Measures of Faith: Science and Belief in Leadership Studies

Abstract

From its inception, leadership studies has embraced the positivist tradition of hypothesis testing. In this tradition, psychometric instruments are meant to ward off belief from scientific practice by testing theories against empirical facts. While leadership scholars purport to conform to the standards of value-neutral science, this paper tells a different story. Drawing on qualitative interviews with 39 positivist leadership researchers, we argue that leadership studies is heavily invested with faith in two main ways: (a) faith in leadership concepts, even when their accompanying measures fall short of methodological standards and (b) faith in leadership studies as a science, even when it is tainted by commercial interests and professional rewards. Ultimately, we suggest that positivist epistemology is accepted in leadership studies as an article of faith. By exploring the interconnection between science and belief in the business school, we draw attention to the “secular religion” of scientism in leadership studies.

Introduction

The German sociologist Max Weber was a staunch defender of the value-neutrality of sociology (Weber, 2004). The social scientist, for Weber, should refrain from expressing personal or political views in their work in order to eliminate possible bias. In this way, the social scientist is much like the bureaucrat: formal rules and methods ought to guide one’s behavior, not values and convictions. However, Weber worried that an administered society run by bureaucrats and scientists would result in a disenchanted world, so he proposed charismatic leadership as the much-needed counterweight (Weber, 1978). The charismatic leader, driven solely by their inner convictions, is everything the social scientist is not: heroic, visionary, and endowed with awesome powers of transformation. Whereas science is a matter of impersonal reason, leadership is tied to the whims of a single individual. Yet while he claimed that charismatic leadership is a value-neutral ideal type, Weber also expressed strong belief in its ability to liberate society from the dangers of routinization and rationalization. Far from being guided by formal rules and methods, Weber’s view of leadership resonates with the romanticized leadership cult that held sway in Germany at the time (Mommsen, 1974). This illustrates how difficult it is to be a value-neutral social scientist, especially when leadership is the object of study. As we show in this paper, the same is true of leadership scholars today: although they purport to conform to the standards of value-neutral science, they also put their faith in forms of leadership that are said to redeem business and academia from recurrent crisis and scandal.

Since its inception in the United States after the Second World War (Grint, 2011), leadership studies has embraced the positivist tradition of hypothesis testing. The basic assumption of this approach is that research should formulate a theory first and then deduce hypotheses from the theory, which are subsequently tested against empirical facts. Typically gathered by means of psychometric instruments, the facts either corroborate or falsify the theory-derived hypotheses—which can be used, ultimately, to predict leadership outcomes in a rigorous and generalizable way (Antonakis, et al., 2004). The testing of hypotheses is supposed to be uncontaminated by any subjective contents, such as biases that derive from the personal convictions of the researcher, because the scientific method excludes value-laden belief from the process. This appeal to value-neutral science in leadership studies cannot be overstated: for leadership researchers, the study of leadership seeks to emulate the natural sciences in its reliability and replicability of results (Atwater, et al., 2014).

Yet our own research—based on semi-structured interviews with leadership scholars— shows that leadership studies is heavily invested with faith, contrary to positivist assumptions. This faith takes two main forms. First, leadership scholars express faith in leadership concepts (e.g., transformational leadership and authentic leadership) even when their accompanying quantitative measures (e.g., Multi-Factor Leadership Questionnaire and Authentic Leadership Questionnaire) are said to fall short of strict methodological standards. Second, leadership scholars express faith in leadership studies as a scientific endeavour even when it is said to be tainted by commercial interests and professional rewards. Leadership studies is thus characterized by an awareness that actual scientific practice has been debased while, at the same time, there exists a strong belief in an uncontaminated—yet practically unattainable—realm of value-neutral science. On this basis, we argue that leadership scholars relate to the hypothetico-deductive method in a similar way as Christianity relates to its creed (from the Latin credere, “to believe”), that is, the positivist doctrine is accepted in leadership studies as an article of faith rather than one that is open to critical scrutiny. Our study therefore draws attention to the interconnection between science and belief in the business school.

This paper contributes to ongoing debates within critical leadership studies (CLS) about the nature and limitations of the scientific study of leadership. As an academic field, CLS takes aim at the romanticization, essentialism, and positivism at the heart of leadership studies, and offers an alternative set of theoretical perspectives that subject the phenomenon of leadership to a broader sociological and philosophical analysis (Collinson, 2011). In particular, CLS scholars have sought to demonstrate that, for all its claims to adhere to scientific methods, leadership concepts such as transformational leadership and authentic leadership are characterized by a seductive power that gives rise to fantasy and wish fulfillment, often tinged with a “strong religious, messianic overtone” (Alvesson & Kärreman, 2016, p. 143; see also Gemmill & Oakley, 1992; Tourish, 2013). This indicates that, for mainstream leadership scholars, such concepts are endowed with the preternatural ability to transform individuals, organizations, and society for the better, despite overwhelming evidence to the contrary (e.g., Enron accounting fraud, Volkswagen emissions scandal, and Facebook data privacy scandal) (Abreu Pederzini, 2018; Collinson, 2012; Tourish & Vatcha, 2005). What’s more, positivist leadership research categorically excludes questions about “tacit personal beliefs and values that drive meaning-making” (Latham, 2014, p. 124), thus concealing the fact that the study of leadership is frequently shaped by subjective interests and concerns (Barker, 2001). In this paper, we aim to extend these discussions by exploring the “attribution of faith” (Alvesson & Kärreman, 2016, p. 142) to concepts and methods among leadership scholars. While previous CLS research focuses primarily on a theoretical critique of positivist epistemology in leadership studies, we develop a qualitative empirical analysis to support our claim that leadership scholars are adherents to the “secular religion” (Ashforth & Vaidyanath, 2002) of scientism.

The paper is structured as follows. In the first sections, we examine the relation between science and belief in general, and the role of faith in leadership studies more specifically. After outlining our method, we turn to our empirical material. Here, we explore how leadership scholars relate to the leadership creed. Finally, we conclude with a discussion of creedal and non-creedal forms of faith in leadership studies.

