Keywords

Introduction

Values resound in all areas of society, particularly in times of crisis and political and social tension. Many politicians and leaders believe that good, common values can keep together communities, societies, and even transnational organisations like the European Union (EU). This makes values not only a frequent element in political speeches but also a popular focus area in research calls aimed at helping politicians solve complex problems in increasingly diverse societies. Values research has been and still is politically prioritised and one field where researchers in social sciences and humanities can at least get closer to the funding sums that are common in natural sciences. Therefore, it is not surprising that there are numerous projects, both bigger and smaller, often focusing on values in relation to something else, like welfare, leadership, public services, civil society, and much more. This would suggest that the last decades have refined conceptual discussions about values that are important to consider in future empirical values research.

Alvesson and Kärreman (2011) ask researchers to look for mysteries or paradoxes when theorising empirical (qualitative) data. This challenge is particularly interesting in the field of values research, which has shifted its focus from normative to empirical research in line with what has been described as the empirical turn (e.g., Thome’s 2003 research). By the middle of the last century, it would have been natural to delve into philosophical, theological, or even phenomenological reflections and to contribute with refined normative approaches when doing a values study. Since then, it has become increasingly common to explore values in empirical studies. This can easily be tested by searching the concept of “values” in the EU research result database Cordis (https://cordis.europa.eu/en). One of the most well-known examples is the huge longitudinal World Value Survey (WVS), which was initiated in the 1980s and has now completed its seventh wave with the prospect of continuing with more waves during the coming decades (see, e.g., Inglehart, 2018). However, empirical studies are not a byword of a lack of conceptualisation. Even the most inductive studies presuppose basic theoretical understandings of the studied phenomenon.

The interest in having a closer look into the use of value definitions in organisational and leadership studies arose from a retrospective amazement about research projects on values that the author was involved herself. Some directed surprisingly little attention to defining values, even though values were one of the key concepts in these empirical projects. Following Alvesson and Kärreman’s (2011) approach, this observation can be made a starting point for mystery solving in at least two respects. On the one hand, one could look for the mysteries in empirical values research that trigger such a potential lack of definitions. On the other hand, one could try to understand the mystery of disinterest in conceptualisations and definitions in empirical studies about values. Based on this, the overarching question of this chapter is, thus, how can value definitions affect the methodological design of empirical research about values in the field of organisational and leadership studies. The chapter contributes to the discussion about the preconditions for empirical value research that is also raised in Beate Jelstad Løvaas’ (2021) chapter about the connection between research purpose and choice of methods and in Thomas Andersson’s (2021) chapter that raises the perspectives of intentionality and agency in relation to values and ponders how these can be taken into consideration in empirical values research.

A Plea for More Definitions

As a master’s or PhD student or a researcher working with an empirical study, one will probably recognise their supervisor’s or colleagues’ demand to make a theoretical contribution. But theorising in social empirical research is a complex endeavour (Merton, 2007). Alvesson and Kärreman (2011) agree with other social researchers that restricting theory use to hypothesis testing, which is common in the natural sciences, is problematic in empirical social research. They base their claim on one of the main criticisms of positivistic research approaches, namely that all scientific conceptualisations are necessarily insufficient due to limitations in language and in the imagination and consciousness of single researchers when studying social phenomena. At the same time, Alvesson and Kärreman argue that the limitations of theory should not result in a helpless retreat into the impossibility of theorising empirical social research and suggest that researchers should see the empirical material “as a potential dialogue partner, leading to questioning, doubting, and problematizing the existing/dominant expectations and frameworks” (Alvesson & Kärreman, 2011, p. 119). This quote illustrates how cautiously they describe theorising activities in qualitative social research and avoid all technical terms that could imply the risk of being lumped together with naive empiricist research that tends to use elementary definitions and hypotheses.

