Keywords

Introduction

This chapter is based on—and contains adapted parts of—our previous empirical research on values (e.g. De Graaf, 2003; De Graaf, 2021; De Graaf & Meijer, 2019; De Graaf & Van Exel, 2009; Paanakker, 2020a, 2020b).

Values are notorious for beinghighly abstract, essentially contested, hard to define and even harder to make operational and measure in empirical research (see Chap. 1 in this volume by Espedal, Løvaas, Sirris and Wæraas). You cannot point out a value; they are neither here nor there. ‘Inherent in institutional arrangement, values are core constructs of normative structures and thus taken for granted’ (Askeland et al., 2020, p. 16). The best we can say is that values never come just by themselves; they never appear unaccompanied. Rather, values are always attached to people, processes and practices, and they express qualities. Here we define values as qualities appreciated for, contributing to or constituting what is good, right, beautiful or worthy of praise and admiration (de Graaf & van der Wal, 2008).Footnote 1

Here it is our aim to unearth values and value conflicts in the public realm. By elaborating two different and related research strategies that have successfully been adopted in different empirical studies, we describe how we can study these tricky things called values in theircontext. More specifically, the goal of this chapter is to uncover values and value conflicts through two (related) strategies: (1) by studying dilemmas and (2) by studying craftsmanship practices.

Having defined values as qualities, it is important to avoid objectifying these qualities. Quality is clearly not an object that can be pointed out (De Graaf, 2003). Values are not part of a transcendental realm within or behind reality, as Platonists argue. In that sense, they cannot be objectified when using our definition. But if values are seen as stable qualities existing in reality, they are still, in a sense, objectified. That would be the case when someone argues that the (a priori) value of something exists. To us, however, values are attributed within a specific practice (which is always within the confines of a specific context) (see Paanakker & Reynaers, 2020); the quality is not already out there. Values are qualities-in-use. It is very well possible that different qualities are attributed to the same phenomenon. We use the expression ‘attributed within a specific practice’ and not ‘attributed by someone’ because values are not private. Just as a private language does not exist (as Wittgenstein has shown us), private values do not exist. This shows how we position our take on values in an interpretivist tradition, using inductive approaches to look for meaning in the subjective experiences of individuals engaging in social interaction. This way, we find general value patterns that take contextual dependence and subjective nuance, variability and meanings into account.

Theoretical Background: Value Pluralism

We take the theoretical stance of value pluralism, which acknowledges reality as comprising many different, co-existing values that are neither compatible nor commensurable (Paanakker et al., 2020). This is depicted in Fig. 6.1. With respect to the incompatibility of values, Stephan Lukes (1989, p. 125) offers an elegant description: ‘There is no single currency or scale on which conflicting values can be measured, and that where a conflict occurs no rationally compelling appeal can be made to some value that will resolve it. Neither is superior to the other, nor are they equal in value’. Perhaps the most famous definition of value pluralism was given by Isaiah Berlin (1982, p. 69): ‘[T]here might exist ends—ends in themselves in terms of which alone everything else was justified—which were equally ultimate, but incompatible with one another, that there might exist no single universal overarching standard that would enable a man to choose rationally between them’. The idea that values inherently conflict or are in some situations incompatible is hardly new; many social scientists have researched it (e.g. Brecht, 1959).

Fig. 6.1
An illustration of value pluralism. The multiple coexisting values are inherently in conflict because of value incommensurability and value incompatibility. The values for both are provided in the circle.

Explaining value pluralism as a starting point for research

Full size image

From the standpoint of value pluralism, however, values are also incommensurable. Value incommensurability, simply put, means that ‘the pursuit of certain values must inevitably comprise or limit our ability to pursue certain other values. The more we see to attain some of these values the less able we are to attain the others’ (Spicer, 2001, p. 509). This does not mean that agents cannot make choices or give reasons for them, ‘rather, it means that some of the reasons we might offer in support of making a particular choice are incommensurable with other reasons we might offer were we to make an alternative choice’ (Spicer, 2001, p. 512). This means that values make themselves heard and felt precisely at the cross points of such value decisions. When actors have to opt for (more of) the one or (more of) the other, they attempt to cognitively rationalise, and to organisationally justify, their weighing options. By doing so, the interpretative repertoire they use to make sense of values comes to the forefront and materialises in specific conscious or unconscious value choices.

