Teacher Work Motivation in the Era of Extrinsic Incentives : Performance Goals and Pro-social Commitments in the Service of Equity

Mindful of the withering of high-stakes accountability and disappointing data from pay for performance evaluations in the US, we ask why management by extrinsic incentives and organizational goal setting may have been far less powerful than designers of accountability and extrinsic incentive systems had expected. We explore how system-generated motives (e.g., attaining specific organizational goals, preventing sanctions, or garnering rewards) stack up against autonomously generated, intrinsic, or service motives.? We found through both quantitative and qualitative data that for teachers in the charter schools a constellation of public service motives predominated: diffuse pro-social commitments, ideologies of fairness and equity, a belief in the moral deservingness of deprived student populations in opposition to societal neglect, and identification with one’s work as a personal calling. By comparison, monetary rewards were embraced as already deserved. Neither rewards, nor accountability, seemed to regulate behavior in a deep way. Prestige Education Policy Analysis Archives Vol. 25 No. 44 was not bestowed by official performance statuses within the accountability system, but flowed from judgments, personally communicated, by students, parents, or colleagues who had direct contact with teachers’ work.


Introduction
At the end of the year 2015, after 14 years in existence, the federal No Child Left Behind accountability system became defunct in the United States.While the successor law, the Every Student Succeeds Act or ESSA (http://www.ed.gov/essa) retains some of the language of high stakes accountability applied to schools performing at the bottom 5%, we contend that the end of NCLB may mark the end of an era, an era during which the belief in the extraordinary power of extrinsic incentives, performance measurement, and organizational goal setting in managing educators' work held sway, and during which these management tools have been a prime mechanism for improving schools serving marginalized students.
Parallel to the end of NCLB, another plank of the incentive paradigm has become shaky.Thanks to consistent evaluations, funded by the U.S. federal government in parallel with Teacher Incentive Fund grants to a wide swath of educational jurisdictions across the country, we now have a corpus of research that fairly conclusively demonstrates the ineffectiveness of pay for performance as means to motivate improvements in teachers' practices (Yuan et al., 2012).This is not to say that goal setting and extrinsic incentives, especially the NCLB sanctions regime, did not have an impact on the operation of schools and teachers' work (see Mintrop & Sunderman, 2009, for a review).High stakes test scores rose substantially in many states, and pedagogy was adapted to the demands of the system.But neither did achievement gaps narrow in expected ways, nor did the systems, for the most part, motivate the expected upgrading of instructional practices in schools under pressure (Anagnostopoulos, 2003;Au, 2007;Diamond & Spillane, 2004;Mintrop, 2004;Valli & Buese, 2007).
In this paper, we ask why management by extrinsic incentives and organizational goal setting may have been far less powerful than designers of accountability and extrinsic incentive systems had expected.Given our research design, our answers are partial.We draw them from a set of charter schools that, on the face of it, could have been considered fertile soil for extrinsic incentives.But we found, figuratively speaking, that incentives and organizational goals merely sent down shallow roots because, below the surface, they never found the nutrients in the deeper layers of diffuse work motivation and commitment without which they could not take off.
There can be no question that educating students who have been consigned to the margins of American society requires motivation and effort that goes beyond what has to be expended in more secure middle class circumstances.Schools with large numbers of ethnically stratified lowincome or poor immigrant students face unique challenges not encountered in other settings.Social, economic, political, and cultural disadvantage and deprivation may make students' connections to school and engagement in learning tentative (Bingham & Okagaki, 2012;Nasir, 2011), and grossly unequal distribution of resources within the educational system exacerbates these disadvantages.A distinct commitment on the part of educators to high achievement and success in the face of social adversities (Gu & Day, 2007) seems to be required.
There are currently two motivational dynamics advocated by reformers of American schools that are to shore up this commitment.Roughly speaking, one dynamic draws from intrinsic service commitments, the other banks on extrinsic incentives and organizational goal setting.This paper investigates the relationship between the two in a unique work environment: public charter schools with a distinct social justice orientation.

Pro-social Commitment and Service
According to Grant (2008, p. 49), pro-social motivation is the desire to expend effort to benefit other people.It manifests itself in empathy, helpfulness, and values of concern for others (Batson, 1987).These dispositions may be relatively stable states and commitments (Shamir, 1990) or they may be activated in situations of contact with others who need help (Grant, 2007).
Teachers in schools serving students living in challenging socio-economic circumstances may generate work motivation by developing a conscious appreciation of students' needs and realities (Ladson-Billings, 2009), a strong ethic of care and service (Noddings, 1988;Talbert & McLaughlin, 1994), a sense of duty to bring about high academic achievement in the face of obstacles, and a feeling of owning, or identifying with, their work (Gagné & Deci, 2005).It is often assumed in educational discourses that this kind of commitment is shorn up by ideals, values, meanings, and goals that make teaching more akin to a calling (Lortie, 1975;Rosso, Dekas, & Wrzesniewski, 2010;Shamir, 1991).We invoke a sense of calling when we refer to those who are highly motivated and committed to work with marginalized students as "dream keepers" (Ladson-Billings, 2009) or "social justice leaders" (Theoharis, 2007).Educators of this type exhibit an unwillingness to abandon their values or goals in the face of numerous distractions, obstacles, or challenges.Sometimes explicit ideologies around social justice or democratic citizenship may bolster these commitments (Westheimer, 1998).

Self-Interest
But teachers, like all employees, private or public, are also self-interested.As such, they may be concerned about workload, status or prestige, and monetary remuneration.Carrying out one's work under difficult circumstances may heighten these concerns (Ingersoll, 2001).Extrinsic incentives tap into these sentiments.Growing out of the "new" public management (NPM) reforms of the 1980s and 90s that introduced private sector management principles into public services (Aucoin, 1990;Hood, 1991;Osborne & Gaebler, 1992;Pollitt, 1990), goal setting and incentives have become pivotal motivational tools.In the incentive-driven dynamic, the combined power of quantitative quality indicators, performance measurement, evaluation, goal setting, rewards, sanctions, accountability, and market competition is to shore up work effort by mobilizing public employees' self-interest in serving their clients well (Christensen & Laegreid, 2011).
But self-interest can also play a less sanguine role as Lipsky (2010) in his analysis of streetlevel bureaucrats has shown.This research shows how human-service workers, perennially battling a mismatch between client needs for care and lack of time and energy to satisfy these needs, cope with work load by rationing services, insulating themselves from personal closeness to clients, and excluding from consideration those deemed undeserving of their care due to lower social status or some presumed moral deficiency."Deficit-thinking" (Valencia, 2012) props up service workers' precarious status by devaluing vulnerable clients, with harmful consequences for teachers' expectations for students (Weinstein, 2002).
Extrinsic incentive and accountability systems, such as the NCLB design, are often explicitly designed to prevent these harmful consequences.Against deficit thinking they presumably marshal evidence by making the most effective solutions, organizations, or individual teachers visible and by creating a pragmatic can-do attitude (Skrla & Scheurich, 2003).Against the tendency of educators to doubt the deservingness of certain groups or categories, the system creates performance categories for the most disregarded groups and punishes neglect (Kim & Sunderman, 2005).Motivating educators in this dynamic would be a matter of commitment to organizational performance goals, extrinsic incentives, evaluation, and appeal to self-interest.

Service and Self-Interest
Service and self-interest are not mutually exclusive.In the reality of work organizations, motivations and behaviors that benefit self and others occur simultaneously, but with varying weights.They can be synergistic, but also antagonistic (Grant, 2009).On one hand, organizations can reward behavior or motives that benefit others (Bolino & Grant, 2016;Frey & Osterloh, 2002) and employees can internalize organizational goals (Gagné & Deci, 2005).On the other hand, organizations can raise the cost of service or pro-social behavior and compel employees or citizens to be primarily self-interested (Mansbridge, 1990), and extrinsic incentives can crowd out intrinsic motives or pro-social commitments (Frey & Osterloh, 2002).We will discuss these dynamics further.

