Conceptual Bases of Employee Accountability: A Psychological Approach

Abstract

The external control of public organizations and their members, commonly referred to as accountability, is an enduring theme in public administration. This article shifts attention from a traditional focus on accountability as a macro-institutional matter to the psychology of accountability, that is, whether and how employees internalize accountability systems. The internalization of rules and expectations varies by individual, which, in turn, has significant consequences for accountability outcomes. We theorize that the micro-foundations of employee accountability are affected by five factors: attributability, observability, evaluability, answerability, and consequentiality. These five dimensions are conceptually distinct but interrelated, representing a deeper common psychological construct of employee accountability. Incorporating the psychological approach of accountability advances the potential of public accountability research. We conclude with a discussion of future research and practical implications.

Introduction

Accountability—a central concept in public administration (Romzek and Dubnick 1987)—is the leverage that maintains social systems and is the hallmark of democratic governance (Bovens 2005; Tetlock 1992). Consequently, holding public employees accountable for their actions has received much attention as a fundamental public management principle (Dubnick and Frederickson 2011a; Kettl and Kelman 2007; Roberts 2002), especially in organizing and organizations (Ferris et al. 1995; Romzek and Dubnick 1994). Accountability directs and guides people in their work, and, in turn, is enabled by the presence of rules and the notion that people are agents of their own actions (Frink and Klimoski 1998). To induce accountability, governments and organizations establish practices, such as oversight committees, employment contracts, disciplinary procedures, and reporting systems.

Accountability is a meaningful concept at many analytic levels—from institutions, to organizations, to individuals (Dubnick 2005; Frink et al. 2008; Han and Hong 2019; Lerner and Tetlock 1999; Peters 2014; Romzek and Dubnick 1987; Romzek, LeRoux, and Blackmar 2012). In the past three decades, public administration research has made significant progress in understanding ubiquitous accountability phenomena (e.g., Bardach and Lesser 1996; Behn 2001; Bovens 2010; Han 2019; Koppell 2005; Mulgan 2000). Nevertheless, accountability remains a vexing problem: why employees feel accountable for their actions remains unclear. To understand employees’ subjective experience of being accountable and develop predictability about the effects of accountability on individual outcomes, we propose a shift in analytic focus: from macro to micro. Evaluating micro-foundations of public administration theories by integrating a psychological approach—behavioral public administration—has recently been emphasized for strengthening public administration knowledge (Grimmelikhuijsen et al. 2017; Kelman 2007). Based on growing streams of disciplinary research, we seek to extend the potential of accountability research by improving the explanatory power of how public accountability systems can make a difference to employees.

A popular belief in modern governance is that holding employees accountable for their actions is an effective means to control behavioral outcomes. Public organizations require that their employees perceive themselves as accountable for their actions by internalizing rules and expectations from external sources (Dubnick and Yang 2011; Velotti and Justice 2016; Yang 2012). Accountability-inducing practices are not uniformly effective, however, because internalization varies by individual: when exposed to the same situation, different employees can hold quite different expectations about their accountability (e.g., Frink and Klimoski 1998; Tetlock 1985, 1992). This discrepancy exists because individual interpretations of external systems are variable. The subjective experience of being accountable is a different phenomenon than subjective ex post facto evaluation of accountability systems that has been a traditional public accountability approach. An understanding of subjective individual experiences is required for research and practice regarding public accountability (Romzek 2014, 317). Integrating micro- and macro-perspectives, therefore, can add measurably to our understanding of accountability.

The psychology and management literatures refer to accountability at the individual level interchangeably as “individual accountability,” “individual-level accountability,” “felt accountability,” or “employee accountability” (e.g., Frink and Klimoski 1998; Hall et al. 2003; Lerner and Tetlock 1999), clearly distinguishing individual accountability from other concepts such as organizational/institutional accountability and responsibility (e.g., Koch and Wüstemann 2014; Morrison and Robinson 1997; Schlenker et al. 1994; Tetlock 1985). Scholars have shown individual-level accountability has significant effects on performance, behaviors, attitudes, and judgments (Hall, Frink, and Buckley 2017; Lerner and Tetlock 1999). We will use the term employee accountability throughout this article, especially when we are referring exclusively to employees in organizations. The exception will be when we are discussing the original empirical studies, where we will defer to the concept name used in the research.

The employee accountability concept is relevant to advancing public administration and management in two ways. First, an employee’s psychological state, as psychology and management research shows, directly affects a public employee’s work outcomes. In addition, the micro lens offers information different from what we already know based on traditional macro accountability research, subjective ex post facto evaluation of accountability mechanisms: although a goal of macro accountability structures is to change individual behavior, we do not know the result unless we measure variations of macro-accountability and the effects of the macro-changes on micro-behavior. The micro lens allows us to capture accountability’s variations and its outcomes (Hall, Frink, and Buckley 2017; Tetlock 1985, 1992) when individuals are exposed to a common accountability practice. Thus, the micro approach enhances predictability about effects of accountability structures on individuals. The micro-perspective also helps us to understand individual accountability in a broader environment of action, which may include multiple accountability arrangements that may be received differently across individuals (Romzek and Dubnick 1987), creating potentially conflicting or unintended influences on people exposed to them. Therefore, public administration and public management needs to integrate the micro lens when examining public accountability.

Management and psychology scholars generally agree that individual-level accountability consists of multiple conceptual dimensions (e.g., Lerner and Tetlock 1999). However, previous research simply identifies dimensions (e.g., Lerner and Tetlock 1999, 255) rather than fully develops the construct in ways that explain how each dimension can be applied to work settings in the public sector. Moreover, multiple dimensions have been studied inconsistently in empirical studies (Dubnick 2011; Hall, Frink, and Buckley 2017). Thus, the absence of a conceptual model that incorporates multiple dimensions of employee accountability limits rigorous analysis and comparison of each dimension, creating inconsistencies in study results.