Science and Belief

Science and religious belief are commonly viewed as antithetical: the two just do not mix. This idea has been fueled by the popularity of authors such as Richard Dawkins (2006) and Christopher Hitchens (2008), who claim to have debunked religious belief using scientific reasoning. The assertion that science and religion are fundamentally at odds is part of the secularization thesis, which holds that religion is increasingly losing its cultural significance due to the progress of science (Harrison, 2017). The simple and intuitive idea here is that science provides certainty, which makes belief unnecessary; scientific evidence settles the matter once and for all. For instance, a rainbow in Greco-Roman mythology was understood as a path connecting heaven and earth. Today, rainbows are explained in scientific terms as the dispersion of light in water droplets, which makes a spiritual explanation superfluous. For advocates of the secularization thesis, the battle between science and religion is therefore a zero-sum game: if science advances, religion must by necessity retreat.

There are good reasons, however, to question the secularization thesis. In particular, we can challenge the assumption that science is about certainty and religion is about doubt.

For instance, the notion that science is the only system that produces certain, or verified, knowledge is itself an article of faith. As philosophers of science have pointed out, science does not produce certainty; rather, it produces provisional knowledge that is subject to revision or refutation (e.g., Kuhn, 1970; Popper, 1959). More broadly, the secularization thesis involves a leap of faith of its own—namely, a belief that the scientific method is applicable to all phenomena, and that science is the ultimate arbiter of all knowledge claims (Stenmark, 2001).

We can also question the secularization thesis from the reverse perspective: just as science cannot be reduced to certainty, neither can religion be reduced to doubt. The utterance “I believe Harry will join the party tonight” is quite different from the utterance “I believe God exists.” The first expresses a probability, and therefore an element of doubt, whereas the second expresses a conviction. A religious conviction is not mere belief in the existence of God, but a feeling of certainty in the face of hardship and adversity. As Eagleton (2009, p. 37) puts it:

Christian faith. . .is not primarily a matter of signing on for the proposition that there exists a Supreme Being, but the kind of commitment made manifest by a human being at the end of his tether, floundering in darkness, pain, and bewilderment, who nevertheless remains faithful to the promise of a transformative love.

On this view, faith is not based on any observable proof; rather, it derives—paradoxically—from “the evidence of things not seen” (Hebrews 11:1), from the invisible sphere of God. This is why empirical reality, captured so vividly by Eagleton, is a test of faith. Faith thus expresses intuitive knowledge, rather than proven knowledge, which in the Christian tradition is revealed to the believer. Most practicing Christians in the United States subscribe to what Boyd (2013, p. 13) refers to as “certainty-seeking faith,” which is rooted in the idea that “a person’s faith is as strong as it is certain”: the more truth is revealed, the more certain one is. We can find similar ideas in the Islamic tradition, such as the three stages of knowledge in Sufism, of which the highest form is absolute certainty (haqqu-l’yaqin) (ad-Din, 1992).

For proponents of the secularization thesis, religious faith depends on the recognition of a higher authority whereas science does not. As Stephen Hawking puts it, “there is a fundamental difference between religion, which is based on authority, [and] science, which is based on observation and reason” (cited in Harrison, 2017, p. 48). But this attempt to demarcate scientific objectivity from religious belief is less clear-cut than it may initially appear. For example, authority relations play an indissoluble role in scientific practice in a number of ways, such as belief in a particular theory or methodological approach, belief in an individual scientist, or belief in science as such. Faith in authority also manifests itself in academic citation practices: an in-text reference to a journal article is often an articulation of belief in its method, its author, and its research tradition (Latour, 1987). Moreover, for Kuhn (1963), pseudo-religious dogmatism in science is not necessarily an obstacle but rather a prerequisite for scientific progress: faith in a particular method or set of theories creates the dedication that is needed for “normal science” to function.

If we reject the idea that science and belief are entirely separate spheres, a different interpretation of the secularization thesis opens up. Instead of saying that religion makes way for science, we ought to say that religion has entered science. This view draws on Agamben’s (2007) understanding of secularization as the transfer of the sacred to spheres that are not typically associated with organized religion. Secularization, for Agamben (2007, p. 77), “leaves intact the forces it deals with by simply moving them from one place to the other.” Take the example of politics. Rather than viewing politics as a secular realm devoid of religion, we should see political theory—following Carl Schmitt (2007)—as “secularized theology”: its central concepts (e.g., sovereignty) stem from theology, but are placed in a non-religious context. In this way, sacred motifs present themselves in secular environments. Secularization does not therefore designate the diminishing influence of religion, but rather the transfer of power from the religious sphere to other realms. This underscores the way that powers associated with the sacred do not necessarily disappear over time, but emerge in different guises in modernity—often in complex and contradictory ways. We might think, for example, about how the scientific method of hypothesis testing establishes a “God’s eye” view of nature and seeks to render the world knowable and predictable while simultaneously disavowing spiritual explanations of the world. From this perspective, we can view the tenets of positivist epistemology outlined above as part of the “secular religion” (Ashforth & Vaidyanath, 2002) of scientism.

This view of secularization, as the transfer of religious ideas to secular contexts, also applies to science in relation to leadership studies. As we will see in the next section, leadership studies is not a sphere that is free from belief, but is in fact infused with the kinds of faith typically associated with organized religion.

Leadership Studies and Religion

Many concepts within the academic field of leadership studies are infused with religious connotations, often in subtle or oblique ways (Spoelstra, 2018). A paradigmatic example is charismatic leadership, which derives from the Greek kharisma (“divine gift” or “gift of grace”). According to Weber (1978), the charismatic leader is allowed to exercise influence because he—and for Weber it is always a he—is considered to have extraordinary or divine qualities. The charismatic leader is treated as a demi-god, or at least as someone who has access to a sublime realm that is closed off to ordinary people. Following this understanding, charismatic leaders are often seen as figures who are capable of steering their organization to a higher level, which could mean making it more profitable, more ethical, or both. This suggests that leadership studies, although it tries to ward off belief from its scientific practice by testing theories against empirical facts, relies on concepts that derive from theological contexts. Sometimes this is partially hidden, as in the case of transformational leadership (Bass & Steidlmeier, 1999), while at other times it is openly acknowledged, as in the case of servant leadership (e.g., Van Dierendonck, 2011) or spiritual leadership (e.g., Fry, 2003).

