Abstract
Ripple Effects Mapping (REM) has emerged as a distinctive approach for capturing the intended and unintended impacts of programs and initiatives. Despite this growing interest, the literature remains fragmented and underdeveloped. We identified three key gaps: the absence of a comprehensive review of REM applications, its application within a narrow set of disciplines – primarily health and wellbeing – and wide variation in methodological rigour and reporting. These gaps limit REM’s development and perpetuate its perception as solely an evaluation tool. To address this, we conducted a systematic review of fifty-two peer-reviewed studies to examine how REM has been applied, adapted and reported. A Scopus database search was conducted in December 2024 and updated in September 2025. Findings confirm REM’s flexibility and capacity to capture both intended and unintended outcomes, amplify diverse voices, and support reflective practice. However, challenges such as participant availability, selection bias, and inconsistent reporting persist. Building on these insights, we propose four pathways for future REM development: 1) continuing REM as an evaluation method, 2) advancing it as a co-production approach, 3) conceptualising rippling as a research framework, and 4) integrating REM with complementary methodologies. Practical recommendations for transparent reporting – including session details, participant numbers, outputs analysed, and analytical approaches – are provided to enhance rigour and replicability of REM. These directions position REM as a versatile methodology capable of strengthening evaluation and research practices across diverse contexts, including those beyond evaluation.
Keywords
1. Introduction
Ripple Effects Mapping (REM) has traditionally been viewed as a collaborative qualitative evaluation method which captures group knowledge and maps the intended and unintended impacts of a project, or initiative. Chazdon et al. (2017) described it as a ‘group participatory evaluation method that engages program and community stakeholders to retrospectively and visually map the chain of effects resulting from a program or complex collaboration’ (pg xi). Originating in the USA, the Community Capitals Framework evaluation (Emery, 2008) and the Horizons Program evaluation, conducted in 2008 by Washington State University, University of Idaho, and North Dakota State, led to the development of this method. REM has four core elements: 1) Appreciative inquiry, 2) A participatory approach, 3) Interactive group interviewing and reflection and 4) Radiant Thinking (mind mapping) (Chazdon et al., 2017).
Appreciative inquiry involves short peer interviews to focus discussions and ‘warm up’ (Chazdon et al., 2017; Harris et al., 2024) participants. The participatory approach treats participants as co-creators, and active participants rather than participants as ‘recipients of evaluation information’ (Chazdon et al., 2017, p. 8). Interactive group interviewing and reflection encourages collective reflection on past events, while Radiant Thinking translates discussions and reflections into visual maps (mind maps) (Chazdon et al., 2017). Session facilitators assist in creating a physical or digital map and encourage participants to reflect and ‘challenge or ask questions about the causal chains reported by others’ (Chazdon et al., 2017, p. 16).
We were drawn to REM while seeking a method to examine the activities of community groups and the co-benefits and impacts they generate. Our aim was to identify an approach that prioritises participant agency, fosters reciprocity (Maiter et al., 2008), and empowers communities to play an active role in research. Although REM traditionally aligns as an evaluation tool, its participatory nature aligns closely with principles of participatory research, which emphasise shared power, engagement, and the co-creation of actional knowledge for social change (Ospina et al., 2021). Unlike conventional research, participatory approaches shift the ‘location of power’ (Cornwall & Jewkes, 1995, p. 1667) by placing participants at the centre and transferring some responsibility for the research process to them (Brown, 2022). Participatory research spans a broad spectrum (Leeds Social Sciences Institute, 2023), often described as a conceptual umbrella (Ospina et al., 2021), and includes co-produced methods that are iterative and collaborative (Mauser et al., 2013; Norström et al., 2020). These methods enable research to be co-designed, co-produced, and co-disseminated (Mauser et al., 2013) by participants.
In exploring REM’s potential beyond evaluation, we identified a significant gap: despite its growing popularity, the literature remains fragmented with limited conceptual development, with most studies focusing exclusively on evaluation. The gap manifests in three key limitations. First, no comprehensive review exists to help researchers understand current applications of REM. Second, its use is largely confined to a few disciplines. Third, existing applications vary widely – from rigorous evaluation methodologies to superficial use of ‘rippling’ terminology – without clear conceptual or methodological guidance, and reporting of REM procedures is inconsistent. Together these gaps constrain REM’s evolution as a research tool and framework, and reinforce its perception as merely an evaluation tool.