Famous examples of incommensurable yet important values in daily life are money and friendship (Lukes, 1989; Raz, 1988; Spicer, 2010). Sometimes we have to choose between spending time to make money and spending time with friends; how do we weigh that? We make such choices, yet we cannot pay for friendship, for then it would not be friendship. ‘Our ordinary experience of the incommensurability among our values denies the monistic claim made by a variety of ethical philosophers, whether deontological or utilitarian, that there is “a common basis…a single reason behind moral claims” (Hampshire, 1983, p. 118)’ (Spicer, 2009, p. 539). For empirical research, this means that the examination of value conflicts and concurrent value choices is key to uncovering the practical role that values play in everyday organisational behaviour and decision making.

Strategy 1: Studying Dilemmas

The first corresponding strategy we discuss to examine values empirically is the studying of dilemmas. In trying to realise intrinsic values in public organisations, intrinsic values conflicts lead to dilemmas. Value conflict in itself is not a problem; value conflicts can bring forth change for the better by prompting alertness and innovation. And, as can be learned from Lipsky’s (1980) seminal study or the later work by Maynard-Moody and Musheno (2003), value conflict is unavoidable—it is a fact of life. Dilemmas are interesting because in studying dilemmas, one studies which values are important in a given context. In a dilemma, there is a conflict between two values that are apparently equally important. If one of them was not, there would not be a dilemma. Here, we are not interested in solving dilemmas, in stating what are the morally right things to do: it is more important to describe what the dilemmas are. Through studying dilemmas, we can uncover the values trail.

Strategy 2: Studying Craftsmanship Practices

The second strategy we discuss to research values and value conflicts is the studying of craftsmanship practices. This strategy focusses on exploring actors’ perceptions and accounts of what constitutes craftsmanship in their work. In our studies on public craftsmanship, we define craftsmanship as ‘the application of concrete skills, knowledge and practices, that, according to public officials, are needed to deliver good work in street-level public service delivery’, for instance, in education, patient care in hospitals or detention (Paanakker, 2021, p. 223). Similarly, craftsmanship could refer to semi-public or private sector professions that share the characteristics of hands-on work, tangible work practices, direct employee-client interaction and a knowledge base that stems from ‘learning-on-the-job’ and ‘learning-while-doing’, in addition to some theoretical knowledge base (Paanakker, 2019). This informal and experiential knowledge is embedded in concrete, everyday craftsmanship practices and offers a wealth of valuable data on how values are enacted and how values conflict in real-life practice.

Actors’ own perceptions of craftsmanship reveal what they deem important in the workcontext, what practical problems they face on the work floor and how well their notion of craftsmanship is facilitated by their organisation. The insights on concrete individual and organisational practices are then used to analytically deduce the values they describe. This offers an interesting method for comparing the manifestation and enactment of values among professionals, but also for comparing different public sector levels (such as public managers and policy makers). Through collective sense-making, it shows how values in the public domain are powerful indicators of public sector behaviours and processes, but also of organisational problems and conflicts between different roles and responsibilities in the organisation. As such, studying craftsmanship practices specifically enables us to uncover the values trail throughout organisational hierarchies.

Below, we elucidate how to apply these two empirical methods to capture and understand the underlying phenomena that shape value dynamics in concrete workcontexts and the more general governance settings that shape organisational settings.

