A critique of the Capability Approach’s potential for application to career guidance

In this article, we provide a balanced critique of Sen’s Capability Approach (CA) with reference to its potential to inform career guidance theory and practice. There are varying understandings and interpretations of the CA. Some see capabilities as universal, whilst others favour a more relativist view. The CA is also vulnerable to misunderstanding. Critiques based on misunderstanding are easily dismissed, so our focus is on substantive conceptual and practical critiques. Three main challenges are explored: conceptual debates about the nature of freedom and justice; limitations arising from the disciplinary origins of the CA; and challenges in operationalising the CA.


Introduction
The Capability Approach (CA) is a way of thinking about the promotion of wellbeing and freedom and originates in the thinking of economist and philosopher Amartya Sen (1985Sen ( , 1990Sen ( , 1999Sen ( , 2003Sen ( , 2009. While it has its origins in welfare economics, it has been applied to many domains, notably educational and economic advancement in developing nations (Kelly, 2012;Norwich, 2014;Orton, 2011;Otto et al., 2015Otto et al., , 2017Picard et al., 2015;Skovhus, 2016). Authors have sought to apply the CA as a theoretical perspective to describe career development experiences or to understand processes of career guidance (e.g. Picard, 2019;Robertson, 2015;Robertson & Egdell, 2018). Indeed, there now exist a wide range of theories of career choice and development with numerous recent contributions to this literature. However, in our view, contemporary career theory has tended to lack criticality (with some notable exceptions that are discussed in the next section). This literature has tended to present multiple theories without critique (e.g. Arthur et al., 2019), or has attempted theoretical integration (e.g. McMahon & Patton, 2018) with the effect that the limitations of one perspective are compensated for by the strengths of another. Criticality in this literature is primarily directed against historic targets, demonstrating the 'superiority' of dynamic contemporary theory over static twentieth century models in addressing issues of the volatility of work (e.g. Pryor, 2016;Savickas et al., 2009). It is relatively rare that contemporary career theorists systematically explore the limitations of their own theories or those of their peers.
Thus, in this article we offer a critical perspective on the CA and its potential for application to career guidance. We are concerned not just with theory but also with practice, by which we mean attempts by practitioners to improve people's lives by intervening in their career development, irrespective of the professional identity or institutional setting. There are a range of authors who have developed the CA, including Nussbaum (1997Nussbaum ( , 2000Nussbaum ( , 2003, so there is some variety of thinking within the literature. Whilst acknowledging this diversity, we focus our critical perspective primarily on the CA as conceived by Sen. A number of critiques and debates are explored. The CA has been vulnerable to misunderstanding, which can result in a superficial reading and dilution in its application (Kremakova, 2013;Sayer, 2012). Critiques based on misunderstanding are easily dismissed, so our focus is on substantive conceptual and practical critiques. This material is organised into three sections. Firstly, the CA can be understood as one of a number of philosophical positions on issues of freedom and justice. Although these debates are relatively abstract, issues of social justice and the freedom to make life choices are of direct relevance to career guidance practice. Secondly, unlike most career theory, which originates in psychology or sociology, the CA has its roots in economics. The limitations of its disciplinary origins can be critiqued from the perspective of other social science disciplines. Thirdly, there are considerable difficulties in operationalising the CA. The distance from the concept to practice is considerable, generating a range of questions to be addressed. In each section we seek to make the debates relevant to the concerns of career guidance, and identify where a reasonable defence of the CA can be made, and where there are limitations in what it has to offer to career theory and guidance. We are also mindful that the CA was not initially intended to be applied, or offer applications, to career theory and guidance. It is easy to identify the limits of a theoretical perspective constructed in another discipline. Consequently, we emphasise that our purpose is not to critique the CA for critiques sake. Rather we seek to explore how the CA may be applied in practice, and areas where it may need to be theoretically and empirically developed to make it relevant for career guidance.