The purpose of this article is two-fold. First, using a multidisciplinary literature review, we identify and describe five theoretical bases of employee accountability to fully disclose its multidimensionality: (1) attributability, (2) observability, (3) evaluability, (4) answerability, and (5) consequentiality. These dimensions represent what employee accountability is and why employees internalize external stimuli and rules. This study proposes a conceptual model of employee accountability consisting of the five bases. The model explains how the five components are related and define employee accountability. Second, the article integrates public sector literature with psychological research by applying the multidimensional psychological aspects of accountability to the public domain, considering differences between the public and private sectors.

We proceed in three stages. We begin by reviewing mainstream accountability research in public administration and formally answer the question: what is employee accountability? Next, we turn to identifying its multiple dimensions, based primarily on a review of multidisciplinary literature including public accountability and experimental research, and offer a conceptual model of employee accountability. Then, we discuss how the model advances existing public administration and public management research. The article concludes with a discussion of the implications of the multidimensional concept for theory and practice.

Conceptualization: What Is Employee Accountability?

Accountability has been a frequent object of attention, not only in public administration, but across a range of disciplines and fields. We first identify the ways the accountability phenomenon is manifest in public administration before discussing research streams outside public administration that are the primary focus of this article.

Public Administration

The most prominent research about accountability grounded in US institutions is Romzek and Dubnick’s (1987) pioneering study focusing on the 1986 Space Shuttle Challenger tragedy. Romzek and Dubnick observe that public accountability was associated historically with the “answerability” of agencies and administrators for their actions, but they caution that public accountability involves much more than answerability. They write: “More broadly conceived, public administration accountability involves the means by which public agencies and their workers manage the diverse expectations generated within and outside the organization” (228, emphasis in the original). Romzek and Dubnick use this definition to identify four accountability systems that represent different sources and degrees of agency control for managing expectations: bureaucratic, legal, professional, and political. Their foundational article has launched much additional research, largely focused on accountability systems and control of conflicting expectations (e.g., Radin 2011; Romzek and Ingraham 2000).

Jonathan Koppell (2005) proposed an alternative accountability model that he contends addresses an overemphasis in the Romzek and Dubnick framework on control and mixing together types of accountability that are substantively different. Koppell proposes five different dimensions of accountability—transparency, liability, controllability, responsibility, and responsiveness. He argues that these distinct dimensions of accountability reduce conceptual fuzziness associated with Romzek and Dubnick’s four accountability systems that focused on different sources of whom to be accountable (Dubnick and Frederickson 2011a). The bottom line for Koppell (2005), similar to that of Romzek and Dubnick, is that “conflicting expectations borne of disparate conceptions of accountability undermine organizational effectiveness” (94), which he calls multiple accountabilities disorder.

A third line of accountability research, cognizant of the two streams above, arose in Europe, partly driven by “accountability deficits” perceived as a consequence of European governance. This accountability model emphasizes social relations between accountable actors and an accountability forum (Bovens 2007). Bovens and others (e.g., Schillemans 2011) view this phenomenon as narrower than the conceptions of Romzek and Dubnick (1987) and Koppell (2003, 2005). Bovens (2007) defines accountability as: “a relationship between an actor and a forum, in which the actor has an obligation to explain and to justify his or her conduct, the forum can pose questions and pass judgement, and the actor may face consequences” (450). Bovens asserts that this definition is not evaluative like the broad definitions used by other public administration scholars, but instead emphasizes social relations that involve a specific actor, obligated to answer to a specific forum, and liable for consequences presumably controlled by the forum.

Psychology and Management: Micro Approaches to Accountability

Public administration scholars and practitioners are likely familiar with at least one and probably more of the lines of accountability research mentioned above: ex post examinations about accountability systems. Existing public administration research, however, is limited about the micro foundations of public employees’ subjective accountability experiences. A line of research less prominent in public administration is about employee or felt accountability, which was first pursued in social psychology. The concept has evolved in the field of management to illuminate accountability phenomena in work settings (Hall et al. 2003). Social psychologists Lerner and Tetlock define individual accountability as the “implicit or explicit expectation that one may be called on to justify one’s beliefs, feelings, and actions to others” (1999, 255). They also emphasize that individuals should expect consequences such as rewards or punishments based on their justifications. We use a modified definition of employee accountability based on their descriptions:

Employee accountability is the expectation that one may be called on to explain an action or inaction to salient audience(s) with the belief of a consequence based on evaluation.

Social Contingency Model of Individual Accountability

Holding individuals accountable via diverse practices is a fundamental goal of organizations. A well-regarded perspective conceptualizing individual accountability in psychology and management is the social contingency model (Hall, Frink, and Buckley 2017; Tetlock 1985, 1992). The model is comprised of three ideas. First is that accountability at the individual level refers to one’s own accountability, not the extent to which an organization is accountable. Applying this focus, scholars have examined the effects of individual accountability (Hall, Frink, and Buckley 2017; Lerner and Tetlock 1999), including job satisfaction (e.g., Breaux et al. 2009), motivation (e.g., Lanivich et al. 2010), and tension/stress (e.g., Hochwarter et al. 2005), along with ethical behavior (e.g., Mackey et al. 2018), job performance (e.g., Frink and Ferris 1998), and organizational citizenship behavior (e.g., Hall et al. 2009). DeZoort, Harrison, and Taylor (2006) found, for example, that the more accountable subjects feel for their actions, the more effort they expend. As such, the model has been utilized to examine outcomes associated with the degree to which employees perceive themselves accountable for their actions.