Methodology

The strategies in this chapter share some important commonalities. For instance, both strategies use an explorative and inductivequalitativeresearch strategy (e.g. De Graaf & Paanakker, 2015) to uncover underlying values and value conflicts. Both strategies adopt a case study approach with a focus on understanding the dynamics present within single settings (Eisenhardt, 1989; Herriott & Firestone, 1983; Yin, 1989) in order to generate theory in the shape of propositions (Gersick, 1988; Harris & Sutton, 1986). This method is appropriate when not much is known about the phenomenon being researched, or when the phenomenon is so complex that neither the variables nor the exact relationship between the variables is fully definable (Hoesel, 1985). Due to their less tangible and somewhat hidden nature, and due to their context dependency, value dynamics and the conflicts rooted within them are eminently suited for case study research.

With both strategies, the main research method used is open-ended interviews. From previous research on values (e.g. De Graaf et al., 2016; Willis & Mastrofski, 2016), it has become clear that many interviewees initially consider the role of values in their profession to be abstract. However, they were able to make the values more concrete for themselves and the researchers when actual (value) dilemmas and (craftsmanship) practices were discussed.

Dilemmas

As stated, dilemmas are interesting because in studying dilemmas, one studies which values are important in a given context. For that reason, questions are asked about the difficult situations or dilemmas experienced. Questions are asked about (1) perceptions of conflicting values; (2) relevant dilemmas experienced, foreseen or known; and (3) how to best deal with dilemmas. The specific (value) conflicts that respondents perceive are important here, as is how they justify (Boltanski & Thévenot, 1999, 2006) and frame (Schön & Rein, 1994) them.

Craftsmanship Practices

Like in the first strategy, context is of essence in the study of craftsmanship practices (Paanakker, 2020a). We stress that craftsmanship practices, due to their hands-on nature, are about practices at the frontline of work and/or service delivery. Such practices are profession-bound, with employees using their own sets of ideas, words, logics and justifications to make sense of values. Having interviewees express their own understandings of key skills and practices, and having them use their own language to do so, prevents important information (and, therefore, important values) from being ignored or overlooked. The questions for interviewees focus on (1) what craftsmanship in their work means to them (or, when not interviewing a frontline employee but a manager or policy maker, what craftsmanship means in frontlinework); (2) perceptions of how their craftsmanship view is facilitated by and in their organisation; (3) how they perceive other hierarchical levels in the organisation to see or endorse craftsmanship on the shop floor and (4) the effects that may result from differences in organisational views on craftsmanship practices.

Table 6.1 shows examples of the interview questions that can be used to gather data. To preserve the richness of the data, all interviews should be taped and transcribed. Especially with explorative, bottom-up strategies, such as the ones described here, preventing data loss is key because the researcher does not know beforehand what patterns may emerge from the data and therefore what data are more important or less important.

Table 6.1 Interview questions: Examples from interview protocols
Full size table

The Follow-up Strategy of Q-Methodology

Up to now we have described how to unearth values and value conflicts. Sometimes researchers are also interested in the value profiles of actors. In order to also find different value profiles of public actors, Q-methodology—a mixed qualitative-quantitative small-sample method—is particularly useful (cf. De Graaf & Van Exel, 2009; Selden et al., 1999). Respondents can be asked at the end of the interview to rank values in a Q-sort in order to get a first impression of the value profiles among the respondents (De Graaf & Van Exel, 2009).

Q-methodology provides a foundation for the systematic study of subjectivity, a person’s viewpoints, opinions, beliefs, attitudes and the like (Brown, 1993). It was introduced by William Stephenson (Stephenson, 1935) when he presented his inversion of the use of intercorrelations so that individuals were measuring themselves rather than being measured by a researcher (Smith, 2001). Stephenson distinguished the method from R methodology (hence the name ‘Q-methodology’), which provides the basis for a science of objectivity in psychology (Brown, 1986). ‘The letter R in R methodology is a generalization of Pearson’s product moment r, which has most often been used in the study of relationships among objective characteristics such as traits, attributes, abilities, and so forth’ (Brown, 1986, p. 57). In contrast to R methodology, Stephenson correlated people rather than test items.

