Service users and academics: collaborators not competitors in welfare research

ABSTRACT

The importance of actively involving users of welfare services in research has increasingly gained ground in both public policy documents and the academic literature. User involvement is increasingly considered a precondition to obtain external funding for research projects. However, a growing opposition to this development among academic researchers appears to increase. The participation of users as co-researchers is seen as a degradation of the researchers’ competence and lacking respect for what creates quality in research. The objections and discussions illustrate the need for a renewed discussion about how knowledge is created and what differences the inclusion of service users in research could bring to the knowledge production process. This article aims to address these issues, and concludes that the participation of users as co-researchers does not mean that academic competence and qualifications are downgraded, but does imply acknowledgement and recognition of the different competencies brought by different actors into the research process. Together, the different competencies will create a broader and more solid basis for knowledge development. The discussion about service user involvement in research should primarily be transformed into a debate about forms of collaboration and how user involvement and co-research could be implemented in ways where both researchers and users could benefit and further improve the development of welfare services.

KEYWORDS:

• User involvement
• co-research
• research quality
• collaboration
• different competencies

Introduction

The importance of actively involving users of welfare services in research has increasingly gained ground in both public policy documents and the academic literature (Hancock et al. 2012; Heaton, Day, and Britten 2016; Macaulay et al. 2011). In the UK, ‘Standards for Public Involvement’ are designed to improve the quality and consistency of public involvement in research (https://sites.google.com/nihr.ac.uk/pi-standards/home). The Social Care Institute for Excellence (SCIE) has developed an approach to what counts as evidence and knowledge that recognizes the scientific value of including the knowledge held by people who are users of the welfare services (Fisher 2016).

User involvement is increasingly considered a precondition to obtain external funding for research projects (Case et al. 2014). In Norway, applications for welfare research programmes financed by the Norwegian Research Council must specifically justify if user involvement is not considered (Norges forskningsråd 2015).

There are two main arguments for involving service users in research. First, there are qualitative arguments. The quality of research will be improved when persons with user experiences participate in the research process. Owing to their background, they will raise questions and interpret data in ways other than academic researchers (Hancock et al. 2012; Mayan and Daum 2016). Second, there are democratic and ethical arguments for involving users. They should have the right to have a say in research that has importance for their lives (Hancock et al. 2012; Rose 2014).

The growing claim for user involvement in research is part of what Nowotny, Gibbons, and Scott (2001) has described as a shift of mode in knowledge production. In the traditional scientific paradigm (mode 1), scientific knowledge is produced within an academic sphere separated from the influence of external society. The new paradigm (mode 2) is described as research practices performed in dialogue with their implementation. Such research practices require researchers to open up to various fields of interest and recognize that knowledge is produced in arenas other than academia. As Hallberg (2007) formulates it, science exists in society, and, conversely, society exists in science. Science and society are coproduced. Consequently, theories, methods and instruments that produce knowledge have a social background and all knowledge has both an epistemological and a social origin.

A consequence of the mode-2 approach is that scientific controversies increase (Hallberg 2007). This development creates tensions and is seen as challenging to more traditional views on research and knowledge production. Consequently, the increasing calls for involving service users in research appears to elicit growing opposition among academic researchers. Much of the resistance seems to be connected to projects wherein service users are actively engaged in different ways as co-researchers. The co-researchers are seen as lacking the necessary skills and qualifications for research, as underestimating the complexity of research activity, and as partisan representatives representing subjectivity and threatening the objectivity of research (Boote, Telford, and Cooper 2002; Case et al. 2014; Hancock et al. 2012; Prior 2003).

