Abstract
In this article, we use original survey data from Wisconsin school board members to determine the extent of conflict present on Wisconsin school boards and the determinants of different types of perceived board conflict. School boards, by virtue of overseeing a public good with readily measurable performance variables, offer an ideal case for testing a relationship between board governance performance and conflict. Using multivariate regression analysis, we find that that the perceived level of conflict on Wisconsin school boards is dependent on the quality of superintendent–board relations, the demographic profile of the school district overseen by the school board, and the age and experience level of school board members. Notably, we do not find a clear relationship between conflict and measures of school district academic performance.
Introduction
Though the study and practice of governance over the delivery of publicly funded services has evolved dramatically since the founding of the field of public administration, the ultimate goal of governance remains the maximization of the performance of public organizations (Osborne, 2010; Stillman, 1999). This goal, however, often raises more questions than it answers. What, for example, constitutes a public organization? Is it an organization subject to government oversight, an organization that is 100% publicly funded, an organization that broadly serves the public, or something else? It is even more challenging to clearly articulate the performance goals of public organizations. Frederickson (1990), Stillman (1999), and others argue that public sector performance measures should include equity, democratic governance, and bureaucratic representation, all of which could arguably come at the expense of efficiency. Conversely, Kettl (2005), Osborne and Gaebler (1993), and Savas (2000) all present frameworks where efficiency, as measured through hard spending and outcome measures, is the ultimate goal of public governance. In this article, we build on the evolving international public governance and performance literature with a critical case analysis of public governance boards in the United States. The analysis brings new evidence to further explain what others have called the “black box” of public management (Andrews and Boyne, 2010: 443). By exploring the behaviors and preferences of actors actually governing public organizations, in other words by examining what goes on in the “black box,” we add context to the broader questions surrounding public organizational structures and goals.
Specifically, our paper examines the determinants of conflict on democratically elected school boards in one American state. The American school board has a long track record as a core public American institution (Alsbury, 2008). In its most romantic form, a school board is the ultimate exercise in local control and democracy; citizens of communities win elections for unpaid positions to gather with other citizens to decide how best to educate the children in their community. The lack of pay ensures people view the position as public service; the desire to do good outweighs the desire for power. But the reality of American school board governance is not so romantic.
Paul Peterson’s (1976) case study of the Chicago school board demonstrated how politics and dysfunction can grip a school board in the same manner as any other political board. More recently, Miller (2008) has questioned the very need for democratically elected school boards. Indeed, sometimes board members disagree on the appropriate direction of a district; both sides convinced they are right. Sometimes, board members form coalitions. Sometimes, board members simply do not like one another. Sometimes, officially nonpartisan board elections are influenced by partisan interest groups (Hess, 2008). Sometimes, board members have radically different constituencies and agendas, i.e. taxpayers versus public employees. Simply put, school boards are comprised of small groups of people exhibiting the same group dynamics as other small groups with a governing responsibility.
One particular group dynamic appearing frequently in the public administration literature is conflict. A variety of scholars across disciplines and eras, including Coser (1956), Deutsch (1973), and Grissom (2010, 2014) have theorized and tested the potential causes and effects of various types of group conflict. More recently, a body of survey-based studies exploring the determinants and effects of group conflict on municipal government bodies has emerged in the United States (Heidbreder et al., 2011; Ihrke and Scott Niederjohn, 2005; Nelson and Nollenberger, 2011). In general, conflict has been shown or assumed, with limited exception, to have a negative impact on the overall performance of public governing boards (Grissom, 2014). Given the generally negative effects of conflict on governing boards, understanding the reasons why conflict appears on boards can provide actionable information that board members can use to lower conflict and improve performance.
However, a notable limitation in the existing literature on municipal board conflict is the absence of clear performance measures in models predicting the extent and type of conflict on boards. Simply, there are few intuitive quantitative measures of organizational performance in many public organizations overseen by elected boards. It is also difficult to determine variation in the level of challenges faced across common types of boards. For example, the governing task of a school board overseeing a school district with a high-income population from well-educated families is arguably less challenging than a board overseeing a school district with a generally low-income population from less education households. Scholars studying board conflict thereby face the basic challenge of identifying the proper variables by which to measure the difficulty of the governance task faced by boards. We theorize that the extent of the challenges faced by boards is likely to influence the level of conflict on a board; if a board has limited governing challenges there are less issues over which to conflict.