Typically, in a Q-methodological study, people are presented with a sample of statements about some topic, the Q-set. Respondents, or the P-set, are asked to rank-order the statements from their individual point of view according to some preference, judgement or feeling about them, mostly using a quasi-normal distribution. By Q-sorting, people give their subjective meanings to the statements, and in doing so they reveal their subjective viewpoints (Smith, 2001) or personal profiles (Brouwer, 1999).

The individual rankings (or viewpoints) are within Q-methodology subjected to factor analysis. If each individual has specific likes and dislikes, Stephenson (1935) argues, their profiles will not correlate; if, however, significant clusters of correlations exist, they could be factorised and described as common viewpoints (or tastes, preferences, dominant accounts, typologies, etc.), and individuals could be measured with respect to them. Brouwer (1999, p. 35) argues that one of the important advantages of Q-methodology is that questions pertaining to the same domain are not analysed as separate items of information but rather in their mutual coherence for the respondent: ‘Subjective feelings and opinions are most fruitfully studied when respondents are encouraged to order a good sample of items from one and the same domain of subjective interest (instead of just replying to single questions)’.

The contextuality of values demands that the quantitative methods used for studying values introduce validity threats: it is hard to know, for example, whether employees who speak of the same value mean the same value. Q-methodology is more suitable because Q-study results are clusters that are functional rather than logical. In other words, the clusters are not logically constructed by the researchers; they result from the empirical data and are operant (De Graaf & Van Exel, 2009). Q-methodology can reveal a characteristic independently of the distribution of that characteristic relative to other characteristics in a population. Unlike surveys, which provide patterns of variables, Q-methodology provides patterns of persons, in this case, public officials and their value profiles. Q-methodology is a mixed qualitative-quantitative small-sample method that provides a scientific foundation for the systematic study of subjectivity, such as people’s opinions, attitudes and preferences (cf. Brown, 1980, 1993; De Graaf, 2011; Twijnstra & De Graaf, 2013; Van Exel et al., 2005; Watts & Stenner, 2005).

In our Q-research in this vein, the specific values that were used were obtained from previous research in public institutions, values that originated from the Dutch governancecode for the public sector, drafted by the Dutch Ministry of the Interior and Kingdom Relations in 2009 (Ministerie_van_Binnenlandse_Zaken_en_Koninkrijksrelaties, 2009) (see Table 6.2).

Table 6.2 The ten values
Full size table

Using the same values from previous research makes comparison with other (public sector) organisations possible. Alternatively, values can be derived from the values that emerged from the analysis of the data on dilemmas and/or craftsmanship practices.

Coding and Research Heuristic

Transcribed interviews produce a great deal of data. Using software programs like MAXQDA or Atlas.ti to help with the text analysis, the interviews can be coded in various steps (Boeije, 2010). The purpose of the coding is to identify the specific values and value conflicts experienced in the case. To accomplish that, first, for the two strategies respectively, all the dilemmas, or with the second strategy craftsmanship practices, can be identified. These steps are based on systematic approaches to coding qualitative material (Schilling, 2006). In the remainder of this section, we provide more detail on how to analyse the data from the perspective of the two different research strategies.

How to Analyse Data on Dilemmas

Once all the dilemmas recorded in the transcripts are coded, the next step is to identify the specific value conflicts experienced. First impressions of overall patterns can be observed and then juxtaposed with the empirical data. This inductive process is clearly not a matter of counting. Respondents in most qualitative research are not randomly selected, and they are usually too small in number for quantitative purposes. But the idea of an explorative study is to consider the nuances and context of value conflicts that are experienced. Constant comparison can be conducted (Boeije, 2010), in which the researchers repeatedly go through the themes to compare results. Thus, it is not just important that a respondent experienced a value conflict: which one, how it was dealt with and how it was worded are important. Each dilemma can be coded simultaneously in software like MAXQDA on the specific value conflict. This inductive analysis process can be repeated many times before the final analysis is written up. Eisenhardt (1989, p. 541) states, ‘The central idea is that researchers constantly compare theory with data—iterating toward a theory which closely fits the data. A close fit is important to building good theory because it takes advantage of the new insights possible from the data and yields an empirically valid theory’.