Furthermore, questions are raised regarding the quality of the research where users are involved as co-researchers. Based on a systematic review of health research from selected databases where patients or relatives were involved as co-researchers, Malterud and Elvebakken (2019) conclude that much of this research is of low quality and has a low impact, and consequently, does not necessarily lead to the advancement of knowledge. They question the relevance of including co-researchers with insufficient training to participate fully in research projects and argue that this subsequently downgrades the theoretical and methodological competence of the researchers. They conclude that the active involvement of co-researchers often has extra costs and signifies a waste of resources, leading to research of low scientific quality. Normative arguments valuing cooperation prevail at the expense of advanced knowledge production. Similarly, the leader of the Department of Health Promotion and Development at the University of Bergen claim that researchers are too timid to assert their qualifications to conduct research (Carlsen 2019). She raises the rhetorical question: ‘Is it not the researchers who know who can do research, who know what is possible to research, how to collect data, draw out the main findings and arrange and publish them?’ She concludes: ‘We must not be so afraid of kicking downwards that we do not dare to speak up and say that in fact we are those who know this best.’ In other words, on the whole, the participation of users as co-researchers is regarded as degrading the researchers’ competence and lacking respect for what creates quality in research.

The objections and discussions illustrate the need for a renewed discussion about how knowledge is created and what differences the inclusion of service users in research could bring to the process of knowledge production. This article aims to address these issues. Such an epistemological debate cannot be separated from a discussion about power and power structures. Who has the power to define what are important issues to investigate and what is valuable and evident knowledge? A discussion of what roles service users should have as co-researchers and what challenges and dilemmas these create for the quality of research must be subject to this epistemological debate. The conclusion of this debate should be that the participation of users as co-researchers does not mean that academic competence and qualifications are downgraded, but implies acknowledgement and recognition of different competencies brought by different actors into the research process, and that together, they will create a broader and more solid basis for knowledge development. The discussion, therefore, should preferably be turned into a debate about forms of collaboration and how user involvement and co-research could be implemented in ways in which both researchers and users could benefit and further improve the development of welfare services.

What is research quality?

Research quality is not an unambiguous concept and cannot be regarded separately from epistemological issues. In the traditional scientific paradigm (mode 1), where scientific knowledge is produced within an academic sphere separated from the influence of external society, quality mainly refers to the scientific process encompassing all aspects of the study design. In particular, it pertains to the judgement regarding the match between the methods and the study questions, the selection of subjects, the measurement of outcomes, and the safeguards against systematic bias, non-systematic bias, and inferential errors (Boaz and Ashby 2003; Lohr 2004). In the paradigm of mode 2, where science and society are seen as deeply intertwined, knowledge production does not occur in a vacuum independent of the context where it is produced, and cannot be reduced to an activity free from societal values and interests. In this way, this paradigm represents a clear opposition to a traditional positivist approach, where phenomena are seen as independent of and unaffected by the behaviour of researchers, who are able to carry out objective research and value-free investigations.

While defining research quality, the Norwegian Research Council refers to three aspects of research that have received general and broad support (Norges forskningsråd 2000): solidity, originality, and relevance. Solidity refers to the complete and systematic documentation of assertions and conclusions of the research. Originality refers to both the value of research outcomes and the innovative implementation of theories and methods. Relevance is related to both the importance and benefits of research for relevant actors or society in general. Separately, the aspects can be seen as legitimating elements for both mode 1 and mode 2, and this document indicates that there might be some disparity between the different dimensions of quality. These disparities might create controversies among researchers who hold different epistemological stances (Hallberg 2007).

The relevance argument has gained increasing importance in research policy (Østensjø and Askheim 2019). In addition, research is increasingly regarded as a tool to meet societal challenges. Consequently, quality is related to how research can create links to societal interests and values and lead to social innovation. ‘Impact’ has become a core concept in valuing the quality and importance of research. A much-used definition of impact among institutions financing research is: ‘An effect on, change or benefit to the economy, society, culture, public policy or services, health, the environment or quality of life, beyond academia.’ (www.hefce.ac.uk/rsrch/REFimpact/).

Even if science and research have always been characterized by discussions and conflicts in matters of knowledge, Hallberg (2007) claims that tensions and controversies will increase as a consequence of the new mode. So-called ‘controversy studies’ have developed as a research area within the theory of science to investigate the processes leading to what is considered as evidence in research (Hallberg and Bragesjö 2003).