Commentators, especially within CLS, have drawn attention to the sacred nature of leadership (see e.g., Grint, 2010; Sliwa, et al., 2013; Tourish & Pinnington, 2002). Less explored, however, is the relation between science and belief among leadership scholars. This is a curious omission since the academic literature is replete with expressions of faith in science, typically framed by “I believe” statements. It is perhaps no coincidence that these statements resemble religious creeds, that is, brief formulations of a religious doctrine. The Nicene Creed, for example, begins: “We believe in one God, the Father almighty, maker of all things visible and invisible”. Compare this with the opening line of an editorial published in The Leadership Quarterly in response to a recent retraction crisis in leadership studies:

Authors who publish in The Leadership Quarterly believe that leadership and leader performance are not a matter of opinion but rather a matter of fact. As such, we believe that leadership research is a scientific discipline. (Atwater, et al., 2014, p. 1174; emphasis added)

This statement represents a declaration of belief in science, which suggests that leadership studies is characterized by the same secular religion of scientism that we encountered in the previous section. Of course, one might object that we are reading too much into this sentence. On this view, “I believe” would mean little more than “I think” or “I hold,” with no obvious connection to religious faith. However, more is at stake than we might initially imagine. Take the following:

I believe that we are more likely to secure responsible leadership in the future if we can demystify its constituent processes. In that sense, enhanced knowledge about leadership may go hand-in-hand with morally desirable forms of leadership (Gardner & Laskin, 1995, pp. 297–298; emphasis added).

We ardently believe that, in order to create the morally responsive forms of leadership that many of us crave, it is vital not only for leaders to demystify these processes but also for followers to do so as we are the people who will guide and influence the leaders. (Jackson & Parry, 2008, p. 6; emphasis added)

These statements of belief articulate the enlightenment ideal of demystification: if we can rid ourselves of illusions about leadership, we will be able to discover the truth about leadership—and save ourselves from a morally bankrupt world. Such expressions of faith are required precisely because of a “fall from grace” in the corporate sector, which typically take the form of leadership failings or corrupt business practices (e.g., Enron accounting scandal, Volkswagen emissions scandal, and Facebook data privacy scandal). What is common to these expressions of faith is the idea that science can redeem organizations from their weaknesses and sins. Such “I believe” statements therefore constitute a creed, or quasi-religious doctrine, about the redemptive power of leadership in organizations.

The leadership creed, however, is broader than a focus solely on corporate misdeeds; it also encompasses academic misconduct within leadership studies itself. In other words, scientism involves a statement of belief in the idea of leadership studies as a science, even when research does not live up to these standards. This point is articulated most forcefully by Antonakis (2017), who—as editor-in-chief of The Leadership Quarterly—offers a roadmap for improving scientific standards in leadership studies in the wake of journal retractions and replication failures (2017, p. 6). He writes:

[B]y improving on how we do our science I believe that what we publish will make a difference and have an impact. [. . .] The editorial team of this journal is passionate about doing good science and ensuring that what we publish matters. . .. If we all follow our passion and do good science, success will find its way to us. (2017, p. 16; emphasis added)

Antonakis expresses faith not only in the science of leadership studies but also in the redemptive power of this science. By practicing “good science” (that is, both rigorous and ethical), leadership researchers will be able to save not only business but also leadership studies itself. The repeated references to “passion” (deriving from the Latin passio, “suffering",” especially in relation to Christ) must not to be overlooked here: scientism is instilled with a quasi-religious fervor, a faith-based certainty that the hypothetico-deductive method is the supreme judge of all knowledge claims. To paraphrase Eagleton (2009, p. 37), if leadership scholars are floundering in darkness, pain, and bewilderment due to retraction scandals and replication crises, they nonetheless remain faithful to the promise of a transformative science of leadership.

In this section, we have argued that leadership studies has become a secularized theology, drawing on Agamben’s understanding of secularization as the transfer of the sacred to spheres outside of organized religion such as science. This becomes clear in the creedal expressions of faith found in the academic literature on leadership. To wit: leadership studies, as a science, has the capacity to redeem our offences, in both the corporate and academic sphere. But this is only a part of the story. Beyond “I believe” statements in editorials and position papers, we can raise a broader question about the quasi-religious doctrine of scientism in the field of leadership studies: How do leadership scholars relate to the leadership creed and what effect does it have on their research practices? This question is worth addressing because it promises to shine a light on the convictions that underpin positivist epistemology, which are normally hidden from view. As we will see, the leadership creed allows leadership scholars to “keep the faith” even—or especially—when leadership research frequently fails to live up to strict scientific standards. Before turning to our empirical material, the following section outlines how we collected and analyzed our data.

Method

The empirical material in this paper is drawn from a broader project on the “life of measures” in leadership studies. The overall purpose of the project is to examine how the development and use of psychometric instruments impacts the social, institutional, and economic relations of a research community. The development of psychometric leadership measures is a multimillion-dollar global industry spanning many institutions including academia. Perhaps the most well-known example of a psychometric measure is the Myers-Briggs Type Indicator (MBTI), which is completed by approximately two million people every year across the world (Emre, 2018). Such instruments are commonly used to measure psychological traits and cognitive abilities in organizations for purposes of profiling, assessing, training, and developing employees. A proportion of these instruments are constructed by leadership scholars in order to advance scientific knowledge about leadership and to aid leadership development in organizations—these are the measures of our interest in this paper. While much has been written on the relative merits and drawbacks of leadership measures (e.g., Barge & Schlueter, 1991; Podsakoff, et al., 2003; Van Knippenberg & Sitkin, 2013), few studies have sought to open up the “black box” (Latour, 1999) of psychometric instruments in leadership studies. This is a shame, because these instruments—as we will see—provide a window into the complex relationship between science and belief in the business school.