How to Analyse Data on Craftsmanship Practices

Analysing data on craftsmanship practices follows a very similar approach (software-supported systematiccontent analysis, using programs such as MAXQDA or Atlas.ti) (Miles & Huberman, 1994). Continuing the same line of bottom-up construction of value patterns is key: the data as presented and worded by the interviewees are put centre stage. Rather than examining how actors translate values to practices, which we commonly see in values research, this strategy follows the reverse direction. It examines perceptions and accounts of craftsmanship practices to see which values they describe, with researchers extracting values from the data and coding values into the data in the analysis stage. Instead of designing the data collection to a fixed mould of predetermined values, researchers go on a quest to inductively deduce values from the data.

To this end, two-stage coding can be applied (Friese, 2012). In the first stage, the data are coded in vivo, with open codes that summarise the types and nature of craftsmanship practices as worded by interviewees. This includes combining data segments with similar content to build a coding scheme of a maximum of 100–150 codes with mutually exclusive codes that ‘reflect the heterogeneity of the data’ (Friese, 2012, pp. 130–131). In the second stage the initial codes are compared, integrated, modified and fine-tuned to inductively aggregate and classify them into the overarching value category they describe (Friese, 2012, pp. 130–131). Table 6.3 shows what this looks like for a study conducted in the prison sector.

Table 6.3 Coding example on craftsmanship practices and values
Full size table

Finally, we advise processing stressors, effects and explanatory variables of value conflict into the data analysis. Examples include codes such as inadequate leadership (a stressor) or work alienation or moral dilemmas (both effects). Examples of different manifestation levels of value conflict are ‘perceived divergence on value identification’ (which types of values are seen to matter) and ‘perceived divergence on value enactment’ (which values are actually emphasised in practice and how) (Paanakker 2020a, pp. 93–117; 133–158). Applying the carefully built and validated coding scheme to the data at large allows for the subtleties of craftsmanship perceptions and emerging value conflicts to be grasped and explained and a comparison between different staff levels made.

How to Analyse Data from Q-Methodology

Individual Q-sorts can be factor analysed using the software program PQMethod 2.11 (extraction method: centroid; rotation method: varimax) in order to reveal the distinct ways in which the values were rank-ordered. For more details on how to analyse a Q-methodology study, see Van Exel and De Graaf (2005).

Examples of Empirical Studies on Value Conflicts

Procedural Versus Performance Values

De Graaf and Paanakker (2015) found that effectiveness versus efficiency—both performance values—is the value conflict most frequently perceived by public officials. Lawfulness versus effectiveness and efficiency is the value conflict between procedural and performance values most frequently perceived by public officials. Transparency versus effectiveness and efficiency is the second most frequently perceived value conflict between procedural and performance values.

A Municipality and a Hospital

The most-found value conflictin a case study conducted by De Graaf et al. (2014) on a municipality and a hospital was between transparency and effectiveness. Sometimes it seems to be in the interest of the municipality not to make something public or to be completely open, but these are tough decisions to make. One public administrator stated, ‘I might be too open sometimes. Because if you are too open, that can sometimes harm the quality of the decision-making process in the interest of the municipality’.

In the hospital, almost all respondents experienced the conflict between transparency and effectiveness. Doctors and nurses find it important to be open with patients and their relatives, but they also have to take into account whether this can damage the effectiveness of treatment if what they have to say is upsetting. On the level of the hospital as an institution, transparency and effectiveness also conflict. Should the hospital be open about things that have gone wrong, or should it save its reputation? A hospital manager stated, ‘I don’t want to have any trouble, so I’d rather choose the safe way out. But that leads to discussion with our medical staff because they do not necessarily agree’.