Social-Justice Oriented Public Charter Schools
In the era of extrinsic incentives, schools serving disadvantaged communities must find ways to accommodate the two motivational dynamics: one more service oriented, the other more selfinterested.The schools we investigate here are in the cross hairs of this tension.Located in poor inner city neighborhoods, the schools are independent public charter schools that were founded with an espoused social justice and progressive mission.We characterize the schools' mission based on publicly available material and initial interviews with school leaders.From both sources we gleaned an explicit commitment to social justice and equity and a pedagogical vision that was student-centered and emphasized relevance for the students' identity and culture.The three schools are similar in this respect.Later, we will see how this espoused mission finds a complement in teachers' beliefs and dispositions.
The schools owe their very existence as charter schools to the state laws that were passed in the wake of public management reform.As independent non-networked schools, they experience the full weight of accountability, managerial independence, and market competition.They exist under the watchful eye of the local school district with which they compete for students and which can make the re-authorization of their charter difficult.Up until the year 2013, their existence depended on performance in the state's test-based and sanctions-driven accountability system that had been in existence for 13 years.A year prior to data collection, the schools also volunteered to participate in the federal Teacher Incentive Fund initiative that gave them additional funds to reward good performance with bonuses that could add up to about 20%of an average teacher salary.This required them to adopt an extensive performance management system for teachers consisting of standardized test scores, value-added measures, teaching evaluations scored by an external agency, and principal ratings of teachers' progress.
Teachers in these schools were non-tenured and non-unionized.Contract renewal from year to year depended on performance and was at the discretion of school administrators and governing boards.At the same time, the schools, to varying degrees, tended to recruit from a pool of young novice teachers who were perhaps drawn to the schools' explicit social justice orientation (see Foote's [2009] anecdotal evidence), but pursued shifting career goals (Johnson, 2004).For teachers in this situation, the uncertainty of reward and employment presumably contrasted with the certainty of educational challenge, ordinarily encountered in schools educating students of the schools' demographic profile.
The TIF grant was overseen and in parts administered by a local provider organization for the three schools.The researchers contracted with this local provider as evaluators of the TIF grant.Though certain federal guidelines had to be followed for the evaluation, the provider did not interfere with the work of the researchers.Evaluation was only one aspect of the research.From the start, we also wanted to contribute to an understanding of teacher work motivation by training a fine-grained lens on motivation patterns under the dramatic circumstances of the three schools.On one hand, performance goals and extrinsic incentives are presumably pervasive and embraced by management in the three secondary schools, and teachers have few labor rights that could buffer their impact.On the other hand, the character of the communities the schools serve calls for an engaged moral stance against the adversities of marginalization and towards social justice, a moral stance that is encouraged by the schools' explicit philosophy.The study explores what motives prevail under these circumstances in the crucible of goals, incentives, service commitments, and teachers' occupational identity.We use the literature on work motivation, teacher work motivation, and public service motivation to conceptualize the study.We describe participants, context, instruments, measures, and protocols for this mixed-methods study.Findings are drawn from analysis of qualitative and quantitative data (Huberman & Miles, 2002;Johnson & Onwuegbuzie, 2004).

Work Motivation Theories
Work motivation has been conceptualized with various theories.We briefly discuss four main ones from which we derived the conceptual framework for this study: expectancy theory, goal setting theory, self-concept theory, and self-determination theory.In the literature on accountability and incentives in education (see literature cited in the next section), the use of expectancy and goal setting theories is frequent and standard.In expectancy theory, workers anticipate rewards and strive to attain the rewards when they expect that increased effort will result in the reward, called expectancy, and when the reward itself promises to fulfill a desired satisfaction, called valence (Lawler, 1973).Valence is of a different quality than values.Valence describes a temporary valuing attached to an action; a value, on the other hand, is, in a classic definition by Rokeach (1973, p. 5), "An enduring belief that a specific mode of conduct or end-state of existence is personally or socially preferable to an opposite or converse mode of conduct or end-state of existence."As an example, substance abusers may have a high valence for the substance they are craving, but may not value, rather even detest, their craving.
In goal setting theory, rewards are secondary.Goals themselves are the primary force.Goal setting theory holds that workers are more strongly motivated in pursuit of specific, worthwhile and attainable goals than in pursuit of "doing their best" (Locke & Latham, 1990).In either expectancy or goal setting models, it does not matter whether goals or rewards come from outside or inside the worker, but rewards or goals need to have valence and be sharp, distinct, or precise enough to be calculable.We posit, in line with previous theorizing (Finnigan & Gross, 2007;Kelley & Protsik, 1997), that no performance accountability regime or pay-for-performance reward program can succeed unless teachers are orienting their work towards relatively precise goals that cut through goal ambiguities endemic in educational work.
Self-concept theory or self-determination theory are less frequently used in research on work motivation in the nexus of externally generated organizational goals, rewards, and sanctions.But they are especially useful when one wants to understand the connection between the latter and motivation that springs from internally generated sources.In self-concept theory (Shamir, 1990(Shamir, , 1991)), the subjective or collectively shared meaningfulness of diffuse goals or aspirations is central.Motivational energy is generated within a dynamic of self-evaluation and dissonance between "Is" and "Ought," one's internalized values in reference to obligations of the collective and the realities of behavior or performance.Specific or precise goals play less of a role in this theory, and deeply internalized values are a premise.This theory may be especially applicable for education if we assume that educational work is strongly moral and embedded in institutional meanings and norms that have motivational power.But the theory gives us few tools to conceptualize the connection between these meanings and norms, on one hand, and something that may be external and extrinsic, such as organizational goals or monetary rewards.In self-determination theory, on the other hand, the process of internalization looms large.Human beings are seen as energized by psychological needs for competence, autonomy, and belonging.And workers are motivated by their work to the degree that they are able to fulfill these needs.Calculative orientations towards precise goals, personal or organizational, are enmeshed in more diffuse internal flows of energy.External organizational goals, expectations, standards, or incentives may be integrated into these internal flows to the degree that they enhance individuals' sense of competence, autonomy, and belonging (Gagné & Deci, 2005;Ryan & Weinstein, 2009), or prevent their diminishment.
Self-determination theory arrays work motives on an internalization continuum from being more externally to being more internally regulated (Gagné & Deci, 2005).Clearly, in work we are exposed to external regulations, no matter what.For most of us, earning a wage is an incontrovertible contingency in order to fulfill needs and desires outside of work.We might accept an external regulation as necessary even though we do not particularly value it.If society wants to examine students on standardized tests, it is something we do as educators whether we think it makes sense or not.We would not be doing our duty in our own eyes and the eyes of others, given that students need to succeed by the standards of society, not our own as educators.In the terminology of self-determination theory, we "introject" this regulation (Gagné & Deci, 2005).Failing to act on introjected expectations may generate feelings of shame.
A deeper form of internalization is identification or integration.Regulations that we have integrated have become part of us.They make up our identity as workers.They motivate autonomous and discretionary action.They radiate out to a larger complex of values, standards of quality, and practices that are meaningful to us.We have integrated standardized tests if we observe ourselves in our work through them in conjunction with other criteria of good practice.And if we could connect monetary rewards for high scores to a job well done by these formal and informal criteria, we may have integrated pay for performance as well.In this case, goals are expressions of more diffuse values (Locke & Latham, 1990), and rewards may be perceived as internally meaningful symbols of achievement (Harackiewicz & Sansone, 2000).
Research on public service motivation (PSM) specifically looks at the interaction between goals and incentives, on one hand, and service commitment, on the other (Grant, 2008(Grant, , 2009;;Perry, 1996;Perry & Hondeghem, 2008;Perry & Wise, 1990;Wright, 2001).In Perry's foundational formulation (1996), originally set out to identify work motives unique to public sector workers, motives such as compassion, civic duty, public interest, pro-social commitments to the benefit of others, or self-sacrifice were seen as a counterpoint to the increasingly dominant emphasis on extrinsic reward calculation with career, status, or money.Researchers found distinctive differences in value orientations between private and public employees, for example lesser value placed on financial rewards than helping others (Boyne, 2002;Crewson, 1997;Moynihan & Pandey, 2007).As the field matured, more encompassing theoretical models were developed that explain public service motivation as a function of values and goals (Grant, 2008(Grant, , 2009;;Perry & Vandenabeele, 2008).
In public service motivation research, goals are discussed in various ways.As we discussed earlier, those who stress goal ambiguity as an ingrained feature of public service work search for more diffuse commitments as main motivators.In Wright's (2001Wright's ( , 2004Wright's ( , 2007) ) PSM models, on the other hand, goal setting is in the center.In his path models, the public service mission of the agency and its incentives and rewards, in combination with employees' values, produce a sense of overall job importance and commitment to organizational goals.Organizational goals, with their specific, attainable, and challenging characteristics, come to regulate work.Goals are central motivators.They are reinforced and aided by organizational mission and personal values.
It is conceivable that when goals are perceived as highly ambiguous, diffuse service commitments fill in as motives, or compensate for, the weakness of precise goal setting.This pattern seems intuitive for education, especially when social environments are volatile and learning outcomes unpredictable, as may be the case for challenging educational environments.But the reverse is also possible: the degree to which specific goals are adopted or available in the organizational context may reinforce public service commitments (Grant, 2008;Jung & Rainey, 2011).Taking its cues from research on public service motivation, the purpose of the current study is to shed light on the connection between rewards, goals, values, and service commitments in an educational setting that is organized as a hybrid between managerialist orientations towards performance and social justice commitments.

