Temporal and spatial trends in study designs used to understand what works in conservation
Author(s)
Type
Journal Article
Abstract
1. Robust study designs are essential for estimating the effects of conservation actions, but a range of evaluation designs are used that differ markedly in their robustness to bias. However, it is unclear whether robust designs are becoming more frequently used or how usage varies across subject areas and geographies.
2. We analysed 7356 English and 1183 non-English language primary studies that tested the effects of conservation actions from the Conservation Evidence database. We classified each study's methodological design to one of five broad types (After, Before–After [BA], Control–Impact [CI], Before–After–Control–Impact [BACI] and randomised experiments) and fitted a Bayesian multinomial logistic regression, with study-level random effects, to test how design usage varies by year, national income, biodiversity intactness, language and subject area.
3. The evidence base has shifted gradually from dominance by simple After designs (1970–2023: 74% to 33%) towards greater use of CI (15% to 39%) and more robust designs: randomised experiments (0% to 15%) and BACI (1% to 6%). Design use was uneven across subjects, with randomised experiments more common in habitat and soil-focused topics (e.g. farmland), and After or BA designs more common in animal-focused topics (e.g. birds and reptiles). Studies in non-English languages and lower-income countries were more likely to use After and BA designs and less likely to use randomised experiments and CI designs.
4. Geographical coverage of robust designs was limited: randomised experiments and BACI studies occurred in few countries, and many biodiverse and environmentally degraded countries had few or no evaluations using any design. Despite modest improvements over time, several structural and contextual constraints (linked to subject, language and resources) continue to shape where rigorous evaluation is feasible.
5. Synthesis and applications: Conservation is shifting towards using more robust designs to evaluate actions, but these remain rare. We suggest embedding study design choice more strategically within adaptive management and evidence-based conservation practice, encouraging cost-efficient, context-dependent upgrades to more robust designs. This will require targeted capacity building, researcher–practitioner partnerships and longer-term funding—especially in biodiverse, lower-income countries with few studies of any design, let alone more robust ones.
2. We analysed 7356 English and 1183 non-English language primary studies that tested the effects of conservation actions from the Conservation Evidence database. We classified each study's methodological design to one of five broad types (After, Before–After [BA], Control–Impact [CI], Before–After–Control–Impact [BACI] and randomised experiments) and fitted a Bayesian multinomial logistic regression, with study-level random effects, to test how design usage varies by year, national income, biodiversity intactness, language and subject area.
3. The evidence base has shifted gradually from dominance by simple After designs (1970–2023: 74% to 33%) towards greater use of CI (15% to 39%) and more robust designs: randomised experiments (0% to 15%) and BACI (1% to 6%). Design use was uneven across subjects, with randomised experiments more common in habitat and soil-focused topics (e.g. farmland), and After or BA designs more common in animal-focused topics (e.g. birds and reptiles). Studies in non-English languages and lower-income countries were more likely to use After and BA designs and less likely to use randomised experiments and CI designs.
4. Geographical coverage of robust designs was limited: randomised experiments and BACI studies occurred in few countries, and many biodiverse and environmentally degraded countries had few or no evaluations using any design. Despite modest improvements over time, several structural and contextual constraints (linked to subject, language and resources) continue to shape where rigorous evaluation is feasible.
5. Synthesis and applications: Conservation is shifting towards using more robust designs to evaluate actions, but these remain rare. We suggest embedding study design choice more strategically within adaptive management and evidence-based conservation practice, encouraging cost-efficient, context-dependent upgrades to more robust designs. This will require targeted capacity building, researcher–practitioner partnerships and longer-term funding—especially in biodiverse, lower-income countries with few studies of any design, let alone more robust ones.
Date Issued
2026-07-01
Date Acceptance
2026-05-24
Citation
Journal of Applied Ecology, 2026, 63 (7)
ISSN
0021-8901
Publisher
Wiley
Journal / Book Title
Journal of Applied Ecology
Volume
63
Issue
7
Copyright Statement
© 2026 The Author(s). Journal of Applied Ecology published by John Wiley & Sons Ltd on behalf of British Ecological Society. This is an open access article under the terms of the Creative Commons Attribution License, which permits use, distribution and reproduction in any medium, provided the original work is properly cited.
License URL
Publication Status
Published
Article Number
e70478
Date Publish Online
2026-07-09