><td>(0.00131)</td><td>(0.00128)</td><td>(0.00129)</td></tr><tr><td>Per capita income</td><td>−0.0744</td><td>−0.0744</td><td>−0.0745</td><td>−0.0744</td><td>−0.0742</td></tr><tr><td /><td>(0.0729)</td><td>(0.0723)</td><td>(0.0724)</td><td>(0.0722)</td><td>(0.0720)</td></tr><tr><td>Unemployment rate</td><td>−174.5*</td><td>−173.7*</td><td>−174.5*</td><td>−173.9*</td><td>−174.5*</td></tr><tr><td /><td>(73.28)</td><td>(72.50)</td><td>(72.80)</td><td>(72.67)</td><td>(72.91)</td></tr><tr><td>State's four-year FTE enrollment</td><td>−0.0294**</td><td>−0.0242*</td><td>−0.0223*</td><td>−0.0262*</td><td>−0.0284**</td></tr><tr><td /><td>(0.0108)</td><td>(0.0103)</td><td>(0.00957)</td><td>(0.0104)</td><td>(0.0102)</td></tr><tr><td>Count of public institutions</td><td>29.10</td><td>28.54</td><td>26.49</td><td>28.69</td><td>27.99</td></tr><tr><td /><td>(148.0)</td><td>(151.6)</td><td>(150.2)</td><td>(151.8)</td><td>(151.5)</td></tr><tr><td>Pct of public institutions that are research universities</td><td>−10.17</td><td>−10.05</td><td>−10.16*</td><td>−10.09</td><td>−10.19</td></tr><tr><td>(5.538)</td><td>(5.239)</td><td>(4.856)</td><td>(5.469)</td><td>(5.773)</td></tr><tr><td>Log-linear time trend</td><td>-</td><td>−241.5</td><td>-</td><td>-</td><td>-</td></tr><tr><td /><td>-</td><td>(132.0)</td><td>-</td><td>-</td><td>-</td></tr><tr><td>Linear time trend</td><td>-</td><td>-</td><td>−42.80</td><td>−106.2</td><td>−24.89</td></tr><tr><td /><td>-</td><td>-</td><td>(22.77)</td><td>(84.42)</td><td>(161.7)</td></tr><tr><td>Square of time trend</td><td>-</td><td>-</td><td>-</td><td>5.542</td><td>−15.49</td></tr><tr><td /><td>-</td><td>-</td><td>-</td><td>(7.161)</td><td>(32.36)</td></tr><tr><td>Cube of time trend</td><td>-</td><td>-</td><td>-</td><td>-</td><td>1.424</td></tr><tr><td /><td>-</td><td>-</td><td>-</td><td>-</td><td>(1.981)</td></tr><tr><td>Constant</td><td>−3,325**</td><td>−2,529*</td><td>−2,492*</td><td>−2,820*</td><td>−3,159**</td></tr><tr><td /><td>(1,236)</td><td>(1,070)</td><td>(1,077)</td><td>(1,125)</td><td>(1,149)</td></tr><tr><td>Observations</td><td>441</td><td>441</td><td>441</td><td>441</td><td>441</td></tr><tr><td>R-squared</td><td>0.068</td><td>0.066</td><td>0.064</td><td>0.067</td><td>0.067</td></tr><tr><td>Number of states</td><td>49</td><td>49</td><td>49</td><td>49</td><td>49</td></tr></tbody></table> </ephtml> </p> <ulist> <item>5 Robust standard errors in parentheses.</item> <item>6 **p <.01, *p <.05.</item> </ulist> <hd id="AN0145086186-18">Appendix 3. Alternative model specifications with dependent variable of total appropriations...</hd> <p></p> <p> <ephtml> <table><thead><tr><td /><td>(1)</td><td>(2)</td><td>(3)</td><td>(4)</td><td>(5)</td></tr><tr><td>VARIABLES</td><td>Dummy variables</td><td>Log-linear time trend</td><td>Linear time trend</td><td>Quadratic time trend</td><td>Cubic time trend</td></tr></thead><tbody><tr><td>White overrepresentation</td><td>−8.482*</td><td>−8.781</td><td>−8.228</td><td>−8.121*</td><td>−8.020*</td></tr><tr><td /><td>(3.276)</td><td>(4.477)</td><td>(4.517)</td><td>(3.609)</td><td>(3.504)</td></tr><tr><td>Republican control of legislature and governorship</td><td>−24.63</td><td>−9.406</td><td>−8.504</td><td>−34.80*</td><td>−36.47*</td></tr><tr><td /><td>(12.48)</td><td>(10.68)</td><td>(10.47)</td><td>(15.20)</td><td>(15.65)</td></tr><tr><td>White overrepresentation X Republican control</td><td>15.76</td><td>15.80*</td><td>12.23</td><td>13.93</td><td>13.49</td></tr><tr><td /><td>(9.116)</td><td>(7.743)</td><td>(7.084)</td><td>(7.369)</td><td>(7.526)</td></tr><tr><td>Are the interacted terms jointly significant?</td><td>Yes</td><td>No</td><td>No</td><td>Yes</td><td>Yes</td></tr><tr><td>(p ≈.0493)</td><td>(p ≈.1533)</td><td>(p ≈.2586)</td><td>(p ≈.0486)</td><td>(p ≈.0431)</td></tr><tr><td>State population</td><td>0.000617</td><td>0.000629</td><td>0.000682</td><td>0.000608</td><td>0.000610</td></tr><tr><td /><td>(0.000752)</td><td>(0.000865)</td><td>(0.000864)</td><td>(0.000735)</td><td>(0.000731)</td></tr><tr><td>Per capita income</td><td>0.00302</td><td>0.00263</td><td>0.00214</td><td>0.00243</td><td>0.00241</td></tr><tr><td /><td>(0.00503)</td><td>(0.00505)</td><td>(0.00453)</td><td>(0.00473)</td><td>(0.00473)</td></tr><tr><td>Unemployment rate</td><td>−14.21</td><td>−15.24</td><td>−15.27</td><td>−14.35</td><td>−14.22</td></tr><tr><td /><td>(8.220)</td><td>(8.319)</td><td>(8.787)</td><td>(8.072)</td><td>(8.094)</td></tr><tr><td>State's four-year FTE enrollment</td><td>−0.00577**</td><td>−0.00662**</td><td>−0.000262</td><td>−0.00589**</td><td>−0.00542*</td></tr><tr><td /><td>(0.00213)</td><td>(0.00151)</td><td>(0.00191)</td><td>(0.00189)</td><td>(0.00206)</td></tr><tr><td>Count of public institutions</td><td>13.81</td><td>11.58</td><td>11.47</td><td>14.70</td><td>14.84</td></tr><tr><td /><td>(28.43)</td><td>(36.88)</td><td>(34.96)</td><td>(34.63)</td><td>(34.85)</td></tr><tr><td>Pct of public institutions that are