In addition, the case of Wisconsin school boards provides a unique opportunity to incorporate the role of organizational performance and challenges in models predicting conflict. Public school districts must publicly report extensive performance and demographic data that can be used to paint a detailed portrait of the governance task faced by elected boards overseeing one of the United States’ almost 14,000 school boards. Wisconsin, the focus of this article, has recently improved its metrics for measuring school district performance as part of its new school and district report card system (Richards, 2013). This system assigns each Wisconsin school and each of Wisconsin’s 424 school districts an accountability score between 1 and 100. The score takes into account student math and reading scores, growth in student math and reading scores over time, student attendance, school and district progress in closing racial and socioeconomic achievement gaps, and measures of student progress toward graduation.
In this article, we combine data from Wisconsin’s new education accountability framework with original survey data collected from Wisconsin school board members to test if the determinants of conflict found in previous case studies of municipal boards are present on Wisconsin school boards, and to consider the role performance and student characteristic variables play in determining the perceived levels of conflict on Wisconsin school boards. The results yield new knowledge on the connections between organizational performance, the challenges faced by boards, and conflict on local governance boards.
Literature review
Stephen Osborne (2010) categorizes the evolution of public governance reforms into three phases. The first phase, Public Administration, focused on bureaucracy, the separation of politics and administration, and professionalism. The second phase, New Public Management, infused private sector management principles into public management, emphasized entrepreneurism, and encouraged the use of market forces and networks in the delivery of public goods. The third and most modern phase, New Public Governance, is a more sprawling phrase that encompasses and builds upon core values of phases one and two. Specifically, New Public Governance is concerned with the fundamentals, architecture, sustainability, values, relational skills, accountability, and evaluation mechanisms pertaining to public service delivery (Osborne, 2010: 11). Inherent in this framework is the importance of the individuals actually involved in governing the delivery of public services, as their policy preferences and professional interactions will impact the ways in which, and quality in which, the key concerns of New Public Governance are addressed. Presumably, the degree to which the key areas of New Public Governance are adequately and appropriately addressed by a board or manager affects overall organizational performance.
Colin Talbot (2009) offers public value (PV), which has relevance to our study, as the next “big thing” in public management. Like Osborne (2010), Talbot attempts to examine the evolution of public management and suggests that PV as a lens has the “advantage of looking simultaneously forward and backward” (169). In looking backward, PV not only includes managerialism, efficiency, and performance that are found in New Public Management but also encompasses the emphasis on trust and legitimacy of traditional public administration. In looking forward, PV explores new forms of governance, networks of policy and implementation, and more agile public services, as indicated by Stoker (2005).
In a previous study, the authors (2015) used the term micro-governance to describe the day-to-day processes of boards overseeing the local delivery of a public service or good. Andrews and Boyne (2010) similarly build on black box theories of public management to determine the relationship between organizational capacity, leadership, and local government performance, concluding that public organizational leadership and capacity are indeed linked to performance. Moynihan (2005), in a study of American state agencies, demonstrates how implementation theory is still very much relevant in explaining the performance of public organizations. According to Moynihan (2005), “group[s] act in a way consistent with [their] motivations, norms, and interests” (p. 221). In summary, the ways in which governance actors, i.e. school board members, answer the questions of New Public Governance can be expected to impact public performance.
