Exploring Community-based Research Values and Principles: Lessons Learned from A Delphi Study

Community-based research (CBR) is a relatively new methodology characterised by the co-generation of knowledge. As CBR is integrated into institutional frameworks, it becomes increasingly important to understand what differentiates CBR from other research. To date, there has been no systematic study of CBR values and principles, which tend to be offered as a list of considerations that are taken as given rather than problematised. Similarly, research has not explored the ways in which understandings of CBR’s underlying values differ among individual researchers compared to the broader research values of a large university. In this article, we report the findings of a Delphi study which addresses these gaps through a systematic, cross-disciplinary survey of CBR researchers at a large Canadian research university. Our findings indicate diverse and complex understandings of both the potentially political nature of CBR and the perceived values of the respondents’ institution.


Introduction
In standard research practice, the scientific obligation to objectivity distances the researcher, as a legitimate producer of knowledge, from the researched, who neither influence the findings nor collaborate in the research process.In response, the past 20 years have witnessed the emergence of participatory approaches that seek to reduce the distance between researchers and the 'subjects' of research by engaging directly with local stakeholders.Communitybased research (CBR) is a relatively new methodology often aligned with critical theory and characterised by co-generation of knowledge and shared decision-making between researchers and community members.As such, CBR may challenge traditional ways of 'doing research'.Supporting CBR has increasingly become a strategic priority for universities due to its potential to enhance research impact (Hall 2009;Speer & Christens 2013).As CBR is integrated into institutional frameworks and a growing number of researchers incorporate CBR into their research practice, it becomes increasingly important to understand CBR research principles and values.
However, the idea of CBR itself can be contestable.In this article, we use CBR as an umbrella term for research that involves community engagement.Other terms that may fall under this umbrella include action research, participatory action research, communitybased participatory research, community-based participatory action research, peer research, (community) engaged research, and inclusion research.In some fields, such as health sciences, it is important to distinguish between CBR, which indicates that research takes place in the community, as opposed to the laboratory, clinic or hospital, and community-based participatory research, in which the community plays an equitable role in every phase of the research (Blumenthal 2011).
Within the literature, scholars have unpacked the terminology frequently associated with CBR, including action (Reid, Tom & Frisby 2006), participation (Cornwall 2008), community (Ross et al. 2010), engagement (Flicker et al. 2008), research (Wells & Jones 2009), peer (Roche, Flicker & Guta 2010) and inclusion (OWHN 2009).Scholarship has investigated methods of knowledge dissemination (Chen et al. 2010), levels of engagement (Flicker et al. 2008) and relationships with institutional ethics review boards (Shore et al. 2011).However, to date there has been no systematic study of CBR values and principles guiding the research process or of how the application of CBR principles differs across academic researchers and community partners in various disciplines in one large university institution.Often, CBR values and principles are provided as a list of ethical considerations that are taken as given rather than negotiated by those directly involved in the research process.Moreover, the means by which particular principles or values are identified is not explained, or is done descriptively, usually by narrating research processes.Similarly, research has not yet explored the ways in which understandings of CBR's underlying values differ with respect to the faculty member's own research compared to the broader research values of a large university with many faculties and departments which may hold rigid ideas of what counts as 'real research'.
In order to address these gaps, our purpose for this study was to provide a forum for discussion of CBR values and principles (VPs) across disciplines for both faculty and community partners.In this article, we report the findings of a systematic cross-disciplinary survey of CBR researchers and community partners at a large Canadian research university.We also explore some common understandings of CBR's defining values and principles among different groups of stakeholders engaged in community-based research.Through the Delphi approach, this study generated a set of community-based research VPs.However, the findings also uncovered diverse and complex understandings among the respondents of the potentially 'political' nature of CBR.We highlight the complexity of defining VPs of CBR in one institution, given the issues of relationality and power reflected in the study.

Literature review
The major themes in the literature on CBR values and principles may be grouped under three broad, interconnected concerns: relationships, power and social change.Relationships refer to the multifaceted relations among community members engaging in research, the community organisations representing community members, university researchers and their institution.Power denotes access and control over resources, data and decision-making as well as over the definition of legitimate academic knowledge production.Social change references the desire of many CBR researchers to better the living conditions of research participants or provide support and capacity-building for greater equity and justice.We explore these issues in more detail below.