Between July 2011 and March 2017, the authors interviewed 82 leadership scholars based in North America, the United Kingdom, Europe, and Australasia, of whom 49 are male and 33 are female. Thirty-nine of our interviewees employ positivist, quantitative methods in their research, 26 use qualitative methods from an interpretivist perspective, and 17 adopt a critical perspective towards leadership using social theory or philosophy (although the distinction between the latter two groups is not clear-cut). Overall, we interviewed 50 professors; 22 associate professors, senior lecturers, and readers; and 10 assistant professors and lecturers. The longest interview was 2 hours and the shortest was 30 minutes, with the average length being 1 hour. All interviews were recorded and fully transcribed. The names of our participants have been changed. The authors were equally involved in data collection.

For the purpose of this paper, we focus solely on leadership scholars with a background in organizational psychology who employ positivist, quantitative methods in their research (a total of 39 respondents). From the outset, we were interested in examining how ideas of methodological “rigor” and practical “relevance” inform the research practices of our participants, especially in relation to the production of psychometric instruments for use in both academic and organizational contexts. Examples of questions we asked are: ‘Why is it important to use leadership measures in your research?’, ‘Do you have any concerns about how measures are created or used?’, and ‘How do commercial interests affect the way leadership measures are developed and used?’ We used these questions as an entry point into a deeper discussion around the nature and purpose of scientific purposes in leadership studies. We soon realized that many leadership scholars were questioning the scientific validity of leadership measures at the same time as they remained loyal to the scientific potential of leadership research. This conundrum came to the fore in the wake of a high-profile retraction scandal in 2014 involving some of the most prominent leadership scholars in the field (Retraction Watch, 2014). In our interviews from February 2014 onward, we therefore included specific questions about how our respondents define scientific “best practice” in leadership studies and how far they feel their own research—and the research of others—meets this standard. Examples of questions we asked are: ‘How do you distinguish between good science and bad science in leadership studies?’, ‘To what extent do you feel that your research lives up to the ideal of science as you understand it?’, and ‘Do you see any problems with the use or development of particular leadership measures?’ Ultimately, this approach allowed us to engage our participants in a conversation about their belief in leadership studies as a science and how their faith is tested by frequent lapses in academic standards. While previous papers from our research project examined the prevalence of questionable research practices in leadership studies (Butler, Delaney and Spoelstra, 2017; Spoelstra, Butler and Delaney, 2016), this paper focuses on narratives around “science” among leadership scholars, especially in relation to doubt and belief in positivist epistemology. It is important to point out that we did not ask specific questions to our respondents about the role of religion in leadership studies or of faith in leadership concepts and measures. During the interviews, we were struck by the way our respondents struggled with doubt about prominent leadership measures at the same time as they expressed a strong belief in the study and practice of leadership. But it only became apparent to us after we had completed the initial process of data collection that our respondents’ reflections on leadership science resonated with ongoing discussions about “secularized religion” in organization studies and philosophy more generally. This led us to analyze our data in light of these theoretical debates.

In line with the principles of interpretivist research (Corbin & Strauss, 2008; Prasad, 2005), we paid close attention to our respondents’ assumptions, meanings, and judgments, focusing in particular on any tensions or contradictions that they articulated in relation to leadership science. Unlike positivist research, which seeks to establish law-like relationships between variables based on hypothetico-deductive methods, we adopt an interpretivist approach that allows us understand how social actors make sense of their professional experiences through narrative accounts of specific practices (Bryman, 2012)—in our case, the production and use of psychometric measures in leadership studies. Starting from these assumptions, we analyzed our empirical material as follows. First, the authors independently read each transcript and manually coded the interview according to the principles of “thematic analysis” (Shank, 2006), which is a way of detecting patterns and regularities within a network of interconnected themes. We then discussed each interview transcript together and soon became aware of a key tension in our respondents’ narrative between the actual practice of leadership science (tainted by monetary and professional rewards) on the one hand and the ideal of science (reliable, replicable, and value-free) on the other. This, in turn, permitted us to zero in on the relation between science and belief in leadership studies and, after several iterations of organizing our empirical material through a process of collective dialogue, we developed three overarching themes that best capture our data: (a) Struggling with doubt; (b) The fall from grace; and (c) Keeping the faith. These themes came to form the structure of our findings section, to which we now turn.

Findings

Struggling with Doubt

In contrast to some of the creedal expressions of faith found in the academic literature (e.g., “We believe that leadership research is a scientific discipline”), the leadership scholars we spoke to articulate “a level of doubt” (Maybell) about the scientific validity of leadership measures. As Queenie puts it:

There [are] a lot of theories of leadership that are really consulting models and they’re very popular. . .but they’re not really based on rigorous research. They’re very elegant in that they communicate really well, and they’re prescriptive. . .they tell a manager just what to do, but [they’re] not really based on good science.

On this view, leadership studies has produced instruments for practitioner engagement that fall short of strict scientific standards. Jacob expresses this point succinctly: “A lot of research falls down because the measures just don’t measure up, ironically.”

What is striking is that these misgivings extend to some of the most popular measures in the field. The preeminent instrument for measuring transformational leadership, the Multifactor Leadership Questionnaire (MLQ), originally developed by Bass and Avolio (1990), came in for particular criticism by our respondents. As Hugo puts it: “Psychometrically, it’s a very problematic instrument.” Transformational leadership is comprised of four dimensions (i.e., idealized influence, inspirational motivation, intellectual stimulation, and individualized consideration), each of which is supposed to measure a different aspect of leadership skill. However, as Abe notes: “Just about every piece of research that uses [the MLQ] reports that it doesn’t factor out into the multi-dimensional structure that it’s supposed to measure.” For this reason, the subscales in the MLQ—which are supposed to measure the four dimensions of transformational leadership—“never seem to capture the construct effectively” (Abe). This leads Erika to conclude that transformational leadership itself “doesn’t have construct validity.” Here, our respondents cast doubt on the idea that the MLQ can in fact measure something called “transformational leadership” in the first place; it may in fact be measuring something else entirely, such as the extent to which employees perceive their manager as likeable in some way.