The second idea in the social contingency model is that individual accountability is a state rather than an unchangeable trait such as personality (Koch and Wüstemann 2014): the formation of individual accountability is contingent on contemporary external social sources. When individuals feel accountable, the theory asserts that the felt accountability occurs because of external sources and can be manipulated situationally. For example, someone may ask me why I feel accountable in my job. I can say, “I feel accountable because I expect my bad performance to result in a penalty,” or “I expect that I will be required to explain my actions.” To demonstrate this, Tetlock and his colleagues (e.g., 2013) manipulated accountability conditions to induce subjects to feel accountable. The model therefore emphasizes that individual (micro-level) accountability is connected to macro-level accountability. Thus, it implies that contextual factors under an organization’s control need to be intentionally managed to influence employee accountability.

The third idea of the social contingency model refers to a phenomenological view that humans perceive and interpret an objective reality (Frink and Klimoski 1998). Looking at the individual, subjective, and internal nature of accountability, the perspective emphasizes accountability as a state of mind, rather than a state of affairs. These individual subjective interpretations of accountability are the focus of interest, rather than the objective accountability mechanisms imposed on that person, because the human psychological state drives individual behaviors. Thus, the social contingency framework proposes that accountability internalized from external sources causes outcomes, such as judgment, decision, and attitudes. These individual subjective interpretations of an objective reality are not identical by person (e.g., Frink and Klimoski 1998). Thus, when a system is designed for holding employees accountable for their actions, the levels of accountability they feel may not be equal and individuals may behave differently (Hall et al. 2009; Mero, Guidice, and Werner 2014).

Similar Concepts of Individual Accountability

Accountability at the individual level, however, is not explained only by contemporary contextual factors. Individual accountability may also can be affected by self-derived expectations such as personal values (Cummings and Anton 1990; Maclagan 1983). For example, a belief that people should bear penalties for wrongdoing is mostly unchangeable and inherent in individuals regardless of organizational systems. Such values are stable, long-lasting, and internalized over the long run through the socialization process (Chaplin, John, and Goldberg 1988). Thus, personally derived values formed through socialization can influence the personal accountability of employees. However, research has shown when external systems, such as a code of conduct, are present, personal values are not strongly related to unethical behavior as their impact is suppressed by the external system (Brief, Dukerich, and Doran 1991; Brief et al. 1996). This result implies that accountability internalized from external sources may be more influential for employee outcomes than accountability as a personal belief. Nevertheless, such individual intrinsic characteristics as well as socially contingent values contribute to the nuance of micro-level accountability.

Accountability is closely related to responsibility, so much so that the two terms have often been used jointly and/or interchangeably in practice and research (Wikhamn and Hall 2014). However, some social scientists note that responsibility is clearly and conceptually distinct from accountability (e.g. Dubnick and Frederickson 2011b; Koppell 2005): others’ existence is key in distinguishing accountability from responsibility, especially at the individual level (Cummings and Anton 1990; Schlenker et al. 1994). The responsibility concept does not cover others’ evaluation, which is at the core of the accountability concept, but responsibility does involve obligations based on self-evaluation. Responsibility connotes personal causal influence on an event, and the observer/judger of the causal behaviors is the relevant performer. Accountability reflects relationships with external, social, and public processes. In this sense, responsibility has been recognized as internal accountability forming individual accountability (Cummings and Anton 1990; Maclagan 1983).

In sum, employee accountability is a psychological construct linking accountability to individual cognitions and behaviors. The literature review suggests that looking only at employee accountability as a socially contingent state is limited in grasping accountability at the individual level. Therefore, we conceptualize employee accountability to include psychological factors affected by both contemporary external contexts and intrinsic factors.

A Multidimensional Conceptual Model of Employee Accountability

Scholars have recognized that employee accountability inherently involves complex and multidimensional characteristics concerning why individuals feel accountable (e.g., Hall, Frink, and Buckley 2017; Lerner and Tetlock 1999). Based on a review of behavioral science research, Lerner and Tetlock argue that the simplest accountability manipulations to make participants feel accountable involve at least four empirically distinguishable dimensions (1999, 255): (1) mere presence of another, (2) identifiability, (3) evaluation, and (4) reason-giving. Previous empirical studies have used these four dimensions in an ad hoc way to capture multiple aspects of employee accountability, typically focusing on one or a few of them (e.g., De Dreu and van Knippenberg 2005; Gelfand and Realo 1999; Green, Visser, and Tetlock 2000).

Lerner and Tetlock left full development of a multidimensional model of individual accountability to future research. Their research does not fully inform scholars of the meanings of the accountability dimensions, how the dimensions are present in the work setting, including the public sector, how the model is integrated into existing public administration and management literature, and what the theoretical and practical implications of individual accountability and its dimensions are for advancing public sector research. A grasp of these issues is fundamental for understanding the multifaceted employee accountability concept for developing accountability theory. Thus, we conducted a multidisciplinary literature review to identify and develop more completely the conceptual bases of employee accountability.

We employed the following process in our review. Articles including “individual accountability,” “individual-level accountability,” “felt accountability,” and “employee accountability” in their titles or abstracts were first identified via a Google Scholar search. All of the articles that used accountability in the title or abstract and covered accountability at the individual level were considered for the present study. The review led to identifying a fifth distinct component of employee accountability. Roch (2005) and Wang, Noe, and Wang (2014) found that a fifth dimension, consequentiality, produced accountability effects distinct from a fourth dimension, evaluability. As a result, our review leads to an employee accountability construct distilled into five dimensions: (1) attributability, (2) observability, (3) evaluability, (4) answerability, and (5) consequentiality.

Employee Accountability’s Five Micro Foundations and Linkages to the Public Sector

Table 1 summarizes a variety of experimental studies that sought to establish causality by manipulating accountability requirements affecting the five dimensions to make participants feel accountable. The discussion of each dimension below includes its definition, relevant experimental results, and connections to public administration and public management.