research universities</td><td>−0.241</td><td>−0.413</td><td>−0.259</td><td>−0.162</td><td>−0.140</td></tr><tr><td /><td>(0.986)</td><td>(0.871)</td><td>(0.758)</td><td>(1.580)</td><td>(1.489)</td></tr><tr><td>Log-linear time trend</td><td>-</td><td>−96.28**</td><td>-</td><td>-</td><td>-</td></tr><tr><td /><td>-</td><td>(18.14)</td><td>-</td><td>-</td><td>-</td></tr><tr><td>Linear time trend</td><td>-</td><td>-</td><td>−0.845</td><td>−93.81**</td><td>−110.7**</td></tr><tr><td /><td>-</td><td>-</td><td>(4.185)</td><td>(18.77)</td><td>(30.94)</td></tr><tr><td>Square of time trend</td><td>-</td><td>-</td><td>-</td><td>8.129**</td><td>12.51*</td></tr><tr><td /><td>-</td><td>-</td><td>-</td><td>(1.645)</td><td>(5.884)</td></tr><tr><td>Cube of time trend</td><td>-</td><td>-</td><td>-</td><td>-</td><td>−0.296</td></tr><tr><td /><td>-</td><td>-</td><td>-</td><td>-</td><td>(0.352)</td></tr><tr><td>Constant</td><td>−603.7*</td><td>−664.5**</td><td>−35.82</td><td>−516.3*</td><td>−445.8</td></tr><tr><td /><td>(241.2)</td><td>(159.3)</td><td>(216.0)</td><td>(196.5)</td><td>(225.5)</td></tr><tr><td>Observations</td><td>441</td><td>441</td><td>441</td><td>441</td><td>441</td></tr><tr><td>R-squared</td><td>0.173</td><td>0.041</td><td>0.017</td><td>0.122</td><td>0.123</td></tr><tr><td>Number of states</td><td>49</td><td>49</td><td>49</td><td>49</td><td>49</td></tr></tbody></table> </ephtml> </p> <ulist> <item>7 Robust standard errors in parentheses.</item> <item>8 **p <.01, *p <.05.</item> </ulist> <hd id="AN0145086186-19">Appendix 4. Alternative model specifications exploring unified Democratic control</hd> <p></p> <p> <ephtml> <table><thead><tr><td /><td>(1)</td><td>(2)</td></tr><tr><td>VARIABLES</td><td>Main effects</td><td>Interaction model</td></tr></thead><tbody><tr><td>White overrepresentation</td><td>−64.96*</td><td>−75.05</td></tr><tr><td /><td>(31.43)</td><td>(43.44)</td></tr><tr><td>Democratic control of legislature and governorship</td><td>58.03</td><td>60.67</td></tr><tr><td /><td>(100.9)</td><td>(100.1)</td></tr><tr><td>White overrepresentation X Democratic control</td><td>-</td><td>27.76</td></tr><tr><td /><td>-</td><td>(56.73)</td></tr><tr><td>State population</td><td>0.00392**</td><td>0.00390**</td></tr><tr><td /><td>(0.00138)</td><td>(0.00139)</td></tr><tr><td>Per capita income</td><td>−0.0763</td><td>−0.0758</td></tr><tr><td /><td>(0.0713)</td><td>(0.0712)</td></tr><tr><td>Unemployment rate</td><td>−177.3*</td><td>−176.1*</td></tr><tr><td /><td>(73.06)</td><td>(73.10)</td></tr><tr><td>State's four-year FTE enrollment</td><td>−0.0275*</td><td>−0.0277*</td></tr><tr><td /><td>(0.0113)</td><td>(0.0111)</td></tr><tr><td>Count of public institutions</td><td>0.696</td><td>0.780</td></tr><tr><td /><td>(148.3)</td><td>(148.9)</td></tr><tr><td>Pct of public institutions that are research universities</td><td>−12.42*</td><td>−12.96*</td></tr><tr><td /><td>(5.733)</td><td>(5.873)</td></tr><tr><td>Constant</td><td>−3,165*</td><td>−3,191*</td></tr><tr><td /><td>(1,292)</td><td>(1,273)</td></tr><tr><td>Observations</td><td>441</td><td>441</td></tr><tr><td>R-squared</td><td>0.062</td><td>0.062</td></tr><tr><td>Number of states</td><td>49</td><td>49</td></tr></tbody></table> </ephtml> </p> <ulist> <item>9 Robust standard errors in parentheses.</item> <item>10 **p <.01, *p <.05.</item> </ulist> <hd id="AN0145086186-20">Appendix 5. Re-estimate of preferred model within the range of −3 to 3 on change in White ove...</hd> <p></p> <p> <ephtml> <table><thead><tr><td /><td>(1)</td></tr><tr><td>VARIABLES</td><td>Change in State Appropriations per FTE</td></tr></thead><tbody><tr><td>White overrepresentation</td><td>−66.26</td></tr><tr><td /><td>(70.67)</td></tr><tr><td>Republican control of legislature and governorship</td><td>−222.6*</td></tr><tr><td /><td>(83.66)</td></tr><tr><td>White overrepresentation X Republican control</td><td>23.85</td></tr><tr><td /><td>(84.92)</td></tr><tr><td>Are the interacted terms jointly significant?</td><td>Yes</td></tr><tr><td>(p ≈.0325)</td></tr><tr><td>State population</td><td>0.00326*</td></tr><tr><td /><td>(0.00148)</td></tr><tr><td>Per capita income</td><td>−0.0617</td></tr><tr><td /><td>(0.0727)</td></tr><tr><td>Unemployment rate</td><td>−131.7</td></tr><tr><td /><td>(75.83)</td></tr><tr><td>State's four-year FTE enrollment</td><td>−0.0320**</td></tr><tr><td /><td>(0.0103)</td></tr><tr><td>Count of public institutions</td><td>24.67</td></tr><tr><td /><td>(140.3)</td></tr><tr><td>Pct of public institutions that are research universities</td><td>−11.88*</td></tr><tr><td /><td>(5.629)</td></tr><tr><td>Constant</td><td>−3,615**</td></tr><tr><td /><td>(1,177)</td></tr><tr><td>Observations</td><td>435</td></tr><tr><td>Number of states</td><td>49</td></tr><tr><td>R-squared</td><td>0.056</td></tr></tbody></table> </ephtml> </p> <ulist> <item>11 Robust standard errors in parentheses.</item> <item>12 **p <.01, *p <.05.