The Authentic Leadership Questionnaire (ALQ), developed by Avolio, Gardner, and Walumbwa (2007), is also viewed with skepticism by our respondents. Part of this criticism is related to the retraction scandal that saw seven articles by Walumbwa and his co-authors struck from the academic record in 2014 (see Retraction Watch, 2014). Since the scandal broke, leadership researchers feel that “the entire construct [of authentic leadership] is under a cloud” (Constantina). However, the problem lies deeper than a handful of articles purged from the leadership literature; the recent retractions may be just “the tip of the iceberg” (Jacob) in terms of poor scientific standards in leadership studies. For example, Maybell suggests that while the retractions shine a spotlight on potentially “unethical” research practices, it is in fact no different from what “everyone is doing” in leadership studies. Similarly, Peter considers the retracted articles to be “a scapegoat for a bigger problem in our field”—namely, a lack of construct validity. As Colin explains, authentic leadership is “not a very valid construct [in terms of] the way it’s been measured and operationalized” (Colin). He continues: “It’s very hard to know if someone is authentic or not. . .Some people are authentic but they’re complete idiots. . .or they have the wrong values” (Colin). In other words, it is a fool’s errand to try to quantitatively measure something as slippery as “authenticity.”

Despite their prominence in the field, MLQ and ALQ are seen as deeply flawed measures. This points to a disturbing question for positivist leadership scholars: What if the most prominent leadership questionnaires in the field are not, in fact, as scientific as they purport to be?

This bleak realization has certainly crossed the minds of some of the leadership scholars we interviewed. Our respondents frequently lamented the dire state of scientific methods in leadership studies, particularly scale development. Constantina, for example, describes a field that is more or less content to create measures by gathering items from pre-existing instruments and “just mashing things together” to see what happens. Harry echoes this concern by acknowledging that, while questionnaires can certainly be “useful,” researchers often “don’t understand the problems with them” and consequently measures tend to be both “overused and misused.” Part of the problem is that psychometric instruments in leadership studies tend to point to “one big lump of good leadership” (Erika) rather than allowing researchers to identify specific constructs such as transformational or authentic leadership. As a result, “a lot of the research is really almost worthless” (Harry), based on “bad science” with hypotheses that “come out of thin air” (Jacob). Taking this thought one step further, Peter estimates that “a large proportion [of leadership research] would have to be retracted” if we were to apply strict statistical standards to these outputs. Given this fact, he wonders whether there is any merit in leadership studies or whether “every now and then you have to say, well, maybe, maybe it’s all a hoax” (Peter).

The Fall from Grace

When pressed to explain why poorly designed measures and invalid leadership constructs are so prevalent in the field, our respondents tended to point to commercial rewards and professional interests. Specifically, our respondents felt that monetary gain and career opportunities discourage researchers from developing rigorous psychometric instruments and are therefore responsible for the scientific “fall from grace.”

Some of the most popular measures in the field are copyright-protected instruments that are sold commercially to practitioners. The MLQ and the ALQ, for example, are available for purchase from the psychological assessment publishing house Mind Garden for potentially hundreds of dollars, which includes computer-generated reports, user manuals, workbooks, and trainer’s guides (although some measures are free or sold at a discount rate to academic researchers who wish to use them for non-profit purposes). Consequently, scholars can “make a lot of additional money” (Astrid) by commercializing their instruments. The ability to sell measures to industry means that “commercial interests behind the scenes. . .have been driving the popularity of these measures and kind of justifying their existence” (Colin). For Harry, the problem is that “people who are marketing their [measures] are much less willing to recognize weaknesses or really test them in ways that would lead to improvements” since this might dent both their scholarly status and commercial viability. For this reason, Woody asserts that researchers who commercialize their instruments “impede the advancement of science and our knowledge about. . .leadership” because “the motivation to make money [trumps] the motivation to push science forward.” One reason for this, Constantina notes, is that “clients just want something easy, they don’t care about the statistics” and so leadership scholars may end up with constructs and measures that are “not empirically valid at all.” For Woody, the path to success and renown in leadership studies arguably depends less on one’s statistical skills as an organizational psychologist and more one’s ability to “promote” one’s constructs and measures as a shrewd “marketer”. He continues: “Like anybody else who is in the market trying to sell something, we’re trying to sell our ideas.”

The popularity of certain instruments is also driven by “politics” (Erika), such as networks of researchers promoting their own constructs and measures for professional advantage. As Peter notes, a “great way to make an academic career” is to “invent a concept, invent a measure related to it, [and] have everybody use that measure.” Of course, every new measure is subject to a degree of methodological scrutiny, as Grace highlights: “The measure you use has to have gone through scale development, a rigorous process, and [it] need[s] to have been published, preferably, in. . .an A journal.” But equally important, she acknowledges, is “the name of the person affiliated with it, and whether they have. . .a track record in the area” (Grace). Heidi also admits that, while “scientific robustness is. . .the basic condition” for the popularity of measures, the “network and visibility of the researchers” as well as “an active strategy to. . .promote certain leadership styles” play an important role in the prominence of certain instruments. Likewise, Renaldo feels that some leadership scholars are “on a mission” to validate their constructs in order to “make their reputation” in leadership studies, often to the detriment of the accompanying measure. As a result, there is a “mantra” (Erika) in leadership research that insists that certain concepts, like transformational leadership and authentic leadership, are “a good thing, full-stop” (Erika), despite evidence to the contrary.