Values of Academic Teachers

In a recent study (De Graaf, 2021), the dilemmas of academic teachers were empirically studied. Thirty-five of the 41 dilemmas found fall within three categories: quality versus efficiency (a value conflict between quality of education and efficiency. The more time you put into teaching, the better the outcome), quality versus equity (this dilemma concerns two aspects: the varying quality of students and the tension between students’ own responsibility and the need for teachers to provide guidance) and equality versus reasonableness (students who ask for exceptions because of special circumstances). The conclusion was that quality and efficiency play an important role in the case of academic teachers.

Values in the Police

Another example is a recent study on the influence of social media on value conflicts in the police (De Graaf & Meijer, 2019). Social media is an important factor triggering new value conflicts in organisations. It turns out that especially the values of lawfulness, transparency and participation conflict with efficiency and effectiveness.

The most frequently perceived conflict in this case study is the classic one between effective governance and efficient governance (working in a more efficient manner might mean that the work is done less effectively). On an operational level, in the case study only a (small) proportion of the detectives are found to be active online. Information about criminal acts increasingly comes up through social media, for example, videos of youths mistreating people. A police officer on the street witnessing such a scene would act immediately. On the digital street, things are less clear.

The conflict between lawfulness and effectiveness is a classic one for the police, the dilemma between law and order. In the case study there was clear evidence that social media cause new conflicts between lawfulness and effectiveness. Technological developments happen quickly while law and regulation lag behind, leading to much uncertainty within police organisations.

Many community police officers in the Dutch case study are active on social media like Twitter in order to build a good relationship with citizens; they want to enhance police transparency and citizen participation in police work. However, it takes time to build a good network, and such networking can generate a large volume of responses from the public, with some community police officers feeling under pressure from social media to react quickly. It is felt that the expectation of always being online is an intrinsic characteristic of the use of social media. On both operational and support levels, conflict was experienced between participation and efficiency.

Values in the Prison Sector

Using the strategy of craftsmanship practices, different studies in the prison sector have found that complex, structural value conflicts exist between professionals, managers and policy makers (Paanakker, 2019; Paanakker, 2020a; Paanakker, 2021). These different levels were found to have remarkably similar notions of key frontline values, which reveals an unexpected value convergence rather than value conflict, but the respondents perceive a large level of value conflict. In practice, each level experiences management above them as prioritising instrumental values over the intrinsic values of good penal service delivery.

The resulting value conflict is between the values of effectiveness and efficiency (the instrumental values in this study) on the one hand and the values of humanity, security, rehabilitation and the ability of adequate and timely task completion (the intrinsic values in this study) on the other. Here, performance-based management doctrines clash with the craftsmanship of professionals. Structural value conflicts produce a toxic value gap that is endemic to the organisation and has far-stretching negative effects on value attainment, employee attitudes, craftsmanship and policy implementation at the frontline.

Conclusion

In this chapter, we elaborated on two distinct research strategies we find useful in research on values and value conflicts: studying dilemmas and studying craftsmanship practices. We explain how, each in their own way, these strategies are designed as flexible yet insightful and instructive methods that leave a lot of room for different value interpretations and meanings. Especially in the public realm, where a vast array of values compete for attention, this allows researchers to understand the complexities and context dependency of values and value conflicts. This makes both the dilemma strategy and the craftsmanship practices strategy particularly suitable to uncover the values trail in organisations, institutions or sectors. This is not top-down and static but bottom-up and focussed on the understanding and sense-making of the individual: of their role, their ideas, their choices, their decisions and courses of action and their work and the challenges it brings. As such, these strategies are helpful for researchers to gain insight into values and value conflicts in everyday decision making and into their effects on citizens, clients, customers and patients, as well as the quality of service delivery. This chapter includes detailed tips, tricks and visualisations for data collection; data analysis and examples of research findings to get researchers started.