RELATIONSHIPS
Most authors agree that the relationship between the researcher and the researched is central to CBR.Accountability, trust, reciprocity, respect, solidarity and collaboration are frequently mentioned in the literature; moreover, for CBR scholars, relationships are part of a process that is at least as important as scholarly outcomes such as publications (Brydon-Miller 2009;Elliott 2012;Israel 2008;Jacklin & Kinoshameg 2008).Several scholars suggest that CBR partners must commit to long-term research relationships and emphasise the iterative nature of the CBR process (CAMH 2011;Horowitz, Robinson & Seifer 2009).Drawing on Kirkness and Barnhardt's (2001) earlier work, Stanton (2014) proposes that CBR should adhere to the 'four Rs' of respect, relevance, reciprocity and responsibility.CBR researchers also stress the importance of open and inclusive processes and acknowledgement of one's social location (Brydon-Miller 2009;Cochran et al. 2008).Stanton (2014) examines the potential for CBR to disrupt mainstream research paradigms that privilege 'individual merit', hierarchical prestige, methodological and discursive norms, and work that culminates in publication, to instead value the lived experiences of individuals and communities and ensure dissemination of knowledge gained to all partners.In this sense, CBR blurs the line between the researcher and the researched by recognising research participants as active 'subjects' rather than passive 'objects;' everyone is an expert (OWHN 2009).For example, St Denis (1992) argues that CBR is for and with rather than about or on research participants.CBR's focus on relationships and accountability creates an affinity with Indigenous research methodologies.As in CBR, Indigenous researchers develop relationships in order to seek knowledge (Wilson 2008).Relationality in Indigenous research is not concerned so much with statistical significance or validity, but rather with accountability to relationships; this requires an unsettling of binaries such as knower/known and subject/object (Wilson 2008).Cautioning that, from the vantage point of the colonised, 'research' has been, and for the most part continues to be, a tool of imperialism and colonialism, Smith (2012) affirms 'research' to be one of the dirtiest words in the Indigenous world's vocabulary, and sets an agenda for research that takes seriously Indigenous ways of knowing and being by posing a series of questions similar to those asked by CBR researchers.These include 'Whose research is it?Who owns it?Whose interests does it serve?Who will benefit from it?Who has designed its questions and framed its scope?Who will carry it out?Who will write it up?How will the results be disseminated?'(p.10; see also Flicker, Roche & Guta 2010;Minkler 2004;OWHN 2009).In these contexts, CBR may go some way to addressing conflict between the Western values of the academic setting and those of marginalised and Indigenous communities (Cochran et al. 2008).

POWER
In conventional research methodology, the 'objects' of research provide data which the researcher ('subject') analyses and owns.Conversely, many CBR scholars share the objective of creating equity in research relationships through attention to social inequities and shared ownership of the project, and findings for the benefit of all partners (Heffner, Zandee & Schwander 2003;Horowitz, Robinson & Seifer 2009;Jacklin & Kinoshameg 2008).Breaking down traditional understandings of research subjects and objects through partnerships based on shared ownership implies disrupting existing power relations.For example, Cochran et al. (2008) argue that conventional research has perpetuated a myth that Indigenous people represent a 'problem' to be examined and solved and that they are passive 'objects' requiring assistance from outside experts.CBR responds to the limitations of traditional research approaches by acknowledging different ways of knowing, valuing the voices of community residents and generating knowledge that meaningfully addresses locally identified problems (Fletcher 2003;Jacobson & Rugeley 2007).
Generating equity in relationships means CBR must challenge power explicitly (Elliott 2012;OWHN 2009).Accordingly, most writing on CBR begins with an assumption that CBR is more openly political (in the sense of naming and unsettling relationships of power) than conventional research aimed at objectivity.For example, Brydon-Miller and Maguire (2009, p. 79) suggest CBR is an 'unapologetically political approach to knowledge creation through and for action'.For many authors, ethics and empowerment are two key pillars of CBR (Blumenthal 2011;CAMH 2011;Elliott 2012;Israel et al. 2001Jacklin & Kinoshameg 2008;Minkler 2004).Building on these ideas, some scholars insist on the importance of anti-oppression principles and an acknowledgement that research is not value free, arguing that claims to objectivity have been used to subordinate research participants (CAMH 2011;OWHN 2009;Savan et al. 2009;Schwartz & van de Sande 2011).CBR scholars assert that CBR is a response to conventional research that has failed to protect or benefit participants and directly or indirectly led to significant harm (Wells & Jones 2009).
Because of CBR's explicit attention to power relations, some critics contend that CBR is unscientific, overly political and susceptible to bias, that community interests supersede theoretical and scientific rigour, and that it constitutes activism rather than research (Hernández 2015;McAreavey & Muir 2011;Ochocka and Janzen 2014).In other words, scholars have identified a perceived tension between the values of scientific rigour and those of community participation (Elliott 2012;Minkler 2004).However, advocates argue that CBR has greater potential for meeting the standards of scientific knowledge creation than conventional social science precisely because researchers are engaged directly in the transformation of the phenomena they study (Brydon-Miller, Greenwood & Maguire 2003).Similarly, feminists have long pointed to the value of acknowledging the situated nature of knowledge (Haraway 1988).