When reading methods literature, the attitude of demarcation from a simplified usage of theory is prevalent among many social researchers. Nevertheless, they do not emphasise definitions either. It is hard to find the keyword “definition” in the headings or index lists of comprehensive textbooks of qualitative social research (cf. Bryman, 2016 or Silverman, 2020). However, protest has recently been lodged against the definition abstinence in social research. US sociologist Richard Swedberg criticises that definitions “seem to be marginal to the sociological enterprise” (Swedberg, 2020, p. 431) and states that there also exists no knowledge about how sociologists use definitions (Swedberg, 2020, p. 441). Without denying the risk of “poor” definitions and the impossibility of creating an ideal definition, he claims that introducing definitions can lead to better analysis. Swedberg’s general claim can easily be applied to the field of value research, which shows both how difficult it is to create proper definitions and how important it is to get a better conceptual understanding of multi-faceted social phenomena like values. As the editors of this book state in the introduction chapter, values are essential to understand but difficult to define. Askeland points to the same dilemma when giving an overview of value constructions in various disciplines and concluding that the conceptualisations of values remain “somewhat ambiguous and taken for granted” (Askeland, 2020, p. 16). Sirris (2020) underlines that values are necessarily in the eye of the beholder and are thus fluid and fixed at the same time when discussing the role of core values in values work. In other words, values are one of those areas where sociological/scientific definitions and everyday definitions overlap or mix. Swedberg considers these kinds of overlaps to be important for the blurriness of key concepts in social research that hinder a clear sociological analysis (Swedberg, 2020, p. 435). All this makes values research a good case for studying the need for definitions in empirical social research.

Of course, this chapter will not consist of an endless chain of (value) definitions. Its aim is to highlight the connection between value definitions and the possible methodological designs of empirical value research in organisational and leadership studies. It starts with the assumption that each values study, whether it be a master’s thesis or an international research project, needs to consider and reconsider its conceptualisation of values. Since the researchers’ preconceptions of values will affect the values research in any case, it is only academically honest to use definitions to make their own understandings explicit and reflect on them. Based on Flick’s (2011) model for the design of qualitative research projects (p. 176), the chapter suggests three steps in the research process that open up work with and the reflection on definitions. Two examples of values research will be used as illustrations of how definitions can influence these three steps and thus the entire research design.

The first example is the GLOBE Project, a large, long-term international empirical research project that puts the connection between values and leadership on the agenda. Since its beginnings as a paper-and-pen-based survey in 70 countries in the late 1960s and 1970s, GLOBE has grown considerably into a hub that offers both material and points of connection for several hundred researchers and a research-based consultancy service for leaders in international and intercultural settings (https://globeproject.com/). The selected publication representing the project in this chapter is one of its large research reports, the second edition of the book Culture’s Consequences (Hofstede, 2001), which was edited for the first time in 1980. This book can be considered to belong to the pre-GLOBE Project phase and has, unlike many later publications from the project, a clear character as a research report. The publication covers the results from the early surveys in the multinational business organisation IBM and relates them to other relevant studies about values, culture, and leadership.

The LIVAP project differs in both size and setup from the GLOBE project. It has been developed within the framework of the master’s programme in values-based leadership at Diakonhjemmet University College (later VID Specialized University) based on the observation that many master’s students were interested in the same issues (leadership in practice) and that new knowledge that transcends the contribution of limited and contextual studies can be gained by analysing these individual studies under a larger theoretical umbrella. This framework was called the Leadership in Practice (LIP) project and comprised about 25 master’s theses. The theoretical canopy was further developed by establishing a research group focusing on leadership and institutional values work in practice (LIVAP) that in the beginning mainly consisted of teachers in the above-mentioned master’s programme (Askeland, 2017). The LIVAP project is closely linked to the editors of this volume, so this chapter will resonate well with other contributions to the edited volume at hand. The major outcomes of the LIVAP project are four PhD theses of Harald Askeland (2016), Stephen Sirris (2019), Gry Espedal (2019), and Tone Lindheim (2021) and an edited volume (Askeland et al., 2020) and a textbook (Aadland & Askeland, 2017). In particular, one of the PhD theses is interesting from the perspective of defining values and using the definitions in different steps of the research process (Espedal, 2019). Espedal studies values work in a faith-based healthcare institution in Norway.

To illuminate the aim of the chapter, namely, to highlight the connection between value definitions and the research design in the field of organisational and leadership studies, the selected two examples will be analysed with the help of the following three sub-questions in the next sections of the chapter: (1) How do the two examples define values? (2) How do the value definitions affect the research design of these two projects and in particular their sampling? (3) In what ways do the value definitions become part of the result presentations of the two examples? The analysis is not meant as research on its own, but as an illumination of the points that the chapter makes as a contribution to methodology in values research.

How to Define Values?

Swedberg’s challenge of paying more attention to definitions in sociological or social phenomena research leaves a value researcher with the question of how to deal with the complicated task of constructing a meaningful value definition that has the potential to improve the quality of the analysis of the empirical material. Swedberg makes clear that stipulative definitions are preferred in sociological research. Unlike lexical or ostensive definitions, which are often based on everyday language, stipulative definitions provide the opportunity to sharpen the focus on and highlight certain aspects, which again can contribute to the development of conceptualisations (Swedberg, 2020, p. 433).