Colin further attests to the importance of reputational factors in accounting for the rise of a particular instrument, but goes one step further:

It depends on the research group that publishes [the leadership measure]. You know, if you have a very famous research team that works on it, it’s going to be much more covered than one that’s not. I think also [it depends on] the ease with which you can measure, how intuitively appealing it feels and sounds. So sometimes something may be popular not because it’s good.

Colin acknowledges that a leadership measure may become prominent in the field if it is developed by well-known researchers. But he also admits that measures have an extra-methodological allure. A leadership measure is valued not solely—or even primarily—in terms of its scientific merit, but also in terms of whether or not it appeals to our emotions (“how intuitively appealing it feels and sounds”). A leadership measure can beguile us with its charms, quite separately from—and perhaps at odds with—its technical rigor. For example, Woody reflects on items in the MLQ: “One of the beautiful things about the MLQ, maybe the only really beautiful thing about it, [is] it really [draws] out emotions as compared to other leadership instruments.” Popular measures such as the MLQ or the ALQ may therefore have an ineffable charisma that draws us towards them, despite their methodological shortcomings. Peter makes the same point about another construct, servant leadership, which “speaks to the imagination” in an indefinable way. He continues: “In some ways, the vagueness of the constructs. . .is the reason why [they’ve] become so popular” (Peter). As Harry puts it, certain constructs are prominent not primarily because they are scientifically valid but because they are seen as “hot” by leaderships scholars, journal editors, and organizational practitioners. This also holds true for new constructs that are proposed by leadership scholars, which are often presented in hyperbolic terms as “the answer to all our prayers” (Erika). Similarly, for Egon, leadership scholars are motivated to develop a new measure because “they believe in their concept,” in spite of any theoretical shortcomings. It is precisely this extra-methodological allure of leadership concepts—to which we will now turn—that also gives rise to some of the strongest expressions of faith among leadership researchers, despite pervasive doubts about the scientific practices in the field.

Keeping the Faith

So far, we have shown that leadership scholars express concerns about the scientific validity of leadership measures tied to a particular construct. Our respondents explain this as a kind of morality tale: the higher aims of leadership research are degraded by the base pursuit of monetary gain and career opportunities. Yet leadership measures and constructs also hold an intrinsic, emotional, supra-scientific fascination for many researchers. Despite their doubts, leadership researchers express a fundamental belief in both leadership concepts in particular and leadership science in general.

Some respondents teetered on the brink of losing their faith in leadership as a science. Maybell, for example, admits: “I don’t know if leadership really can be a science, as much as we like to think we can be. I’m not sure it’s really well suited for that because there’s so much error that can happen in what we try to measure.” It is notable that Maybell expresses her doubt in explicitly religious terms, describing her waning belief in leadership science with wry amusement:

I don’t know if I’m there yet, but I might be eventually. It’s almost like being a borderline atheist, right? I might be—if I were to use that analogy—I might be an agnostic [laughter], not quite atheist. Like, a strong level of suspicion and scepticism, but maybe I’m not quite ready to go that far.

For all her doubt, Maybell is unwilling to concede that leadership studies, as a science, is bunk. She still feels a residual degree of faith in the enterprise, despite her reservations about leadership research in practice.

This was a common view among our respondents. For example, despite the well-known problems associated with authentic leadership, Peter is still hopeful about the future of the concept: “So many people have lost faith in the whole authentic leadership realm. [But] I think more people will buy into it when we demonstrate, like ‘Hey, it’s not perfect’.” The most troubling thought for Peter is that the poor standards of scientific practice in leadership studies will result in a loss of faith among the believers. Paradoxically, the way out of this predicament is to freely admit that concepts like authentic leadership will always fall short in methodological terms. This, he reasons, may yet redeem authentic leadership from the morass of journal retractions and the cloud of potential academic wrongdoing. Peter justifies this by emphasizing that “in the end, I think we’re trying to do something good.” He elaborates: “I do believe that honesty, authenticity, is one way in which we will make a better world.” This hints at the underlying morality among leadership scholars: there is a need to save the concept, despite its many imperfections, precisely because it is good.

This viewpoint is also expressed by Erika in relation to transformational leadership. Despite her misgivings about the concept, she acknowledges that transformational leadership is valuable because ultimately it contributes to “developing your employees and making sure they have an opportunity to thrive and develop in [their] job” (Erika). For this reason, she says: “I believe in the concept as such, I think it is a good thing.” Similarly, despite her concerns about the prevalence of questionable research practices in leadership studies, Portia states that “I haven’t lost faith” in leadership science because “in my heart I believe in” the positive effects of leadership. What is intriguing here is the way that leadership scholars believe in an image of leadership (whether transformational, authentic, or whatever) despite the scientific flaws associated with a construct and its measure. For such leadership scholars, it is crucial to not lose faith in the enterprise—just as one might be tempted to do in light of commercial and professional interests that tempt researchers away from their true calling.

It is not only leadership concepts that our respondents believe in; it is also the nature and purpose of leadership studies as a science. This is articulated most clearly by Jerry, which goes to the heart of his identity as a leadership scholar: “I’m not just a leadership guy who teaches leadership—I’m a guy who believes in assessment in leadership. . .Scientific assessment to help managers, based on science.” On the surface, this expression of faith in psychometric instruments may seem surprising in light of the doubts shared by our respondents about the scientific validity of the most popular constructs and measures in the field. Seen from another perspective, however, we can understand this belief in science as a conviction that leadership studies will eventually transcend the shabby realities of the field. For example, while Hugo feels that much of the research produced on leadership is “esoterica”, he remains optimistic about the advancement of knowledge in the long-run or, as he puts it, “a process of successive iteration toward what we hope, at some point, ultimately is the truth.” From this perspective, Hugo sees himself as a humble “servant of science” rather than, say, a leadership scholar who is interested in seeking lucrative consulting gigs or building professional networks. Hugo goes on to favorably compare leadership studies to the medical sciences:

This is like medicine, in the sense, yes, we’re not killing people, but we’re affecting their lives. There are actually people [whose] quality of work and their quality of life is impacted by what their organizations practice, which in part is informed by the research that we conduct.