Linkage between Action and Self: Attributability

Employees who feel accountable expect their contributions, errors, and activities to be identified and linked to them. This expectation concerns the imputation of causality: “Who did it?” that is, whether a person caused a particular event (Schlenker et al. 1994). Attribution theorists have long noted that in everyday life people are presumed to be agents of their actions as they are responsible for what they do (Kelley 1973; Tetlock 1985). Attributability is about actions or attitudes being properly attributable to—and reflective of—the agent’s self (Shoemaker 2011). In various fields including psychology, politics, business, public administration, and management, accountability debates rage about “who should answer to whom, for what, and under what ground rules” (Bovens 2005; Lerner and Tetlock 1999, 255; Thompson 1980). As such, the factor concerning “who should answer?” is essential for the accountability concept.

Attributability has been specified as a matter of choice in empirical studies. For example, Schlenker et al. (1994) manipulate attributability to induce subjects to intentionally produce consequences without coercion. They conclude that more responsible people are associated with stronger triangular linkages binding actors, prescriptions to guide their behaviors, and behaviors involved in the event. What a person decides or does is linked to her/him personally. One might say, “It is my decision,” and it means direct choice (Fox and Staw 1979). Thus personal, explicit, and unambiguous events increase attributability. Price (1987) experimented by asking participants to place their names on voting ballots. In such an experimental setting, anonymous participants were expected not to feel accountable because individual decision outputs could not be linked to them personally. Attributability is also fostered through group membership when individuals share the group’s goal. As long as employees choose to be a member of an organization and maintain their membership, they share the collective responsibility of the group. Responsible employees conform their actions to organizational rules (Ferris et al. 1997; Tetlock, Skitka, and Boettger 1989). In the case of organizational misconduct, therefore, every organizational member may be held accountable (Bovens 2005).

Role acceptance is another way attributability develops in an organization (Cummings and Anton 1990). Many employees feel responsibility in the workplace because the job assigned to them is their own job and their job-related activities are linked to them. Public administrators entrust skilled employees with tasks relying on their expertise. Public experts manage their expectations via informal and internal accountability mechanisms for which they are responsible rather than close supervision (Romzek, LeRoux, and Blackmar 2012; Finer 1941). Thus, such public employees feel strong attachments to their assigned jobs and professional norms. Such role expectations include vicarious responsibility in which the person is responsible for the action of others due to his/her role (Semin and Manstead 1983). When one accepts the role as head of a department, for example, the head takes on the responsibility for the overall condition of the organizational unit. Further, the head can be accountable for others such as their employees, even though poor decisions may not have been made by the head (Zellars et al. 2011). Conventional wisdom is that as one ascends the organizational hierarchy, there should be an accompanying increase in employee accountability (Ferris et al. 1997). In the public sector, public employees can often confront conflicting expectations or pressures from institutionally different accountability sources (e.g., Romzek and Dubnick 1987). In this confusing situation, employees suffer role conflicts (Rizzo, House, and Lirtzman 1970). The role conflicts negatively influence behaviors associated with accountability expectations (Schlenker et al. 1994; Sims 2000).

Group members’ attributability can be affected by the extent to which their contributions to the achievement of a common goal are identified. To test the effects of this attributability mechanism, scholars have manipulated accountability conditions in experiments: for example, Williams, Harkins, and Latané (1981) set up a soundproof room to create anonymity for an experimental task of group shouting. Under those conditions, the subjects did not expect their linkages to the goal to be identified. Complex environments around public organizations may create uncertainty about who is causally responsible for an action, which could decrease employee attributability (Bovens 2005). Mitchell et al. (1998) found that, in a manipulated setting, clearer standards as to what should be done increased conformity among decision makers. Therefore, monitoring and scrutinizing an employee’s effort or making clearer job descriptions and prescriptions for behavior can increase employee attributability.

Presence of Another: Observability

Employees who perceive a working accountability system expect their job-related activities to be seen by others. The expectation to be observed is distinct from the expectation to be identified (Lerner and Tetlcok 1999, 255). For example, performers in an orchestra can expect audiences to observe them, but the performers might not expect their contribution to the orchestra to be recognized. Social facilitation theory contends that presence of others is a source of arousal (Zajonc and Sales 1966). The observability mechanism works by the mere presence of observers even when there is no direct detection (Frink and Klimoski 1998). Individuals feel accountability because the presence of an audience triggers cognitive mechanisms or feelings that affect individual behavior.

In experimental settings, observability often manifests as the mere presence of another (Lerner and Tetlock 1999). It is an experimental technique often used for testing the observability dimension to examine the responses of subjects in front of people whose eyes are covered. In particular, observability and attributability have often been separately manipulated in the same experimental study: in the blinded study, monitoring effects were removed, but the independent effect of the mere presence not to detect subjects’ activities and decisions was discovered (Guerin 1986). Observability was also manipulated by making the third person (i.e. observer) simply show up or not show up in front of the other subjects (Stone, Yates, and Caruthers 2002) while identifiability was manipulated by making subjects’ submission to an experimenter anonymous or nonanonymous (e.g., Schopler et al. 1995). The former found that exposing subjects’ decision-making environment to a present observer had an independent effect on risk-taking behavior.

The observability dimension has also been specified in experiments as individuals being observed by multiple audiences. A social contingency perspective of accountability (Tetlock 1985, 1992) include audiences’ characteristics (Lerner and Tetlock 1999). For example, Seta, Crisson, et al. (1989) increased audience size to manipulate the observability dimension of accountability and confirmed that the size of audiences with high status positively affected felt anxiety. Within organizations, individuals can expect that their attitudes or efforts can be observed by multiple audiences including supervisors, coworkers, team members, and/or clients (Bovens 2007; Carnevale 1985). As the size of audiences increases, employees generally feel more accountable (Frink and Klimoski 1998) because they expect to be observed. As such, external sources such as audience size can affect employee accountability.