Conceptual debates about the nature of freedom and justice
There is a prominent and long-established social justice and egalitarianism agenda within the career guidance field. The promotion of social equity is seen as one of the prime aims of public policy in relation to career guidance (Watts & Sultana, 2004). Indeed, in recent years there has even been explicit debate regarding issues of social justice in the academic and professional literature (Hooley & Sultana, 2016). This has included a focus on the contested (political and philosophical) nature of social justice. Some contributors have sought inspiration to address modern injustices in the work of early pioneers of vocational guidance, who were social activists with collectivist values (Plant & Kjaergård, 2016). Many have explored ways to respond to the career issues facing specific populations who might be characterised as socio-economically disadvantaged (e.g. Bimrose et al., 2019;Ginevra et al., 2019;Hancock & Taylor, 2019;Newman et al., 2018), or have addressed issues of poverty, class, and precarity (e.g. Ali, 2014;Blustein, 2019;Roberts, 2005). Others have sought to suggest strategies or tools for career development professionals to adopt (e.g. Arthur, 2005;Kenny et al., 2019;Thomsen, 2016) or have offered a call to emancipatory action (Blustein, 2019;Hooley et al., 2019). This diverse literature has explicitly sought to bring a critical perspective to current theories, institutions and practices, and the extent to which they are infused by culturally dominant neoliberal economic assumptions that limit freedom as it is expressed in careers (Hooley et al., 2017).
In exploring the contested relations between social justice and career guidance, scholars have pointed to a range of philosophical approaches. These include the ancient philosophy of Socrates and Plato emphasising social harmony; the Kantian tradition emphasising rights; the ethical/relational perspective of Levinas and Derrida; and 'recognitive justice' inspired by Young that acknowledges oppression of groups in society (Irving, 2004(Irving, , 2020Sultana, 2014). Irving (2004Irving ( , 2020 and Sultana (2014) acknowledge the relevance of John Rawls' theory of justice which offers a 'distributive justice' or 'resourcist' approach; but do not foreground the work of Amartya Sen. This is curious because Sen's CA is located in an arena of conceptual debate regarding egalitarian theories of social justice and the nature of freedom and justice. Sen (1985Sen ( , 1990Sen ( , 2003 evens acknowledges the influence of Rawls on his work. Comparing a full range of social justice concepts is beyond the scope of this article. Instead we highlight a key debate in the literature regarding the nature of freedom and justice.
Egalitarian theory contends that institutions and individual actions should improve the quality of life of those who are worse off in society (Yılmaz, 2016). There are two prominent examples of egalitarian theories of social justice, addressing the key question of what is a proper measure of social justice-and an extensive literature questioning which of these is able to provide the most convincing yardstick of social justice (Pogge, 2002). These two approaches are the resourcist approach (developed by authors including Rawls (1999) and Dworkin (1981aDworkin ( , 1981b and the CA (developed by authors including Sen (1985Sen ( , 1990Sen ( , 2003 and Nussbaum (1997Nussbaum ( , 2000Nussbaum ( , 2003 (Yılmaz, 2016). The criterion presented by Rawls (1999) and Sen (1985Sen ( , 1990Sen ( , 2003 are focused upon here. Rawls ' (1999) theory of justice emphasises equal basic liberties and just equality of opportunity. The Rawlsian principles of justice are that "each person is to have an equal right to the most extensive total system" and that social and economic inequality are arranged so that they are "(a) to the greatest benefit of the least advantaged, consistent with the just savings principle, and (b) attached to offices and positions open to all under conditions of fair equality of opportunity" (Rawls, 1999, p. 266). The focus is on the holding of 'primary goods' (including income, rights, a sense of self-worth and opportunities), defined as things that it is presumed that every "rational man" wants whatever their plans are (Rawls, 1999, p. 54). However, these primary goods answer the needs of individuals rather than their desires and preferences (Rawls, 1999) and attention is not paid to how individuals use these resources (Yılmaz, 2016).