To address these limitations, this paper systematically reviews REM applications, synthesises their benefits and constraints, and proposes four distinct future research directions: 1) continuing REM as an evaluation tool, 2) developing REM as a co-production method, 3) conceptualising ‘rippling’ as a research framework, and 4) integrating REM with complementary approaches. These directions collectively advance REM from its traditional role in evaluation toward a broader participatory research framework. In doing so, we strengthen REM’s methodological foundations for diverse research applications while offering practical guidance for those employing REM in evaluation contexts.
2. Methods
This review follows the PRISMA 2020 guidelines (Page et al., 2021). A Scopus database search was conducted on the 30th December 2024, and updated on 5th September 2025. Focusing on peer-reviewed articles the following search string was used:
TITLE-ABS-KEY (“*ripple effect mapping*” OR “*ripple effects mapping*”) AND (EXCLUDE ( DOCTYPE, “ch”) OR EXCLUDE (DOCTYPE, “cp”) OR EXCLUDE (DOCTYPE, “ed”)).
A total of 52 articles were included in the systematic review (Figure 1). Since the search yielded a small number of results and did not require any decisions to be made about the inclusion or exclusion of papers (after the exclusion of book chapters, conference papers and editorial pieces), it was decided that the lead author would conduct the review. PRISMA chart
2.1. Data Extraction
Data extraction followed a two-step approach. First, key details were recorded: publication information, topic/discipline, aims/objectives/research questions, location where research was conducted, methods used alongside REM, Appreciative Inquiry details, software used in map creation, analysis approach, session length and number, number of participants and limitations. There was also a free text box for researcher notes. Second, upon reviewing findings with the second author, additional data were extracted: online/in-person facilitation, if a map was produced/displayed in the session, map or extract of map inclusion in paper, paper type (empirical or commentary/methods), whether REM modifications have been used and whether the research uses the Community Capitals Framework.
Details of the included articles, along with some of the information extracts are provided in Supplementary Table 1.
3. Result
3.1. Number of Publications
REM publications began in 2012 (Kollock et al., 2012), and have steadily increased, peaking in 2024 with ten papers (Figure 2). By September 2025, eight papers have been published. Number of REM publications split by year of publication
3.2. Authors
Across the 52 articles, 156 authors have contributed. Most authored one paper, but five authors (Chazdon, S., Higgins, L., Nobles, J., Rosas, L.G., Hall, J,) authored three of more.
3.3. Study Location and Language
59% of studies have been conducted in the USA, 19% in the UK, the remainder in Australia, Belgium, Colombia, Ireland, Netherlands, Sweden and Switzerland (Figure 2). Up until 2021, all publications bar Naccarella et al. (2019) were from the USA.
Most REM sessions are assumed to have been conducted in English, as Scopus requirements mean all journals must have English language abstracts and titles. However three papers report Spanish-language sessions: Sadeghzadeh et al. (2022) who conducted one of six REM sessions in Spanish; Rodriguez Espinosa, Martinez Mulet et al. (2023), who conducted REM with Spanish-speaking community health workers and promotoras, and Page-Reeves et al. (2024) who conducted participant research in Spanish.
3.4. Journals
REM studies appeared in 35 different journals (Figure 3). The ‘Journal of Extension’ published the most (n=8), followed by ‘BMC Public Health’ (n=4), ‘Community Development’, and ‘Health Research Policy and System’ (each with n=3) and ‘BMC Medical Research Methodology’, ‘Health Promotion Practice’ and ‘Palliative Care and Social Practice’ each have two publications. All other journals have one REM publication.
As several papers originate from the Journal of Extension, we have included a summary of our understanding of ‘Extension’ to aid readers who, like the authors, may be unfamiliar with ‘Extension’ terminology (Figure 4). Proportion of publications split by country where the REM study was conducted. Unlabelled pie-chart segments represent 2% Extension studies definition.

3.5. Methodology Papers
Eleven papers focus on REM methodology-focused discussions and commentary (n=6) or REM protocols (n=5). Study protocols describe the research plan and methodology, and are common in the fields of health research, with some funding bodies encouraging the publication of protocols (e.g. NHS Health Research Authority, (Al Shakarchi, 2022)). The five protocol papers all appear in medical or methods journals.
The six methodological-focused commentary and discussion papers focus on: 1) providing a summary of REM, the background, benefits and limitations (Kollock et al., 2012) 2) discussing the visual options of REM and what to do with the data (Rome et al., 2020), 3) the different approaches to REM including web mapping, in-depth rippling and theming and rippling (Emery et al., 2015), 4) extending the foundations of REM and integrating it with grounded theory (Peterson & Skolits, 2019), 5) integrating a realist evaluation with REM (Harris et al., 2024) and 6) qualitative reflections of seven REM applications and 10 recommendations for conducting REM (Creaser et al., 2024).