Observability has been associated with another phenomenon: opening or sharing information with others outside a group or organization (Geen 1991). Outside audiences especially relevant for public organizations include courts, elected officials, the media, interest groups, professional organizations, and citizens (Romzek and Dubnick 1987). In many cases, citizens cannot observe civil servants’ job-related activities, but can influence civil servants by acts such as requesting government information or filing a civil complaint (Brewer 2007). Political accountability expectations are managed for more open and representative public agencies as the “public” literally competes against the “private” (Bozeman and Bretschneider 1994; Perry and Rainey 1988; Romzek and Dubnick 1987). This attribute of the public sector is actually reflected in laws or rules: for example, the US Freedom of Information Act. Any civil servant may expect that audiences such as professional groups may freely request their organization’s performance information that may be necessary and interesting to them (Wood and Lewis 2017). Thus, the observability mechanism is notable in public organizations where transparency is emphasized in a democratic society. Transparency has been advanced as a fundamental tool to achieve accountability (Koppell 2005: Piotrowski and Van Ryzin 2007).

Assessment by Another Based on Rules: Evaluability

Employees who expect their performance to be reviewed are more likely to feel accountable. Scholars (Frink and Klimoski 1998; Lerner and Tetlock 1999) identify the expectation for evaluation as a key element of individual accountability. Individuals are recognized as agents of their own actions who are subject to evaluation on their conduct in an accountability relationship (Tetlock 1985). According to social influence theory (Ferris and Mitchell 1987), the arousal of people is striking when their audience gives off cues that they can evaluate the subject (Paulus and Murdoch 1971). The arousal based on evaluation expectations becomes leverage to maintain the social relationship and the evaluative reckoning is at the heart of accountability (Schlenker et al. 1994).

Experimental studies have verified that evaluability increases individual accountability. Harkins and Jackson (1985) manipulated the attributability and evaluability dimensions of accountability separately to induce subjects’ accountability and found independent effects. Furthermore, combinations of different accountability dimensions have synergistic effects on individual accountability. The absence of an audience and low evaluation resulted in low individual accountability, while the mere presence of an audience and a high evaluation expectation led to high individual accountability (Innes and Young 1975). Conducting an evaluation to compare group members’ performance additionally can be more helpful to eliminate social loafing than only identifying their efforts in a group activity to prevent social loafing (Geen 1991; Sanna, Turley, and Mark 1996).

Employees can expect that their activities would be evaluated by audience(s), such as superiors, through an evaluation system (Ouchi and Maguire 1975). Evaluability has been measured by assigning a task to subjects whose performance can be rated and who will receive feedback from a group leader. For example, Frink and Ferris (1998) manipulated evaluability in a laboratory experiment by informing subjects that there would be a group meeting with a team leader. The subjects were expected to formally discuss their goals and performance with the leader. Evaluability in an organizational setting depends on performance measurement/rating, comparison with others, and/or feedback. Information exchanges among principals increase the accountability of public agents (Schillemans 2011). Performance measurement and accountability are also two common terms in public administration and management (Moynihan and Ingraham 2003; Radin 2011). Performance information data have been produced to evaluate ongoing performance for public organizations. Generating feedback increases bureaucrats’ evaluation expectations while standards form expectations against which they can be evaluated (DiIulio 1994). Accordingly, the performance evaluation system is a formal accountability system to generate felt accountability (Ferris et al. 1995; Hall et al. 2003). Agents entering into an accountability relation expect that their performance will be assessed according to some normative ground rules (Lerner and Tetlock 1999). Organizations should establish appropriate and useful criteria in terms of performance definitions, rating tools, and feedback provisions for their employees to feel accountable (Hall et al. 2003).

Evaluability has also been examined by allowing subjects to be assessed by multiple and different types of audiences. As the number of evaluators increases, individuals generally feel more evaluated and therefore more accountable (Bagley 2010; Frink and Klimoski 1998). Employees’ performance can be evaluated by peers, clients, or external professional experts, among others, in addition to traditional assessments by supervisors (Bovens 2005). Frink and Ferris (1998) arranged no evaluating audiences for no accountability, supervisors for medium accountability, and supervisors and peers for high accountability. In particular, more arousal will be produced when the evaluator’s status is high or evaluation results are likely to be unpredictable or negative rather than positive (Seta, Wang, et al. 1989). Thus, evaluability characteristics can influence employee accountability.

Reason-Giving for Actions: Answerability

Accountability has been recognized as the adhesive that binds social systems together (Frink and Klimoski 1998). People should be able to explain and justify their behaviors within the social order of an accountability relationship with others. Otherwise, there would be neither shared expectations nor a basis for the social order. It is a typical mechanism of external accountability systems (Romzek and Dubnick 1987) that employees are expected to explain their work-related decisions and actions with proper justification (Hwang and Han 2019). A common form of such answerability is how employees address a customer’s or citizen’s complaint. Scholars explicated that the answerability expectation is an independent core element of individual accountability (Hall, Frink, and Buckley 2017; Tetlock 1992).

The answerability dimension is examined via the considerable number of experimental studies by asking participants to explain or justify their decision or activities to an experimenter or group members to induce the participants to feel accountable (Scholten et al. 2007; Sedikides et al. 2002; Tetlock and Boettger 1994). For covering a specific work context, especially, Curtis, Harvey, and Ravden (2005) and Klimoski and Inks (1990) manipulated the accountability of supervisors to subordinates. When supervisors were required to provide feedback for subordinates in a laboratory experiment, they were also asked to justify their evaluation decision. The supervisors felt more accountable when they met face-to-face to give feedback to the subordinates than when they provided only written feedback for their subordinates without face-to-face meetings.