SOCIAL CHANGE
Following from the focus on power relations, several authors suggest that a key principle of CBR involves the integration of knowledge and action for social change, with the objective of transforming fundamental structures that sustain inequalities in order to improve the lives of those involved, as they define improvement (Brydon-Miller & Maguire 2009;CAMH 2011;Elliott 2012;Jacklin & Kinoshameg 2008;OWHN 2009;Stanton 2014).For Ross et al. (2010), social justice is a goal of CBR that includes ensuring research priorities respect the needs of marginalised communities and promote self-determination.Similarly, St Denis (1992) argues that CBR must be committed to critiquing the status quo and building a more just society.
As part of promoting positive social change, many authors stress the commitment of CBR researchers to capacity-building, co-learning, and expansion of critical consciousness (Brydon-Miller & Maguire 2009;CAMH 2011;Horowitz, Robinson & Seifer 2009;OWHN 2009;Stanton 2014;Stoudt, Fox & Fine 2012).For example, Freudenberg and Tsui (2014) argue that improvements in health equity requires addressing the social determinants of health; consequently, policy change becomes a public health goal.Power dynamics are woven throughout policy efforts to improve health, and the work of public health researchers is inherently political because it concerns power relations (Freudenberg & Tsui 2014).
In sum, the literature reveals CBR as an ethical research practice that calls for researchers to be reflexive throughout the research process, leading to social transformation.Although the literature speaks to the values and principles of CBR, they have not been clearly articulated.
To address this gap, we conducted a Delphi study among active CBR researchers at a research university in Western Canada.

Methodology
The university in which we conducted this study piloted a CBR initiative a few years ago.A steering committee under the Vice-President Research was set up to increase the visibility of CBR and promote the adoption of best practice at the university.Despite the fact that CBR is widely practised on campus, there has been a concern among CBR researchers that they are disadvantaged in research ethics reviews and tenure and promotion processes by the lack of understanding of the values and principles (VPs) of CBR.To address this concern, we conducted this study to generate a list of VPs that could be used as reference for these reviews and processes.
As a comprehensive research university with over 5000 research faculty members, CBR researchers come from a variety of disciplines, each of which has its own research tradition, stakeholders and understanding of CBR.It is therefore difficult to identify all CBR researchers.Besides, the very essence of CBR entails the participation of community partners; therefore, it was also important that we involve their voices in a study to explore the values and principles of CBR.The actual number and identification of all community partners involved in CBR with this university was hard to determine.Thus, it was technically difficult to generate a frame for sampling via a traditional survey method.
Since broad generalisation was not our goal, we decided to employ the Delphi technique developed by the RAND Corporation in the 1950s (Dalkey 1967) to conduct this study.The Delphi method is a popular approach widely used in different fields to generate agreement through synthesis of a diverse range of expert opinions (Hasson, Keeney & McKenna 2000;Yan & Tsang 2005).As a research tool, Delphi depends on group dynamics rather than statistical authority to achieve consensus among experts (Okoli & Pawlowski 2004).It is a systematic, multiple-step process to solicit and collect information from respondents who are experts in a subject area.The design of a Delphi study is flexible and responsive to the actual data collection process.The number of rounds of data collection is contingent on the emergence of consensus which, although mainly based on majority view, is achieved without respondents feeling they are being judged (Geist 2010).Delphi also allows respondents to respond to emerging ideas during the research process in a time-effective manner (Tersine & Riggs 1976).In the absence of a face-to-face group discussion, respondents can express and exchange ideas freely in a confidential and anonymous fashion (Okoli & Pawlowski 2004).

RESPONDENTS
Following the Delphi tradition, respondents in our study were presumed to be experts in community-based research.Prior to this study, the Steering Committee organised several events to promote CBR on campus.An email list of approximately 200 CBR research practitioners was compiled.Phone, email and in-person invitations to take part in the study were sent to all registered researchers.We also invited people on the list to refer to us other CBR researchers who might be interested in participating, and an email was sent to all university Deans with a request to forward the invitation to members of their faculties who may have been actively involved in CBR.We also invited researchers who confirmed their participation to recommend at least one of their community partners take part in this study.A total of 106 people, including 50 faculty researchers, 37 community partners and 19 staff, who are research staff supporting and working on CBR research projects conducted by faculty researchers, were finally confirmed.They were invited to participate in three rounds of data collection, which were to take place from April to July 2015.Generally, Delphi prefers a stable and small group of respondents throughout the process.However, as it was difficult to monitor this large group of respondents, particularly when their participation was anonymous due to a requirement of the institutional ethics review protocol, ultimately only 70 of the 106 (66 per cent) confirmed participants took part in the first round of the survey.Attrition rate in Round 2 was 38.6 per cent and in Round 3, 48.6 per cent.Despite this, as noted in Table 1, there was a fair representation from faculty, community partners and staff in all three rounds.However, due to the small sample and the purpose of the study, we did not compare the answers from these three groups of respondents.