Research on Teacher Work Motivation
In the literature on teacher work motivation, beginning with Lorie (1975), work motivation is often broadly conceptualized in terms of extrinsic and intrinsic motives.In the seventies, eighties and early nineties, with scholarship by Lortie (1975), Johnson, (1986), Mitchell, Ortiz, and Mitchell (1987), Nias (1989), or Firestone & Pennell (1993), the view prevails that teachers are primarily intrinsically motivated.Echoing this line of scholarship, Mitchell, Ortiz, and Mitchell (1987) summarize the evidence this way: "All available evidence supports two basic conclusions regarding the value or potency of various rewards for educators.First educators generally find intrinsic rewards more meaningful and attractive than extrinsic ones.[…] The second broadly supported conclusion is that educators rely on sharply divergent subjective meaning systems for interpreting their work responsibilities and experiences" (p.188).Thus, work motives are seen as largely internal, culturally and self-determined.
At a time in the late 1990s, when "probation" was still a novel shock to the system, Mintrop (2004), using expectancy and self-concept theories, found that various patterns existed side by side.Teachers in his study tended to disregard targets and sanctions as motivators and instead interpreted them as further signs of unfairness and injustice leveled at schools serving the neediest students.The schools he studied were for the most part located in poor African-American neighborhoods.Rather than compelling new work effort, the threat of sanctions transformed into more energetic pleas for help.Highly engaged teachers were especially critical of new extrinsic incentives and found little educational meaning in accountability goals, yet were also more inclined to worry about prestige, reputation, and autonomy.They were therefore willing to do what needed to be done to "keep the state off [their] backs."So were the principals.In hindsight, this pattern seems to mark a transition from traditional assumptions of service to new ones associated with accountability.Finnigan & Gross (2007) studied Chicago public schools serving similar demographics, several years after Mintrop's study.Using expectancy and incentive theories as their conceptual framework, they found that extrinsic incentives did compel teachers to attach valence to organizational goals and expend effort.After an initial upsurge of motivational energy, however, demoralization spread in schools where targets could not be met in short order and effort did not seem to prevent sanctions.The researchers speculated that this sort of demoralization hinted at motivational dynamics not adequately captured by expectancy logics.According to Frey & Osterloh (2001), demoralization can be interpreted as a typical symptom of intrinsic motivation being crowded out by extrinsic motives."Crowding-out" does not mean that intrinsic or extrinsic work motives are mutually exclusive (see a discussion in Supovitz, 2009), but that at a certain point the strength of one can diminish or overpower the other.Alternatively, the new extrinsic motives could become synergistic with (Milanowski, 2000;Supovitz, 2009), or perhaps internalized into (Skrla & Scheurich, 2003) traditional intrinsic service commitments.
We explore these connections by examining the interplay of four main sources of work motivation: precise goal setting, diffuse pro-social or service commitments, different degrees of identification with one's work, and material benefits or status.

Figure 1. Conceptual Model
These four motivational forces are embedded in the specific charter school setting that in turn is fundamentally shaped by the experience of social adversity and deprivation, given the specific student populations the schools serve, and the potentially trenchant effect of the performance management regime along neo-liberal public management or managerialist lines that the schools labor under or voluntarily adopted.

Cases
The data on which this paper reports were collected in the first two years of a four-year longitudinal study of the responses of three publicly funded charter schools to a new bonus pay system.The three schools are relatively small in size (see Table 1).The schools are secondary schools and urban in character.They are located in distinctly poor sections of a metropolitan area in the state of California.As was mentioned, the schools are similar in their espoused mission of social justice, equity, and progressive pedagogy.Helping disadvantaged students gain college access is a shared orientation as well.But they also differ in some characteristic ways.Apart from differing size and measured performance (see below), School B, the smallest of the three schools, specializes in helping students who have not been successful in regular district schools.Social-emotional support for students in a small 'family' atmosphere is valued greatly.School A sees itself as a communitybased school, guided by a philosophy of critical pedagogy and engaged citizenship.School C is a much larger school than A and B. It houses both an elementary school and a middle and high school.Only the middle and high school is considered in this paper.Equity and social justice are paired with a strong and explicit emphasis on academics.
Within the state accountability system, now defunct, School C performed strongly for its demographic profile.Schools A and B, on the other hand, were poor performers by state standards.None of the three schools met their growth targets in the school years during which data for this paper were collected.The 48 teachers included in our sample are middle or high-school teachers.From the 35 teachers that responded to the survey, we glean the characteristics displayed in Table 2.

Table 2 Characteristics of Responding Teachers
School A School B School C Years teaching (average) 6.3 5.2 7.5 Years teaching at current school (average) 2.4 4.4 3.9 Fully certified teachers 73% 67% 83% Working at least 7 hours beyond contract 82% 83% 88% Expecting to stay in their current school for at least 3 years 63% 16%* 67% * See below in text Average tenure at the schools at the time the survey was conducted was fairly short (3.5 years) and commitment to stay was tenuous, confirming the expectation of shifting career goals.About one-third of teachers in two schools and more than two-thirds in one school expected to leave within three years.Of the three schools, School C is most stable.It has less teacher turn-over, a strong reputation in the community and many more student applicants than admission slots.Admissions were handled with a lottery system.School B was the least stable.It is a very small school and at the time of data collection several highly engaged teachers were considering new personal career moves that were not related to the conditions in the school.
Thus, compared to a typical public school work environment, work motivation in the three schools may be influenced by relative job insecurity, accountability pressures, performance management, and a relatively short expected tenure on the part of relatively novice teachers.In the findings sections, we explore in more detail how the schools' organizational characteristics may have influenced teacher motivation patterns.

Methods and Data
The study is a mixed methods study of the three schools and the individuals working in them (Creswell, 2013).In particular, we selected a qualitative dominant mixed method approach, which relies in large part on qualitative data to understand relationships among factors, complemented by quantitative data to augment rough quantifications (Johnson, Onwuegbuzie, & Turner, 2007).We aimed at both the precise identification of different sources of motivation as well as a deeper understanding of how teachers allow for those motives in their work lives.Towards these ends, we administered a questionnaire to all teachers in the three schools and delved more deeply into work motives during four rounds of interviews conducted over a period of two years.

Quantitative Data
A teacher questionnaire was administered to 48 teachers in the three small high schools during the Spring of 2012.With a response rate of 73%, 35 teachers returned completed questionnaires.The questionnaire contains 256 items.For this paper, we report on a subset of data extracted from 88 items that directly pertain to teacher motivation and commitment, goal setting, rewards, and organizational culture and leadership.Other items capture perceptions of the details of the TIF performance management system and professional development dynamics that are the topic of other papers.Wherever possible, we used items and scales that have previously shown good psychometric properties, are widely used in the field, or were created in previous research by the first author.Constructs represented by the scales were tested for internal consistency for this small sample with Cronbach's alpha.Scales with a coefficient over 0.75 were eligible for analysis, with the exception of the two-item scale "goal clarity" for which we could not achieve better consistency with more items.Most items and scales are constructed in a 5-point Likert scale format with the scale mid-point defined as neutral.
We use the Work Extrinsic and Intrinsic Motivation Scale (WEIMS) to explore work motives.The instrument was developed within self-determination theory (Tremblay, Blanchard, Taylor, Pelletier, & Villeneuve, 2009).The WEIMS asks respondents why they do their work and arrays motives on a continuum from most externally regulated to most internally regulated.At the internal end of the continuum, we find intrinsic motivation (i.e.satisfaction and pleasure with the work itself) and integrated regulation (i.e.identification with, and ownership of the work).1 Moving to the external end of the continuum, we find introjected regulation (i.e.duty generated by guilt or shame) and external regulation (i.e.garnering an income or monetary rewards).
We added a scale that specifically captures broader ideals encapsulated in pro-social commitment to others and the common good (Grant, 2009). 2 These kinds of commitments can be more externally (duty) or internally (identity) motivated (Grant, 2009).We further added a scale developed in the first author's previous research that captures the concern for external validation (i.e.prestige, reputation, standing, and professional pride in the eyes of evaluating or auditing authorities) which can be either externally or internally regulated (Buunk et al., 2005).Thus, the spectrum of diffuse work motives gives us a range: material interests, status, duty, identification with work or ownership, work-inherent pleasure, and service to others or the common good.
Task-specific motives are captured with goal-related scales.Three scales, specificity, difficulty (Wright, 2004), and commitment (Hollenbeck et el. 1989) inquire about the degree to which respondents adopt or embrace specific and challenging goals.Lastly, to capture baseline attitudes toward extrinsic incentives we used a scale and a set of items developed by Ballou & Podgursky, (1993) that explores respondents' acceptance of the idea of bonus pay for differential performance.
Quantitative data are analyzed with univariate and bivariate statistics.In this paper we present frequencies (scale means) and, on occasion, correlations (Pearson's r) that appear solid given the small sample size.Again, the purpose of the quantitative analysis is limited: to understand the broad quantitative contours of patterns that rest on further exploration with more qualitative detail.In the quantitative portion on teacher motivation patterns, we did not find distinct school differences (see below in the findings section).Given the small sample size, we therefore treat respondents from the three schools as one universe.