</item> </ulist> <ref id="AN0145086186-21"> <title> Notes </title> <blist> <bibl id="bib1" idref="ref15" type="bt">1</bibl> <bibtext> Some scholars have disputed whether data from a single election can support such weighty interpretations. It is worth nothing, however, that these critiques have not dismissed the role of race in the Republican Party, but rather have argued that it is difficult to disentangle race from other factors using existing evidence (Morgan, [53]).</bibtext> </blist> <blist> <bibl id="bib2" idref="ref72" type="bt">2</bibl> <bibtext> The most common such technique, a likelihood-ratio text, is unavailable in models such as ours that employ clustered standard errors to address serial correlation. We therefore rely upon a test of joint significance for the three variables included in the interaction term (Brambor et al., [10]).</bibtext> </blist> <blist> <bibl id="bib3" idref="ref64" type="bt">3</bibl> <bibtext> As noted in "robustness checks and limitation," we limit interpretation to the range of −3 to 3 on the variable of White overrepresentation. This avoids extending our inferences beyond the region that contains the majority of observed data. State-years that fall outside this range are unusual (see Figure 4), and their behavior may not be explained particularly well by our model.</bibtext> </blist> </ref> <ref id="AN0145086186-22"> <title> References </title> <blist> <bibtext> Abramowitz, A. (2018). The great alignment: Race, party transformation, and the rise of Donald Trump. New Haven, CT : Yale University Press.</bibtext> </blist> <blist> <bibtext> Abramowitz, A., & Webster, S. (2016). The rise of negative partisanship and the nationalization of U.S. elections in the 21st century. Electoral Studies, 41, 12 – 22. doi: 10.1016/j.electstud.2015.11.001</bibtext> </blist> <blist> <bibtext> Abramowitz, A., & Webster, S. (2018). Negative partisanship: Why Americans dislike partisanship but behave like rabid partisans. Advances in Political Psychology, 39 (1), 119 – 135. doi: 10.1111/pops.12479</bibtext> </blist> <blist> <bibl id="bib4" idref="ref135" type="bt">4</bibl> <bibtext> Allen, D. (2016). Education and equality. Chicago, IL : University of Chicago Press.</bibtext> </blist> <blist> <bibl id="bib5" idref="ref78" type="bt">5</bibl> <bibtext> Bail, C. A., Argyle, L. P., Brown, T. W., Bumpus, J. P., Chen, H., Hunzaker, F., ... Volfovsky, A. (2018). Exposure to opposing views on social media can increase political polarization. Proceedings of the National Academy of Sciences, 115 (37), 9216 – 9221. doi: 10.1073/pnas.1804840115</bibtext> </blist> <blist> <bibl id="bib6" idref="ref45" type="bt">6</bibl> <bibtext> Baker, D. (2019). Pathways to racial equity in higher education: Modeling the antecedents of state affirmative action bans. American Educational Research Journal, 56, 1861 – 1895. doi: 10.3102/0002831219833918</bibtext> </blist> <blist> <bibl id="bib7" idref="ref57" type="bt">7</bibl> <bibtext> Barber, M., & Pope, J. C. (2019a). Does party Trump ideology? Disentangling party and ideology in America. American Political Science Review, 113 (1), 38 – 54. doi: 10.1017/S0003055418000795</bibtext> </blist> <blist> <bibl id="bib8" idref="ref16" type="bt">8</bibl> <bibtext> Barber, M., & Pope, J. C. (2019b). Conservatism in the era of Trump. Perspectives on Politics, 17 (3), 719 – 736. doi: 10.1017/S153759271900077X</bibtext> </blist> <blist> <bibl id="bib9" idref="ref79" type="bt">9</bibl> <bibtext> Benkler, Y., Faris, R., & Roberts, H. (2018). Network propaganda: Manipulation, disinformation and radicalization in American politics. New York, NY : Oxford University Press.</bibtext> </blist> <blist> <bibtext> Brambor, T., Clark, W. R., & Golder, M. (2006). Understanding interaction models: Improving empirical analyses. Political Analysis, 14 (1), 63 – 82. doi: 10.1093/pan/mpi014</bibtext> </blist> <blist> <bibtext> Brunner, E. J., & Johnson, E. B. (2016). Intergenerational conflict and the political economy of higher education funding. Journal of Urban Economics, 91, 73 – 87. doi: 10.1016/j.jue.2015.11.006</bibtext> </blist> <blist> <bibtext> Cantwell, B. (2018). Broad access and steep stratification in the first mass system: High participation in the United States of America. In B. Cantwell, S. Marginson, & A. Smolentsiva (Eds.), High participation systems of higher education (pp. 227 – 265). Oxford, UK : Oxford University Press.</bibtext> </blist> <blist> <bibtext> College Board. (2014). Trends in college pricing. Retrieved from https://trends.collegeboard.org/college-pricing/figures-tables/percentage-first-time-students-public-four-year-institutions-who-are-state-residents-fall</bibtext> </blist> <blist> <bibtext> Cramer, K. J. (2016). The politics of resentment: Rural consciousness in Wisconsin and the rise of Scott Walker. Chicago, IL : University of Chicago Press.