However, definitions of values tend to be not only at the intersection between everyday language and scientific-analytical language but also between normative and empirical approaches to research. Against this background, the distinction between substantive and functional definitions that are common in the field of religious studies and theology could also be useful for value research (Verwiebe, 2019). Substantial definitions concentrate on what values are, while functional definitions explore how values work (Guhin, 2014, p. 586). In the analysis of the two projects, we will investigate whether value definitions are used and what character they do have.

The GLOBE project is based on an elaborated theoretical model of the differences between cultures and how they can be understood and made visible in large international surveys. Both values and cultures are defined as belonging to what Hofstede calls for the mental programme of people that expresses itself on universal, collective, and individual levels. Hofstede defines that values “can refer to the desired or to thedesirable” (2001, p. 1) and measures them by asking for preferences in quantitative questionnaires. Hofstede’s main theoretical interest in values is to create a model that helps him measure culture. Values thus become central in his model of cultural manifestations, which ranges from symbols, over heroes, to rituals and values, which are situated at the very core of the model. At first glance, this definition seems rather unspecific and broad and neither functional nor substantial. At the same time, such a simple definition can serve as a plain operationalisation of values as manifestations of culture that can be formulated as questions to which the respondents of questionnaires can agree and disagree. Based on this, he developed a complex system of questions related to leadershipcultures, with which he measured values by asking respondents to evaluate statements as desirable or not desirable. Hofstede’s definition is not substantial as it does not differentiate between what makes values good or bad. In contrast, it could be understood as functional, as it starts from the preunderstanding that values function as preferences or wishes regarding how leadership should be conducted.

Espedal is inspired by institutional theory when she defines values as “the sayings and doings in organizations that articulate and accomplish thedesirable in relevance to right and wrong action and behavior” (2019, p. 47). Her definition has both substantial and functional elements. The functional part of her definition understands values as “the desirable” that can be captured in written or oral expression and actions of organisations. By adding the categories of “right or wrong,” a normative element is included in Espedal’svalue definition. This opens up a reflection about the substance or content of the values that she empirically observes in her study and for a normative evaluation of them.

The two examples illustrate that different value definitions make different research designs and different analyses possible. Stipulative value definitions facilitate empirically based reflections about what aspects should be taken into consideration when researching values. A certain degree of elaboration regarding the definition enables a potentially richer reflection about values. Only definitions with substantive elements allow for normative considerations.

Contact Sites for Definitions in Value Research

All research, if normative or descriptive, if quantitative or qualitative, is related to earlier knowledge. In presentations of empirical research, the theory chapter usually has the task of indicating and explaining this connectedness. Moreover, theoretical and conceptual preunderstandings typically affect all the stages and processes of social research, irrespective of whether the studies have chosen a deductive, inductive, or abductive approach. This can, for example, be illustrated by Flick’s model of components of research designs for qualitative research consisting of eight elements (translated by the author): research aim, research question, selection (or sampling), presentation, resources, methods, theoretical frame, and generalisation (2011, p. 177). Transferred to our two example studies, each of Flick’s components could be discussed regarding how it is affected by the implicit or explicit value definition on which the study is based. Below, we will discuss the relevance of value definitions for the methodological design by having a closer look into how the two example studies handle two of Flick’s components. More specifically, we will ask how they relate the selection of data or sampling and the generalisations (i.e., the presentation of the results at the end of the study) to the value definitions that they present. The example of the two studies can be easily translated into any empirical values study with which one might work.

The Connection Between Value Definition and Sampling

Bryman claims that all sampling in (qualitative) research has to be purposive or purposeful but that there is a difference between those research designs in which the criteria for the sampling units are decided on and fixed before the data collection starts and those in which these criteria can be negotiated and changed during the process of data collection (2016, p. 410). Based on the distinctions of Teddlie and Yu (2007) and Hood (2007), Bryman illustrates how fixed or open criteria can affect the research process.

In Teddlie and Yu’s “Taxonomy of sampling techniques for social and behavioural sciences” (2007, p. 78), the differentiation between sequential and non-sequential sampling is a sub-category in the purposive sampling type, while they also mention three other types of sampling. Non-sequential purposive samplingstrategies are, according to them, samplings that aim at achieving “representativeness or comparability” or at studying “special or unique cases,” while the four sub-categories of purposive sampling are “theoretical sampling,” “confirming or disconfirming cases,” “opportunistic sampling,” and “snowball sampling” (p. 81). Purposive, sequential sampling approaches are often put at the same level as qualitativeresearch designs, but Teddlie and Yu’s typology clarifies how quantitative, non-sequential approaches are purposeful as well. While values (work) research ranges from large, representative studies to small data sets based on limited case studies or opportunistic sampling, purposeful sampling applies to all these different research designs.