Qualitative Data
We conducted several rounds of semi-structured interviews with teachers across the three schools.The interview method allows to access the subjective perspective of individuals and inquire into the categories by which they observe their experiences, world-views, and motivations (Patton, 1990;Weiss, 1995).Across the three schools we selected 33 teachers.In the two smaller schools, we interviewed almost all teachers.In the larger school (C), we interviewed almost two-thirds of the teachers, with an emphasis on teachers teaching academic subjects (Math, Science, English, and Social Studies).For this paper, we analyzed data from four rounds of interviews collected through the 2011-2012 and 2012-2013 schools years.Each round of interviews informed the successive one.Through each round we expanded our understanding of the motivational dynamic among teachers, which allowed us to refine our instruments.Most explicitly, issues of work motivation were addressed in the first, second, and fourth rounds.The results of the survey conducted in the spring of 2012 were used for the development of the instrument for round 4. In the first round conducted in Fall 2011, we broadly inquired about teaching philosophy, school mission and culture, and overall diffuse pro-social motivation and commitment.In the second one, we went more deeply into the tensions between moral commitments/ pro-social motives and extrinsic incentives.In the third round, we interviewed teachers about their ways of learning about instruction.Although this round did not explicitly address teacher motivation in the protocol, relevant information on motivation was nevertheless obtained.Finally in round 4, we were ready to develop an interview guide that explored the finer interactions between service commitments, pro-social motivation, goal setting, and intrinsic and extrinsic motives, bringing together all relevant topics of inquiry.In total, we conducted 56 interviews, lasting between 25 and 60 minutes.Thirteen teachers were interviewed more than once.Table 3 summarizes the interviews collected organized by schools and round of data collection.Codes for analysis were developed along the dimensions of the conceptual model (Miles, Huberman, & Saldaña, 2013).Initially, we developed seven distinct theory-guided codes.Additional codes emerged from the data, most notably the code of "deservingness" and the distinctions between various objects of commitment as well as more detailed aspects of goal setting.The codes are listed and further described in Table 5. Interviews were coded using a qualitative data analysis software (Dedoose).We defined, operationalized, and illustrated the codes with representative quotes from interviews..We analyzed qualitative data in three rounds: (1) we ascertained the concrete meanings that interviewees attach to pro-social commitments, pleasure, duty, goals, prestige, and rewards, etc. in contradistinction to other concepts; (2) cross-referenced these meanings to search for overlap, tension, contradiction, or synergy; (3) looked for patterned differences across various organizational and individual teachers' profiles.

Goal
Orientations that direct or concentrate work effort.These orientations can take the shape of specific goals, objectives, or broader purposes.They can originate from outside the worker, i.e. be externally regulated, in the form of organizational expectations or system demands, or from inside the worker, i.e. be internally regulated.There are learning or performance goals.Goals can have different degrees of precision.

Prestige
Sense of status or self-worth based on the perceptions of others.
Prestige can be derived from material goods, symbols, or moral esteem.

Money
Motives that are connected to material benefits.The work is performed to attain the desired reward derived from the activity.
After analyzing quantitative and qualitative data, we analyzed school specific patterns and individual cases, looking for organizational or individual variation.For the quantitative part of this analysis, we relied on radar charts as a way to illustrate and analyze motivational profiles.The radar chart is a graphical method that allows displaying multiple variables in a simultaneous fashion in a two-dimensional representation.This analysis proceeded as follows.First, based on the quantitative data, we created 35 radar charts based on the motivation scales from the survey.Second, based on the means, we identified a modal pattern for the entire sample.Third, with the purpose of looking for differences between schools, we clustered individual profiles by school and contrasted them with the modal pattern.We did not find identifiable patterns that clearly differed across organizations.
Fourth, taking the modal pattern as a point of reference, we identified differences among individual profiles.We clustered individual cases into three categories: modal, more pragmatic, and more idealistic.More 'pragmatic' teachers showed a higher orientation towards goal-setting motives and less emphasis on pro-social commitment or pleasure.More 'idealistic' teachers showed a lower orientation towards goal-setting but a heightened importance of pro-social commitments, identification with work, and pleasure.From these two clusters, we selected two prototypical cases that illustrate differences quite well.For each of these two in-depth cases, we developed a narrative that draws from all available data we had from these individuals up to the end of the fourth round of data collection.

Findings
We begin the report of findings with a display of the relative strength of the various motives under investigation indicated by scale means.The scales are derived from self-determination and goal setting theories.In 2012, when these data were collected, test-based accountability had been established for 13 years in the state, and federal No Child Left Behind demographic subgroup calculations had been in existence for over ten years.Also, a relatively generous pay for performance scheme had been adopted in the previous year.It stood to reason that this should have left a strong mark on teachers' work motives, reinforcing the relative importance of specific goals and externally regulated performance expectations and duties relative to more internal or self-adopted motives.The quantitative data show that this was not the case.lowest with means below the neutral mark.Thus, in reference to the literature on teacher work motivation, teachers on the average in the three-school sample appear to be rather traditional: driven by broad pro-social commitments, a sense of identification with one's work, and work-inherent "psychic rewards" (Lortie, 1975) while acknowledging the enormous challenges that their work poses to them.By comparison, specific goals and commitment to organizational goals receive mean ratings of less than moderate strength.Material incentives were perceived as a very weak or negligible force for work motivation, with mean ratings close to a rating of 1 (not at all a reason for one's work motivation).But this contrasts with the finding that 69% of respondents found the idea of monetary bonuses for better performance acceptable.School-wide bonuses were considered acceptable by an overwhelming 83% of respondents.There seems to be a gap between being open to receiving money and being regulated by it in one's work.
It is quite conceivable that the self-reported means across the array of motives captured by the WEIMS and other scales reflect desired ideologies, rather than a true motivational pattern.As Dan Lortie (1975) already noted, traditional occupational ideologies of teachers in the United States have tended to privilege moral purposes over material gain, so the quantitative data need to be triangulated with qualitative data that prompt and challenge teachers to explain how in their minds the various motives interact with each other.
Mirroring quantitative scales, qualitative codes were developed in four main complexes: service and pro-social commitments, degree of identification with the work, goal setting, and reward related to material benefits and status.Codes related to service or pro-social commitments and identification with work refer to self-determination and self-concept theories as previously discussed.Codes related to goal setting and reward refer to goal setting and expectancy theories.Given the nature of the interview material, the complex of service and pro-social commitment was subdivided into commitment to students, society, and school.As was described in the section on methods, the code of deservingness was developed from the material.It captured the many statements that explicitly stressed the moral worthiness of students in contradistinction to a society that seems to value them less.The complex identification with work followed the distinctions of pleasure, ownership, and duty.Given the diffuse nature of narratives related to goals, we first captured the complex of goal setting with a broad code of "goals" which we subsequently subdivided into various kinds of goals and the commitments that interviewees attach to them.Following goal setting theory and public service motivation studies, we coded goals according to their degree of specificity or precision versus diffuseness, as well as the degree to which organizational goals and personal goals overlapped.Pro-social and Service Commitments "I want to help" was the most frequent reason given by teachers when they were asked why they came to work every day.In almost every interview, teachers expressed a high awareness of their students' living circumstances or social position.They saw their students as in need, underserved, diverse, and "without enough adult support."They described them as low-income, second language learners, who have had negative school experiences, who have struggled with school, and are in danger of dropping out.These descriptions attach themselves to intense feelings: It makes me feel even more motivated as a teacher knowing that I'm serving students who would not be getting as many adults who believe in them as they do here Though challenges were daunting, it was also "inspiring" to "make a difference in students' lives:" My inspiration has always been to work for, with students who don't have that support somewhere else, who don't have someone at home necessarily who is pushing them and who even knows how to prepare them for the future (School C, 17) In the face of adversity, interviewees intended to give students "an alternative", a "second chance," and "a caring relationship that pushes them to meet the standards."They voiced a sense of responsibility to do a good job and to teach good lessons.They felt rewarded when students overcome adversity, make progress, learn, grow, or accomplish to get into college.These rewards were called up when the work would get frustrating, something that was also on teachers' minds.
Commitment to students and commitment to society flowed from a similar root sentiment: disapproval of inequality, injustice, and educational disadvantage associated with American society."Social change" was a central idea in the interviews, and teaching "underserved students" was the contribution.Teachers insisted that their students were as deserving as students in wealthy areas: You know, there's no reason why our kids shouldn't have the same opportunities as a kid in [a nearby wealthy district].It doesn't really matter what home they were born into; they should have the same opportunity.So, we wanna give them that chance.(School C, 32) Thus, service to students and society was imbued with a distinct commitment to leveling the playing field, to restore the moral deservingness of underserved poor students and to offer them an education that was as good as that of more advantaged students.It is not surprising in light of this interview material that challenge and pro-social commitments were rated so prominently on the survey (see figure 2).