Research quality and power

When relevance and impact become core concepts of research quality, a logical consequence will be closer cooperation between researchers and other actors and their inclusion as collaborators in the research process. Nevertheless, within the mode 2 paradigm two main theoretical traditions can be identified in the research literature (Kindon, Pain, and Kesby 2010; Wallerstein and Duran 2008). One tradition involves a pragmatic, consensus-oriented approach based in action research that was developed by Kurt Lewin in the late 1940s. This approach considers the development of theory and knowledge as something that occurs in cyclic processes through dialogue between different actors, where actors’ experiences contribute to the creation of practical solutions to complex problems. The increasing interest of public authorities in user participation can largely be related to this pragmatic tradition and how research can create positive innovation and benefit society (Glasby and Beresford 2006; Heaton, Day, and Britten 2016; Rose, Thornicroft, and Slade 2006). By involving users in research, better evidence for what leads to positive effects can be obtained. As the research will be relevant, it will have better quality; it will be ‘more accurate and precise,’ as formulated in the Norwegian government’s national research and innovation strategy, ‘HelseOmsorg21’ (Health Care 21) (https://www.helseomsorg21.no/).

The other tradition has an explicit starting point in social constructivism, stating that the social reality is not a neutral entity but construed by humans (Moses and Knutsen 2012), which implies that knowledge is always situated in a context. This means that the background, context, culture, and historical period of which we are a part will influence our interpretations of the world. It is formed through interactions with others and the historical, social, and cultural norms that shape people’s lives. A constructivist approach is the background for what is called ‘standpoint epistemology’, developed by feminist researchers (Harding 1986, 1993; Hartsock 1983). This approach has an explicit power perspective. Individuals and groups in society are stratified into hierarchies of privilege and deprivation, domination, and subordination. Societal power relations play out epistemologically. Groups with societal power also have the power to perpetuate their dominant perception of reality and convert it into an ‘objective’ truth. In their approach, the standpoint epistemologists combine a radical feminism and Marxism and argue that, in the same way as a proletarian view run out of the antagonism between labour and capital, the feminist standpoint has its roots in the different systematic structure of life for men and women in the Western world. From such a point of view, it will accordingly be urgent to include the voices and standpoints of silent and oppressed groups in research. Consequently, there will be a widening of the knowledge base, and research will be democratized. For oppressed groups, it will mean increased opportunities for liberation and justice.

A standpoint approach to research is also inspired by the Latin American thinkers Paulo Freire (1996) and Orlando Fals Borda (Fals Borda & Rahman 1991). They consider research and knowledge development as tools for colonization and the elites that have contributed to marginalization, oppression, and injustice. From their viewpoint, the intention of research should be to break dominance of the elites and contribute to mobilization, liberation, and justice for marginalized and oppressed groups. In other words, the goal of research should be to contribute to both knowledge production and social change.

Standpoint theory has more recently been adopted in welfare research with the term ‘service-user standpoint theory,’ which argues that service users occupy a particular standpoint on the basis of their experiences of being service users and that they facilitate insights, perspectives and understandings based on lived experience (McLaughlin 2010; Duffy and Beresford 2021). The approach especially has influenced research on disability issues and mental health. The emphasis is on valuing the direct experiences of people, including both their experiences with welfare services and their general experience of their life situation (Sweeny et al. 2009; Beresford 2013). In the so-called ‘social model of disability’, disability is considered a result of marginalization and discrimination of persons with impairments (Oliver and Barnes 1998; Thomas 1999). As early as 1992, the concept of ‘emancipatory research’ was introduced, where the goal of research was to contribute to liberation and eliminate marginalization and oppression (Oliver 1992). In the mental health area, the term ‘survivor research’ was established in a similar way, emphasizing the close connection to user movements and regarding research as an important tool in the fight against the marginalization and oppression of people with mental health problems (Sweeny et al. 2009; Rose 2017).

The standpoint perspective argues that so-called ‘mainstream’ knowledge that bases its claim for knowledge production on the belief that it is ‘value free’ and objective cannot escape the fact that it is socially situated, and that the knowledge produced will be contingent and partial. Harding (1993) consequently argues that knowledge that is both self-reflectively partial and aware that it is grounded socially is more transparent and ‘objective’ and is characterized by what she calls ‘strong objectivity,’ whereas ‘mainstream research’ is characterized by ‘weak objectivity’ because it hides its values and epistemological bases to naturalize and normalize the position of marginalized groups, thereby making the status quo unquestionable.