A body of research by Gabris and Nelson (2013), Gabris et al. (2001), Ihrke et al. (2003), and Nelson et al. (2011) give insight into the group characteristics that are most conducive to answering key governance questions, such as those proposed by Osborne (2010). Generally, boards that minimize destructive conflict create regenerative relationships, exhibit collegial relations with each other and the board executive, and exhibit a high degree of cooperation which can be expected to perform better than those that do not. Conflict, however, should not inherently be assumed to be negative. The sociologist Lewis Coser (1956) laid out a basic framework for understanding the positive and negative functions of social conflict in small groups, arguing that not all conflict is bad, that in fact conflict can often be a way for people to blow off steam, say in an argument, without damaging the quality of work at the task at-hand. Over 40 years after Coser, Jehn (1997, 1999) engaged in extensive qualitative research that identified three specific types of conflict: relationship, task, and process. In the school board context, an example of process conflict might be the means by which a board determines their board president. An example of task conflict might be a disagreement over which superintendent to hire. An example of relationship conflict might be a case of two board members that simply do not get along.
The three conflict types identified by Jehn were shown by Heidbreder et al. (2011) and Ihrke and Scott Niederjohn (2005) to have slightly different impacts on city councils in Michigan and Wisconsin. Heidbreder et al. (2011) found that conflict in Michigan municipalities resulted from the demographics of a municipality, the structure of city government, the demographics of municipal board members and executives, and the perceived importance of specific policies. Ihrke and Scott Niederjohn (2005), in a survey of Wisconsin municipalities, found that conflict is predicted by board member age, the leadership credentials of the executive, and the specific behavior of board members. Notably, Ihrke and Scott Niederjohn (2005) do not find a relationship between government structure and conflict; this finding runs counter to the arguments of Svara (1990) and the findings of Nelson and Nollenberger (2011).
The single known study of the determinants of school board conflict in California found that school district location, the percentage of minority pupils in a school district, urban location, the interest group environment, and measures of professionalism and goal coherence influence board conflict (Grissom, 2010). The study did not, however, consider measures of district performance. Other scholars have more generally emphasized the importance of good relations between school boards and hired superintendents (Bjork, 2008; Mountford, 2008; Peterson and Fusarelli, 2008). Mary Delagardelle (2008) has also found a connection between board governance and student achievement. Though she does not explicitly discuss the role of conflict on school boards Delagardelle does stress the importance of agreement over common goals on effective school boards.
Overall, the existing literature on local elected government boards and conflict inform our approach to measuring determinants of conflict on elected school boards (Heidbreder et al., 2011; Ihrke and Scott Niederjohn, 2005). Accordingly, we build on this existing body of research by considering the connection between measures of school district performance and demographics, in addition to variables measuring role of board structure, superintendent–board relations, and board member demographics, and school board conflict.
Why Wisconsin?
The determinants of conflict on governing boards are of general interest to practitioners, as evidenced by the routine inclusion of recommendations for minimizing group conflict in the governance best practice literature (see Carver, 2006; Houle, 1997; Smoley, 1999; Walser, 2009), and academics, as evidenced by the previously discussed literature on determinants of board conflict (Grissom, 2010; Heidbreder et al., 2011; Ihrke and Scott Niederjohn, 2005; Johnson and Ihrke, 2004; Nelson and Nollenberger, 2011). However, what is it about the case of Wisconsin that makes it worthy of analysis? In and of themselves, Wisconsin school board members are not particularly unique in comparison to school board members serving other U.S. states. However, the use of Wisconsin school board members to further understanding of the nature of conflict and board governance makes it a critical case that adds value to the public administration literature.
John Gerring (2004) defines a case study as “an intensive study of a single unit for the purpose of understanding a larger class of (similar) units” (p. 342). In this analysis, Wisconsin school boards serve as one of a diverse group of public governing boards. The group dynamics literature, including Gabris and Nelson (2013), Golembiewski (1995), and Nelson et al. (2011) argues that governance itself can be used to improve public performance, and that one important way to do this is the reduction of board conflict. For this argument to be valid, common findings relating to governance must hold across cases.
Consider that the particular subjects of conflict are inherently different on a school board and municipal board. A municipal board would never conflict over student curriculum just like a school board would never conflict over a function like public works. The key argument is that the dynamic of conflict, and the not the subject of that conflict, is the key variable. Thus far, case studies relating to conflict on city councils in Michigan and Wisconsin (Heidbreder et al., 2011; Ihrke and Scott Niederjohn, 2005), medium-sized municipalities in the United States (Nelson and Nollenberger, 2011), and California school boards (Grissom, 2010) have all shown how board member characteristics, relations with the executive, and population served, connect to conflict. The purpose of our research is to demonstrate in another arena, and one which performance measures are considered, that the determinants of conflict are consistent (Gerring, 2004, 2007).