Identification with Work
In self-determination theory, workers who identify with their work may express that the work is part of who they are or that the work is fused with a sense of identity.This sense of identity could be related to a number of supporting motives: the results of the work could be socially meaningful, the work itself could be pleasurable, stimulating, and suited to the person, identification with the work could be related to a sense of guilt associated with duty or external obligation, or the recognition and social status one derives from the work could constitute an important part of a person's identity.In this section we examine the relationship among these motives which can be distinct, but also overlap.(Rewards and status are examined in a separate section further below).
"Oh, I love it!This is my favorite thing to do."This was one of the most common replies when teachers were asked to describe their work.Most interviewees characterized their work as challenging but also interesting.They liked preparing lessons, developing curriculum, thinking about how students learn, learning new ways of teaching, or seeing students learn and grow.They described classroom interactions as joyful and uncertain at the same time.They saw teenagers as "fun people" who "make them laugh" and "bring mystery to the classroom" everyday.They enjoyed creating relationships with students, some of them lasting even past graduation.
A high proportion of teachers in the three charter schools described teaching as a part of who they are, something that naturally flowed from them, an intrinsic desire to share, a vocation.In the following quote, a teacher vividly communicates how joy, stimulation, identification, and moral purpose intertwine to motivate her to be a teacher: I don't know, when something calls you like that, I don't know that you can always have a why, like I would call it a calling.So I could say because it felt like a magnet pulling me to it, so that's like feeling fulfilled.It was interesting, every day's different.You get to build relationships, it's challenging, and it feels also like doing something good in the world.Like creating more balance or more justice where there have been imbalances.But I feel, like, on the deepest levels it was just what my heart wanted to do.(School B, 1) Quantitatively, identification with work is strongly associated with pro-social and service commitments.In survey data from the three schools, pro-social commitments were strongly and significantly correlated with ownership, or self-identification with work (.76), less so with pleasure and duty, and not at all with material reward.Speaking with self-determination theory, pro-social commitments to students and the common good can be considered largely internalized, i.e. sensed as something that is part of oneself.But feelings of guilt and the resulting sense of obligation or duty are blended in as well.Some interviewees saw themselves in a more privileged position relative to their poor and immigrant students of color, and this sometimes generated guilt.A teacher in School C expressed the mélange of feelings this way: I mean I think we just have a responsibility to equalize […] When I think about my initial curiosity and sense of guilt and just discomfort with the disparity.This feels like a way to kind of put that into practice, rather than just kind of looking from -I could stand over there and say, "Wow, it's really too bad that those brown children don't get the education they deserve," but this feels a little bit more authentic and honest to say, "Yup, they don't get the education they deserve.Let me try and do at least my little piece to address that."And I do think I have a different relationship with teaching, having come up through that lens as a teacher than I would have if I had stayed in rural Massachusetts and taught people who looked like me and were from middle class families.So I'm happy for it.(School C,22) But this sort of happiness, composed of identity, service, pleasure, and duty went hand in hand with a sense that the work was indeed very hard and not always satisfying.In the interviews the work is also associated with feelings of being overwhelmed by the difficulties of the job, being underpaid, being under pressure, not getting enough support, being overlooked, and being "frustrated."At first sight, these frustrations may seem to contradict the intrinsic love for the work itself.But in the face of these difficulties, the mostly young teaching force expressed that they were "challenged but loving it;" that they were "passionate to learn" to try new things and to improve their practice.

Goal Setting
In examining goals, we are interested in finding out how teachers make sense of their goals in the midst of goal ambiguity and goal uncertainty that are characteristic for teachers' work in general.Organizational performance goals enshrined in features such as standardized tests, accountability targets, teaching evaluation scores, etc. presumably help reduce this goal ambiguity and concentrate teachers' efforts.Following goal setting theory, we wanted to know if teachers pursued a relatively limited set of goals that help them prioritize, to what degree these goals are precise and specific, and to what degree they are informed by organizational goals or goals which are espoused by the performance management system.Following self-determination theory, we wanted to know if externally generated organizational goals were internalized, i.e. had become important personal goals used to orient action.
Quantitatively, the strength of motives in Figure 2 showed that the high ratings for prosocial commitments were only rivaled by the high ratings for work challenge, i.e. the perception that work objectives are quite demanding and require a high degree of skill and knowledge.By contrast, perceptions of goal clarity or specificity, i.e. knowing with precision what needs to be done, and sensing that job responsibilities are clear, was much weaker.Commitment to espoused organizational goals, seen as worthwhile and obtainable, was weaker still.
In the previous sections, we saw that the teachers were strongly motivated by a sense of service, pleasure, and identification with their work, yet also high challenge.Here we ask how these relatively strong, but diffuse forces might be channeled into more precise goals and directed activities that might help teachers concentrate effort and move them beyond a diffuse quest to do their best (Locke & Latham, 1990).For example, a diffuse quest for equity might become goal oriented if equity is more clearly defined in terms of outcomes.Or equity goals may be oriented towards organizational goals, for example, if outcomes are defined in terms of accountability targets that may over time narrow achievement gaps.
When we asked the teachers about their goals, we did not presuppose a specific definition of that term.We let interviewees describe what they considered to be their goals.We assumed that organizational goals would be relevant and meaningful to them if they talked about the latter and saw them as influencing their work in a positive or motivating way.Once interviewees had described their own goals, we prompted for degree of precision of personal goals and the role of organizational goals in their work.
When asked about their goals, most teachers made spontaneous reference to a wide variety of orientations, not all of them goals by our definition.On the whole, when interviewees named goals they tended to be vague purposes or intentions to carry out activities, such as helping students to become better readers, writers, critical thinkers, or long-life learners, or getting students ready for college: …giving students the tools and opportunities they need to be able to go to college, to be able to go to a training school, to be able to be successful, to be able to do politics and social things, to not to be screwed over by other forces.(School B, 3) By contrast, precise and measurable performance goals for either students or teachers were rarely mentioned despite repeated prompting.On occasions when specificity comes into the picture, it is reflected in immediate learning targets or objectives associated with particular units or lessons: Well, what comes to my mind is when I'm in instruction, we set learning targets for our students.And so then any time we're doing something, it's specifically to help them get to this learning target.Often I'll have a learning target that's kind of a bigger idea, and then I'll break it into sub kind of skills or mini goals.Because one big thing can be kind of complex.And so then I'll focus on like one of those things for students.And when I talk about learning targets, you know, two-thirds of them are content focused, but a third of them are process and skill focused.So some of them are like solving problems.Whereas some of them are explaining the gas laws or something like that…So we're always working towards those goals.(School C, 21) Even though the great majority of interviewees did not state quantitative performance targets, such as test or evaluation scores, as goals, some teachers were not unfamiliar with the logic of formulating specific achievement goals and some even identified precise organizational performance goals, for instance, "We set school wide goals, so this year our big goal… is to have an API of 800 or 70% An important source of goals was teachers' own learning.Goal statements in this area refer to interviewees' personal growth.Most often these goals revolve around specific instructional strategies or formats, for example "getting better in asking questions to students," or "organizing more effective lesson openings or closings."Some of these goals seem to have been formulated in collaboration with colleagues, instructional coaches, or instructional supervisors: "I set goals when I do a [peer] review with [my instructional coach]."Continuous improvement appears to be an important feature for teachers in all three schools.While goals may be a marker of this process, the fluidity of the learning process and the actual experience of organic growth was a more compelling motivational force: Right, yeah I mean, in general to be perfectly honest, because I know there's not going to be any follow up, I don't really work very hard to achieve the goals.
[…] I think the goals are fine but that a person, you know, goals alone don't make a person better, right?What I really, you know, need is help in planning.How I'm going to get better at that and, you know, somebody observing me attempting to implement it and giving me feedback about how it looked and, you know, there are a lot of steps that are kind of missing.(School C,17) In sum, goals are plentiful in the three charter schools and they come in a variety of formats, some quite general, others quite specific.With regard to student learning outcomes, stated goals tended to be diffuse or they referred to unit objectives.With regard to instruction, goals were formulated around specific strategies or practices.With regard to teacher professional learning, goals were formulated as personal and formative growth goals, not as summative goals subject to evaluation.
A connection between teachers' stated goals and organizational performance goals that may link system quality indicators and measurement of performance to a sense of being motivated was largely absent in the interviews.Precise performance goals that could lend themselves to evaluation did not seem to loom large as motivators in any of the three schools.
The relative weakness of setting precise personal goals and of internalizing organizational goals is corroborated when teachers rate the importance and usefulness of various performance indicators that play a role in the three schools' performance management system.Despite highstakes accountability with its presumably precise performance targets, in force for the entire duration of the schools' existence, high skepticism prevailed.Scales that measure the perceived clarity and usefulness of the State Standards Test (CST) for one's work show means in the negative range.Only 8 out of 31 respondents rated the usefulness of the State Tests as positive.For externally scored teaching evaluations (SET=Summative Evaluation of Teaching), a relatively novel feature for the schools, the ratio is 7 out of 26 respondents.Internal measures of performance, such as principal supervision (FET=Formative Evaluation of Teaching) and peer evaluations, as well as an internally administered writing test received better ratings.The bonus pay formula was seen as least useful.insignificant.Given the small sample size, this pattern should not be over interpreted lest one commits a Type 2 error and states "false negatives" where perhaps an association exists.In light of the qualitative data, however, the link between diffuse service commitments and goal setting seems doubtful.In repeated interviews, we failed to find teachers making a clear connection between their work challenges and goal clarity, and between their goals and pro-social or service commitments.Plentiful, multi-faceted, and shifting "goals" (in the language of the teachers) were imbued in the teachers' descriptions of their work, but these "goals" rarely seem to rise to the level of concentrating one's work effort in terms of specific goals described in goal setting theory.
Research in the tradition of public service motivation discussed above asks how goal setting might be embedded in more diffuse service commitments and identification with one's work.For schools, this interaction could potentially be powerful.When precise performance goals are reinforced by the energy that identification with the work, pro-social commitments, pleasure, or duty can generate, they may be pursued with more vigor.But there is little indication in the qualitative data from the three charter schools, and (with all statistical precaution) no contradictory evidence from the quantitative data, that such synergistic confluence of motivational energy took place.This finding has serious consequences for the power of organizational goal setting and extrinsic incentives.In a logic of management by extrinsic incentives and organizational goals, some precision in goals and some internalization of organizational goals as personal ones ought to occur (Kreps, 1997).In the three charter schools, however, goal ambiguity and uncertainty coupled with skepticism towards organizational performance goals prevailed.