As a consequence, a standpoint approach will be explicitly political. Research is regarded as an essential political activity rather than a ‘neutral fact-finding mission’ (Beresford 2013, 145). Research cannot be detached from power structures and those who have the power to define and conceptualize and analyse issues and problems.

Garrow and Hasenfeld (2017) argue that a standpoint epistemology should direct research into social work more strongly. They refer to the ‘global definition of social work’ developed by the International Federation of Social Workers (IFSW), which states that the main tasks for social work are to identify and combat causes of oppression, exploitation, and social inequality, to eliminate conditions that lead to human suffering, and to advocate for social rights. They argue that much of the ‘normal’ science in social work fails to meet these objectives. Much of the research focuses on individual-level risks and protective factors in coping with environmental adversities. Consequently, both the policy and practice implications of such research are to promote personal transformation via individual-level intervention modalities, rather than to inform structural change. Nevertheless, although they state that such a standpoint epistemology must imply that users have an active voice in research, they do not explicitly argue for the active participation of users in the research process.

Co-research – different approaches

When service users are actively involved in research, a common term is to name them as co-researchers. However, what constitutes co-research has yet to be clearly defined. A common distinction is between consultation, collaboration, and user or consumer control referring to different types and degrees of user involvement (McLaughlin 2009; Cooklin and Oldenburg 2016). Consultation implies that users are asked about their views to influence decision-making regarding research, for instance, as members of a user panel or reference group or are asked to provide input in different phases of the research process. In collaboration, an ongoing partnership between researchers and users occurs in some or all of the steps of the research process. User – or consumer-controlled research implies that users design, undertake, and disseminate research results with researchers involved only at the request of the users. The International Collaboration for Participatory Health Research (ICPHR) network (2013) states that the goal is to maximize the participation of users in all parts of the research process, including the formulation of research issues and questions, design development and method selection, data analysis, interpretation of the results, publication and dissemination of the results, and implementation. Similarly, this is the ideal in ‘participatory action research,’ where the action criterion is of primary importance.

In research based on a social model of disability and mental health problems under headings such as ‘emancipatory research’ and ‘survivor research,’ a main point is that control should be placed in the hands of the researched, not the researchers. Users should have control over both the process and the resources. As formulated by Oliver (1997, 17), ‘It is not possible to research opposition in an objective or scientific way.’ In a special issue of the journal Disability & Society, the authors raise the need for a renewed alliance between activists and researchers, and state that the gap between activism and academia should be bridged (Moore 2019; Oliver 2019)

Based on a review of articles that have summarized how user participation is implemented in research and/or its potential impact in different ways, Østensjø and Askheim (2019) conclude that consultative strategies dominate in projects initiated by researchers. Cornwall (2008) states that what she calls ‘functional participation’ is the most common method of participation by non-researchers. This means that they are invited to participate in projects without being included in developing goals, but with the aim to meet these goals in a more effective way. An illustrative example is the 5-year project ‘The University and College Social Services Office’ implemented by the Norwegian Directorate of Health and Social Services. The overriding goal of the project was to reinforce knowledge and quality in municipal social services and the Labour and Welfare Service (NAV) through developing and expanding new forms of relationships between research, education, practice, and users. Even if a main objective was to empower the users, a main conclusion from the project was that the researchers with a traditional view on research controlled the project. User representatives took part, but their influence was limited, and their role was diffuse (Gjernes and Bliksvær 2011).