Conversely capabilities scholars focus on fundamental diversity; understanding that while two individuals may hold the same primary goods, they have different freedoms to live a life that they have reason to value (Sen 1985(Sen , 1990(Sen , 2003. Sen (1990Sen ( , 2003 argues that to judge equality in terms of primary goods (i.e. a Rawlsian principle of justice) gives priority to the 'means' of freedom rather than the 'extents' of freedom. As such capabilities are the combinations of functionings and the real (not just formal and/or 'in principle' freedoms) opportunity to achieve them (Sen, 2003). This distinction is crucial to understand the contribution of conversion factors including social conditioning which may make an individual unable to choose (Robertson & Egdell, 2018;Sen, 2003). In acknowledging the influence of social relations as well as the societal and institutional structures, the CA rejects methodological/ontological individualism (Robeyns, 2005). The CA does however embrace ethical individualism i.e. claims "about who or what should count in our evaluative exercises and decisions. It postulates that individuals, and only individuals, are the units of moral concern. In other words, when evaluating different states of social affairs, we are only interested in the (direct and indirect) effects of those states on individuals" (Robeyns, 2005, p. 107).
For career guidance, this may be an important distinction when considering gender inequalities for example. The existence of rights to equal pay supported by legislation has not removed gender inequality in pay (Fortin et al., 2017;O'Reilly et al., 2015). Similarly, the right to pursue any occupation without discrimination does not remove the possibility that in practice women may be subject to sexual harassment from both colleagues and customers (Equality and Human Rights Commission, 2018). Thus, a CA to career guidance theory and practice goes beyond a resourcist position that justice in women's careers requires only equal rights and adopts a stance more consistent with a stronger feminist position. It also acknowledges that real life practical barriers must be overcome for capabilities to be equal.
Taking this example of workplace gender equality further, the CA might stop short of requiring that justice necessarily means symmetrical functioning in all situations, if it was the case that individual men and women had reason to value different lives and careers, and in consequence, made different choices. The emphasis of the CA is upon the freedom that individuals have to make choices that they value, and the importance of individuals not having values imposed on them regarding what is a 'good' life (Sen, 1985(Sen, , 1990(Sen, , 2003. Equality of freedom cannot be achieved through the equal distribution of primary goods, nor can it measure well-being when it does not acknowledge interpersonal variations in the transformation of primary into capabilities (Sen, 1985(Sen, , 1990(Sen, , 2003. As such the CA frames justice in terms of capabilities to function i.e. ends rather than means and reflects substantive freedom to achieve well-being (Sen, 2003;Yılmaz, 2016). A distinction is drawn between the functionings space (beings and doings) and the capabilities space (all that the individual can be and do) (Sen, 2003).
Thus, while resourcists argue that the resources that people need should be distributed, proponents of the CA emphasise the equalisation of people's capabilities (Yılmaz, 2016). A distinction is also made in terms of whether "alternative feasible institutional schemes [should] be assessed in terms of their participants' access to available resources or in terms of their participants' capabilities, that is, access to valuable functionings" (Pogge, 2002, p. 178). As such the metric of justice proposed by Rawls (1999) are primary goods. The emphasis is on an understanding of 'standard' needs rather than the value attached to these goods by reference to the specific needs and attributes of the individual (Pogge, 2002). The Rawlsian approach is focused on just institutions and underpinned with ideas of public consensus (although privately citizens may hold opposing views). Conversely the CA emphasises just societies and shared conceptions of good in society (Yılmaz, 2016). Yet, the distinction between the CA and resourcists may not be so stark as first appears, with some arguing that resourcists focus not on the goods persons actually have or consume, but on the goods persons can have or consume (Pogge, 2002). Certainly, as mentioned previously, Sen (1985Sen ( , 1990Sen ( , 2003 acknowledges the influence of Rawls on his work. Equally, resourcist and capability theorists agree regarding the purpose of the distribution of means in terms of satisfying human needs and the of positive freedom (Yılmaz, 2016).