In other words, the case of Wisconsin school boards is critical in establishing that the dynamics of governing boards are predicable across governance functions and geographic locations (Gerring, 2007). We chose to specifically study the case of Wisconsin school boards because they are similar in structure but different in function to the municipal governance boards studied by Ihrke and Scott Niederjohn (2005) and others, and similar in function and structure but different in location from the boards studied by Grissom (2010). While the Wisconsin case alone is not generalizable across public governing boards, we believe its link to a growing body of diverse research on board conflict makes it a critical case in establishing generalizable knowledge (Seawright and Gerring, 2008).
Data and survey description
Our survey instrument was developed and fine-tuned over approximately nine months, from April 2012 to February 2013. The original draft survey contained over 130 questions and was gradually reduced to 39 questions. The survey questions were adapted from a previous survey conducted by the National School Boards Association (Hess and Meeks, 2011), and a Michigan municipal governance survey (Heidbreder et al., 2011). The unit of analysis targeted by the survey was individual school board members. Some survey questions were worded to provide basic information about board members (i.e. sex, ideological leanings, years of service), while others were worded to obtain information from individual school board members on their perceptions of the board as a whole. A similar strategy was used in previous school board surveys by Grissom (2010) and Hess and Meeks (2011). Given this unit of analysis, the goal of the survey was to maximize the number of individual school board member respondents. We note that in our analysis we use both board member-level data and aggregated board-level data. The use of board member-level data, which was used in similar work by Heidbreder et al. (2011), Ihrke and Scott Niederjohn (2005), and others, allows for the inclusion of board member characteristics in our models. However, averaging out the responses of members serving together on a single board, as done by Grissom (2010), provides a more intuitive approach to understanding board-level conflict.
Surveys were sent electronically to the population of school board members in Wisconsin with publicly listed e-mail addresses. A total of 1379 school board members representing Wisconsin school districts received a survey. The response rate was 23.3% for individual board members; however, responses rates did vary somewhat by question. The response rate was similar to the 23.6% response rate obtained in a national school board member survey conducted by Hess and Meeks (2011). As a check against possible response bias, the demographic and performance characteristics of districts represented in our sample were compared against the population of school districts in Wisconsin. The comparison showed both groups had similar demographics and performance indicators.
As can be seen in Tables 1 and 2, all variables included in the model come from one of three sources:
the previously described survey instrument; the Wisconsin Department of Public Instruction; Continuous variable summary statistics and source DPI: Department of Public Instruction; NCES: National Center for Education Statistics. Categorical variable summary statistics and source
Many of the survey questions used in the analysis were measured on a 5-point Likert scale where 1 equals no agreement or description and 5 equals total agreement or description. The variables measured on the Likert scale are identified in Table 1.
Dependent variable summary statistics
Conflict index variables
Alpha coefficient: 0.87 for members, 0.82 for board level data.
Hypotheses and methods
Five hypotheses on the determinants of conflict on Wisconsin school boards are tested in this paper. Those hypotheses, as well as our rationale for testing them, are discussed below.
Hypothesis 1: School boards overseeing lower performing school systems are more prone to conflict.
Lewis Coser (1956) posits that in-group conflict can be particularly intense because individuals who closely identify with a group naturally have a strong sense of ownership over the group. A natural outgrowth from Coser is the proposition that people serving on boards overseeing organizations with a demonstrated deficiency or need for performance improvement will experience particularly intense conflict with board members who have differing ideas of how best to improve organizational performance. The particularly entrenched and politicized problem of low academic achievement levels in school districts makes it plausible to think that school boards with low levels of achievement are prone to intense levels of conflict (Delagardelle, 2008; Peterson, 1976).