</bibtext> </blist> <blist> <bibtext> Delaney, J. A., & Doyle, W. R. (2007). The role of higher education in state budgets. In K. M. Shaw & D. E. Heller (Eds.), State postsecondary education research: New methods to inform policy and practice (pp. 55 – 76). Sterling, VA : Stylus.</bibtext> </blist> <blist> <bibtext> Douglass, J. A. (2007). The conditions for admission: Access, equity, and the social contract of public universities. Palo Alto, CA : Stanford University Press.</bibtext> </blist> <blist> <bibtext> Doyle, W. R., & Delaney, J. A. (2009). Higher education funding : The new normal. Change, 41 (4), 60 – 62</bibtext> </blist> <blist> <bibtext> Flores, S. M., & Shepherd, J. C. (2014). Pricing out the disadvantaged? The effect of tuition deregulation in Texas public four-year institutions. The ANNALS of the American Academy of Political and Social Science, 655 (1), 99 – 122. doi: 10.1177/0002716214539096</bibtext> </blist> <blist> <bibtext> Foster, J. M., & Fowles, J. (2018). Ethnic heterogeneity, group affinity, and state higher education spending. Research in Higher Education, 59 (1), 1 – 28. doi: 10.1007/s11162-017-9453-3</bibtext> </blist> <blist> <bibtext> Gallup. (2017). Why are Republicans down on higher education? Washington, DC : Author.</bibtext> </blist> <blist> <bibtext> Gándara, D. (2019a). Does evidence matter? An analysis of evidence use in performance-funding policy design. The Review of Higher Education, 42 (3), 991 – 1022. doi: 10.1353/rhe.2019.0027</bibtext> </blist> <blist> <bibtext> Gándara, D. (2019b). How the sausage is made: An examination of a state funding model design process. The Journal of Higher Education, 1 – 30. doi: 10.1080/00221546.2019.1618782</bibtext> </blist> <blist> <bibtext> Gándara, D., Rippner, J. A., & Ness, E. C. (2017). Exploring the "how" in policy diffusion: National intermediary organizations' roles in facilitating the spread of performance-based funding policies in the states. The Journal of Higher Education, 88 (5), 701 – 725. doi: 10.1080/00221546.2016.1272089</bibtext> </blist> <blist> <bibtext> Grogan, C. M., & Park, S. (2017). The racial divide in state Medicaid expansions. Journal of Health Politics, Policy and Law, 42 (3), 539 – 572. doi: 10.1215/03616878-3802977</bibtext> </blist> <blist> <bibtext> Grossman, M., & Hopkins, D. A. (2016). Asymmetric politics: Ideological Republicans and group interest Democrats. New York, NY : Oxford University Press.</bibtext> </blist> <blist> <bibtext> Haddad, N., & Reckhow, S. (2018). The shifting role of higher education philanthropy: A network analysis of philanthropic policy strategies. Philanthropy & Education, 2 (1), 25 –52.</bibtext> </blist> <blist> <bibtext> Hainmueller, J., Mummolo, J., & Xu, Y. (2019). How much should we trust estimates from multiplicative interaction models? Simple tools to improve empirical practice. Political Analysis, 27 (2), 163 – 192. doi: 10.1017/pan.2018.46</bibtext> </blist> <blist> <bibtext> Harper, S. R. (2012). Race without racism: How higher education researchers minimize racist institutional norms. The Review of Higher Education, 36 (1), 9 – 29. doi: 10.1353/rhe.2012.0047</bibtext> </blist> <blist> <bibtext> Harper, S. R., & Simmons, I. (2018). Black students at public colleges and universities. Los Angeles : University of Southern California.</bibtext> </blist> <blist> <bibtext> Hertel-Fernandez, A. (2019). State capture: How conservative activists, big businesses, and wealthy donors reshaped the American states—And the nation. Oxford, UK : Oxford University Press.</bibtext> </blist> <blist> <bibtext> Hill, A. J., & Jones, D. B. (2017). Does partisan affiliation impact the distribution of spending? Evidence from state governments' expenditures on education. Journal of Economic Behavior & Organization, 143, 58 – 77.</bibtext> </blist> <blist> <bibtext> Hillman, N. W., & Crespín-Trujillo, V. (2018). State accountability policies: Can performance funding be equitable? In G. Orfield & N. W. Hillman (Eds.), Accountability and opportunity in higher education (pp. 45 – 60). Cambridge, MA : Harvard Education Press.</bibtext> </blist> <blist> <bibtext> Hillman, N. W., Fryar, A. H., & Crespín-Trujillo, V. (2017). Evaluating the impact of performance funding in Ohio and Tennessee. American Educational Research Journal, 55 (1), 144 – 170. doi: 10.3102/0002831217732951</bibtext> </blist> <blist> <bibtext> Hooghe, M., & Dassonneville, R. (2018). Explaining the Trump vote: The effect of racist resentment and anti-immigration sentiments. PS: Political Science and Politics, 51 (3), 528 – 534.</bibtext> </blist> <blist> <bibtext> Hopkins, D. L. (2018). The increasingly United States: How and why American political behavior nationalized. Chicago, IL : University of Chicago Press.</bibtext> </blist> <blist> <bibtext> Hovey, H. A. (1999). State spending for higher education in the next decade. San Jose, CA: The National Center for Public Policy and Higher Education. Retrieved from ERIC database. (ED439633).