PROCEDURES
The Delphi method is a stepwise process.The first step involved creating a draft list of values and principles.Based on an extensive literature review, which included records generated from previous CBR activities organised by the Steering Committee, 13 major categories (see Table 2) with a total of 150 itemised VPs were generated.The list was reviewed and discussed by a working group which had been set up to advise the Steering Committee on ethical issues related to CBR.Minor adjustments were made based on the discussion.In view of the diverse terminologies used by researchers from different disciplines across the campus, the working group also recommended not to provide an operational definition of CBR in order to allow respondents to describe their practice in an open-ended way.An online survey tool was employed in the three rounds of data collection.The aim of the first round of the survey was to refine a list of VPs for community-based research rooted in the experiences of the researchers and community members involved in the study.This list then formed the basis of subsequent rounds of Delphi.Respondents were asked to indicate whether each category, and its itemised values and principles, was significant for their CBR practice and therefore could remain on the list in subsequent rounds.A comment box was provided under each VP for additional comments.The final question asked respondents to describe their CBR practice.Forty-three respondents provided a total of 252 comments in the first section, and 52 respondents provided descriptions of their CBR practice.
Delphi generates consensus largely and arbitrarily based on a majority rule.However, determining a reasonable cut-off point for sufficient consensus can be controversial (Yan & Tsang 2005).According to the literature, the minimum cut-off is 51 per cent, and some Delphi studies employ up to 80 per cent.Following completion of Round 1, a workshop was held to discuss the desirable cut-off point.All survey participants were invited.Twelve people attended the workshop (five faculty members, three staff, three community partners and the project RA).Following previous Delphi studies reported in the literature, attendees at the workshop decided to adopt a two-third majority rule, i.e. 67 per cent, as the cut-off point; the same figure was used for the Round 2 and 3 surveys.Although the 67 per cent cut-off was in fact arbitrary, it was considered by workshop attendees to represent a reasonable figure that was neither too restrictive nor too open.In Round 1, respondents were asked to select items from the provided list of VPs.VPs chosen by two-thirds or more of respondents were included.In Rounds 2 and 3, the Delphi survey questions were about importance and relevance as they perceived them.Answers were arranged on a Likert scale from 1 (least) to 5 (most).Only VP items for which two-thirds or more of respondents checked '4' or '5' were retained.As a result, the total number of itemised VPs was reduced to 88 from 150.Workshop attendees and the study researchers also discussed rewording some VP categories and itemised values and principles, for example, 'Participation' became 'Equitable participation'.Further, since only one itemised principle under 'Transformation of fundamental structures' passed the cut-off, it was decided that we would eliminate this category and move the remaining item to the category of 'Long-term relationship', where it was more appropriately located.
The Round 2 survey had two parts.In the first part, the central question concerned the relevance of the remaining itemised VPs in each category.Specifically, we asked, 'On a scale of 1-5 (with 5 being "most relevant") how relevant is this itemized VP to: a) your CBR practice?b) the university?'Each section included a space for respondents to share additional comments.Part 2 was an optional question that asked whether any of the VPs or refined meanings that had fallen below the 67 per cent cut-off in Round 1 were crucial to respondents' research.
Round 3 focused on the importance of the 12 remaining VP categories.Respondents were asked, 'On a scale of 1-5 (with 5 being "most important") how important is this VP to a) your CBR practice?b) the university?'Each section included a comment box for respondents to share additional comments.Since the intention of this round of the survey was to prioritise these 12 categories based on their importance, no cut-off point was used.Instead, totals were derived from the sum of the percentage of respondents who indicated the importance to be either four or five.
At the end of the Round 3 survey, we provided an opportunity for respondents to share their thoughts on possible uses of the results from this study as well as any other comments they wished to share.Twenty people provided a total of 23 comments.

LIMITATIONS
This study has its limitations.It was hard to provide a universally accepted definition of such a heavily laden term as CBR.By not providing an operational definition, we were able to include diverse opinions, but the respondents might have answered the questions from different or even contradictory perspectives.Due to the nature of the research design, we were not able to compare these diverse perspectives in the answers of the three major stakeholder groups.Further, the Delphi method is meant to solicit opinions from a group of experts through a methodologically 'neutral' medium, a survey in this case.However, the 'majority rule' in determining criteria unexpectedly raised some of the same political challenges that many CBR researchers have already experienced with regard to their CBR practice vis-a-vis the university and their departmental colleagues, in that some respondents found some of the VPs most important to them were not held in the same regard by other respondents and therefore not included in the final list.