In summary, Sen's (1985Sen's ( , 1990Sen's ( , 2003 CA entered an arena of conceptual debate about the nature of freedom and justice. Competing philosophies offer alternatives to the CA. Career guidance theory and practice needs to navigate these relatively abstract debates; including the adequacy of resources in achieving freedom and the tensions between individual and collective or community values, and the implications of this problem for freedom. Contemporary debates of the nature of social justice in the career guidance literature has acknowledged the work of Rawls but not that of Sen (e.g. Irving, 2004, 2020Sultana, 2014). While the differences between Sen's approach and a Rawlsian perspective should not be overstated, the CA does ensure that the pragmatic challenges of converting resources into lifestyle outcomes are not neglected.

Limitations arising from the origins of the Capability Approach
The disciplinary origins of the CA potentially limit its contribution to understanding careers and career guidance. Sen's CA emerged from critiques of mainstream economics focusing solely on growth measures, resources, and the actions of rational individuals (e.g. Sen, 1999). Its roots are thus primarily in economics. In contrast, the study of careers is best understood as a transdisciplinary field in which a wide range of perspectives intersect. Whilst (labour market) economics is one discipline; it is not the prime source for theory and practice. Psychology is the dominant discipline in career theory, while sociology makes an important contribution to career theory and to career-related policy analysis. There are few explicit links to psychology and sociology in Sen's work, with the CA's reputation for a being a normative tool for policy research and practice, rather than an explanatory tool for critical social theory, partly explaining this (Kremakova, 2013). The CA's neglect of sociology is rooted in the troubled relationship between the discipline and economics (Holmwood, 2013). However, some sociologists such as Hart (2012) demonstrate congurence between the work of Sen and Bourdieu. In terms of the practice of delivering career services, education, counselling, social work and youth work are strong influences.
There can be no doubt that Sen's work demonstrates a central concern for those living in poverty and seeks to extend the opportunities disadvantaged groups have to implement choices about their lives according to their own values. While Sen (2003Sen ( , 2009) distances the CA from utilitarianism and mainstream economics, critics suggest that Sen's work is infused with liberal-individualism (e.g. Carpenter, 2009). It has been argued that the 'elephant in the room' is that the CA does not address the exploitative nature of capitalism. The individual is abstracted and removed from relations of power and capitalist structures of domination that give meaning to individual freedom (Dean, 2009;Gasper, 2002;Sayer, 2012). Welfare economics is rooted in ideas that are market-oriented and evangelises market-based solutions. It is therefore at odds with sociology which defends against the tyranny of the market (Burawoy, 2005;Holmwood, 2013).
These concerns are mirrored in the career guidance literature, with authors such as Hooley et al. (2017) raising objections to the way in which neoliberal thinking has infused career guidance policy. Neoliberal welfare states focuses upon raising selfinitiative and do not account for either contextual constraints, nor individual choice (Egdell & McQuaid, 2016;Wright, 2016)-which is at odds with the CA. Irving (2017Irving ( , 2020 characterises the neoliberal perspective (in the tradition of Friedman) as 'retributive justice' in which economic freedom and property rights underpin a free market. This author advocates a critical perspective of the status quo, combined with a radical educational approach to raising consciousness in response to injustice and inequality in careers. Whether a CA could serve to meet these calls can be questioned. There is a tendency for the CA to focus on 'minimum thresholds' and limited state intervention, rather than more radical equality of outcomes (Carpenter, 2009). However, it is argued that the CA could be combined with understandings of power and class in ways that would provide support for conclusions so radical that "policy makers and advisors who favour the Capability Approach would find alarming" (Sayer, 2012, p. 583). Albeit the structural external constraints on the achievement of inequalities need to be addressed in order that these radical implications are realised (Sayer, 2012). The individualist tendencies perhaps reflect a mismatch between the bold aims of the CA and means to realise them, with Carpenter (2009) suggesting developing a radical CA through a connection with fuller political economic and social analysis.