Most methodology papers do not include empirical findings. However, there are three exceptions. Peterson and Skolits (2019) and Kollock et al. (2012) presented REM maps from fieldwork, but they do not discuss these, the maps are used to further their discussions, and Harris et al. (2024), illustrated a realistic REM approach with a case study.
3.6. Papers from the Same Project
Four papers reported on the same project. De Pooter et al. (2025) and Luna Pinzon et al. (2025), reported on the LIKE Programme, and Darger et al. (2017) and Darger (2014) reported on the University of Minnesota’s Extension model. As each of these papers offered distinct insights into REM applications, the decision was taken to keep them in the review.
The decision was taken to keep these four papers in the review as they provide different empirical insights, and reported on different aspects of the REM application. De Pooter et al. (2025) reported on REM applications used within the project, while Luna Pinzon et al. (2025) reported on a different application of REM, when it was used to reflect on the LIKE project and outcomes, with researchers who worked on the project. Darger (2014) focused on REM and its application, while Darger et al. (2017), discussed the wider application and findings of REM as part of a broader programme evaluation, including a greater number of REM sessions than Darger (2014).
3.7. Empirical Papers
This section focuses on the remaining thirty-one papers and excludes the eleven methodological papers.
3.7.1. Themes
All the papers used REM as either a tool for evaluating, and/or identifying the impacts, of a program, project or initiative. Most related to the topics of health and fitness by either a) directly by assessing programmes and initiatives which aim to improve health, fitness and wellbeing which can include topics of mental health, social isolation and bereavement (for example: de Pooter et al., 2025; Eneslätt et al., 2021; Ernst et al., 2024; Gold et al., 2020; Huiberts et al., 2024; Nobles, Fox, et al., 2022; Olfert et al., 2019; Rafie et al., 2024; Rubio et al., 2022; Welborn et al., 2016; Wilkie et al., 2025; Wilson & Bates, 2025) b) indirectly by assessing programmes created for practitioners involved in improving health and/or fitness (for example Bailes et al., 2023; Crowther et al., 2024; Dowlen et al., 2024; Keaver et al., 2023) or c) in the assessment of public health planning and partnership activities (for example:Haggerty et al., 2018).
Nine papers fall outside health, fitness and wellbeing topics. Moore and Hatcher (2019) focused on journalism and used REM to evaluate a storytelling project designed to ‘disrupt traditional news routines and foster collaboration between media professionals and citizen storytellers’ (pg 753). Daniels et al. (2016) evaluated the impact of a University Extension Garden Team, Vettern and Flage (2018), evaluated the changes in schools and communities after introducing a Youth and Families mentoring program. Darger (2014) examined the impacts of a community-driven Business Retention and Expansion program, Bhattacharyya et al. (2017) assessed a community tourism development program, and Hallinan et al. (2025) used REM to examine ‘long-term process of community building, self-advocacy, network and alliance building’, (pg. 2) focusing on a Deafblind Hub. McGladrey et al. (2025) used REM to understand practitioners’ experiences of using a technological decision tool in a psychiatric facility. Karnik et al. (2025) evaluated a Public Health Leadership programme which sought to provide participants with ‘skills and perspectives to address evolving public health challenges effectively’ (pg 592). Even though this is public health-related, it has not been grouped in the earlier health, fitness and wellbeing grouping as the focus is on the evaluation of an educational leadership program. Similarly, Luna Pinzon et al. (2025)’s REM application is linked to an LIKE obesity prevention program by de Pooter et al. (2025) but Luna Pinzon et al. (2025) focused on the effect of the whole LIKE project with and on consortium members, rather than LIKE participants.
3.7.2. Practical Applications and Facilitation Details
This section acts as a summary, and more details are found in Supplementary Table 1.
3.7.2.1. Number of Participants per REM Session, and Number of REM Sessions
Chazdon et al. (2017) recommend that the ideal number of participants per REM session is 12 to 20. However, this review has found sessions ranged from 2 (Bloom, 2021) and 30 people (Moore & Hatcher, 2019). Most studies report on findings from less than five REM sessions. The maximum reported sessions in one paper is fourteen sessions by Gold et al. (2020).