Material Rewards and Prestige
In the quantitative ratings, acceptability of performance-contingent rewards was fairly high (see above), but material rewards and prestige were named as weak motivators.We allowed that this pattern may be a reflection of occupational ideology rather than a reflection of one's work motives.The qualitative material, however, suggests the latter.Interviewed teachers were clearly sensitive to material benefits, but almost all of them were adamant that extra money did not "help them to do a better job," or "try harder or change practices."In their eyes, extra pay was a good idea and legitimate mechanism to reward teachers, and they felt positive about receiving bonuses.But for most, the bonuses were already earned before they were disbursed.They were justified as rewards for hard work, effort, and service continuously expended in these trying school contexts, a pattern that is consistent with other studies (Glazerman & Seifullah, 2012;Marsh et al., 2011).The quotes from two teachers stand for many more: I mean, you stick with it, you keep working to grow, to improve your teaching practice.You keep focusing on self-improvement.There should be remuneration for that, for meeting those mile markers (School A, 9) Obviously money is not the motivator of why I do what I'm doing, but it certainly would be nice to be recognized for my consistent work.(School C, 9) Several teachers stated that if the bonus money was there, they would "happily take it."Since theirs was an underpaid line of work to begin with, "educators are better off with a little extra" or some "extra appreciation."The clear link between performance indicators, measured test and evaluation scores, and monetary amounts was doubted.Two main reasons were given.One had to do with selfworth, the other with validity.Interviewees did not want monetary rewards to interfere with their sense of ownership and commitment; money would devalue these commitments which, by themselves, already motivated them to work with maximum effort.And skepticism abounded that any discrete performance indicators could validly capture the full scope of their work engagement.
As two teachers explained: I think the reason that educators are paid so poorly is because we are not deemed professionals and it is supposedly an altruistic or martyr system.Like, I am a teacher because I want to create change not because I want money.I think there is a danger in paying for success because how do you gauge that success.Do you know what I mean?That is going to really make many people not want to work at lower income schools because it is much more difficult to create improvement.Many times rewards based on money are based on test scores and that is problematic.I mean of course you are going to feel rewarded if you get money, but I would have a really hard time working in that system unless I knew that system was actually assessing teacher ability, not just teacher ability but teacher dedication.(School A, 3).
So, I mean, I got a bonus check, so I know how much money I got, but I don't really know why or where that came from, just because, personally -I don't know -for me, this grant has been, like, I think it's amazing and, of course, I'm grateful for it.But, at the same time, I don't know − for me, I don't want to be motivated by money.And so I just decided not to look.(School C, 12) Thus many interviewed teachers described monetary rewards, in the language of self-determination theory, as something external to their motivation, something they accept, but not something that regulates them or energizes them in their work.
Prestige or reputation could be more externally or internally sourced.Among survey respondents prestige is modestly and significantly correlated with duty (.5), a more outward orientation than ownership or identification with one's work.In the interviews, prestige or reputation was associated with the need to be seen as "hard workers" who are "willing to try new things," being seen as such by members of the school community.Being recognized by parents as committed and engaged and receiving positive feedback from students were described as desirable: "I'm held in the esteem of my students because they are kind of my main -they're my main critics."In School C, there was a pervasive pride of working in a highly effective and popular school that succeeded in sending almost all students to post-secondary institutions.In School B, there was pride that the school managed to educate students that had been rejected from more mainstream district environments.In the eyes of interviewees, another important source of prestige was colleagues recognizing one's hard work and tireless engagement in learning and improving.Thus, the sources of prestige were primarily school-internal, diffuse, and derived from judgments of those with whom teachers directly interacted.Interviewees did not make any connections between prestige or reputation and specific performance indicators or monetary rewards, save the pride in moving large numbers of disadvantaged students from high school into post-secondary placements.

Summary of Modal Pattern
Qualitative material and rough quantifications speak to a wide-spread pattern in these charter schools.Serving disadvantaged students, doing social justice work, insisting on the deservingness of marginalized students and providing them opportunities through personal engagement and connection that society denies them make up the substance of work motivation.Narratives of personal experiences and life course decisions brought many of these young teachers to see their work as part of their identity.But the fulfillment of precise personal or organizational performance goals is not an important source of identification.Personal goals are aplenty, but in their diffuseness and ambiguity lose the motivational punch of specific goal setting.Instead, the fluidity of personal growth for both students and teachers, and achieving tangible opportunities for life (high school graduation, college) generate energy.System or organizational performance measures of the day are acknowledged in their existence, but they do not go "deep" in motivating teachers.In the eyes of most interviewees, money in the form of bonus rewards is already deserved for strenuous effort and committed service given the perception that theirs is a line of underpaid and undervalued work.If anything, it is concern for prestige, looking good in the eyes of those who are immediately knowledgeable of one's work, that perhaps instills a concern for meeting organizational goals.

Organization Level Differences
Organizational differences with respect to motivation patterns were not pronounced.To begin with, the schools explored through quantitative data looked fairly similar.Organizational culture and leadership were rated quite similarly across the three sites.In addition, we interviewed school leaders.We asked them about school mission, their own values, and emphases in leading their schools.The schools had different aspirations, for example the difference between a stress on academics in School C and social-emotional support in B. When we asked about organizational goal setting, they all made reference to the state's Academic Performance Index and the desire to have large proportions of students entering post-secondary college.But none of them pointed to precise or quantitative goals that they pursued with their staff and none of them described efforts in their schools that were driven by these kinds of goals, though they embraced the performance management system.In this organizational context, none of the motivation-related scales or items showed noteworthy discrepancies across cases.In a next step, we took the 35 individual profiles displayed in the radar charts.In our clustering exercise, we grouped them into school level clusters in order to see if we could find organization specific patterns.We could not.Perhaps this is not surprising given how similar the schools are in their organizational culture.In the goal setting dimension, organizations should make the biggest difference.They did not.We did, however, see individual level differences in the spider web profiles.