Findings
Throughout the Delphi exercise process, we did not define CBR for respondents.Instead, in Round 1, we deliberately asked respondents to describe their CBR practice.The descriptions that we received set the context for understanding the Delphi results.According to the descriptions provided, respondents' research practices involved an extensive range of types and intensity of collaboration with community partners/researchers. Specific methods also varied widely and included both qualitative and quantitative methods.For some respondents, CBR starts with community needs, and the research questions, methods and actions taken are defined by community members.For other respondents, research questions and methods originate from the researcher and there is no expectation that action will be taken on the findings.

DICHOTOMISED VIEWS
Some respondents acknowledged that addressing power relations was important but resisted the characterisation of such efforts as 'political'.For example, one participant stated: The research itself need not be (perhaps, should not be) political or politically motivated...However, the issue of power relationships between researchers, community groups and community members is important and must be consciously and overtly addressed.
Others found descriptors such as 'anti-oppressive' and 'empowering' to be too negative and/ or pathologising, preferring more positive framings, e.g.'social justice'.The controversy around the political nature of CBR was well reflected in participants' narratives, particularly regarding the adherence to Indigenous epistemologies and anti-racist, anti-oppressive, and anti-colonial or decolonising perspectives.
For some respondents, these overtly politicised perspectives were critical to their work.As one participant aptly noted: With the dropping of the above VPs [related to these perspectives], academic researchers maintain their privileged ability to define, design, and implement…Ideally we should all be aiming to protect the most vulnerable and be committed to praxis that contributes to decolonizing and anti-oppressive methodologies and theoretical frameworks.
Through the process, we also heard a strong voice from a few respondents who repeatedly pointed out the relevance of Indigenous and anti-colonial frameworks to CBR.For example, one respondent stated in the second round: I feel strongly that the values related to Indigenous and anti-colonial frameworks need to be included, otherwise we will continue to conduct research that is colonial and creates harm.I am not sure how many Indigenous communities or partners participated in the first round, but it might be important to offer this again if the N is low.
However, not all participants agreed that CBR is always conducted with marginalised groups or that all ways of knowing should be respected.For example, as one participant suggested, 'Community collaborative research is not always about isms and oppression.'These dichotomised views were also evident in the results of the survey.Indeed, as reflected in the narrative data, we can see a divergent perspective on the 'political' nature of CBR (Table 3).The two VP categories, ' Addressing power relations' and 'Transformation of fundamental structures', which were perceived by some respondents as 'political', were trimmed down significantly in terms of the number of itemised VPs and, in the latter case, removed entirely.As indicated above, many of the more overtly political itemised VPs did not reach a majority consensus in the first round of the survey; for instance, 'Problematizes systematic relations of power in the social construction of knowledge' (44 per cent), 'Based on an anti-oppression framework' (41 per cent) and 'Fundamentally challenges the structures of oppression' (54 per cent).CBR seeks to identify and address power imbalances…by paying attention to discrimination, systemic inequalities, history, social location and how status and privilege impact partnerships.
The question of whether 'action' is an objective of CBR also provoked disagreement (Table 4).While a number of respondents characterised their research as 'community-based participatory action research', others did not suggest any action after the research, if at all.Related to this question, participants also varied in their perspectives on whether CBR questioned the status quo.For example, one commented, 'Sometimes the status quo is not that bad.' Table 4 Divergent perspectives on action as a goal of CBR

Action is not a goal of CBR Action is a goal of CBR
CBR should not be tied to action -it is about discovery, not necessarily action.This suggests removal of individual and community selfdetermination.
[CBR] moves research to action to improve and transform colonial structures, relations, practices, and dynamics.
[CBR involves] a commitment to social action according to the needs expressed by the community-based participants.

VALUES AND PRINCIPLES
A central purpose of this study was to generate a list of VPs that different stakeholders could use as a reference for research ethics applications, tenure and promotion reviews, and formal collaboration agreements (readers who would like to view the complete list of itemised VPs should contact the lead author).However, when formulating this study, we were reminded at the outset that CBR researchers have frequently perceived discrepancies between the research values that are important to them as individual researchers and the priorities of the university as an institutional collective.Therefore, in both Rounds 2 and 3 we asked respondents to rank the relevance and importance of VPs for both their own CBR practice and that of the university community.The discrepancy between individual and institutional research values is indeed reflected in our findings.