Answerability has often been specified in experimental conditions as not allowing excuses as justifications for one’s own actions and decisions. If employees work in an organization that allows excuses and waivers, then their expectation to explain and justify job-related issues is affected and their accountability is diminished (Schlenker et al. 1994). Willingness to allow excuses weakens answerability and, in turn, reduces accountability (Schlenker et al. 1994; Thompson 1980). Even when officials’ choices are deeply constrained might not become grounds to exempt an official from responsibility, especially in situations that involve the many hands of public officials and politicians (Bovens 2005; Thompson 1980). Under a weak accountability system that tolerates false justifications without treatments based on ground rules, the expectation to give reasons and answers decreases, which in turn negatively affects ethical behavior (Mackey et al. 2018). Furthermore, leaders may be held legally liable for their subordinates’ misconduct and therefore may defend them because leaders accept roles assigned to their positions (Zellars et al. 2011). Thus, the answerability of the leader may be high because of the role mechanism (Cummings and Anton 1990).

Scholars have argued that multiple answerability sources affect accountability mechanisms (e.g., Frink and Klimoski 1998; Romzek and Dubnick 1987). Experimental studies have examined answerability by establishing situations of reporting and justifying their work to multiple supervisors or customers. For example, DeZoort, Harrison, and Taylor (2006) and Bagley (2010) manipulated answerability using “no accountability,” “single accountability,” or “multiple accountabilities” conditions targeting auditors of public accounting firms to assess subjects’ feelings of accountability. Recently, various accountability systems, including the Program Assessment Rating Tool (PART), which was used during the George W. Bush Administration, have been introduced to public organizations. Scholars and practitioners have recognized that recursive and multiple accountability demands can improve public accountability deficits and bureaucratic legitimacy (e.g., Schillemans 2011). These new accountability requirements for public organizations can increase reporting and answerability to different forums (Radin 2011).

Outcomes Accompanying Actions: Consequentiality

Employees working under effective accountability systems expect their actions to result in consequences. Long ago, Greek philosophers, including Plato and Aristotle, used the accountability concept as punishment for individual wrongdoing for deviations below expectations (Schlenker et al. 1994). It continues to be used synonymously with punishment or sanctions (Mansbridge 2014) and managing outcomes can increase accountability for the public sector (Heinrich 2002). When somebody says that they will be accountable for any event, it can indicate that he or she would bear penalties, such as blame or resignation. Applying Vroom’s (1964) expectancy theory, people are generally motivated by expectations to gain positive rewards from people to whom they are accountable. Employees can expect positive consequences ranging from recognition to promotion for deviations above expectations. The expectation of potential consequences is the impetus for individual reactions to accountability (Frink and Klimoski 1998). This accountability mechanism has been effective in controlling employees and preventing acts detrimental to the organization (Beu and Buckley 2004).

Experimental studies have manipulated consequentiality to induce individual accountability by informing participants that they would be evaluated by audiences enforcing an extrinsic reward/sanction. For example, Wang, Noe, and Wang (2014) manipulated two conceptually distinct dimensions in their field experiment: (1) evaluation only and (2) evaluation plus reward. They found that evaluation plus positive reward was more effective on knowledge-sharing behavior than the evaluation only condition. Roch (2005) also separated evaluation and consequences in a laboratory experiment. The experimental results showed that providing an external incentive to raters, plus evaluating the rating of raters, was more effective in rating quality than only comparing ratings of raters. Thus, the two dimensions are not only conceptually distinct, but also empirically distinct.

Rewards involve not only positive and negative consequences, but also extrinsic and intrinsic attributes, such as self-realization. The evaluation does not necessarily lead to extrinsic rewards or sanctions. For example, student advisors can review students’ work without providing extrinsic reward or sanction. Without such extrinsic reward, students can feel self-achievement as an intrinsic reward. Also, in a work setting, extrinsic rewards are likely to be less effective for highly conscientious individuals because these employees enjoy the job itself, are attracted to the job, and are dutiful in their performance (Fong and Tosi 2007), especially in strong-culture public organizations (DiIulio 1994). Such employees may be less influenced by external than internal rewards (Schillemans and Busuioc 2014). Research on pay-for-performance systems for the public sector has also shown that the system has often failed in motivating employees pursuing public service (e.g., Perry, Engbers, and Jun 2009) because the reward system was not as powerful as intrinsic motives.

Consequentiality has also been specified in studies examining whether accountability systems work based on predetermined sets of expectations. According to equity theory (Adams 1965), individuals are motivated when they feel the reward is fair when compared with others. Employees’ beliefs that their organization will fulfill contractual obligations such as pay/promotion can vary, which affect their expectation that rewards would follow (Morrison and Robinson 1997). Accordingly, if they think that the compensation system is unfair to them, they would feel less accountable for their work. For example, Hall et al. (2003) found that the relevance and fairness of pay and promotion policies positively influenced employees’ felt accountability. Factors other than pay, such as job characteristics, could be understood as a fairer compensation standard for public servants if the pay-for-performance system does not work well for government (Rainey 2009). Instead of increased pay, if public employees have more opportunities to propose policy as the consequence of high performance, they may feel more accountable. Thus, when consequences are perceived as appropriate based on preexisting norms or expectations, employees may feel accountable.

A Five-Dimension Micro Model of Employee Accountability

The literature review concludes that when employees feel accountability, they commonly expect their actions to: (1) be attributed to them; (2) be observed by others; (3) be evaluated; (4) require answering for them; and (5) carry consequences such as rewards and punishments. Employee accountability is represented and influenced by five distinct dimensions at the micro level. Although not previously studied collectively, each of five dimensions is associated with a deeper common psychological meaning—employee accountability. Employee accountability is likely an additive function of the five factors. Therefore, all five factors may need to be present either to maximize felt accountability or for scholars to fully map and assess the accountability landscape at the individual level. In sum, employee accountability conceptually consists of five distinct dimensions that are inter-related, as depicted in figure 1. The five dimensions additively contribute to the higher-order construct, employee accountability, as depicted in figure 2.