The stakes of leadership research, for Hugo, are exactly the same as medicine in the sense that it has the potential to improve the world, even to “save lives”’ (Joseph). This points to the “moral imperative” (Colin) of leadership scholars to rise above petty concerns (such as wealth and recognition) for the greater good because leadership studies has the capacity to serve the “interest of humanity” (Colin). For this reason, Colin continues, “we should be studying leadership with the same self-interested intensity that we study diabetes or other life-threatening diseases.” Cornelius directly echoes these concerns:

The work we do influences real people in real organizations who could potentially go bankrupt. . .It’s not as direct as killing someone. . .[but] we have a huge potential for creating misery. [. . .] If you do your job and. . .publish the right data. . .the world becomes a better place. (Cornelius)

While expressing deep reservations about the current state of research in the field, our respondents also hold the hope that leadership studies will one day become a science on a par with medicine—with the same capacity to positively impact people’s lives. Indeed, for Cornelius, the retractions are – counterintuitively—a “sign of health” in leadership studies because it implies that “the field is mature” in the same way as the natural sciences.

Here we encounter an apparent paradox. On the one hand, leadership scholars express faith in leadership concepts (such as transformational or authentic leadership) even when their accompanying measures are said to fall short of scientific standards. This faith is framed in explicitly moral terms: because the concept is good, it must be salvaged despite its methodological flaws. On the other hand, leadership scholars express faith in leadership studies as a scientific endeavor that has the capacity to rise to the same status as medicine even though it is contaminated by money and careerism. This second faith is also framed in moral terms: leadership scholars articulate a conviction in leadership science (however debased it has become) because it has the capacity to positively affect organizations. In the following section, we explain this paradox by examining the leadership creed in relation to our empirical material.

Discussion

In line with Stenmark’s (2001) description of scientism, our respondents expressed faith in scientific methods. Rarely did our respondents question whether hypothetico-deductive testing as such is the most appropriate way of doing leadership research, despite the lamentable state of scientific standards in the field. In other words, our respondents continue to believe in the basic tenets of positivist epistemology that underpin leadership studies even when the evidence—notably, a lack of construct validity and poorly designed measures—seems to point in the opposite direction. But what allows leadership scholars to “keep the faith”? The answer, we suggest, lies in the leadership creed, which takes the form of “I believe” statements in the leadership literature. This creed serves to hold the field together and prevents it from fragmenting in the wake of questionable research practices, replication crises, and retraction scandals.

To understand how leadership scholars relate to this creed, let us turn once again to religion. In Christian theology, the nature and purpose of religious creeds have been the subject of debate for centuries. However, two functions are particularly relevant for our own argument: (a) the creed summarizes the central truth of a religion and (b) the creed is a pathway (or “rule of faith”) for living this truth.

First, the leadership creed summarizes the central truth of leadership studies, namely, its reliance on positivist epistemology. This aspect of the leadership creed is captured in the phrase “we believe that leadership research is a scientific discipline” (Atwater, et al., 2014, p. 1174), repeated with minor variations elsewhere in the literature. Such an expression of faith reminds leadership researchers of fundamental truths that are core to their discipline, such as the unity of the sciences, the progress of knowledge, the possibility of replication, and the practical relevance of scientific knowledge. The leadership creed therefore articulates the dogmatic belief of leadership scholars, that is, a belief that is itself beyond questioning. We find traces of this dogmatic belief in our interviews. For example, Maybell expresses a “level of doubt” about the scientific validity of leadership measures, yet acknowledges that she is “not quite ready” to lose her faith entirely. Peter, Erika, and Portia all express similar sentiments. In short, our respondents acknowledge that leadership research often fails to live up to basic scientific standards, but they are unwilling to seriously entertain the possibility that the problem lies with applying positivist principles to the phenomenon of leadership. In other words, for all the poor scientific practices in the field, leadership scholars nonetheless express a feeling of certainty that leadership studies is, irrefutably, scientific. Our respondents’ faith in science is therefore not based on any observable proof, but instead—to return to Hebrews 11:1—derives from “the evidence of things not seen.” This is precisely the conviction that the leadership creed conveys.

Second, the leadership creed is a pathway that allows scholars to live this dogmatic belief. In the case of leadership studies, this pathway is a methodological rule of faith. In other words, the leadership creed asserts that the scientific nature of leadership studies is demonstrated by hypothetico-deductive testing. The pathway offered by the leadership creed ensures that the researcher does not take a wrong turn or get lost. The leadership creed thus provides scholars with direction and enables them to stay on track: they know what route to take (e.g., develop measures and test theories) and what paths to avoid (e.g., qualitative research and conceptual analysis). Again, traces of this rule of faith can be found in our interviews. For example, respondents like Cornelius, Hugo, and Colin extoll the virtues of publishing “the right data” (read: quantitative and measure-based) in order to increase “quality of work [and] life” and therefore serve the “interest of humanity.” The path to knowledge does not therefore branch out towards alternative approaches, but leads researchers towards a single unitary method that involves “a process of successive iteration toward. . .the truth.”

In the same way that a religious person’s beliefs are sometimes challenged by lived experience (e.g., illness and misfortune), leadership scholars’ faith is also tested by real life. For our respondents, measures such as the MLQ and the ALQ are rarely said to reach typical benchmarks of science (e.g., replicable and value-free). Notably, the explanation for such lapses resemble cardinal sins, namely, greed and pride. For example, the money to be made from measures are said to have a deleterious effect on the field, not least by serving to “impede the advancement of science.” Or again, a “famous research team” that wishes to preserve its reputation may strategically “promote certain leadership styles” over other, perhaps more methodologically sound, constructs and measures. Put simply, if leadership studies is awash with invalid constructs and faulty measures due to the glittering temptations on offer, this is seen by our respondents as a test of faith that must be withstood rather than, say, concrete evidence that psychometric instruments used in the field are junk science.