In addition, there is a considerable debate over the ability of public school systems to overcome the problems of poverty that exist outside the walls of schools. From the point of view of school officials, these problems are intractable (Ravitch, 2013). If indeed many of the pressing problems within school districts cannot actually be adequately addressed by school officials, a certain level of frustration from school board members is plausible. It is equally plausible that this frustration manifests as board conflict.
Hypothesis 2: Stability as indicated by low levels of board turnover lowers perceived levels of conflict.
The longer a group maintains consistent membership the more time there is for the group to develop established norms, established channels for conflict resolution, become comfortable with each other, and able to be productive through acceptance of each other’s strengths and weaknesses (Deutsch, 1973; Houle, 1997). In addition to a growing acceptance with each other over time, a stable board also, by virtue of the same members being continuously reelected, should enjoy a level of acceptance from the public at large (Lutz and Iannoccone, 2008). Accordingly, we think that board members indicating a low level of turnover on their board also experience low levels of conflict.
Hypothesis 3: Positive board member–superintendent relations are an indicator of low conflict.
Arguably the most important thing a school board does is hire a superintendent to implement its vision (Bjork, 2008; Eliot, 1959; Viteritti, 2008). Poor relations between a board and its superintendent likely means that a board is either in disagreement over what it wants from its superintendent, or paralyzed to the extent that it cannot take decisive action to replace an ineffective leader. Either way, there is good reason to suspect that poor board–executive relations on school boards have a similar effect on conflict as poor board–executive relations do on other municipal boards.
Hypothesis 4: Ideological diversity increases levels of conflict.
Paul Peterson’s (1976) case study of the Chicago school board demonstrates the explosive nature of school board politics. Indeed, the differing education policy preferences and priorities on the liberal and conservative side of the political aisle are well documented in both scholarly works and the editorial pages of America’s newspapers (Ravitch, 2013). It is logical to suspect that the partisan political debates of a state as ideologically diverse as Wisconsin will lead to conflict over policy preferences on school boards.
Hypothesis 5: Boards overseeing more challenging student populations exhibit higher levels of perceived conflict.
Despite years of trying there remains little consensus among educators and advocates on how to close stubborn racial and socioeconomic achievement gaps in the United States’ public education systems (Hanushek et al., 2013; Kozol, 1991, 2005; Ravitch, 2011, 2013). This lack of agreement in the education community can conceivably result in conflict on school boards exhibiting a similar lack of agreement over the best way to proceed with a challenging student population. In addition, boards overseeing districts with higher percentages of special needs and low-income students are forced to make policy decisions that are more of a social service nature rather than an education nature. We suspect, in part due to the previously mentioned findings of Grissom (2010), that the larger scope of decisions faced by board members overseeing more challenging students present more opportunities for board conflict.
All five of these hypotheses are tested using multivariate regression models with different measures of conflict as the dependent variable. The same methodology was used in similar studies of conflict in municipalities (Heidbreder et al., 2011; Ihrke and Scott Niederjohn, 2005; Nelson and Nollenberger, 2011). All four models include several focal variables of interests, and a set of board and board member characteristic variables, and school district characteristic variables. Lastly, diagnostic tests were run on each model during the specification process and no significant multicollinearity or heteroscedasticity issues were found.
Results
Multivariate OLS regression results: Board members
Standard errors in parentheses.
p < 0.05, **p < 0.01, ***p < 0.001.
Multivariate OLS regression results: Board averages
Standard errors in parentheses.
p < 0.05, **p < 0.01, ***p < 0.001.
Models 1 and 5 predict the level of total perceived conflict on a school board as measured by our conflict index variable. Both models reveal a negative relationship between viewing the superintendent as a governing partner, and having a higher percentage of students designated as special needs, and the conflict index. Having a higher percentage of minority students is positively related to total conflict in both the board member and board-level models. Models 2 and 3, which predict the perception of personalized disagreements and predictable coalitions, respectively, reveal findings consistent with the total conflict index. Viewing the superintendent as a governing partner reduces perceived personalized disagreements and predictable coalitions, while serving a higher number of minority pupils increases both personalized disagreements and predictable coalitions. Serving a higher percentage of special needs students reduces the perception of predictable coalitions but has no significant relationship with perceived personalized disagreements. The board-level model predicting the perception of predicable coalitions, Model 7, yields similar findings to the board-member level model. However, Model 6 shows that no relationships exist between our focal dependent variables and perceptions of personalized disagreement when aggregate-level data is used. Model 6 also has the weakest predicative power of all our models.