Advancing Public Administration Research

The five-dimension model of employee accountability provides a foundation for transforming public accountability research. Experimental research informs us that employee accountability is contingent on situations external to the individual. The five-dimension employee accountability construct identified in the literature review for the first time gives us a micro-tool that can be used in conjunction with macro-tools (e.g., legal, bureaucratic, professional and political) traditionally used in public administration research (Romzek and Dubnick 1987).

One example of a substantive topic where the employee accountability construct may be helpful is understanding the behavior of street-level bureaucrats who exercise discretion about diverse accountability requirements (Maynard-Moody and Musheno 2000). When street-level bureaucrats need to manage policy expectations, they sometimes break rules or apply the rules inconsistently, adhering to their own judgment. Such behavioral outcomes constitute avoiding accountability. Physical distance and information asymmetry between workers and their supervisors can lower attributability. At the same time, the efforts of street-level workers to help clients can be significant. Despite low attributability, their effort may result from the consequentiality of their actions, which has meaningful impacts on their organizations, clients, and society. Thus, even if street-level workers’ attributability may be low, the consequentiality of their actions may be high, and their efforts and performance may be high, too.

Public administration research has shown that public agencies with too many rules are often ineffective for their employees (Bozeman 1993) including street-level bureaucrats. Too much red-tape may confuse the street-level employees, and negatively influence work’s speed or administrative costs. On the other hand, functions of an accountability system can be defective due to its loose constraints (Schillemans 2011). As such, consistently applied rules, written requirements, and optimal control can be useful to front-line workers (DeHart-Davis 2008) or redundancy can improve accountability deficits. The multiple requirements can increase accountability feelings such as evaluability and answerability, which would contribute to the improvement of the street-level workers’ efforts and workload, ultimately advancing performance. The conceptual model of employee accountability says that their subjective experience of being accountable based on rules varies by individual. Thus, impacts of rule characteristics on employees rely on accountability feeling. It will be meaningful to examine how rule characteristics influence front-line employees’ accountability feeling and their behaviors. As such, the employee accountability model supplements predictability on effects of rules in the public sector.

Public employees, including street-level workers, often face conflicting multiple requirements among institutional superiors such as managers and professional entities. Public administration research has found ex post facto that such conflicts in answerability to superiors led to poor decision making and performance (e.g., Romzek and Dubnick 1987). The conceptual model of employee accountability explains that conflicts may aggravate uncertainty on reward or situational complexity, reducing the efficacy of professional norms or organizational rules. The institutional conflicts among accountability sources can negatively influence the effects of multiple answerability or evaluability expectations on work outcomes. The employee accountability model is informative for predicting the effects of such contexts on individual outcomes.

In sum, various macro factors including conflicting pressures, multiple demands, and characteristics of accountability sources can positively or negatively influence employee accountability and its impacts in public organizations. Integrating the psychological approach into public administration accountability research creates a foundation for middle-range accountability theory that explains why and how public accountability systems make a difference to individuals.

Next Steps

Scholars argue that accountability is ubiquitous as its mechanisms are manifest at all levels of an organization. Accountability research shows that there are fundamentally two different types of accountability: accountability at the micro (i.e., individual) level and accountability at the macro (i.e., organizational/institutional) level. Public administration has focused on accountability at the macro level and the emphasis on institutions has made considerable progress in understanding public accountability (Behn 2001; Bovens 2007; Koppell 2005; Romzek and Dubnick 1987). However, recent public accountability research acknowledges that examining variations across individuals is also necessary (Olsen 2015; Romzek 2014; Schillemans and Busuioc 2014).

Previous macro-analyses focusing on examinations about existing ex post accountability systems are limited in explaining individual roles concerning how accountability systems can make a difference to employees: in other words, their subjective experience of being accountable—why employees feel accountable for their actions, as a key driver to predict individual behaviors. Improving the predictability of attitudes or behaviors requires tracing micro foundations of accountability to the individual level rather than seeking modifications arising from macro considerations. We propose that public administration research needs to be augmented by psychologically focused studies that assess the extent to which people internalize the rules and norms imposed by overhead institutions: “employee accountability.”

Theoretical Implications

Developing a Nomological Network for Employee Accountability

Empirically examining the nomological network of the employee accountability construct is a logical next step. It is useful to develop a multidimensional scale using convergent and discriminant validity tests to capture specifications of all five dimensions of employee accountability.1 For future empirical studies, variables of obvious interest are traditional macro-accountability systems, including institutional instruments or rules that can stimulate or activate the five components of employee accountability. Relationships between diverse public accountability systems (e.g., transparency practices, resident participation systems, such as public hearings and participatory budgeting) and public employee accountability can be tested to validate the five-dimensional model. Additional research could compare differential effects of public accountability systems and generic accountability systems on the five dimensions—which macro systems are related to each dimension. Theses future studies will be informative, especially to discover determinants of the five dimensions of employee accountability. Changes of organizational structure (e.g., traditional hierarchy, hybrid, and horizontal patterns) or civil service systems can be studied as antecedent variables for the five dimensions of employee accountability.

Other meaningful constructs distinct from the employee accountability construct need to be examined to develop the focal concept. They will be relevant to individual characteristics associated with or affecting employee accountability, ranging from values, to dispositions or predispositions (e.g., public service motivation), to personality attributes (e.g., conscientiousness), and gender. Because the perceptions of employees about their contractual obligation with their organization/employer can gradually change over time, the psychological-contract variable can be expected to influence employee accountability (Guzzo and Noonan 1994). To advance understanding of how the employee accountability construct works in its nomological network, relationships among employee accountability, other constructs, and work outcomes need to be studied. An example of a relationship that might be studied is employee accountability’s mediating role between public service motivation and job performance. Furthermore, various work outcomes resulting from employee accountability, for example, ethical behavior and mental health, expand its theoretical network. Accordingly, it will be meaningful to examine whether overall employee accountability influences those work outcomes as well as specifically which dimensions of the employee accountability construct most influence them.