Nevertheless, the link between theory and sampling is clearest in the purposive sampling technique of theoretical sampling. Theoretical sampling is mostly discussed as an approach that has emerged from the grounded theory tradition and is characterised as having a contingent process of data collection, in contrast to a priori definitions that guide the sampling (Hood, 2007, p. 157). Flick describes theoretical sampling as research that aims to generate theory through empirical data (Flick, 2011, p. 158). It is a process where data collection,coding, and analysis take place in parallel. What data will be collected next depends on the analysis of the data that has been collected earlier. In theoretical sampling, both the scope and the characteristics of the collected data are not defined beforehand. The data collection is completed when a kind of theoretical saturation has been achieved. For the field of values research, this would mean that the analysis of the empirical material is not beforehand based on a specific concept or definition of values but rather develops new, empirically based concepts and definitions. One way of interpreting such a research process would be that value definitions must be the result of an empirical study that started without any conceptualisation when selecting the empirical material. However, one could ask if it is possible to select any empirical data about values without having any preliminary value definitions in mind. At least Alvesson and Kärreman (2011) seem to have a type of pre-conscious definition in mind when asking the researchers to go into dialogue with the empirical research material, since any dialogue presupposes a kind of pre-conception for both dialogue partners.

Therefore, we must explore the connection between sampling and value definitions by analysing the ways in which the data collection in the GLOBE study and Espedal’s PhD project are purposeful and based on their earlier introduced value definitions.

Within the model of the GLOBE study, the value definition does not primarily affect the selection of who is chosen as the provider of empirical data (i.e., of the respondents), but how of the empirical data is collected (i.e., through agree and not agree questions in a questionnaire). The theoretical definition of values permeates the values-related questions in the questionnaire that was created in the 1960s and developed further after this (Hofstede, 2001, p. 467ff.). Seen from Alvesson and Kärreman’s critical reflections about the connection between theoretical concepts in the form of simplified hypotheses or definitions on the one hand and empirical data on the other hand, Hofstede’s study can be described as a typical example of quantitativeempirical research that, in similarity to other quantitative values studies, tests hypotheses about values that are in turn based on a simplified functional value definition.

Espedal’s methodological approach is closely related to her value definition, even though her main interest is values work and not values in themselves. Her sampling follows in many ways what Flick describes as theoretical sampling. She describes her research design as a triangulation of observation, interviews, and document analysis conducted in different phases and linked with a narrative approach. The whole research process is motivated by the aim of getting a better understanding of values according to her initial definition, with both functional and substantial elements. The sample of the empirical material is thus not specified beforehand, but it is open-ended and influenced by the research. Completing her data collection does not mean for her to accomplish a certain number of interviews and observations but to reach a level where she concludes that she has enough empirical material to answer her research question. There is a direct connection between the open-ended sampling of the project and her value definition, which integrates functional and substantial elements, thus allowing her to consider both the function and content of the values that she finds in her material.

The GLOBE study and Espedal’s PhD project are two distinctive research projects with very different value definitions. While Hofstede’s research starts from a simple functional value definition that allows him to study values through preferences in agree-not-agree questionnaires, Espedal’s definition combines both functional and substantial elements to capture both the normative and the descriptive dimensions that are constitutive for values work, that is, values forpractice or values inpractice (Askeland & Aadland, 2017). Even though their value definitions obviously differ, the two studies have in common the sampling process is closely linked to the value definitions. This indicates that when planning research about values, it is very likely that the sampling will be affected by (conscious or unconscious) value definitions.

Value Definition in Writing Up the Research: The Final Section

Making a research project accessible to other researchers by writing up its framework, decisions, and results and discussing them is not a favourite topic in textbooks in social research. While dozens of pages are devoted to single approaches for data collection and analysis, the issue of presenting the whole study in a consistent text, be it a report, a thesis, or an article, is often only given comparably few pages at the end of the books; see, for example, Bryman (2016, pp. 661–684) or Silverman (2020, pp. 421–489). Even less attention is given to writing the conclusion or final section of a research publication. This is noteworthy since academic readers usually start by reading the introduction and the conclusion when accessing new research.