Critique of the CA's liberal-individualist focus also draws attention to the limited acknowledgement of social solidarity and belonging. Several authors argue that the CA is flawed as it does not account for the interdependency of humans, and overemphasises rational cognitive action (e.g. Dean, 2009;Deneulin & McGregor, 2010;Iversen, 2003;Zimmermann, 2006). Individuals only exist because they are members of networks of care and responsibility, dependent on others (Kittay, 2001;Sevenhuijsen, 1998). While Nussbaum (1997Nussbaum ( , 2000Nussbaum ( , 2003 develops a list of central interrelated capabilities which includes 'affiliation'; this has been argued to be abstract, with the 'person' and 'the other' framed as abstract bearers of capabilities (Dean, 2009, p. 268). This critique has direct relevance, as some authors highlight the relational nature of career development and place this aspect as central to their conception of career guidance (Blustein, 2011(Blustein, , 2019Richardson, 2009). These perspectives do not privilege formal paid employment over other kinds of work and relationships. Others have pointed to the potential for a community level conception of career guidance (notably Thomsen, 2012). The CA is not inconsistent with this project. Whilst ultimately concerned with the well-being and freedom of individuals, it does allow for community level applications (Alkire, 2005;Alkire & Deneulin, 2009).
Taking this issue of interdependency further, the ways in which we care for each other are socially negotiated and consolidated over time (Connidis, 2001;Finch & Mason, 1993) but the time dimension is lacking in the CA (Gasper, 1997). This is problematic as biographies develop over a working life, and careers can only be understood looking backwards and projecting forwards (Robertson, 2015;Zimmerman, 2006). Here we move to a psychological critique. Perspectives on maturation derived from developmental psychology (notably Gottfredson (1981) and Super (1957)) conclusively demonstrate the inadequacy of considering career choice as anything other than a process which unfolds over time. Yet this dynamic conception is entirely absent the CA. Joncas and Pilote (2018) suggest a solution to this problem-viewing the conversion of career resources into capabilities and functionings as an iterative process, rather than as a single point in time.
There is also critique of the CA in terms of it not providing any substantive discussion of choice processes, what brings individuals satisfaction or any theory of 'good' (Gasper, 1997(Gasper, , 2002. The skills, personality formation and social supports required in making decisions about what is valuable is not elaborated upon (Gasper 2002) which present challenges for those wishing to apply the CA to career guidance practice. Sen (2009) does question the concept of rationality within economics. Unlike its parent discipline, the CA does not routinely adopt money as a proxy for 'good', or view maximising income as the only rational strategy. In rejecting methodological/ontological individualism the links between individuals and society are highlighted (Robeyns, 2005). Neither does the CA see values purely as an individual difference variable as might be the case in some psychological conceptions (e.g. Schein, 1990). It places human values as central to choice and allows for thoughtful trade-offs between them. 'Reason to value' implies a process of reasoning prior to choice but no insight as to how that process might work is provided.
A key challenge here for career guidance is that, the freedom of individuals and groups to reason about the life chances open to them, is powerfully shaped by a range of social, cultural and economic factors as demonstrated by structural theorists (notably Roberts (2009)). Furthermore, these are not simply experienced as external barriers, but become internalised as constraints on choices in the thinking of individuals (Betz & Hackett, 2006;Gottfredson, 1981;Hodkinson & Sparkes, 1997;Willis, 1977). To an extent, Sen recognises this problem in that he is critical of the use of subjective measures of well-being (at least if used in isolation) because people living in relative deprivation habituate to their circumstances (Binder, 2013;Sen, 1987). It is less clear if he recognises the impact of deprivation on choice reasoning processes, and the implications that this has for just outcomes. This raises questions as to whether at the empirical level Sen is in fact fixated with 'generic' and 'rational' individuals (Zimmermann, 2006, p. 474). Adding more choices may be undesirable as excessive choice may distract, distort and divert (Gasper, 2002). Sen's (2009) argument for the role of public deliberation in developing capabilities lists, may only to serve to exacerbate the problem, and be framed by normative liberal assumptions about who the 'public' are (Dean, 2009). Even Nussbaum's (1997Nussbaum's ( , 2000Nussbaum's ( , 2003 richer elaboration of the CA is incomplete compared to understandings from psychology as to what brings people satisfaction and well-being, nor does it differentiate between types of pleasure and happiness (Gasper, 2002).