Relevance of itemised VPs
As reflected in the list, respondents place great emphasis on a few key VPs that have been discussed in the literature, such as dialogue, togetherness, reciprocity, respect for local knowledge, accountability to the community, and the importance of iterative processes, to name a few.However, our findings also show that the respondents hold different perspectives in terms of the relevance of the VPs to their own CBR practice versus the practice of the university as an institution.In Table 5, we summarise the number of itemised VPs relevant to both respondents.Looking into each VP category, we notice that, with one exception, most itemised VPs within the categories that are perceived to be relevant to the institution are also on the top of the list for researchers' own CBR practice.For example, respondents placed significant emphasis on 'Accountability' for both their own CBR practice and the practice of the institutional community as a whole.Most researchers suggested that, in terms of their own practice, accountability was primarily to their community partners.The excepted item was 'Researchers are accountable to the university' in the category of 'Accountability'.While 74.4 per cent of respondents perceived this principle to be relevant or very relevant to the institution, it did not pass the cut-off point for respondents' own CBR practice.Several participants also emphasised that researchers were accountable to outside funders; as one stated, 'In the excitement to collaborate we sometimes forget who is the funder and it is the funder who ultimately pulls the strings.'Although there are two VPs ('Equitable participation' and 'Self-determination') for which respondents indicated equal numbers of itemised VPs as relevant to both their research and that of the university as an institution, there are major differences in most categories.The greatest differences were in the categories of 'Reciprocity' and 'Reflexivity'.Filtered by the cut-off point (67 per cent), 67 out of the 88 itemised VPs included in the survey were thought to be relevant or very relevant to respondents' own CBR practice and only 33 to the practice of the institution.

IMPORTANCE OF VP CATEGORIES
Many VPs inform CBR practice; however, not all bear the same importance.In Round 3 of the survey, we asked respondents to rank the perceived importance of the 11 VP categories to their own CBR practice and that of the institution.Comparing the percentage ranking of importance for almost all VP categories, respondents tended to assign a lower importance to the work of the institution than to their own work (Table 6).'Collaboration/partnership' and 'Accountability' topped both lists, albeit in different order.To illustrate, one participant asserted: 'CBR respects diverse epistemologies and ontologies'; another stated, 'All who are actively involved in the research are accountable to each other and to an ethical research process'.The difference in perceived importance was greatest for 'Equitable participation', 'Empowerment' and 'Self-determination'.These differences may reflect many CBR researchers' uncertainty about the extent to which the university, as an institution, respects and supports their CBR work.As suggested by one respondent:

I believe as an organization [our university] is most interested in outcomes rather [than process]
. This stems from the discourse about research (and other) excellence wherein funding dollars, prestige and numbers of publications still appear to be valued most highly.Next, we present our reflections on the findings outlined above and suggest some implications for CBR researchers.

Discussion and implications
Our findings indicate that researchers engaging in CBR have diverse understandings of the nature of CBR.Despite this diversity, the findings show some consensus among respondents on VPs that are central to the practice of CBR.The final list of itemised VPs may fill a gap in the literature.Here, we highlight a few observations on the discrepancies we identified with respect to participants' perspectives on relationships, power and action for social change.We then interweave these observations with reflections on the Delphi research process, especially with regard to the political nature of CBR.