3.7.2.2. Length of REM Session
Approximately two-thirds of papers report the duration of the REM sessions. Most REM sessions take between 90 minutes and 120 minutes to complete. The shortest being 50 minutes (Karnik et al., 2025), and the longest is 4 hours (Spain et al., 2021).
3.7.2.3. Online or In-Person Facilitation
While most REM sessions are conducted in-person, ten papers reported online or hybrid formats, often prompted by the COVID-19 pandemic. Greene et al. (2022) adapted the REM approach using online surveys (further details in Section 3.8). Dowlen et al. (2024) conducted a three-hour Zoom session with ten dance artists, noting that although participants were familiar with the Zoom platform, technical issues still arose. Similarly, Keaver et al. (2023) facilitated a one-hour Zoom session with undergraduate students, and Karnik et al. (2025) used Zoom to engage professionals across six US states. Rafie et al. (2024) also used Zoom for a research team and project stakeholders session, while Tuepker et al. (2024) and McGladrey et al. (2025) conducted online REM, though the platforms were not specified.
De Pooter et al. (2025) opted for using Microsoft Teams, citing pandemic restrictions as the reason for moving online. Huiberts et al. (2024) conducted eight REM sessions online, initially due to the pandemic but continued with this format post-pandemic for consistency. They identified online facilitation as both a strength – allowing greater flexibility for participants – and a limitation as it can be difficult to stimulate discussions online. Luna Pinzon et al., 2025. adopted for a hybrid format for their REM workshop, but the platform was not disclosed.
In contrast, some authors chose not to transition REM online. Eneslätt et al. (2021), for example, avoided online REM sessions due to participants’ limited computer literacy, and instead opted for a modified REM approach, detailed in Section 3.8.
3.7.3. Appreciative Inquiry Process
Appreciative Inquiry (AI) Questions
Among the = papers that did not mention appreciative inquiry, several included questions similar to those in Table 1, though these were often posed to the group rather than in paired interviews. For instance, Darger (2014) asked participants to respond aloud to the questions such as ‘what outcomes have happened as a result of the Gteam formation?’ with responses recorded on post-it notes by the facilitator. Darger et al. (2017) presented similarly worded questions in a table, but did not describe the use of appreciative inquiry or any pairing of participants.
Four papers report activities conducted before the appreciative inquiry process. Olfert et al. (2018) include an ice-breaker activity while Spain et al. (2021) introduced a framework and encouraged participants to think about the program in relation to the framework prior to appreciative inquiry. Ernst et al. (2024) incorporated a shared meal and opportunity to review research notes, and Crowther et al. (2024) embedded REM within quarterly workshops, integrating pre and post-session discussions around the REM activity.
3.7.4. Radiant Thinking – Creating Maps
Radiant Thinking, another core element of REM (Chazdon et al., 2017), involves the co-creation of a map during an interactive group session. Most papers in this review reported that participants, supported by a facilitator, created such maps. However, ten papers deviate from this. Five of these, Eneslätt et al. (2021); Greene et al. (2022); Hallinan et al. (2025); Naccarella et al. (2019); Page-Reeves et al. (2024), used adapted REM methodologies, which are discussed further in Section 3.8. Four papers, Karnik et al. (2025); Luna Pinzon et al. (2025); Moore and Hatcher (2019); Rodriguez Espinosa, Martinez Mulet et al. (2023), indicated that a map was produced, or that XMind software was used, suggesting a map creation. However, it remains unclear whether these maps were generated in real-time or retrospectively from audio recordings, and whether participants had visibility of the maps during the session. One paper, Haggerty et al. (2018), provided no information on map production, however REM data collection is not the central focus of the paper.
XMind is the most commonly used software for creating REM maps, either in real-time during the sessions or when converting physical maps into digital maps. Other less frequently used tools include Vensim used by Nobles, Fox, et al. (2022), ATLAS.ti for mac used by Wilkie et al. (2022, 2025), Mindjet Mind Manager used by Washburn et al. (2020), Adobe Illustrator and Indesign used by Sadeghzadeh et al. (2022), Miro Board used by Crowther et al. (2024); Huiberts et al. (2024); Wilson and Bates (2025), Google Jam board used by Dowlen et al. (2024), Mind Meister used by Rafie et al. (2024) and Microsoft Word and PowerPoint, used de Pooter et al. (2025).