In summary, while it could be argued that the CA offers a fresh perspective on careers by virtue of deriving from economics, economics is suffused with liberal-individualism. To a certain extent, the same critique can be levelled at the contemporary policy and practice of career guidance (Hooley et al., 2017;Irving, 2017Irving, , 2020. While Sen (2009) does problematise the concept of rationality, the CA nevertheless does not account for human interdependency (Dean, 2009;Deneulin & McGregor, 2010;Iversen, 2003;Zimmermann, 2006) and does not offer the temporal dimension required to understand how biographies unfold (Robertson, 2015;Gasper, 1997;Zimmerman, 2006). As such the CA may not be sufficiently equipped to provide a sociological analysis of inequality in careers, or a psychological description of career choice processes.

Challenges in operationalising the Capability Approach for career guidance practice
The abstract nature of the CA (some of which is highlighted in the previous sections) presents challenges to career guidance practitioners seeking to operationalise it for the development and evaluation of their interventions. For the purposes of career guidance, it is necessary to specify what kinds of resources people may have. There are existing resource models for career development (notably Hirschi, 2012) but these tend to focus primarily on individual psychological factors, whilst the CA would also require consideration of wider economic, social and environmental resources. It would also require identifying how resources may be converted into realistically attainable opportunities, how choices are made, and the kinds of outcomes that may be obtained. To achieve this, the processes of assessment and measurement need to be made explicit. However, the operationalisation of the CA in this way is challenging because of the complexity, context dependency and its under-specificity Robertson & Egdell, 2018;Robeyns, 2008;Roemer, 1996;Srinivasan, 1994).
A key critique of levelled at Sen is that in his presentation of the CA no defined central human capabilities and functionings, or explanation of how these should be weighted against each other, are offered. Sen's rationale for this is that any definition of what people value should be open to diverse conceptions of good, justice and advantage, and reflect that capabilities lists are used for different purposes (Robeyns, 2005;Sen, 2009). Equally, there is no such agreed framework to identify 'results' and evaluate career guidance practice; outcomes are conceptualised in a variety of ways (Killeen, 1996). Guidance is provided for the development of context specific lists in the application of the CA (Burchardt & Vizard, 2007). Sen has, since the first inceptions, developed his argument as to how the CA might be applied, moving from a 'thin view' which solely argues the case for capability and functioning as evaluative spaces, to a 'thick view' which accounts for the evaluative implications of applying the CA (Qizilbash, 2011;Sen, 1999Sen, , 2009). However, the lack of specified capabilities, functionings and weightings mean that Sen's CA remains underspecified and incomplete (Qizilbash, 2011;Robeyns, 2008). This makes it flexible in its application across contexts, but is arguably a relativist position, and one that can be adapted to justify a range of actions.