Concluding a study usually means returning to its starting points and evaluating how the research project ended. Silverman writes, for example, about giving the research an own “twist to the wider implications” (2020, p. 432). In values studies, this would mean that the values researchers discuss in the conclusion if they can ascertain any new knowledge about values. This would automatically bring them back to the issue of definitions and conceptualisations of values. Consequently, the conclusion, discussion, and maybe also generalisations would be a natural place to learn about the conceptualisations/definitions of values in a research project and how they were applied throughout the study. This means that we should be able to learn about the value definitions of research projects in organisational and leadership studies by scrutinising their conclusion chapters or sections. This might also be the case for our two examples.

When Hofstede in the GLOBE project draws his conclusions from the survey, it becomes obvious that what he is interested in is culture, not values. His conceptualisation and definition of values are tools for developing the theoretical concept of culture. This means that values are a prominent concept in the discussion of concluding chapters of Hofstede’s book (see, in particular, Chaps. 8-10) but they are not the focus of the theoretical reflections. As a tool for deepening theoretical discussions about cultures, it is not necessary for Hofstede to pay immersed attention to the concept of values. This explains why Hofstede uses a rather simplistic functional definition of values.

In Espedal’s PhD thesis, the discussion of values is closely connected to the discussion of values work. This also becomes evident in her conclusion, where she points out theorising the process of value inquiry as her main theoretical contribution (2019, p. 98). Even though she starts her project with a value definition that would have provided an interesting vantage point for theoretical considerations about values, she does not return to this definition in the conclusion. She states nevertheless that her study showed that “values reside partly in the narrative unconscious and partly in the pre-reflective corporeal action” (2019, p. 99). This is a finding that could have developed her value definition.

Common to the conclusion in both studies is that they touch on the issue of a theoretical conceptualisation of values but delve into other theoretical debates that are related to the concept of values. It thus seems difficult to enter deeper discussions about such a basic and at the same time comprehensive concept as values, with results from both quantitative or limited contextual empirical studies.

Conclusion

It is not the conclusion of this chapter that all empirical values researchers should enter into the large, overarching theoretical debates about what values are. But avoiding conceptual reflections about values can create mysteries regarding what the empirical values research actually is about. It would mean leaving the field of theoretical value conceptualisations in empirical value research to quantitative, hypothesis-testing studies and thus impoverishing the empirically based theoretical discourse about values. Working with definitions always implies a decision about which discourse researchers want to participate in. Analysing two empirical research projects about values in the field of organisational and leadership studies showed that value definitions in (qualitative) empirical research are complicated, but give more starting points for fundamental reflections the more stipulative and elaborated the definitions are.

In particular, from the perspective of contextual and limited qualitative studies, it seems difficult to feel qualified to contribute to the large and often normatively influenced theoretical debates about values. Our two examples also show that empirical value research might instead contribute to other theoretical debates, such as the leadershipculture debate in the case of the GLOBE study or the values work discussion in the case of Espedal’s study. Against this background, the chapter aims to encourage empirical researchers in the field of values to take definitions of values into consideration in their work in at least one of the three dimensions below.

First, when working with values research, it is almost impossible not to come across the question of defining values. Values researchers will meet value definitions and examples of how value definitions have been integrated into research designs when reviewing earlier research in the field. It could therefore become natural for values researchers to include their own reflections about value definitions when considering their own contribution to the research field.

Second, it can be claimed that it is part of the academic fidelity that all values researchers attest to the research community how they understand values and what consequences this understanding has for the selection of their empirical material. It is important that these considerations are not made too implicit, but are made explicit for each step of the research process, particularly when justifying the samplingstrategy.

Third, as long as values have been important for an empirical research project, it is appropriate to spend some time reflecting on the (theoretical) understanding of values at the conclusion of the project. Only if conceptualisations and definitions of values become visible in the conclusion can the study of values have the potential to influence the wider theoretical debates about values. These debates benefit from each contribution, not at least if they are based on an empirical study. Therefore, it is worthwhile that empirical values researchers make this additional effort in their conclusion, even though their empirical results might be limited and contextual or guide them into other theoretical discourses.

When using these opportunities for conceptual reflections about values that the research design offers, (qualitative) empirical studies in the field of organisational and leadership studies—and likeliest also your study—can give important inputs to the debates about how values should be conceptualised and defined theoretically. They can thus contribute to keeping the mysteries of empirical value research alive and, at the same time, make suggestions for solving them.