3.7.5. Analytical Approach
In terms of analysis, most studies follow a similar process. Initially, facilitator notes are compiled and used to update the REM map with any additional information not captured during the session. Researchers then review the map and, where necessary, conduct follow-up interviews to fill gaps or enrich the data. Qualitative analysis is subsequently undertaken, typically using thematic approaches or framework-based coding. In cases where XMind is used, maps are often exported into spreadsheet format to facilitate coding. A final step, reported inconsistently across studies, involves sharing the map with participants to gather feedback and make revisions (further details in section 3.7.8).
Several studies incorporate analytical frameworks, with the Community Capitals Framework (CCF) (Flora et al., 2016) being the most prevalent. CCF offers a systems-based approach to evaluating community change, comprising seven interrelated capitals: built, financial, political, social, human, cultural, and natural (Flora et al., 2016). Emery and Flora (2006) argued that growth in one capital can catalyse changes in others, enabling communities to systematically assess the impacts of interventions and identify indicators across all seven capitals. REM was developed from CCF to visualise how ripples in one capital influence others (Chazdon et al., 2017).
Given this foundational relationship, it is unsurprising that thirteen REM studies integrated CCF into either the session design or analytical approach. For example, Vettern and Flage (2018) and Bhattacharyya et al. (2017) coded REM map content according to CCF themes. Sadeghzadeh et al. (2022) and Welborn et al. (2016) provided participants with CCF handouts, allowing them to code their own contributions. Spain et al. (2021) introduced the framework during sessions, displaying it on the wall to guide participant thinking. In a modified REM approach, Greene et al. (2022) used survey responses coded to the CCF and structured the resulting map around the seven capitals.
3.7.6. Advantages of REM
Chazdon et al. (2017) identified several advantages of REM; it is cost-effective, easy to implement, engaging, iterative, conclusive, and adaptable to diverse contexts. While previous sections have shown the range of REM applications, this section also examines the benefits reported across the studies.
A key strength is REM’s ability to help participants reflect on their activities and conceptualise the impacts of programmes/initiatives. Welborn et al. (2016) described REM as community-centred, and Bhattacharyya et al. (2017), showed how it helped contextualise tourism efforts. Daniels et al. (2016) noted that REM enabled participants to see both their individual contributions and the broader program impact. Dowlen et al. (2024) highlighted how REM gave dance artists, who typically work alone, a rare opportunity to reflect collectively. Similar benefits are reported by Ernst et al. (2024); Huiberts et al. (2024); Sadeghzadeh et al. (2022); Washburn et al. (2020) who added that REM helped participants recognise the project’s significance. These findings support Chazdon et al. (2017) claim that REM reveals unexpected outcomes, which may be missed by traditional evaluations.
REM also amplifies diverse voices, including those who may feel unheard (Rubio et al., 2022; Wilkie et al., 2022), and participants often express appreciation for the opportunity to reflect (Ernst et al., 2024). Its capacity to map both intended and unintended consequences is well documented. Bakelants et al. (2024) emphasised REM’s value in identifying where ripples may lead to “dead ends,” making impact tracking more complex. Their example of a grief infographic illustrates how REM can capture the early stages of long-term effects.
The map produced through REM is not only useful for researchers and evaluators but also for broader communication. Darger et al. (2017) reported that communities used REM outputs to showcase programme impact and even submitted them for international awards, which they won. These successes generated further enthusiasm and engagement among community leaders. The energising nature of REM is echoed by Dowlen et al. (2024); Ernst et al. (2024); Kollock et al. (2012) and Taylor et al. (2020), who described how REM sessions sparked new ideas, reinforced achievements, and deepened motivation. This suggests that a REM session can create further ripples.
Several studies also note that REM sessions themselves can generate ripples. Darger et al. (2017) found that communities initiated their own REM sessions years later. Dowlen et al. (2024) reported that dance artists created maps to support funding applications, while Taylor et al. (2020) described participants requesting newsletters to maintain stakeholder engagement.
REM’s adaptability is another advantage. It can be combined with other methods, such as semi-structured interviews (Nobles, Fox, et al., 2022), or used longitudinally, as in Crowther et al. (2024)’s quarterly workshops. Welborn et al. (2016) suggested REM maps can evolve over time, and both Huiberts et al. (2024) and Nobles, Fox, et al. (2022) highlighted its utility in capturing impacts systematically across project phases.
Finally, REM is a relatively low-resource method. Sadeghzadeh et al. (2022) noted it is less burdensome than surveys or interviews, and Rodriguez Espinosa, King, et al. (2023) emphasised its inclusivity, enabling diverse stakeholder input without the time demands of other approaches.