Models 4 and 8 predict the ability of a board to rally around the flag and move on from the decision process by supporting decisions once they are made. The findings in Model 4 are consistent with Models 1–3. Viewing the superintendent as a governing partner and overseeing a district with higher percentages of special needs students increases the likelihood of a board supporting decisions once they are made. When board-level aggregate data are used the results are similar, with the notable exception of the report card score having a negative relationship with a board supporting decisions once they are made. However, the substantive significance of this counterintuitive finding is minimal, as a full unit increase in the report card score results in just a .068 reduction in the Likert scale measured dependent variable.
Several of the control variables used in our models also reveal relationships of interest with the dependent conflict variables. Board member age has a positive relationship with conflict in Models 1, 3, and 7, meaning older board members are likely to have higher perceptions of total conflict and the formation of predictable coalitions. Board member experience, in contrast, has a negative relationship with conflict; board members with more than three years of experience have lower levels of total conflict and personalized disagreement and are more likely to serve on boards that support decisions once they are made. Finally, district size is a statistically significant predictor of conflict; the larger the district the lower the levels of perceived total conflict and the presence of predictable coalitions. However, the coefficients for district size are extremely small making this relationship of little substantive importance.
Discussion and conclusions
In the following section, we revisit our hypotheses and discuss the implications of our findings. First, we find little evidence to support the hypothesis that school boards overseeing lower performing school systems are more prone to conflict. A district’s overall accountability score was a substantively small predictor of just one of the dependent variables in just one of the models. The absence of a relationship was surprising to us and a clear indication that the nature of the conflict–performance relationship needs further examination in future research. It is possible, and somewhat supported in our findings, that the district racial and socioeconomic variables commonly found to predict academic performance in school districts are the focus of board debates and conflicts (Hanushek, 1997). In other words, boards may conflict over the root cause of low district academic performance rather than the low performance itself.
Our second hypothesis, that stability as indicated by low levels of board turnover lowers perceived levels of conflict, is also not supported by our analysis of Wisconsin school board members. There is no significant relationship between having low board turnover and indicators of perceived conflict. This finding is perhaps an indicator that a high degree of board stability can be a double-edged sword. Whilst, it may allow for the establishment of norms, but it may also enable the formation of predictable coalitions. Interestingly, there is a significant relationship between board members having three or more years of experience and lower levels of perceived total conflict and personalized disagreements. So while overall board stability is not necessarily a net positive or negative, having a board consisting of more experienced members is likely to have a positive influence on board relations.
Hypothesis three, that positive perceived board member–superintendent relations are an indicator of low conflict, is well supported. There is a relationship between viewing the superintendent as a partner and lowering perceived levels of negative conflict types in all but one of the presented models. This finding is consistent with the municipal governance literature’s findings on the link between good executive–board relations and low conflict (Ihrke and Scott Niederjohn, 2005). Clearly the most obvious thing a school board seeking to lower its levels of board conflict should do is address any strains in, and make efforts to improve, the board–superintendent relationship.
Our fourth hypothesis, that ideological diversity increases levels of conflict, is not supported by our findings. The perceived level of shared political beliefs on Wisconsin school boards has no significant relationship with conflict. Similarly, board member self-identification as either liberal or conservative has no significant relationship with any of our conflict variables. The lone significant finding in regards to ideology is that board members who self-identify as moderate are less likely to support board decisions once they are made. This could be evidence that most boards tend to swing to one ideological side or the other leaving moderates to struggle with accepting any ideological board position. More likely, the minimal evidence of a link between board member ideology and conflict is a sign that Wisconsin’s nonpartisan school board election format is successful at minimizing ideological conflict on school boards. The finding is also consistent with local government literature showing local office holders, like school board members, tend not to seek higher partisan political office (Sokolow, 1989).