Serious consideration would need to be made if a capabilities list were developed for career guidance. Nussbaum (1997Nussbaum ( , 2000Nussbaum ( , 2003 develops a list of ten central interrelated capabilities (bodily health; bodily integrity; senses, imagination and thought; emotions; practical reason; affiliation; other species; play; and control over one's environment) which are in turn categorised into three capability types: (1) basic capabilities (essential requirements to develop more advanced capabilities); (2) internal capabilities (conditions of the individual); and (3) combined capabilities (internal capabilities and the external conditions in the exercise of functionings) (Nussbaum, 2000, pp. 78-80, 84-85). However, Pogge (2002, p. 210) questions the value of capability lists, and the equitability of their application, as it would involve "grading all citizens for their natural aptitudes toward each of the capabilities on the list, determining their specific deficits, and ensuring that these deficits are duly neutralized through suitable compensatory benefits". Specificity is also required by career development practitioners rather than broad-brush higher-level descriptive frameworks like Nussbaum's (Robertson & Egdell, 2018). Context sensitivity would also be important. In the generation and/or selection of such a distinct career focused list, public discussion and debate would be required to understand what people do have reason to value (Sen, 2009). This public deliberation is achieved through the 'informational basis of the judgement of justice' (IBJJ), that is the information framed as relevant by actors when considering a situation (Sen, 1985(Sen, , 1990(Sen, , 2003. This deliberation is open ended and multiple IBBJs may co-exist (Kremakova 2013). Public discussion and debate could serve as a way in which to address the problem of adaptive preference formation (Qizilbash, 2011); that is situations where inequality in circumstances results in individuals internalising conceptions of their own self-worth (Nussbaum, 1997(Nussbaum, , 2000. However, there are concerns that public discussion and debate could result in paternalism, with democratic reasoning overriding individual preferences and placing restrictions on individual liberty (Sugden, 2006). Such a view would present career guidance practitioners with complex questions surrounding the appropriateness of enhancing any capability. A society must acknowledge various distinct liberties and freedoms, but must prioritise them, in a democratic way, when they conflict (Qizilbash, 2011). As such individual client judgement needs to be developed so that they can value in which way it is appropriate and avoid any abuse of capabilities (Saito, 2003). Added to this, even if a distinct career focused list were agreed, the list and the weightings given to the value of each functioning, would not be static to reflect changing societal priorities over time (Qizilbash, 2011). However, the public deliberation aspect makes the CA distinct and offers a way in which to operationalise the approach (Sen, 2009). In this work, service user capability for voice would need to be developed; that is "their effective possibility to express their concerns with regard to the choice of the informational basis" (Bonvin & Farvaque, 2005, p. 269). The capability for voice concept bridges the gap between subjective preferences and objective assessment in evaluation.
In terms of evaluating a CA to career guidance there are further limitations that would need to be overcome. Evaluation underpins evidence-based practice in career guidance: it is therefore important, but it is also problematic. A range of approaches are available (Hooley, 2014;Killeen, 1996). Some use objective measures of career outcomes, such as salary; others use subjective measures, such as job satisfaction. Criteria may be economically or employment focused, but they may also conceptualise career development as an educational experience and therefore adopt learning outcomes (Killeen & Kidd, 1991). Some prefer psychological measures of attitudinal change or quasi-clinical measures of well-being related variables. Easy to measure variables tend to be preferred, although they may not offer the most insightful approach, and service user's perspectives are often neglected (Plant, 2012;Plant & Haug, 2018), particularly in the consideration of deeper and strategic issues, such as how service outcomes are defined and prioritised. The challenges of evaluating career guidance are considerable, and the CA implies that career capabilities are the desirable focus. On the one hand this opens-up a fresh way of conceptualising the outcomes of career guidance-by assessing the capability set: the range of realistically attainable lives/careers that can genuinely be accessed (Sen, 1985(Sen, , 1990(Sen, , 2003. On the other hand, this creates new problems. Whilst unrealised potentialities are intriguing, they are difficult if not impossible to capture for the purposes of evaluation. Furthermore, career guidance is not necessarily complete if a range of opportunities have been opened-up. It also serves to help people choose one path among many i.e. in the language of the CA, to select the functioning that they have reason to value (Sen, 1985(Sen, , 1990(Sen, , 2003. Sen (2003) does however offer the practical compromise, whereby observations of capability are constructed based on presumptions, relating well-being to the achieved and observed functionings, rather than the capability set.