Despite these strengths, REM has limitations. It offers only a snapshot of a specific context, and as Moore and Hatcher (2019) caution, the same approach may yield different outcomes in different settings.
3.7.7. Challenges/Limitations of REM
In reviewing the literature on REM, it was noted that several papers did not report any specific challenges or limitations. These generally fell into two categories: papers published in the Journal of Extension, likely constrained by its 5,000-word limit, and papers that discussed broader project limitations without addressing REM directly (for example: Haggerty et al., 2018; Nobles, Fox, et al., 2022; Page-Reeves et al., 2024; Rodriguez Espinosa, Martinez Mulet et al., 2023).
Among those that did report challenges, three recurring themes emerged: participant availability, selection bias, and sample size. Additional limitations were also identified but did not fall into clear thematic categories.
3.7.7.1. Availability of Participants
Participant availability was frequently cited. Bailes et al. (2023) noted issues related to travel and scheduling, while Sadeghzadeh et al. (2022) reported that key stakeholders were unable to attend. Dowlen et al. (2024) and Rodriguez Espinosa, King, et al. (2023) also faced attendance challenges. To address this, Dowlen et al. (2024) and Huiberts et al. (2024) held multiple sessions, though this raised methodological questions about whether to merge or separate the resulting maps. Welborn et al. (2016) attributed low attendance to the timing of sessions, suggesting REM should be conducted within 6–12 months of an initiative’s conclusion to maintain engagement.
3.7.7.2. Who Participates? Selection and Non-Response Bias
Selection and non-response bias were also common concerns. REM participants are often those already involved or interested in the initiative, resulting in a self-selected sample (Dowlen et al., 2024). Those who dropped out or were less engaged are frequently excluded, as noted by Huiberts et al. (2024); Keaver et al. (2023); Olfert et al. (2019); Rodriguez Espinosa, Martinez Mulet et al. (2023); Rubio et al. (2022); Spain et al. (2021); Wilkie et al. (2025); and Workman and Browder (2020).
3.7.7.3. Sample Size
Small sample sizes were highlighted by Bloom (2021); Ernst et al. (2024); and Gold et al. (2020). While typical in qualitative research, this limitation has been widely debated (For example: Bryman, 2008; Sandelowski, 1995), though a full discussion is beyond the scope of this paper.
3.7.7.4. Other Challenges/Limitations
Other reported challenges include potential bias from appreciative inquiry questions (Workman & Browder, 2020), technical difficulties with online REM sessions (Dowlen et al., 2024), and limited transferability of findings to other contexts (Dowlen et al., 2024; Karnik et al., 2025; Workman & Browder, 2020). Dowlen et al. (2024) also noted that REM maps may vary depending on the stage of the project. As previously discussed, REM captures a snapshot in time (Sadeghzadeh et al., 2022) and place (Moore & Hatcher, 2019), and describes influences rather than causality (Ernst et al., 2024; Harris et al., 2024).
Some limitations were project-specific. Bailes et al. (2023) raised concerns about confirmation bias when researchers involved in programme development also participated in REM sessions. Bloom (2021) questioned the scalability of REM in volunteer-based programmes, where volunteers may not feel equipped to facilitate. Welborn et al. (2016) reported difficulties applying the CCF, with participants understanding the themes but struggling to articulate their impacts. Karnik et al. (2025) and Washburn et al. (2020) noted that REM tends to highlight positive aspects, with participants less likely to share negative experiences.
3.7.8. Reporting of REM Maps Back to Community Groups
Chazdon et al. (2017) note that follow-up with participants may be necessary after REM sessions to ensure the map accurately reflects individual contributions. Several studies report taking additional steps to share the map with participants or group leaders for verification. For example, Bloom (2021); Dowlen et al. (2024); Greene et al. (2022); Rodriguez Espinosa, Martinez Mulet et al. (2023); Taylor et al. (2020); Washburn et al. (2020); and Workman and Browder (2020) shared maps with participants, while Darger (2014); Darger et al. (2017); and Eneslätt et al. (2021) shared them with group leaders.
Some studies formalised this process. Nobles, Fox, et al. (2022) referencing Nobles, Wheeler, et al. (2022), incorporated follow-up workshops to allow participants to reflect on and review REM outputs. Ernst et al. (2024) conducted a similar second session to gather further feedback. In Harris et al. (2024)’s Realist REM approach, discussing the final map is a key stage of the process.
