Translating research for policy: the importance of equivalence, function, and loyalty

The paper addresses the long-standing challenge of making academic social science research more useful to government. Despite growing expectations for research use and impact, and institutionalised links between academia and policy, visible failures of evidence to influence policy persist across time and contexts. The authors argue that barriers are systemic and not fully explained by prevailing models. They propose reframing research use through concepts from Translation Studies to analyze what is carried over from research into policy, beyond linear notions of transfer and beyond approaches that either abandon translation or emphasize total transformation. The research question is how to conceptualize and improve research translation so that useful content is preserved while meeting policy functionality. The paper develops a middle-ground perspective using three Translation Studies concepts—equivalence, function, and loyalty—and explores them empirically via two projects on the use of academic research within a UK central government ministry.

The literature on research use consistently identifies enablers and barriers including political/institutional context, relevance of evidence, and relationships between communities. Linear, rational models of evidence use dominate policy self-understanding but are widely critiqued as empirically weak; alternative accounts stress complexity, politics, and multiple non-instrumental uses (enlightenment, political, tactical). The "two communities" metaphor emphasizes intercultural barriers and solutions like knowledge brokers, boundary spanners, partnerships, and embedded researchers. Risks of individual brokers (advocacy, inefficiency) motivate organizational models and coproduction. However, most scholarship focuses on interaction processes rather than the cognitive content of evidence and how it is transformed. The concept of translation is used in two conflicting ways: (1) linear transfer of "what works"; (2) ANT/STS-inspired views emphasizing transformation and "betrayal" (translation as displacement). Some scholars reject the translation metaphor altogether, favoring practice and interaction frameworks. The authors argue both responses are insufficient, since meaningful content must endure for research use to make sense. Translation Studies offers relevant insights: its historical debates over equivalence (word-for-word vs sense-for-sense), audience effects, and cultural contexts; the functionalist turn (skopos theory) prioritizing utility and norms; and Nord’s concept of translator loyalty as a moral constraint balancing function and fidelity.

Two linked projects funded by the UK Research Councils’ Connected Communities programme were conducted collaboratively with researchers from the UK Department for Communities and Local Government (DCLG). The main project examined the impact of academically authored policy briefings and broader practices of research production and use by civil servants. Team: academics (including some briefing authors) and a DCLG social researcher. Methods: interpretive, ethnographic approach; semi-structured interviews with 11 civil servants (GSR analysts and policy staff) and all 11 briefing authors; three separate weeks of participant observation and interviewing within DCLG; four joint workshops with civil servants and academics. All interviews were recorded and transcribed; detailed field notes were kept for workshops and observations. A follow-on action research project by Connelly and Vanderhoven brokered connections between DCLG policy teams and 25 Connected Communities academics, resulting in four recorded face-to-face meetings; relevant on-the-record email communications were also analyzed. Ethics: focus on confidentiality of individuals and sensitive policy processes; informed consent for interviews and emails; collective agreement for meetings/observation; transcripts shared only among academics; University of Sheffield ethics approval obtained. Analysis: initial inductive thematic coding of transcripts, field notes, and emails to elucidate conceptualizations and practices of research use and constraints. Subsequently, a directed content analysis reframed the data using Translation Studies concepts (equivalence, function, loyalty) as core themes, with inductively developed sub-codes (e.g., handling academic texts, judging research quality). Representative but confidentiality-preserving quotations were selected to illustrate points; observational data were presented sparingly due to sensitivity.