On the other hand, in the integrating strategy, there are projects in which research is initiated by or developed in cooperation between researchers and other actors and makes use of different ways of cooperation with the goal of integrating different kinds of knowledge. Walker (2010) characterizes such forms of cooperation as ‘strong co-production of knowledge’ with the intention of producing knowledge by integrating different actors’ perspectives and experiences. Blackstock, Kelly, and Horsey (2007) refers to this as a substantial form of cooperation that encourages exploration of the issue under investigation from different perspectives. Askheim and Raak Høiseth (2019) emphasize the importance of adopting such an approach in the early phase of a research project. This is the phase in which the research questions and issue under investigation are elaborated. In other words, it is in this phase that the premises for the project are established and the roles of the co-researchers defined. In a review with the goal of documenting the impact of involving patients and other users in health and welfare research, Brett et al. (2014) confirm the importance of involving users in the early phase of a project. They further show that integrative approaches wherein users are more involved in the initial and planning phases took better care of their needs and interests in the later phases of the project, thereby strengthening the quality of the research. User involvement in the interviewing process contributed to deeper insight into the issues investigated. The users also contributed additional perspectives from those of the researchers in the data analysis and interpretation and contributed to more extensive spread of the results. Involving users also makes it easier to recruit informants in the research process, especially from groups that might otherwise be difficult to gain access to (Ennis and Wykes 2013; Case et al. 2014; Mayan and Daum 2016). Based on a literature review and broader studies of public and patient involvement Hughes and Duffy (2018, 1189), conclude that ‘the greater and more meaningful the level of involvement the more likely there are to be positive outcomes for all involved.’

Alm Andreassen, Romsland, and Lundgaard Søberg (2019) argues, however, that the divisions between the different approaches can be rather fluid. A consultative user panel could, for instance, function as a kind of public, where the researchers become responsible for the users and must justify their choices and actions to them. Kindon, Pain, and Kesby (2010) also state that in practice, the degree of participation will vary, and that co-produced research should not necessarily imply that everyone participates in the same way or to the same extent (Fleming et al. 2014; Frankham 2009). Different kinds of participation could also be valid in different parts of the research process. McLaughlin (2010) states that it would not be right to ask co-researchers to perform tasks for which they have neither skills nor qualifications. For instance, it would be wrong and unrealistic to expect co-researchers with a service-user background would to be competent in advanced statistical analysis, questioning quantitative research data and utilizing advanced qualitative research techniques. All co-researchers might also not be interested or have time to participate in all phases or parts of the research process. McIntyre (2008) argues that it is the quality, not the degree of participation, that is decisive, and that the best strategy is for researchers and co-researchers to discuss and identify the most expedient form of cooperation together. This is about developing a ‘complementary collaboration, based on a complementary expertise’ (Fleming et al. 2014, 711).

Academic researchers and service users as co-researchers: complementary competencies

Belcher et al. (2016) have identified four main principles with attached definitions and criteria to consider research quality within what they define as a co-production paradigm in research: relevance, credibility, legitimacy, and efficiency. Relevance refers to the necessity, importance, and impact that the research has for the actual issue under investigation and greater society. This principle includes the research issue and questions, as well as the results. Credibility refers to the degree to which the research results are scientifically robust and credible according to established scientific criteria, supplemented with criteria for considering transparency and reflexivity through the research process. Legitimacy refers to whether the research process is seen as impartial and ethically proper by the recipients and users of the knowledge. This principle includes acceptable representation of users, an account of how they are involved, a consciousness connected to the diversity of perspectives, values, and interests, and how they can be best taken care of. The question of representation is often raised in research when service users are involved as co-researchers. Kylberg et al. (2015) claim that unlike in qualitative studies where a main goal is to let the informants share their deep personal experiences, the users as co-researchers should represent broader perspectives than just their individual experiences. Alm Andreassen, Romsland, and Lundgaard Søberg (2019) argues that instead of asking if representation should represent individual traits with the users, the main question should be whether they bring in alternative questions and issues. However, just as important is how the researchers understand the input the co-researchers contribute with in the research.

The last principle according to Belcher et al. (2016), efficiency, relates to how the goals for collaboration are achieved. They state that traditional goals, such as the publication of a certain number of scientific articles, will not be sufficient to evaluate the efficiency of research that aims to contribute to changes of understandings and/or practices. Such changes could be measured indirectly as changes of behaviour, skills, or technology. However, they indicate that more fundamental changes in the society represent a complicated challenge to evaluate.