We are now in a position to understand the riddle we encountered in the previous section and how it relates to the leadership creed. On the one hand, leadership scholars express faith in leadership concepts even when their accompanying measures are said to fall short of scientific standards. On the other hand, leadership scholars express faith in leadership studies as a scientific endeavor even when it is seen as tainted by money and careerism. Once we understand that empirical reality is experienced by leadership scholars as a test of faith in science, rather than an obstacle to positivist epistemology, then the paradox unravels. The leadership creed—as a summary of the central truth of leadership studies as a science and a pathway for living this truth—prevents leadership studies from coming apart at the seams by allowing scholars to maintain a belief in the positive effects of leadership constructs and measures, despite frequent lapses in scientific standards. Or, more accurately, the leadership creed functions precisely because leadership studies fails to live up to its own ideals.

In the Christian tradition, humans are said to have fallen from grace, which means that they are tempted by earthly attractions, including money, power, and desires of the flesh. Since God has created humans in His own image (the imago Dei), however, they have faculties such as reason and intuition that they may use to overcome the fall, either in this life or the next. We can recognize much of the same narrative in relation to leadership scientism: there is said to exist a pure realm of science, yet in real life, leadership scholars do not manage to live up to this ideal. While the field is “sinful,” one’s faith is all the stronger for it. Questionable research practices, replication crises, and retractions scandals remind leadership scholars—to paraphrase Matthew 7:14—that the paths to temptation are manifold and the way to salvation is narrow, and open to the few; it is only by keeping the faith and strictly following the hypothetico-deductive method, even when it is unrealistic and unattainable, that leadership scholars will be able to redeem both organizations and academia from the sins of avarice and hubris. In this way, the leadership creed serves to convert doubt into conviction.

Conclusion

The question is, does the leadership creed make matters better or worse for leadership studies? In our view, the dogmatic belief that leadership studies is a science—despite indications to the contrary—is counterproductive for the prospects of a more valuable, “non-denominational,” leadership studies that is no longer in thrall to the quasi-religious doctrine of scientism. It would surely be beneficial to challenge the dogma, instead of slavishly following it, given the poor scholarly standards in leadership studies (Spoelstra, Butler & Delaney, 2016; Tourish, 2019).

Much of the problem stems from the idea of a single, unitary method (deriving from the Greek methodos, “following after a path,” especially in pursuit of knowledge) in the first place. Of course, knowledge about leadership is not an unattainable goal per se. However, it is an endeavor that raises a number of insoluble dilemmas for leadership scholars. For example, is it possible to quantitatively measure charisma, understood as the “gift of grace,” when the phenomenon is so elusive and intangible? Can we determine the extent to which a leader is “authentic” on the basis of self-rated questionnaires such as the ALQ when its items fail to entertain the possibility of self-delusion (e.g., “I admit my mistakes to others”) or duplicity (“I rarely present a ‘false’ front to others”) (see Avolio, Gardner & Walumbwa, 2007)? Is it even feasible to study “leadership” if it is treated as a pseudo-religious concept that cannot, by definition, be grasped by scientific study? One need only think of spiritual leadership, servant leadership, charismatic leadership, and visionary leadership to realize that leadership studies is inexorably drawn to a higher, extraordinary realm that lies beyond the ordinary, mundane activities of managers (Spoelstra, 2018). The hypothetico-deductive method is unable to capture these nuances or raise these questions for the simple reason that it excludes the possibility of forking paths towards multiple answers (or better yet, further questions), insisting instead upon the one true way of hypothesis-testing and theory-building.

We are not suggesting an “anything goes” attitude towards leadership studies. Quite the contrary: precisely because of its impossibility to live up to the ideal of value-neutral science, in practice too much is already tolerated in leadership studies. Consider, for example, the prevalence of questionable research practices in the field (Butler, Delaney & Spoelstra, 2017). Grey zone practices such as HARKing and p-hacking suggest that leadership studies is already characterized, despite its denials, by an “anything goes” approach to knowledge. On this view, the leadership creed functions much like Žižek’s (1989, p. 34) account of the Tibetan prayer wheel:

You write a prayer on a paper, put the rolled paper into the wheel, and turn it automatically without thinking. . . In this way, the wheel itself is praying for me, instead of me—or, more precisely, I myself am praying through the medium of the wheel. The beauty of it all is that in my psychological interiority I can think about whatever I want, I can yield to the most dirty and obscene fantasies, and it does not matter because. . .whatever I am thinking, objectively I am praying.

To relate this example to our own case, if the leadership creed expresses faith in leadership studies as a science, then one’s actual practices do not matter; despite all its flaws, leadership studies is still science. In this way, creedal expressions of faith allow leadership scholars to imagine that what they are doing aspires to the ideal of science (scientism’s version of the divine), even if the field is plagued by invalid constructs, dubious measures, and questionable research practices (scientism’s version of the fall from grace).

We may conclude by raising the question about what a non-creedal leadership studies might look like. This does not necessarily mean that we have to give up on our beliefs altogether, or somehow “cleanse” measures of their faith and belief. A non-creedal approach to leadership would mean, however, that we no longer subscribe to one belief in particular. Within Christianity, faith can be either creedal or non-creedal. A creedal form of Christianity is based on doctrines of belief that are beyond dispute—that is, closed to critical scrutiny. A non-creedal form of Christianity, by contrast, considers doctrines of belief (such as the Nicene Creed) as hindrances to establishing a personal relationship to God. Applying this to leadership studies, we can see how a non-creedal approach would allow leadership scholars to abandon the quest for a unitary method and instead embrace a pluralistic, or “interfaith,” approach to studying leadership. To some extent, this project is underway under the umbrella of “CLS” (Collinson, 2011), but so far this has not led to any rigorous questioning of the leadership creed within the positivist tradition in the business school, whose prayer wheel—in the face of recurrent crisis and scandal—just keeps on spinning.