• Equivalence: Policymakers often expect translation to preserve research richness and nuance without loss. Translation Studies argues full equivalence is impossible; some change and loss is inevitable and meaning is context-dependent. For non-literary research texts, equivalent effect entails comprehension of core ideas and warranted trust; accuracy must be balanced with target audience norms. Simple quantitative indicators may transfer readily (e.g., counts of neighborhood planning projects), but their interpretation is audience-specific (success metric for policy vs analytic input for academia). A priori definitions of equivalence are unworkable; decisions about what to preserve must be contextual. - Functional translation (skopos): Functionalist theory prioritizes the end user’s purposes. In UK central government, Government Social Researchers (GSRs) serve as internal brokers/translators with creative agency, aligning research with policy needs, synthesizing, and reframing content. A broadly shared skopos existed: informing policy with relevant evidence. Yet utility requires specificity; effective translation needs detailed briefs, which are often absent or hard to formulate cross-culturally. Interaction helps: starting meetings with policy teams articulating current concerns improved academics’ ability to translate their knowledge in real time; academics often struggle to align language/register and to distill complex arguments into concise, policy-relevant messages. - Balancing substance, accessibility, and relevance: Ideal translations combine demonstrable evidential/theoretical grounding, digestible presentation, and clear implications for action. Translations need not be exhaustive; a succinct “tip of the iceberg” can be valid if underpinned by robust foundations that can be accessed when needed. - Loyalty and trust: Functionalism risks excessive tailoring to user needs. Nord’s loyalty reframes quality as a moral responsibility to both authors and users, constraining translator freedom and anchoring translations in the author’s intentions and legitimate interests. Trust is foundational: GSRs assess trust via academic credentials, methodological transparency, and perceived neutrality; academics weigh risks of misrepresentation and political use. Relationships, sensitivity to each other’s contexts/risks, and a sense of shared endeavour foster mutual loyalty. - Practical implications observed: GSRs’ embedded broker role is pivotal; interaction structures matter (e.g., sequencing meetings to elicit policy needs first); academics benefit from briefs and from learning policy language; not all academics are well-placed to do translational work; investment in relationships and cross-boundary understanding is essential.

The findings address the central question of how to make research more useful to policy by reframing translation as an inherently transformative yet morally constrained practice. Equivalence highlights what is carried across and why research use is meaningful, but cannot be prescribed independent of context. Functionalism emphasizes adaptation to user needs, explaining why interaction and detailed contextual knowledge are crucial for utility; it also clarifies why non-interactive push/pull modes often falter. Introducing loyalty reconciles the tension: good translations are those that are sufficiently equivalent to preserve the author’s intentions and evidential warrants while being functional for policy. This moral, interpersonal framing explains the consistent importance of relationships, trust, and mutual respect reported in the research use literature and observed empirically (e.g., GSRs’ gatekeeping, academics’ concerns about misuse). The analysis offers a middle path between linear transfer and radical transformation views, foregrounding the work translators (often GSRs) do with substantive content, and making visible translation dynamics across both interactive and non-interactive pathways.

The paper argues that the translation metaphor should be retained and deepened using Translation Studies. A dilemma emerges between equivalence (what endures) and function (utility for users); both alone are inadequate. Nord’s loyalty offers a resolution by morally binding translators to both source authors and users, yielding translations that are functional while respecting original intentions and evidential bases. Practically, improving research translation requires: mutual, detailed understanding of needs, institutional contexts, and risks; shared fluency in each other’s languages; careful micro-organization of interactions (e.g., start with policy needs); accessible briefs outlining pressing policy questions; and valuing/recognizing the specialized translational work within government (e.g., GSRs). Investment should create more opportunities for authors, translators, and users to meet, build trust, and cultivate mutual loyalty around a shared endeavour to inform better policy. These insights are contextually grounded yet broadly applicable as middle-range concepts (equivalence, function, loyalty) to study and guide research translation across settings.

• Case specificity: Empirical work focused on one division of a UK central government ministry; findings may not capture variations across departments, policy fields, or countries. - Sensitivity and confidentiality: Data concerning ongoing policy initiatives limited the presentation of observational details and quotations, potentially constraining transparency. - Non-generalizability of translation moments: Translation is situated practice; there is no general theory for every instance, which limits prescriptive precision. - Resource and structural constraints: Limited academic capacity to engage interactively, and challenges in producing clear briefs, affect the feasibility of proposed improvements. - Qualitative design: No statistical outcomes; insights are interpretive, reliant on interviews, observations, and directed content analysis.