Individual Cases
The individual cases narrated here enrich the qualitative analysis with individual teachers' personal reflection about their situation and their motivational preferences, and they shed light on the range of individual motives within the modal pattern.Towards this end, we selected two cases that differ according to service commitments, ownership or identification with work, goal setting, and extrinsic rewards on quantitative variables, displayed in radar charts or spider web diagrams below.Both cases fill in personal detail.In our sample, Adarsh exemplifies the idealistic side of the modal motivational pattern, while Jane is much more goal-oriented than the majority of her colleagues, yet both cases, on the whole, are variations of the main pattern.For Ardash, goal setting, monetary reward, and concern for prestige pale in comparison to pro-social commitment, identification with, and pleasure in, her work.Inspired by her grandmother's activism in Gandhi's struggle for freedom in India, and especially for rural education, Adarsh wanted to be a teacher since she was three.It was important for Adarsh to communicate to us that she was educated in "the third world."She describes herself as "being a second language learner herself, many times overlooked by her teachers," and that "she struggled with English."This previous experience motivated her to become an English teacher, one with empathy for her students' struggles through adversity.She shared that her main commitment is "to help students to become proficient in English."In her view, being proficient in English is the "ticket to freedom and seeing the world." Adarsh goes to work everyday motivated by "the faces of the kids."In her view, students are not just students but "her kids."She connects with her students in an emotional way.She says that her workload sometimes gets overwhelming and she feels stressed, unappreciated, and unsupported.Yet, she still puts her students first no matter what: "My work is Justin and Marvin and Jessica and Sylvia."Adarsh's desire to help students makes her start at 4 am in the morning and begin her work of planning, teaching, daily grading, and giving feedback to students.When "a student moves from barely reading and writing to reading and writing at grade level" she gets her biggest reward.In her classroom she is committed to do what she calls "social justice work" which consists of creating, what she calls, "unthinkable literacy" for kids who "do not wish to read."She is careful not to make her students feel threatened or ashamed, emphasizing non-cognitive skills along with cognitive ones.The cultivation of relationships with students is paramount.
Adarsh feels that her "work is her life."Her standards of quality, the way she sees it, far surpasses what standardized measures can capture.She does not feel recognized for the good work that she believes she puts out.But despite some bitter experiences, she will do this work for "the rest of my career as an adult living in the world." When asked about her specific goals she, first, mentions college readiness, complex literacy, emotional intelligence, artistic intelligence, and creativity as important goals for her work.She also wants to pursue improvement goals defined with her supervisor in her annual growth plan, and she refers to lesson objectives.Wrapping up she says: "The short-term goal is to work with a text or work with a small unit plan.The long-term goal is getting you ready for the world."Adarsh is willing to face conflict when school-wide measures do not match what she perceives as her students' needs.She is willing to "push back" against standards and evaluation criteria that are not sensitive to the reality of her students.State assessments, for example, are in no way, holistic enough to embrace "the kind of work that is going on in [her] classroom." When it comes to monetary rewards she is clear: "I don't care for the monetary rewards, I work for the kids."She believes that "money is stupid" because money "makes people do stupid things."She thinks that money distorts peoples' behaviors, distracts the attention from "the work that we do for the kids", and "encourages competition among the staff."Though Adarsh cannot deny appreciating bonus money, she sees money as a violation of her ethic of work and as a distraction from the main purpose of her work: kids.
In our sample, Jane is less moved by pro-social commitments, ownership, and pleasure as she is moved by goals and prestige, relative to other study participants.She does exhibit the 'classical' service motivation.She always had "the expectation to do things for others."She believes this expectation was instilled in her early involvement in her church and in community service.Since high school and college, she wanted to "advance things that she felt important" and "would make a difference in society."This desire brought her to serve as a volunteer in teaching.But Jane did not go into teaching right away after college.She wandered in different careers that she later considered "pointless."She describes it this way: "A lot of motivation was wanting to do something where I really felt I was changing society." Coming from a privileged background, she knew that students from wealthier families "will go to college whether [she] work[s] with them or not."Working with this kind of student she is "not going to achieve social change."That is why she felt inspired to teach "students who don't have that support somewhere else, who don't have someone at home pushing them and who even knows how to prepare them for the future."Jane knows if she does "good and hard work," she "will influence students' opportunities in the future." Her pro-social engagement is tempered by pragmatic pursuits.Jane connects to her work on a daily basis because she "really likes what she does".She explains to us that she "look[s] forward to students interacting with" her, when she creates good lessons.She "feel[s] excited to teach things that students are interested in."She feels most rewarded when "students who struggled figure out a lesson or achieve something" beyond her expectations.It is at that moment when she feels proud of her work.Jane cares about what her colleagues and administrators think about her work.Also, she likes to feel that she is a high performer, not for external recognition, she says, but for herself.Yet, getting signals of good performance from others "is a matter of personal pride."For her, external rankings of performance are "not an incentive to work harder," but when she feels "dissatisfied with her own performance", she becomes activated.
Compared to Ardash and the modal pattern across all interviewees, Jane is more driven by specific goals.But specificity is provided, not by longer term student performance goals or any kind of teacher performance ratings, but by goals directly related to her teaching.She says, "For any unit, activity, or assignment, I would know what my objectives are."Some examples of these goals are: getting better at explaining how a piece of evidence proves a point; making a persuasive logical argument; getting better at understanding the nuance of a text.Sometimes she finds it difficult to accomplish her goals.Sometimes students "get" the lesson objective, but do not apply the skill later.Her goals are all about student mastery of content and skill.
Jane thinks that monetary rewards are "a good idea."But she also sees clear boundaries.She does not think that her colleagues at School C require any money to motivate them.They already do all they can.She is convinced that anyone "who is incentivized solely by money is not going to be good at anything in any profession."But money, she contends, has a role to play for those who are not motivated to work as professionals.

Discussion
In the wake of educational reform in the United States over the last decade or so, weighty performance goals and incentives were introduced into the school system that are extrinsic, sharp, explicit, and high-stakes in nature, relative to earlier eras.State and federal accountability systems, with standardized assessments, quantitative growth targets, and sanctions for not meeting those targets, had been a reality since 1999 in the state where the three schools are located.Spawned by the federal Teacher Incentive Fund initiative and following on the heels of high stakes accountability, more refined performance management systems added value-added measures, standardized teaching evaluations, bonus pay, and career ladders.
Presumably, the schools whose account is given here should be fertile soil for incentivedriven performance management.At the time of conducting the study, the schools had been used to state and federal accountability expectations for 13 years, and they volunteered to participate in the pay for performance pilot program funded by the US-federal government.Compared to typical public schools, the three schools were accustomed to charter self-management, market competition, and deregulated labor-management relations.These conditions, coupled with the relative youth of the teaching force, should have been conducive for extrinsic incentives to take hold and to influence work motivation in substantial ways.Indeed, while the schools had no choice participating in the accountability system, and while competition with regular district schools were a fact of life, the performance management system, which brought the researchers into the schools, was the schools' own choice.And indeed, we confirmed Ballou & Podgursky's (1993) finding that acceptance of performance-contingent pay among teachers might be much higher than one would expect.But as we saw, accepting an incentive as a good idea does not mean that one is regulated by it in one's intent or actions.
Following the body of literature on teacher work motivation that developed contemporaneously with incentive policies, one could expect a motivational pattern in which systemgenerated motives, such as attaining specific organizational goals, preventing sanctions, or garnering rewards, might play an important role in shaping teacher work motivation.We wondered if this was indeed the case, and if so, what would happen to autonomously generated, intrinsic, or service motives that research from a previous era believed to be dominant work motives.Did intrinsic motives endure or prevail despite the presumed power of extrinsic incentives?Were extrinsic and intrinsic motives juxtaposed with equal strengths?Did extrinsic incentives and goals manage to become internalized and self-determined?Or were intrinsic orientations reduced to second place or crowded out by more rational calculations of self-interest?
We tested the relative strength of several distinct work motives: pro-social and service commitments, identification with work (ownership, pleasure, duty), prestige, material benefits, and task specific goal setting.These motives array from more externally to more internally regulated ones, from more specific to more diffuse ones.From the outset, we steered clear of all notions of dichotomy along the dimensions of intrinsic versus extrinsic, internal versus external, bureaucratic versus professional, and so on.Instead, being mindful of the teacher work motivation research conducted in the pre-incentive era and concurrent with the era of extrinsic incentives, we explored a spectrum of motives and assumed that regardless of where a motive might have originated, for example stemming from system or organizational demands or stemming from an individual's personal life course, internalization occurs on a continuum so that what erstwhile may have appeared as alien and controlling to teachers, for example an externally generated evaluation score of their teaching, may over time become integrated into the fullness of their varied work motives and may subjectively be experienced as motivating autonomous action.
Public service motivation (PSM) research has shown that despite a strong momentum in many countries during the current era to manage public sector work along private business principles, public sector workers have maintained a pattern of public service motivation that is distinguished from the private sector.Service, altruism, citizenship, and notions of the common good are values that regulate public sector workers even when their performance is increasingly managed through goal setting, competition, and appeal to self-interested material rewards.
The three semi-public organizations we investigated are an interesting hybrid.On one hand, they are open to the public management principles advocated by neo-liberal public management doctrines, on the other hand their work is carried out in a societal environment characterized by social deprivation, adversity, and inequity that pushes to the forefront issues of fairness, justice, and service, and that makes the accrual of self-interested rewards a very uncertain calculus relative to the likelihood of sanctions and frustrations.Like many public and charter schools whose role it is to educate populations that are consigned to the bottom rungs of society's stratification ladder, the three schools examined here responded to their challenge with a distinct organizational mission of social justice and service.
In a nutshell, we found through both quantitative and qualitative data that for teachers in the three charter schools a constellation of public service motives pre-dominated.This constellation mostly consisted of diffuse pro-social commitments, ideologies of fairness and equity, a belief in the moral deservingness of deprived student populations in opposition to societal neglect and compensating for weak family support, the need for personal engagement in social-political change, identification with one's work as a personal calling, and the pleasure of constantly learning, growing, and improving as instructors and persons in relationship with students and colleagues.
By comparison, monetary rewards were embraced as already deserved.Even the idealists who ideologically objected to reward money would gladly take it when it was offered.But neither rewards, nor accountability, seemed to regulate behavior in a deep way.They were in the picture, and they were taken into consideration, but they did not incentivize specific behaviors or performances, save the opportunistic acceptance of available benefits for whatever behavior the system was willing to pay out.Prestige was not bestowed by official performance statuses within the accountability system, but came as a result of personal judgment of those who had an immediate insight into one's work.
We began this paper by asking why management by extrinsic incentives and organizational goal setting may have been far less powerful than designers of accountability and extrinsic reward systems had expected.In this study, we deliberately looked at schools, public charter schools with a social justice mission, that could have perhaps been fertile grounds for extrinsic incentives to connect to service commitments.It is perhaps not as surprising that in these schools rather traditional patterns of intrinsic motivation and internalized pro-social commitments prevailed.It is perhaps more surprising that motivation patterns that Dan Lortie may have been familiar with in the 1970's showed their enduring power in the schools, despite the great institutional convulsions of public management reform of the 1990s and 2000s that the schools were part of.But the great surprise finding came in the goal dimension.For any kind of performance management system banking on extrinsic incentives, goal setting is the fulcrum.If rewards are to regulate work, they need to go through goals.No clear goals guiding teachers' work --and the rewards, or sanctions, become arbitrary in their regulative function.In public service motivation research, goals are discussed in various ways.As we discussed earlier, those who stress goal ambiguity as an ingrained feature of public service work search for more diffuse commitments as main motivators.In other models, goal setting is in the center.Public service mission of the agency (for example schools) and its incentives and rewards, in combination with employees' values, produce a sense of overall job importance and commitment to organizational goals.Organizational goals, with their specific, attainable, and challenging characteristics, come to regulate work.Goals are central motivators.They are reinforced and aided by organizational mission and personal values.
There was no evidence in the data of such a dynamic among the teachers we studied.Instead it was striking that teachers understood "goals" as an enormously fluid and multi-faceted concept.No discrete performance goals stuck out that were clearly marked by either interviewees or survey respondents as guiding intent or actions.Goals were plentiful.There were learning and growth goals.There were goals to carry out certain activities.There were the objectives for today's lesson or this week's unit.There were broad images of what students ought to be able to do after having been taught.There were life goals, relationship goals, goals for citizenship, and goals for all participants in the educative process to feel good about themselves and each other.The most important quantitative performance goals that were watched carefully were the percentage of students who graduated from high school and went to post-secondary institutions.These goals were seen as reflective of the schools' mission to equalize educational opportunity and to open up tangible life chances for students.
Overall, the findings suggest that public management principles centering on goals and extrinsic incentives have had a fairly weak influence on teachers' work motivation in the three charter schools in the past and would have a difficult time connecting to teachers' deeper work motivation in the future.What the research uncovered instead is a pronounced public service motivation that most strongly responds to the challenging social context and to the adversities and needs that students in this environment expose teachers to.To speak with Frey & Osterloh (2001), acceptance of extrinsic incentives, freedom to manage, deregulated labor, etc. were far from crowding out the power of this public service motivation.Nor did the tools of incentive-driven performance management "crowd in" and become internalized into the deeper strata of teachers' work motivation.
It is important to keep in mind that this research was not intended to explore responses to policy or test the theory of action of accountability or pay for performance incentives.Our main approach in the survey was to give teachers a menu of possible motives that they believe guide their work as teachers.Our main questions to teachers in the interviews was not: how do accountability goals, monetary rewards, or incentives influence your work.Rather we asked, what gets you out of bed every morning to come to this school; what goals do you pursue when you do your work.We found that the language of "goals" came easy to the interviewed teachers and goal setting was pervasive, but in its pervasiveness it lost the specificity and clarity that goal setting theory presupposes as the motivational force of goals.