Our fifth and final hypothesis, that boards overseeing more challenging student populations exhibit higher levels of perceived conflict, is supported with a significant caveat. Overall, there is a statistically significant relationship between the percentage of minority students served by a school district and negative conflict types: The higher the percentage of minority pupils, the higher the perceived level of board conflict. However, boards overseeing districts with higher percentages of special needs pupils actually exhibit lower levels of conflict. These findings beg the question; why would one set of challenging students increase conflict while another decrease conflict?
We believe the answer has to do with the existence of clearly outlined federal laws and procedures for special needs students in American states. Under federal law, pupils suspected of having special needs are evaluated, and, if the existence of special needs is confirmed, given an Individual Education Plan (IEP) in consultation with their parents and teachers. Annually that IEP is evaluated and modified as necessary, and a variety of data regarding student performance and placement is reported as required under federal law. No doubt school boards face challenging decisions regarding the hiring of special needs teachers and proper placement of resources for special needs pupils, but there are a clear set of guidelines for the logistics of serving special needs students that are created outside of the authority of the school board. Though a board may disagree with state and federal special education policies, they are unlikely to conflict with each other over these disagreements with an outside body. In contrast, no clear set of policies from the state or federal government dictates how school boards are to deal with problems of racial achievement gaps in schools. As discussed earlier there is clear disagreement among policy makers on how best to deal with low minority achievement in Wisconsin (and elsewhere), making this a likely area for school board members to conflict.
Finally, we note that the presented model with the lowest explanatory power, Model 6, yielded no statistically significant relationships. That model used board level-aggregate data to predict personalized disagreement and does not include any board member characteristics. This may mean that relationship conflict is a function of the personalities serving on a board rather than the overall board structures and dynamics. Future research on relationship conflict and the personal characteristics of members of public governing boards could yield additional insight into this finding.
In general, we find that the best course of action for Wisconsin school boards looking to lower levels of board conflict is to make efforts to improve board–superintendent relations. While other factors, most notably the percentage of minority pupils and age and experience of board members, influence perceived levels of board conflict, there is little a board can do to change the makeup of their student body and age and experience levels of their fellow board members. However, boards can take action, including engagement in joint strategic planning and/or board–superintendent development exercises, to cultivate the relationship between board members and the superintendent. While some of these steps may seem simple, the creation of safe forums in which a partnership between the board and the executive can be fostered has potential, as demonstrated, to reduce board conflict and improve performance (Grissom, 2014).
We began this article with a discussion of the public performance literature and its increasing focus on the need to better understand what goes on in the black box of public organizational governance. We argue that the modern-day public governance challenges articulated by Osborne (2010) and others can only be met if researchers can make better sense of how the people and processes involved in public governance impact its effectiveness. The presented case of Wisconsin school board members adds to existing scholarship by incorporating organizational performance variables, and indicators of the extent of a board’s governance challenges in models predicting board conflict. More importantly, this study adds a new case study that builds on the existing diverse body of literature studying the determinants of board conflict. This new case, when connecting to those in other functional areas and jurisdictions, moves the literature closer to a place where scholars can separate the generalizable aspects of governance in public organizations from those that are function and place specific. Future research should focus on whether our findings hold true in other states, particularly those with structural and demographic differences with Wisconsin. In addition, a greater understanding of the relationship between small group dynamics on school boards and measurable academic performance can be obtained by further questioning the relationship (or absence of) between conflict and performance. Does conflict impact performance rather than the other way around? Do school boards focus on the root causes of poor performance rather than the performance itself? Are key measures of academic performance missing something? Answering these questions will further strengthen the findings presented here, and further the ability of public organizations to take informed steps to improve their governance behaviors.
Footnotes
Declaration of conflicting interests
The author(s) declared no potential conflicts of interest with respect to the research, authorship, and/or publication of this article.
Funding
The author(s) received no financial support for the research, authorship, and/or publication of this article.