Belcher et al.’s (2016) principles indicate that different competencies are important and necessary to fulfill the goal of the co-production paradigm, or what Nowotny, Gibbons, and Scott (2001) describe as mode 2 in research. Both the established academic quality criteria and the quality that users bring with their direct experiences with the issue under investigation are important and necessary, but cover different principles in Belcher et al.’s model. Credibility and legitimacy are both well-established principles that the research should follow, and for which academia provides the researchers robust training. The relevance and efficiency principles point to the importance of involving users in the research process. At the same time, the relevance principle acts as a reminder of the political aspects of research, and that it cannot be separated from a discussion about power and power structures, i.e., for whom is the research relevant? The benefit of combining the competencies illustrates the importance of implementing the integrating approach in research and involving users in the early phase, when the research issues are being defined and questions formulated.

Adding to Belcher et al.’s (2016) principles with a concrete content leads us to conclude that the user perspective need not compete with the competence of academic researchers. Service users bring in and seek alternative perspectives on reality and different understandings of problems and solutions. Including the users as co-researchers therefore, does not mean that the competence of academic researchers is downgraded and devalued. The competence that they bring through their training and experiences will be fundamental to the quality of the research. What it means is that together, different competencies should create a broader and more solid basis for knowledge development.

However, such a conclusion does not imply that user involvement is a guarantee for quality in research. At the same time as they proclaim that the primary goal is to place control of the research into the hands of the researched, researchers who are profiled spokespersons for strong and significant user participation also emphasize that user involvement is no guarantee for quality and that the research must be reflexive and self-critical. McLaughlin (2010, 1600) states that ‘to merely argue that the involvement of service user co-research will naturally improve a research project is as misguided as believing only academic researchers can undertake “real” research.’ We should therefore pay research that involves service users as co-researchers ‘the same respect as we pay other research and examine its methods and claims critically and avoid giving a “softer touch” just because it is undertaken in collaboration with or by service user researchers’ (McLaughlin 2010, 1064). A self-critical approach also appears to have gained a more distinct position among the proponents of user involvement in research in recent years. Beresford and Rose (2009) state that ‘survivor research’ needs to be as systematic, rigorous, and committed to quality and reliability as mainstream research. The main point is that this research is based on a value base distinct from a traditional positivist research approach.

McLaughlin (2009) refers to Nolan et al. (2007), who notes that privileging user-controlled research itself could potentially represent a form of oppression because it assumes that all service users want to exercise this level of control, even when the evidence does not support this viewpoint. He also refers to Barnes and Mercer (2006), who comment, in relation to disability research, that while some researchers would claim the need to select a disabled person as an interviewer of other disabled people, it is not clear how far they would take the matching process. Does the matching process also need to cover traits such as age, gender, ethnicity, sexuality, social class, religion, and type of impairment? Being a service user does not automatically provide someone with an affinity for other service users or necessarily make someone a good researcher. If we accept that service users have an important contribution to make to knowledge creation as co-researchers, we need to be able to assess the validity of the knowledge claims: ‘The criterion for a research identity thus becomes dependent on research skills, not one’s identity as a service user or not’ (McLaughlin 2009, 82).

It is also important to point out that in order to take user involvement in research seriously, resources with respect to both time and economy are required. The importance of being conscious of resources when projects are planned and implemented is emphasized (Case et al. 2014; Nolan et al. 2007; Minkler and Corage Baden 2008). The question of resources is involved in, for instance, not only the clarification of expectations regarding the research, training and support, and building trustful relations, but also the acceptable remuneration to the co-researchers. If these costs are not addressed, the research will easily be undermined and in danger of becoming tokenistic (McLaughlin 2006).

Recognizing these issues might create tensions for academic researchers who often operate with scarce economical resources. However, as professionals and producers of knowledge they should have little to fear. Instead of downgrading service-user involvement in the research they should approve it and show respect and humility to the knowledge, perspectives and understandings the users bring to the research process. Together, the different competencies create a broader and stronger basis for knowledge development. As McLaughlin (2009, 39–40) has put it: ‘Failing to take up this opportunity is to ensure an incomplete picture and to have ignored the possibility of improving the nature and quality of research’.

Disclosure statement

No potential conflict of interest was reported by the author.

Correction Statement

This article has been republished with minor changes. These changes do not impact the academic content of the article.