Bhattacharyya et al. (2017); Rubio et al. (2022); and Sadeghzadeh et al. (2022) also reported sharing maps with participants, though it is unclear whether this was for verification or to disseminate the final output.
3.8. Adaptations of REM
Fourteen papers in this review describe or propose adaptations to the traditional REM approach. These modifications often respond to contextual constraints or aim to enhance REM’s flexibility.
3.8.1. Survey and Interview-Based Adaptations
Survey and interview-based adaptations emerged during the COVID-19 pandemic. Eneslätt et al. (2021) used interviews to explore ripples from the use of DoBra Cards in end-of-life discussions. Instead of a group session, they created a map from interview data and presented it to a broader project group. The authors note that a traditional REM would have been preferable, but pandemic restrictions and digital literacy barriers limited options. Greene et al. (2022) adapted REM to an online survey format. Participants were introduced to REM via a Microsoft Teams meeting, then completed a survey structured around Appreciative Inquiry and the Community Capitals Framework. The resulting data were used to create a map resembling a traditional REM output. Bakelants et al. (2024) supplemented REM sessions with interviews due to participant availability. Appreciative Inquiry questions were asked, and interviewees reviewed the map. Additional data from policy documents, field notes, and meeting minutes were used to identify further ripples.
3.8.2. Other Adaptations
Other adaptations include temporal mapping and iterative workshops. Nobles, Fox, et al. (2022) incorporated a timeline into the REM map and held follow-up workshops every 3–4 months for reflection and updates. Crowther et al. (2024) embedded REM into quarterly workshops, each lasting one hour, allowing participants to revise and expand maps over time. Two physical maps were later merged into a single digital version. Hallinan et al. (2025) created two maps during their REM session: one charting the development of the DeafBlind Hub on a timeline, and another using a traditional REM format. A third map was later developed to show broader impacts of capacity-building. Ernst et al. (2024) introduced reflective diaries, completed weekly for two months prior to the REM session. Thirty empathy-related ripples were displayed during the session, and participants added comments before engaging in two rounds of Appreciative Inquiry and map development. The diaries also informed the final map.
Some authors modified the Appreciative Inquiry process. Huiberts et al. (2024) used small groups instead of pairs, while Bailes et al. (2023) asked participants to write personal stories before sharing them during paired Appreciative Inquiry. Finally, Harris et al. (2024) proposed a Realist REM (RREM) approach, combining REM with realist evaluation to explore causation and understand the mechanisms behind observed impacts.
3.8.3. REM or REM Terminology
Use of REM terminology without full application was noted in several papers. Gold et al. (2020); Hallinan et al. (2025); Naccarella et al. (2019); and Page-Reeves et al. (2024) referred to REM but omitted core elements such as Appreciative Inquiry, participatory group reflection, and Radiant Thinking. Gold et al. (2020) described pre-defined branches used in a focus group, but it is unclear whether these were participant-led. Gold et al. (2020) cited Peters et al. (2016) for the study protocol, but again, this provides limited information on REM. Naccarella et al. (2019) conducted focus groups using ripple terminology but did not report map creation or participant reflection. Hallinan et al. (2025) also omit Appreciative Inquiry and do not present maps. Page-Reeves et al. (2024) created a map from evaluation notes and transcripts, suggesting an analytical rather than participatory approach. Further commentary on REM approaches is provided in the next section.
4. Discussion
This systematic review highlights that while REM is an emerging and increasingly popular method, its application remains relatively narrow, with most studies focused on evaluating health, wellbeing, and medical programmes. The reviewed literature falls into three broad categories. First, studies that applied all four core principles of REM—Appreciative Inquiry, participatory approach, interactive group reflection, and Radiant Thinking (Chazdon et al., 2017)—and produced maps in real time with participants. Second, studies that use modified REM approaches, such as survey-based methods or those incorporating a temporal dimension (For example: Dowlen et al., 2024; Greene et al., 2022). Third, studies that appeared to use ripple terminology and mind-mapping techniques but do not fully implement REM. This third group underscores the conceptual appeal of REM, particularly its use of ripple metaphors and visualisation, even when the full methodology is not applied.
Welborn et al. (2016) argued that REM’s strength lies in its adaptability. Building on this, our review proposes four potential pathways for future REM development: (1) continuing REM as an evaluation method, (2) developing and applying REM as a co-production method, (3) conceptualising ‘rippling’ as a framework to guide research activities, and (4) integrating REM with complementary approaches. We argue that pathways 2, 3 and 4 advance REM applications beyond traditional evaluation methodologies towards a more research oriented approach, with applications beyond evaluation.