Testing a multilevel (macro-micro) framework of employee accountability can improve understanding of how accountability systems are connected to micro accountability. For example, to understand the relationships between agents and principals, previous literature has explained agents’ problems that arise when agents do not follow the interests and requirements of the principal. However, the literature has not recognized another important link in accountability phenomenon, namely that agents internalize external systems differently and perceive themselves as accountable differently (e.g., Mero, Guidice, and Werner 2014). It is rational to consider that accountability-inducing practices used by organizations would be internalized by employees, and this internalized accountability would contribute to work outcomes. Therefore, the indirect impact of employee accountability can be tested to effectively examine the total impact of institutional instruments on behaviors in addition to the direct impact of the instruments on behaviors. Consequently, examining the mediating perceptions of agents can supplement previous empirical findings to explain diverse and complex phenomena between agents and principals.

Examining Diverse Public Contexts

Research demonstrates that contextual characteristics, such as occupation, location, and task affect individuals’ attitudes and behaviors (Johns 2006). Public employee accountability is manifest in contexts distinct from the private sector. Bookkeepers and street-level workers can be distinguished in terms of the five dimensions of employee accountability based on their contexts. Street-level employees have many chances to mix with citizens and are easily visible to citizens’ eyes (Maynard-Moody and Musheno 2000); bookkeepers do not directly interact with citizens and citizens cannot watch bookkeepers in backroom activities. Unlike the five-dimensional structure of the employee accountability construct for the public sector, recent private sector literature shows that a three-dimensional structure emerges to capture private employees’ accountability (Frink et al. 2018). The observability dimension may be more meaningful for street-level public workers than private employees or bookkeepers. Public employees in stronger-culture organizations would feel more consequentiality from the moral, social, or symbolic rewards (DiIulio 1994) from their context. Accordingly, testing the five dimensions of employee accountability collectively will be informative for theorizing about how the five dimensions come together for overall employee accountability in the public sector. Future studies need to consider diverse external conditions (e.g., organizational culture, accountability arrangements) that can influence each of the five dimensions of employee accountability.

Utilizing Experiments

Much of the experimental research to date has focused on individual dimensions, giving most attention to some of the five dimensions identified in the present study. The employee accountability construct defined here opens up new vistas for experimental research. One vista is a line of research that looks to employee accountability in a public context, which has largely been ignored in prior experimental research. The experimental research could, for example, focus on ambiguities in public contexts that may undercut predictable accountability responses from public employees. The experimental research could also focus on specific dimensions (e.g., consequentiality) in the conceptual model that have not received significant attention in prior experimental research. Examining a nomological network of employee accountability via experiments, especially relationships between macro and micro accountability, will create a deeper integration of public accountability and the psychological approach.

Practical Implications

Identifying Best Practices

The conceptual model implies that effectively managing public employee accountability is important because public employees with low accountability can perform badly and be less motivated and engaged in the organization. Best practices associated with high levels of each dimension of employee accountability and with the construct as a whole should be established for better outcomes. For example, to improve the attributability and evaluability dimensions of employees, job descriptions and assessment standards will have to be more explicit. To increase observability, communication networks between organizational members and people outside the organization will have to effectively work. Using social media or releasing employee information such as profile or contact information would advance their accessibility and influence for the public sector. More frequent employee reviews would foster their answerability by preparing them to report and justify their activities including good or poor performance. Frequent reviews may also increase job stress and require more organizational resources, increasing administrative costs (Lavertu, Lewis, and Moynihan 2013). Therefore, for the whole organization, public leaders must compare the negative consequences of too much accountability with the positive effects of employee accountability, such as quality decisions, improved job performance, ethical behaviors, and engagement.

Training and Development

The case of the 1986 explosion of the Space Shuttle Challenger gives a lesson of how crucial training and development can be for enhancing employee accountability. What if NASA officials involved in the event had significantly expected their wrong decisions to be identified or to affect their organization? Accountability expectations of employees can be nurtured by training and development programs: for example, organizations can use educational programs that emphasize how employee activities and decisions can affect the organization and society. Through such training in public health, for instance, nurses can better recognize that their work may save citizens’ lives.

Conclusion

The behavioral public administration movement seeks to more deeply integrate psychology and public administration (Grimmelikhuijsen et al. 2017). Psychological theories help in explaining a universal phenomenon that employees perceive themselves as accountable for their actions by internalizing rules and expectations from external sources. Accountability-inducing systems that are internalized by employees can lead to desirable work attitudes and behaviors. By applying the psychological approach to public accountability, this article developed an employee accountability concept represented by five micro dimensions, including an observability dimension that reflect a distinctness from most private sector research. This study is important because the conceptual model improves the explanatory power and understanding of systems established to hold public employees accountable can make a difference to individuals and why public employees feel accountable.

This article proposes a significant change in public accountability research, seeking to couple public administration’s traditional macro-orientation to psychology’s micro-orientation. It implies a major shift in the way public accountability has been discussed and studied. This study provides a meaningful foundation to extend public accountability research, especially into examination of individual accountability of public employees and its link to macro-level characteristics (e.g., government transparency) of public sector accountability.

In that employee accountability is affected by five dimensions, public leaders must examine all the dimensions rather than focus on a specific dimension for their employees. Employee accountability contributes to various outcomes including ethical behavior, engagement, and performance. Organizations become more and more productive, trustworthy, and competent as such positive outcomes of their employees accrue. Therefore, public organizations must effectively manage employee accountability among their members and within their work units to generate better outcomes for their workforce.

Funding

This work was supported by the Ministry of Education of the Republic of Korea and the National Research Foundation of Korea (NRF-2018S1A3A2075609).

Acknowledgments

We thank Barbara Romzek, Trent Engbers, and Elizabeth Benson for their comments on an earlier version of this article.

References

Footnotes