A final key issue in evaluating career guidance is consideration of the level of analysis to be used. A 'good' result might be understood at a micro (individual), meso (group/organisation), or macro (societal) level (Killeen, 1996). Whilst individual level analysis tends to be given primacy, it is important to recognise the potential for career guidance to be a community level intervention (Thomsen, 2012), and the potential for conceptualising capabilities at this meso level. In pursuing this route, the issue of public deliberation becomes particularly salient. All the issues explored in this section have ethical implications for practitioners. The CA to career guidance would suggest that autonomy is a central ethical principle for practice (Robertson & Egdell, 2018), and that career choices are to be understood as valuesbased. Ethical problems and dilemmas may arise where there are clashes between the priorities of individuals, communities, practitioners, and career services. Whilst debate and discussion may go some way towards reconciling these differences, some conflict of values may be unavoidable.
In summary, the key question for career guidance, arising from the issue of evaluation, is whether capabilities can be measured. There are arguments that only functionings can be measured, as substantive freedoms and opportunities cannot be observed, only deduced Verd & Andreu, 2011;Zimmermann, 2006). Indeed, at the theoretical application level, the blurring of the boundaries in the CA between options and capabilities are highlighted (Gasper, 2002). As such there are difficulties when operationalising the CA for the real world in measuring capabilities separately from functionings (Walby, 2012). The mutual dependency of capabilities, functioning and conversion factors add another layer of complexity (Kremakova, 2013, p. 403). Added to this service users may not be used to expressing their views and may find it hard to answer the open questions required to identify the benefits of capability informed career guidance interventions .

Conclusions
Having examined these three groups of challenges, it is concluded that the CA stands up well to conceptual critiques, provided the claims made for it are moderate. It offers a distinctive philosophical approach to social justice, but clearly not the only possible approach. The differences between Sen's approach and a Rawlsian perspective should not be overstated (Pogge, 2002;Rawls, 1999;Sen, 1985Sen, , 1990Sen, , 2003Yılmaz, 2016), but the CA does ensure that the pragmatic challenges of converting resources into lifestyle outcomes are not neglected. It offers a fresh approach to thinking about careers by virtue of deriving from economics, but in common with its parent discipline, it is not equipped to provide an adequate sociological analysis of inequality in careers, or a psychological description of career choice processes (Carpenter, 2009;Dean, 2009;Gasper, 2002;Kremakova, 2013;Robertson, 2015;Sayer, 2012;Zimmerman, 2006). Nonetheless, the CA may provide a frame within which transdisciplinary dialogue about career development can take place.
It presents practical challenges in making the leap from theory to practice, leaving many difficult choices for the practitioner in terms of identifying resources, conversion factors, and outcomes Robertson & Egdell, 2018;Verd & Andreu, 2011;Walby, 2012;Zimmermann, 2006). It offers a new angle on assessment, evaluation, and service user involvement, but offers few guidelines on meeting these challenges. "The problem is that even though such an approach develops a challenging attempt to conceptualize freedom with regard to public action, it provides neither a theory of society nor a methodology of inquiry, dimensions that are both required for sociological investigation" (Zimmermann, 2006, p. 469).
Its fundamental limitation as an approach to career guidance is its incompleteness. It can stimulate thinking, but only by combining the approach with psycho-social perspectives on careers can an adequate understanding be achieved. Given the diversity of career development concepts now available it seems likely that a variety of hybrid approaches are possible. Thus, there may be no single CA to career guidance, but the potential for a diversity of models for practice sharing a common starting point, inspired by the work of Sen.
Funding The authors received no specific funding for this work.

Compliance with ethical standards
Conflict of interest The authors declare that they have no conflict of interest.
Open Access This article is licensed under a Creative Commons Attribution 4.0 International License, which permits use, sharing, adaptation, distribution and reproduction in any medium or format, as long as you give appropriate credit to the original author(s) and the source, provide a link to the Creative Commons licence, and indicate if changes were made. The images or other third party material in this article are included in the article's Creative Commons licence, unless indicated otherwise in a credit line to the material. If material is not included in the article's Creative Commons licence and your intended use is not permitted by statutory regulation or exceeds the permitted use, you will need to obtain permission directly from the copyright holder. To view a copy of this licence, visit http://creat iveco mmons .org/licen ses/by/4.0/.