</bibtext> </blist> <blist> <bibtext> Jaquette, O., & Parra, E. (2016). The problem with the Delta Cost Project database. Research in Higher Education, 57 (5), 630 – 651. doi: 10.1007/s11162-015-9399-2</bibtext> </blist> <blist> <bibtext> Jardina, A. (2019). White identity politics. Cambridge, UK : Cambridge University Press.</bibtext> </blist> <blist> <bibtext> Johnson, D. R., & Peifer, J. L. (2017). How public confidence in higher education varies by social context. The Journal of Higher Education, 88 (4), 619 – 644. doi: 10.1080/00221546.2017.1291256</bibtext> </blist> <blist> <bibtext> Kramer, D. A., Ortagus, J. C., & Lacy, T. A. (2018). Tuition-setting authority and broad-based merit aid: The effect of policy intersection on pricing strategies. Research in Higher Education, 59 (4), 489 – 518. doi: 10.1007/s11162-017-9475-x</bibtext> </blist> <blist> <bibtext> Ladson-Billings, G. (1998). Just what is critical race theory and what's it doing in a nice field like education? International Journal of Qualitative Studies in Education, 11 (1), 7 – 24. doi: 10.1080/095183998236863</bibtext> </blist> <blist> <bibtext> Li, A. (2017). Dramatic declines in higher education appropriations: State conditions for budget punctuations. Research in Higher Education, 58 (4), 395 – 429. doi: 10.1007/s11162-016-9432-0</bibtext> </blist> <blist> <bibtext> Li, A. Y. (2019). The weight of the metric: Performance funding and the retention of historically underserved students. The Journal of Higher Education, 1 – 27. doi: 10.1080/00221546.2019.1602391</bibtext> </blist> <blist> <bibtext> Lupton, R. N., Myers, W. M., & Thornton, J. R. (2017). Party animals: Asymmetric ideological constraint among Democratic and Republican party activists. Political Research Quarterly, 70 (4), 889 – 904. doi: 10.1177/1065912917718960</bibtext> </blist> <blist> <bibtext> Mason, L. (2015). "I disrespectfully agree:" The differential effects of partisan sorting on social and issue polarization. American Journal of Political Science, 59 (1), 128 – 145. doi: 10.1111/ajps.2015.59.issue-1</bibtext> </blist> <blist> <bibtext> Mason, L. (2018). Uncivil agreement: How politics became our identity. Chicago, IL : University of Chicago Press.</bibtext> </blist> <blist> <bibtext> McLendon, M. K., Deaton, R., & Hearn, J. C. (2007). The enactment of reforms in state governance of higher education: Testing the political instability hypothesis. The Journal of Higher Education, 78 (6), 645 – 675. doi: 10.1353/jhe.2007.0041</bibtext> </blist> <blist> <bibtext> McLendon, M. K., Hearn, J. C., & Mokher, C. G. (2009). Partisans, professionals, and power: The role of political factors in state higher education funding. The Journal of Higher Education, 80 (6), 686 – 713. doi: 10.1080/00221546.2009.11779040</bibtext> </blist> <blist> <bibtext> McLendon, M. K., Tandberg, D. A., & Hillman, N. (2014). Financing college opportunity: Factors influencing state spending on student financial aid and campus appropriations, 1990 through 2010. The ANNALS of the American Academy of Political and Social Science, 655 (1), 143 – 162. doi: 10.1177/0002716214540849</bibtext> </blist> <blist> <bibtext> Mettler, S. (2011). The submerged state: How invisible government policies undermine American democracy. Chicago, IL : University of Chicago Press.</bibtext> </blist> <blist> <bibtext> Mettler, S. (2014). Degrees of inequality: How the politics of higher education sabotaged the American dream. New York, NY : Basic Books.</bibtext> </blist> <blist> <bibtext> Miller, G. N. S., & Morphew, C. C. (2017). Merchants of optimism: Agenda-setting organizations and the framing of performance based funding for higher education. The Journal of Higher Education, 88 (5), 754 – 784. doi: 10.1080/00221546.2017.1313084</bibtext> </blist> <blist> <bibtext> Morgan, S. L. (2018). Fake news: Status threat does not explain the 2016 Presidential vote. Baltimore, MD : The Johns Hopkins University.</bibtext> </blist> <blist> <bibtext> Murnane, R. J., & Willett, J. B. (2011). Methods matter: Improving causal inference in educational and social science research. New York, NY : Oxford University Press.</bibtext> </blist> <blist> <bibtext> Mutz, D. C. (2018). Status threat, not economic hardship, explains the 2016 Presidential vote. Proceedings of the National Academy of Sciences, 115 (19), E4330 – E4339. doi: 10.1073/pnas.1718155115</bibtext> </blist> <blist> <bibtext> National Center for Education Statistics [NCES]. (2017). Digest of education statistics. Washington, DC : US Department of Education.</bibtext> </blist> <blist> <bibtext> Ness, E. C., Tandberg, D. A., & McLendon, M. K. (2014). Interest groups and state policy for higher education: New conceptual understandings and future research directions. In M. B. Paulsen (Ed.), Higher education: Handbook of theory and research (pp. 151 – 186). Dordrecht, The Netherlands : Springer.