Limitation
Three small charter schools do not make a pattern from which one can derive direct practical lessons for a "system" of performance management.In this regard our study is limited.It is quite conceivable that there are teachers, schools, districts, or charter networks in which extrinsic incentives, most notably the threat of sanctions, crowd out service motivations with the result that work motivation is high as long as self-interest motives continue to be satisfied, or demoralization sets in when results are not forthcoming (see Finnigan & Gross, 2007).Or that there are organizations that successfully fuse self interest and social justice commitment.In such organizations, a strong self-interested drive to garner the rewards and prestige that the system's extrinsic incentives have on offer may also be seen as expressive of ones' social justice and service commitments, a pattern that we initially conjectured we might find in the three charters schools we studied.

Relevance
The relevance of this study is two-fold.Theoretically we wanted to show that teacher work motivation research, especially in the era of extrinsic incentives, benefits when multiple motivational dynamics are explored.These dynamics ought to be informed by work motivation theories that train the lens on varied conceptualizations of what constitutes core energetic forces under given circumstances of work.We wanted to show that expectancy, goal setting, self-concept, and selfdetermination theories embedded in an established line of research on public service motivation, can potentially yield understandings of the deep and multifaceted layers of teacher motivation under adverse circumstances.An eye to a critical sociology of status and deservingness helps understand the social context of work motivation.
If we extrapolated the pattern found here and imagined a suitable performance management, incentive, or accountability system for the type of teachers or organizations that came into view in this study, we would have to question the idea that systems banking on the power of extrinsic incentives, such as the now-faded NCLB system, could get us very far.If after 13 years of high stakes accountability, managerial freedoms for charter schools, and voluntary participation in a pay for performance system, we find the shallowest roots of incentives in the fertile soil of public service motivation, we need to search for new system designs.Here is not the space to get into design details.But two design principles are suggested by this research.We need to find ways for any sort of performance management to tap into, reinforce, and crowd in, positive and affirming commitments for equity, justice, and moral deservingness of marginalized students and their families, in contradistinction to subjecting teachers to a blanket system of corrective action.We further need to think hard about ways to develop the goal function in schools in a way that delivers on precision and leaves room for ambiguity and the fluidity of personal growth.There is a great deal of cooperative effort among staff here.I can count on colleagues here when I feel down about my teaching or my students.

Leadership and School Culture Scales
In this school, the faculty discusses major decisions and sees to it that they are carried out.Supportive Leadership (Cronbach's α=.68) The principal usually consults with staff members before s/he makes decisions that affect teachers.
Staff members are recognized for a job well done.
The school administration's behavior toward the staff is supportive and encouraging.Moral Leadership (Cronbach's α=.79) Places the needs of children ahead of personal and political interests.Models the kind of school they want to create.School Management (Cronbach's α=.82)This school is well managed.Overall this school functions well.This school is disorganized.a Autonomy (Cronbach's α=.81) Teachers' expertise in the classroom domain is respected here.In this school, I am encouraged to be creative in my classroom.
In this school, I am given the space to exercise my professional judgment as to what is best for my students.Open Communication (Cronbach's α=.80) Faculty gatherings provide a forum to discuss different perspectives on school improvement.It is okay to speak up when you disagree with the powers that be.
Open discussions about the criteria for good performance are encouraged.a. Values are reversed.

Figure 2 .
Figure 2. Strength of Work Motives *=Number of cases (n)Figure2compares the relative strengths of an array of work motives.Expressed in scale means, teachers in the three-school sample named as their main motivations pro-social commitments, feelings of identification with or ownership of the work, intrinsic pleasure associated with the work, and the difficulty and challenges their work poses to them.Specific goals were found in the neutral range.Sense of duty, concern for prestige, and garnering material benefits ranked

Figure 3 .
Figure 3. Usefulness of Performance Management Criteria *=Number of cases (n)

Figure 4 .
Figure 4. Adarsh's Motivational Profile Figure 5. Jane's Motivational Profile ItemsCollegiality (Cronbach's α=.82)Most of my colleagues share my beliefs and values about what the central mission of the school should be.

Table 1
Demographics and Performance 2012 Note: C means the school had significant demographic changes and will not have any growth or target information.* School C enrollment in secondary school grades

Table 4
Summary of Interviews per School and Survey

Table 5
DutyObligations derived from external sources.Duties are carried out irrespective of needs fulfillment (pleasure) or self-interest.They appeal to internal values or goals, but are clearly seen as external to oneself.Organizational goals can be one source of duties, more diffuse institutional obligations and responsibilities are another source.

Table 6
Meta-Matrix for Main Codes

Table 6
. I love it.I don't want to work in any other environment.I love it.(SchoolB, 9) You can't succeed in a population like this without empathy, without love.(SchoolA, 9)We're seeing all of the breakdown of some of these problems in our country in front of us.I love it.I love helping people help themselves.I love the challenge of what many different persons may bring.
(School C, 1)While difficulty was taken for granted, the majority of educators chose voluntarily "to work with this student population."Intheireyes, this work could only be carried out by a "special person," with particular sensitivity, who "knows how to relate with students and parents" and is strong enough not to give up:It takes, like, a special person to, like, teach like at our types of schools.Like, you have to, like, be strong.AndI was, like, whoa, and so it gets me a little emotional because I thought about it and I was, like, you know what, like, people can easily give up working with, like, our youth, you know.And so a part of me was, like, you know what -I do belong here and it's a challenge and I love it and I want to come back.(SchoolB, 8) Others doubted the energizing power of precise goals preferring to be carried forth by the "bigger picture:" I feel like I know how to go through the motions of setting that goal, taking steps to work toward it, and then reflecting on what I've done.But that doesn't feel like what really is driving me.It feels a little bit bigger picture than any one of those kind of specific instructional goals, if that makes sense.(School C, 20) proficiency [in math]."But these types of goals seemed rather irrelevant as motivational force.Some interviewees mentioned these goals as sources of negative pressure or stress: "The reason I don't remember [the quantitative targets] is because it stressed me out so much that I said, you know what?Ignore it" (School B, 8).

Table 7
Leadership and School Culture