THE ETHICS OF CBR
As shown in the literature, CBR is a value-driven research approach.However, while there was broad on the importance of trust, respect, collaboration, partnership and dialogue across disciplines, each CBR researcher tended to adhere to different VPs.
Values and principles related to Indigeneity raised some concerns among participants.Research is viewed negatively by many members of Indigenous groups because it has been used as a tool of exploitation and colonialism (Smith 2012).Some researchers see CBR as a potential means of overcoming these issues and addressing past harms, but it is not possible to simply insert an Indigenous worldview into the dominant research paradigm, which is based on the belief that knowledge is an individual entity.Conversely, Indigenous paradigms arise from the fundamental belief that knowledge is relational and experiential (Wilson 2008).These ideas raise the question of what it means to take Indigenous worldviews seriously when some researchers do not see decolonisation as a meaningful research objective.There was a sense among some respondents that anti-colonial frameworks and respect for Indigenous epistemologies were only relevant when Indigenous people were directly involved in the research.From a decolonial perspective, this is problematic as we are all (settlers and Indigenous) negatively affected by colonial structures and, arguably, collectively share responsibility to address these structures in society.
Participants pointed to the importance of respecting diverse ontologies and epistemologies, but also emphasised that not all worldviews should be respected (e.g.Nazism).Still, most participants agreed that CBR is an 'ethical research practice'.Although 'ethical' was not defined precisely, such comments seem to imply that not all research is or has been ethical, a point which is also made strongly in the literature (e.g.Smith 2012).The high ranking accorded collaboration/partnership in the survey is consistent with the major discussion in the literature, reflecting the nature of CBR as a collaborative project (Brydon-Miller & Maguire 2009;CAMH 2011;Elliott 2012).Indeed, this is the VP that was least contested.In other words, there was a lot of agreement on the means (collaboration) and less on the ends (e.g.decolonisation, production of new knowledge) of CBR, and also whether power relations should form a consideration.Questions were also raised about the nature of the relationship in terms of whether solidarity was a desirable element.It is somewhat surprising to us that many researchers who are engaged in collaborative partnerships with community partners resisted characterisation of their research as 'political', when equitable inclusion (which all respondents acknowledged as important) is an overtly political intervention that challenges university hierarchies of knowledge production (Stanton 2014).On the other hand, if CBR is being conducted with powerful groups who hold reprehensible worldviews, ensuring equitable inclusion may be problematic for researchers committed to both CBR and social justice.While we are cognisant that the final list of VPs is a result of consensus based on the majority rule, i.e. an artificial cut-off point, it has provided a common base from which CBR researchers can engage in dialogue among themselves as well as with their stakeholders, particularly the university administration.The following observations may be useful for CBR researchers who have to negotiate their work constantly with their affiliated institution.It is not uncommon to hear CBR researchers complain that their work is not treasured, particularly under the current neo-liberal managerial atmosphere of the academic setting.
First, in terms of importance, respondents proposed a similar ranking of the categorical VPs for both their own CBR practice and their expectation of their institutions.However, 67 VPs were considered to be relevant or very relevant to respondents' own CBR practice compared to 33 for the institution; among the 33 itemised VPs deemed relevant to the university community, eight (24 per cent) concern 'Accountability'.In other words, there is a perception that the institution is concerned most with accountability and publications and less with relationship building, which accords with the broader neoliberal context.CBR researchers, even those who resist the characterisation of CBR as political, seem to want to insert ethical considerations into research processes, while the university is more concerned with measurable outcomes in terms of publications.Finally, principles such as 'Values process and outcomes', 'Long-term relationships' and 'Reflexivity' seem to matter to respondents' own practice more than to that of the university as an institution.This may reflect respondents' perception of the administrative emphasis of the university as an institutional organisation or a cynical attitude on the part of researchers who feel their research is not held in high regard by their institution.Indeed, it is this perceived lack of regard that lay behind the creation of the CBR initiative at the university in question.

THE POLITICS OF CBR
Questions of power are inherently political because, in broad terms, politics concerns the distribution of power and resources in society.Coming from social work, education and geography disciplinary backgrounds, we had understood CBR and indeed all research to be 'political'.However, some of the responses we received to the survey reveal that this is not the view of all CBR researchers.Here we explore the implications of respondents' differing perspectives on equality/equity, anti-oppression and objectivity.
It became evident early in the study that, when we tried to define CBR, many tensions emerged amongst faculty members in various disciplines, between those doing more quantitative research than qualitative research, and between faculty and community members and researchers and their institution.Interestingly, the main tensions seemed to be rooted in the epistemological, ontological and axiological positions of the respondents, which were closely tied to their discipline and institutional context.This was further complicated by the reality that, despite our efforts as researchers to be as inclusive as possible through various recruitment methods, including institution-wide invitations, faculty-wide invitations and personal invitations, our participants were inevitably only partially representative of the faculty, staff and university community.The absence of many voices led us to question the ethical nature of the research that we were undertaking, especially when we read many of the comments on the study in Round 3 concerning the importance of respecting diverse epistemologies, addressing power imbalances and accountability to an 'ethical research process'.
Admittedly, the consensus-seeking nature of the Delphi approach might have further marginalised some political views held by CBR researchers from some disciplines and, as a result, in many 'political' VPs being eliminated.Most participants agreed that building equality or equity in relationships means addressing power explicitly; however, based on the 67 per cent cut-off, itemised VPs that included openly political terms such as power, anti-oppression, Indigenous and anti-colonial were dropped following the first round.In other words, at least among the respondents to this study, most did not agree with the 'political' nature of CBR.However, some respondents in the second round expressed concerns with this result.We realised that we did not have the means, given that the consensus of the group determined the final list of VPs, to deeply address the many tensions and systemic inequities that seemed to mark the texts of the survey responses.Fortunately, some members of the study spoke up during our workshop after Round 1 and consensus was reached to reinstate several VPs that otherwise would have been eliminated from the final list, due mainly, in our view, to the absence of certain marginalised voices, disciplines and non-mainstream approaches to research in the survey process.This was due in part to systemic inequities and institutional absences.Removing the most overtly politicised VPs was perceived by some respondents to leave academic researchers in the privileged position that many scholars claim CBR is supposed to redress, and perhaps to undermine decolonial and anti-oppressive methodologies.In short, the tendency of the majority of respondents to opt for a relatively objective and apolitical position was viewed by others as masking what were fundamental issues of injustice which have significant impact on institutional practice of tenure, promotion and ethical approaches to CBR.These findings raise many questions.What does it mean to suggest that CBR (or, indeed, any research) is non-or apolitical?What are the implications of resisting acknowledgement of the political nature of research?One of the critiques of objectivity in the literature is that it has been used to subordinate research subjects within specific projects as well as CBR researchers in the academy (Absolon & Willett 2005;Deloria 1997;Wells & Jones 2009).Is it possible or desirable to acknowledge one's positionality and simultaneously claim objectivity?Why do some researchers resist designating their research anti-oppressive or anti-colonial?What are the effects of this resistance for researchers, research participants, and CBR more broadly?