</bibtext> </blist> <blist> <bibtext> Ness, E. C., & Gándara, D. (2014). Ideological think tanks in the states: An inventory of their prevalence, networks, and higher education policy activity. Educational Policy, 28 (2), 258 – 280. doi: 10.1177/0895904813515328</bibtext> </blist> <blist> <bibtext> Ostfield, M. C. (2018). The new White flight? The effects of political appeals to Latinos on White Democrats. Political Behavior. doi: 10.1007/s11109-018-9462-8</bibtext> </blist> <blist> <bibtext> Paulsen, M. B., & Toutkoushian, R. K. (2008). Economic models and policy analysis in higher education: A diagrammatic exposition. In J. C. Smart (Ed.), Higher education: Handbook of theory and research (pp. 1 – 48). Dordrecht, The Netherlands : Springer.</bibtext> </blist> <blist> <bibtext> Pew Research Center. (2017). Sharp partisan divisions in views of national institutions. Washington, DC : The Pew Foundation.</bibtext> </blist> <blist> <bibtext> Pew Research Center. (2018a). Most Americans say higher ed is headed in wrong direction, but partisans disagree on why. Philadelphia, PA : Pew Foundation.</bibtext> </blist> <blist> <bibtext> Pew Research Center. (2018b). Wide gender gap, growing educational divide in voters' party identification. Washington, DC : The Pew Foundation.</bibtext> </blist> <blist> <bibtext> Pusser, B. (2004). Burning down the house: Politics, governance, and affirmative action at the University of California. Albany, NY : SUNY University Press.</bibtext> </blist> <blist> <bibtext> Pusser, B. (2018). The state and civil society in the scholarship of higher education. In B. Cantwell, H. Coates, & R. King (Eds.), Handbook on the politics of higher education (pp. 11 – 29). Cheltenham, UK : Edward Elgar Publishing.</bibtext> </blist> <blist> <bibtext> Rabe-Hesketh, S., & Skrondal, A. (2012). Multilevel and longitudinal modeling using Stata. College Station, TX : Stata Press.</bibtext> </blist> <blist> <bibtext> Rocha, R. R., & Matsubayashi, T. (2014). The politics of race and voter ID laws in the states: The return of Jim Crow? Political Research Quarterly, 67 (3), 666 – 679. doi: 10.1177/1065912913514854</bibtext> </blist> <blist> <bibtext> Rojas, F. (2007). From black power to black studies: How a radical social movement became an academic discipline. Baltimore, MD : The Johns Hopkins University.</bibtext> </blist> <blist> <bibtext> Schaffner, B. F., MacWilliams, M., & Nteta, T. (2018). Understanding White polarization in the 2016 vote for President: The sobering role of racism and sexism. Political Science Quarterly, 133 (1). doi: 10.1002/polq.12737</bibtext> </blist> <blist> <bibtext> Sides, J. (2017). Race, religion and immigration in 2016: How the debate over American identity shaped the election and what it means for a Trump Presidency. Washington, DC : The Democracy Fund, Voter Study Group. Retrieved from https://<ulink href="http://www.voterstudygroup.org/publications/2016-elections/race-religion-immigration-2016">www.voterstudygroup.org/publications/2016-elections/race-religion-immigration-2016</ulink></bibtext> </blist> <blist> <bibtext> Sides, J., Tesler, M., & Vavreck, L. (2018). Identity crisis: The 2016 Presidential campaign and the battle for the meaning of America. Princeton, NJ : Princeton University Press.</bibtext> </blist> <blist> <bibtext> Soss, J., Schram, S. F., Vartanian, T. P., & O'Brien, E. (2001). Setting the terms of relief: Explaining state policy choices in the devolution revolution. American Journal of Political Science, 45 (2), 378 – 395. doi: 10.2307/2669347</bibtext> </blist> <blist> <bibtext> State Higher Education Executive Officers Association [SHEEO]. (2018). State higher education finance: 2017. Denver, CO : SHEEO.</bibtext> </blist> <blist> <bibtext> Tandberg, D. A. (2009). Interest groups and governmental institutions: The politics of state funding of public higher education. Educational Policy, 24 (5), 735 – 778. doi: 10.1177/0895904809339163</bibtext> </blist> <blist> <bibtext> Tandberg, D. A. (2010). Politics, interest groups and state funding of public higher education. Research in Higher Education, 51 (5), 416 – 450. doi: 10.1007/s11162-010-9164-5</bibtext> </blist> <blist> <bibtext> Taylor, B. J., & Cantwell, B. (2019). Unequal higher education: Wealth, status and student opportunity. New Brunswick, NJ : Rutgers University Press.</bibtext> </blist> <blist> <bibtext> Toutkoushian, R. K., & Paulsen, M. B. (2016). Economics of higher education. Dordrecht, the Netherlands : Springer.</bibtext> </blist> <blist> <bibtext> Wilder, C. S. (2013). Ebony and ivy: Race, slavery, and the troubled history of America's universities. New York, NY : Bloomsbury Press.</bibtext> </blist> </ref> <aug> <p>By Barrett J. Taylor; Brendan Cantwell; Kimberly Watts and Olivia Wood</p> <p>Reported by Author; Author; Author; Author</p> </aug> <nolink nlid="nl1" bibid="bib20" firstref="ref1"></nolink> <nolink nlid="nl2" bibid="bib61" firstref="ref2"></nolink> <nolink nlid="nl3" bibid="bib73" firstref="ref3"></nolink> <nolink nlid="nl4" bibid="bib56" firstref="ref4"></nolink> <nolink nlid="nl5" bibid="bib51" firstref="ref5"></nolink> <nolink nlid="nl6" bibid="bib46" firstref="ref6"></nolink> <nolink nlid="nl7" bibid="bib25" firstref="ref7"></nolink> <nolink nlid="nl8" bibid="bib35" firstref="ref8"></nolink> <nolink nlid="nl9" bibid="bib44" firstref="ref13"></nolink> <nolink nlid="nl10" bibid="bib63" firstref="ref14"></nolink> <nolink nlid="nl11" bibid="bib14" firstref="ref17"></nolink> <nolink nlid="nl12" bibid="bib45" firstref="ref18"></nolink> <nolink nlid="nl13" bibid="bib55" firstref="ref19"></nolink> <nolink nlid="nl14" bibid="bib71" firstref="ref20"></nolink> <nolink nlid="nl15" bibid="bib11" firstref="ref21"></nolink> <nolink nlid="nl16" bibid="bib31" firstref="ref22"></nolink> <nolink nlid="nl17" bibid="bib49" firstref="ref23"></nolink> <nolink nlid="nl18" bibid="bib74" firstref="ref24"></nolink> <nolink nlid="nl19" bibid="bib75" firstref="ref25"></nolink> <nolink nlid="nl20" bibid="bib60" firstref="ref26"></nolink> <nolink nlid="nl21" bibid="bib15" firstref="ref27"></nolink> <nolink nlid="nl22" bibid="bib36" firstref="ref28"></nolink> <nolink nlid="nl23" bibid="bib47" firstref="ref29"></nolink> <nolink nlid="nl24" bibid="bib48" firstref="ref30"></nolink> <nolink nlid="nl25" bibid="bib21" firstref="ref33"></nolink> <nolink nlid="nl26" bibid="bib22" firstref="ref34"></nolink> <nolink nlid="nl27" bibid="bib23" firstref="ref35"></nolink> <nolink nlid="nl28" bibid="bib58" firstref="ref36"></nolink> <nolink nlid="nl29" bibid="bib57" firstref="ref37"></nolink> <nolink nlid="nl30" bibid="bib19" firstref="ref38"></nolink> <nolink nlid="nl31" bibid="bib28" firstref="ref39"></nolink> <nolink nlid="nl32" bibid="bib41" firstref="ref40"></nolink> <nolink nlid="nl33" bibid="bib38" firstref="ref41"></nolink> <nolink nlid="nl34" bibid="bib72" firstref="ref42"></nolink> <nolink nlid="nl35" bibid="bib64" firstref="ref43"></nolink> <nolink nlid="nl36" bibid="bib24" firstref="ref44"></nolink> <nolink nlid="nl37" bibid="bib67" firstref="ref46"></nolink> <nolink nlid="nl38" bibid="bib43" firstref="ref49"></nolink> <nolink nlid="nl39" bibid="bib32" firstref="ref50"></nolink> <nolink nlid="nl40" bibid="bib30" firstref="ref55"></nolink> <nolink nlid="nl41" bibid="bib52" firstref="ref56"></nolink> <nolink nlid="nl42" bibid="bib39" firstref="ref65"></nolink> <nolink nlid="nl43" bibid="bib34" firstref="ref80"></nolink> <nolink nlid="nl44" bibid="bib69" firstref="ref82"></nolink> <nolink nlid="nl45" bibid="bib70" firstref="ref83"></nolink> <nolink nlid="nl46" bibid="bib16" firstref="ref89"></nolink> <nolink nlid="nl47" bibid="bib68" firstref="ref91"></nolink> <nolink nlid="nl48" bibid="bib62" firstref="ref93"></nolink> <nolink nlid="nl49" bibid="bib78" firstref="ref94"></nolink> <nolink nlid="nl50" bibid="bib29" firstref="ref95"></nolink> <nolink nlid="nl51" bibid="bib59" firstref="ref96"></nolink> <nolink nlid="nl52" bibid="bib50" firstref="ref99"></nolink> <nolink nlid="nl53" bibid="bib37" firstref="ref101"></nolink> <nolink nlid="nl54" bibid="bib54" firstref="ref104"></nolink> <nolink nlid="nl55" bibid="bib42" firstref="ref105"></nolink> <nolink nlid="nl56" bibid="bib10" firstref="ref109"></nolink> <nolink nlid="nl57" bibid="bib33" firstref="ref110"></nolink> <nolink nlid="nl58" bibid="bib27" firstref="ref111"></nolink> <nolink nlid="nl59" bibid="bib13" firstref="ref112"></nolink> <nolink nlid="nl60" bibid="bib76" firstref="ref114"></nolink> <nolink nlid="nl61" bibid="bib17" firstref="ref115"></nolink> <nolink nlid="nl62" bibid="bib66" firstref="ref116"></nolink> <nolink nlid="nl63" bibid="bib26" firstref="ref127"></nolink> <nolink nlid="nl64" bibid="bib40" firstref="ref130"></nolink> <nolink nlid="nl65" bibid="bib18" firstref="ref131"></nolink> <nolink nlid="nl66" bibid="bib77" firstref="ref133"></nolink> <nolink nlid="nl67" bibid="bib65" firstref="ref134"></nolink> <nolink nlid="nl68" bibid="bib12" firstref="ref138"></nolink>
Header DbId: eric
DbLabel: ERIC
An: EJ1264420
AccessLevel: 3
PubType: Academic Journal
PubTypeId: academicJournal
PreciseRelevancyScore: 0
IllustrationInfo
Items – Name: Title
  Label: Title
  Group: Ti
  Data: Partisanship, White Racial Resentment, and State Support for Higher Education
– Name: Language
  Label: Language
  Group: Lang
  Data: English
– Name: Author
  Label: Authors
  Group: Au
  Data: <searchLink fieldCode="AR" term="%22Taylor%2C+Barrett+J%2E%22">Taylor, Barrett J.</searchLink><br /><searchLink fieldCode="AR" term="%22Cantwell%2C+Brendan%22">Cantwell, Brendan</searchLink><br /><searchLink fieldCode="AR" term="%22Watts%2C+Kimberly%22">Watts, Kimberly</searchLink><br /><searchLink fieldCode="AR" term="%22Wood%2C+Olivia%22">Wood, Olivia</searchLink>
– Name: TitleSource
  Label: Source
  Group: Src
  Data: <searchLink fieldCode="SO" term="%22Journal+of+Higher+Education%22"><i>Journal of Higher Education</i></searchLink>. 2020 91(6):858-887.