WHAT IS CBR FOR?
Building on the debate over the political nature of CBR, the question of whether positive social change was a meaningful research objective was also contested by participants.Although respondents agreed that CBR results should benefit all participants, there was less agreement on whether improving lives was a desirable or reasonable goal of CBR.It is interesting to note in this context that no itemised VP from the category of 'Empowerment' passed the cut-off point.
Related to these questions, for some researchers critiquing the (presumably inequitable) status quo was crucial to their practice, while others argued that the status quo was not always in need of critique and that the goal of CBR should be discovery and knowledge creation.Yet, we wonder if CBR is simply aimed at the creation of new knowledge, how can researchers avoid reinscribing colonial relations or repeating the mistakes of past research that mined community members for their 'data' without improving their lives?To be critical of power relationships implies the desire for change.We expected to see these concerns reflected in our findings.Although 'Action for positive social change' remained important for many participants, 'Transformation of fundamental structures' was removed after the first round.Once again, the more overtly political actions tended to be rejected.In other words, there was some agreement that action is an important principle of CBR, but much less agreement on the nature of the action, for example, whether the goal of action is to further decolonisation or something more mundane (e.g.publication of a report).This goes to the heart of the disagreement among participants: is CBR a political research approach aimed at action to improve lives, or is it an objective research approach that seeks to create new knowledge?Can it be both?

Implications for future research
Based on the findings and our related reflections, we propose the following additional questions about CBR may be worth exploring further: -What does 'political' mean in the context of CBR, and how political should CBR be? -As a research method, should CBR have a 'predetermined' outcome?-Does CBR require different forms of accountability compared to other methodologies?-Is 'action' an objective of CBR?What is the relationship between CBR, action and justice?-Is CBR only for marginalised/colonised groups?To what extent should CBR be informed by a particular discourse?-Is there any element that distinguishes CBR from other research approaches on which all CBR researchers could agree?Should CBR be defined?To conclude, CBR is a growing research approach increasingly being adopted by researchers from diverse disciplines.While the findings reported here may fill a gap in the literature on which values and principles matter to CBR, they also raise additional questions for further exploration.The diverse perspectives on the political and action-oriented nature of CBR comprise an important issue that researchers and community members whose work comes under the CBR banner should address as more and more academic institutions begin to emphasise the importance of community-based research.NOTE: Readers who would like to view the complete list of itemised VPs should contact the lead author.

Francis,
Yan and Gill Gateways: International Journal of Community Research and Engagement, Vol.11, No. 1, May 2018

Francis,
Yan and Gill Gateways: International Journal of Community Research and Engagement, Vol.11, No. 1, May 2018 Exploring Community-based Research Values and Principles Gateways: International Journal of Community Research and Engagement, Vol.11, No. 1, May 2018 Exploring Community-based Research Values and Principles Gateways: International Journal of Community Research and Engagement, Vol.11, No. 1, May 2018

Francis,
Yan and Gill Gateways: International Journal of Community Research and Engagement, Vol.11, No. 1, May 2018

Francis,
Yan and Gill Gateways: International Journal of Community Research and Engagement, Vol.11, No. 1, May 2018 Exploring Community-based Research Values and Principles Gateways: International Journal of Community Research and Engagement, Vol.11, No. 1, May 2018

Francis
Journal of Community Research and Engagement, Vol.11, No. 1, May 2018 Exploring Community-based Research Values and Principles Gateways: International Journal of Community Research and Engagement, Vol.11, No. 1, May 2018

Francis,
Yan and Gill Gateways: International Journal of Community Research and Engagement, Vol.11, No. 1, May 2018 Exploring Community-based Research Values and Principles Gateways: International Journal of Community Research and Engagement, Vol.11, No. 1, May 2018

Table 3
Divergent perspectives on the political nature of CBR

Table 6
Percentage ranking of importance for VP categories *Total percentage of respondents who assigned 4 or 5 to this item.