– Name: Avail
  Label: Availability
  Group: Avail
  Data: Routledge. Available from: Taylor & Francis, Ltd. 530 Walnut Street Suite 850, Philadelphia, PA 19106. Tel: 800-354-1420; Tel: 215-625-8900; Fax: 215-207-0050; Web site: http://www.tandf.co.uk/journals
– Name: PeerReviewed
  Label: Peer Reviewed
  Group: SrcInfo
  Data: Y
– Name: Pages
  Label: Page Count
  Group: Src
  Data: 30
– Name: DatePubCY
  Label: Publication Date
  Group: Date
  Data: 2020
– Name: TypeDocument
  Label: Document Type
  Group: TypDoc
  Data: Journal Articles<br />Reports - Research
– Name: Audience
  Label: Education Level
  Group: Audnce
  Data: <searchLink fieldCode="EL" term="%22Higher+Education%22">Higher Education</searchLink><br /><searchLink fieldCode="EL" term="%22Postsecondary+Education%22">Postsecondary Education</searchLink>
– Name: Subject
  Label: Descriptors
  Group: Su
  Data: <searchLink fieldCode="DE" term="%22Politics+of+Education%22">Politics of Education</searchLink><br /><searchLink fieldCode="DE" term="%22Higher+Education%22">Higher Education</searchLink><br /><searchLink fieldCode="DE" term="%22Racial+Bias%22">Racial Bias</searchLink><br /><searchLink fieldCode="DE" term="%22Racial+Attitudes%22">Racial Attitudes</searchLink><br /><searchLink fieldCode="DE" term="%22Whites%22">Whites</searchLink><br /><searchLink fieldCode="DE" term="%22Educational+Policy%22">Educational Policy</searchLink><br /><searchLink fieldCode="DE" term="%22State+Policy%22">State Policy</searchLink><br /><searchLink fieldCode="DE" term="%22Disproportionate+Representation%22">Disproportionate Representation</searchLink><br /><searchLink fieldCode="DE" term="%22State+Government%22">State Government</searchLink><br /><searchLink fieldCode="DE" term="%22Educational+Finance%22">Educational Finance</searchLink><br /><searchLink fieldCode="DE" term="%22State+Aid%22">State Aid</searchLink><br /><searchLink fieldCode="DE" term="%22Political+Attitudes%22">Political Attitudes</searchLink>
– Name: DOI
  Label: DOI
  Group: ID
  Data: 10.1080/00221546.2019.1706016
– Name: ISSN
  Label: ISSN
  Group: ISSN
  Data: 0022-1546
– Name: Abstract
  Label: Abstract
  Group: Ab
  Data: Dominant explanations of state higher education policy tend to emphasize economic models that foreground the business cycle or political approaches that cast ideology as fairly fixed. We instead foreground changing social context to conceptualize state appropriations as predicted not only by these classic explanations, but also by the interplay of racial representation and political party control. Drawing on the racial backlash hypothesis and quantitative analyses, we show that party control of state government and racial representation in higher education jointly explain state appropriations. Unified Republican governments spent more than Democratic or divided governments when White students were overrepresented. Republicans spent less otherwise. These results suggest that partisan attitudes toward racial representation in higher education may shape state government support for colleges and universities.
– Name: AbstractInfo
  Label: Abstractor
  Group: Ab
  Data: As Provided
– Name: DateEntry
  Label: Entry Date
  Group: Date
  Data: 2020
– Name: AN
  Label: Accession Number
  Group: ID
  Data: EJ1264420
PLink https://search.ebscohost.com/login.aspx?direct=true&site=eds-live&db=eric&AN=EJ1264420
RecordInfo BibRecord:
  BibEntity:
    Identifiers:
      – Type: doi
        Value: 10.1080/00221546.2019.1706016
    Languages:
      – Text: English
    PhysicalDescription:
      Pagination:
        PageCount: 30
        StartPage: 858
    Subjects:
      – SubjectFull: Politics of Education
        Type: general
      – SubjectFull: Higher Education
        Type: general
      – SubjectFull: Racial Bias
        Type: general
      – SubjectFull: Racial Attitudes
        Type: general
      – SubjectFull: Whites
        Type: general
      – SubjectFull: Educational Policy
        Type: general
      – SubjectFull: State Policy
        Type: general
      – SubjectFull: Disproportionate Representation
        Type: general
      – SubjectFull: State Government
        Type: general
      – SubjectFull: Educational Finance
        Type: general
      – SubjectFull: State Aid
        Type: general
      – SubjectFull: Political Attitudes
        Type: general
    Titles:
      – TitleFull: Partisanship, White Racial Resentment, and State Support for Higher Education
        Type: main
  BibRelationships:
    HasContributorRelationships:
      – PersonEntity:
          Name:
            NameFull: Taylor, Barrett J.
      – PersonEntity:
          Name:
            NameFull: Cantwell, Brendan
      – PersonEntity:
          Name:
            NameFull: Watts, Kimberly
      – PersonEntity:
          Name:
            NameFull: Wood, Olivia
    IsPartOfRelationships:
      – BibEntity:
          Dates:
            – D: 01
              M: 01
              Type: published
              Y: 2020
          Identifiers:
            – Type: issn-print
              Value: 0022-1546
          Numbering:
            – Type: volume
              Value: 91
            – Type: issue
              Value: 6
          Titles:
            – TitleFull: Journal of Higher Education
              Type: main
ResultId 1