4.1. REM as an Evaluation Method
In its purest form, REM is an evaluation data collection method that applies its four core elements: 1) Appreciative Inquiry, 2) Participatory Approach, 3) Interactive group interviewing and reflection, and 4) Radiant Thinking (mind mapping). Most studies in this review follow this approach. As discussed in Section 3.7.6 REM’s ability to identify and map both intended and unintended consequences is a key strength, offering rich insights into programme impacts.
4.2. REM as a Co-Production Method
REM can also be positioned within the co-production spectrum and the co-production umbrella (Verschuere et al., 2012), where research is collaboratively designed, conducted, and disseminated. Co-production is a well-established approach (For example: Bell & Pahl, 2018; Mauser et al., 2013; Norström et al., 2020; Parks et al., 1981; Turnhout et al., 2020; Verschuere et al., 2012; Voorberg et al., 2015) involving stakeholders throughout the research process. REM lends itself naturally to co-production, as participants co-create the ripple effects map and, in some cases, engage in iterative feedback and verification. Future REM applications could deepen this by involving participants in designing appreciative inquiry questions, selecting session formats, and even facilitating sessions themselves (For example: Keaver et al., 2023).
Dissemination is another area where co-production could be further developed. While most REM studies focus on academic outputs, Darger et al., 2017 reported that REM maps were used in award submissions, and Harris et al., 2024 included dissemination as a final phase in their Realist REM approach. Co-dissemination, where participants are involved in shaping outputs during the reflection and/or verification stages, or in publication outputs, could enhance the relevance and impact of REM.
While co-production may not suit all research agendas, its integration into REM offers a promising avenue for participatory and inclusive evaluation, particularly when researching community-based or stakeholder-driven contexts.
4.3. Rippling as a Framework
While REM is typically applied as a participatory method, this review identifies an alternative use; rippling as a framework. This approach shifts the focus from participant-led data collection to a broader evaluation lens, where ripple effects are mapped holistically by researchers or practitioners. Unlike REM as a method, this framework does not rely on Appreciative Inquiry or group interaction but retains the principle of Radiant Thinking.
Undertaken as a desk-based exercise, researchers or practitioners could create their own REM map, where the program or initiative is in the centre of the map, and they think about and map potential ripples radiating from this. Researchers or practitioners could also identify where they could obtain data on the ripples they have suggested. For example, in the early stages, it can help anticipate potential outcomes and guide data collection planning. Later, it could inform decisions about who to involve in evaluation REM sessions and where to gather additional evidence. At project completion, it can support reflection on the overall impact and understandings of a project, similar to the approach by Luna Pinzon et al. (2025).
This framework can also extend the utility of REM maps created through participatory sessions. As shown in studies by Bakelants et al. (2024); and Hallinan et al. (2025) post-session analysis may involve integrating additional data sources—such as interviews, surveys, or quantitative metrics—into the original map. Ernst et al. (2024) suggested this could help trace mechanisms and pathways of impact. While detailed data may not always be presented within the map itself, supplementary materials or references can support transparency and depth. In this approach, the participant map could be seen as the foundation map upon which researchers or practitioners build.
This approach maintains the native and interconnected nature of ripple effects, while allowing for an opportunity to incorporate additional evidence that may help evaluators map the impact of a programme or initiative, or researchers gain a further understanding of the participants, activities or group they are examining. The original REM map remains central, with participants contributing to its development and refinements. Researchers can overlay questions or prompts to guide further data collection, ensuring the map evolves as a meaningful evaluation or research tool.
4.4. Merging REM with Other Approaches
This review also highlights the evolving nature of REM through its integration with other methodologies, and beyond the traditional evaluation context. A notable example is Harris et al. (2024) Realist REM (RREM), which combined REM with realist evaluation to explore causality and deepen understanding of impact pathways. RREM introduces several key modifications: a pre-REM discovery workshop to clarify evaluation aims; prompts during REM sessions to explore causal mechanisms; development of a causal connections map; and a feedback session where participants critically reflect on both the REM and causal maps. This is followed by a testing phase using mixed methods to validate causal links, and a final workshop for stakeholder deliberation. Such hybrid approaches offer promising avenues for researchers seeking a more systemic understanding of programme impacts, or applying REM to non-evaluation settings. By merging REM’s participatory strengths with the analytical rigour of realist evaluation, RREM exemplifies how REM can evolve to meet complex evaluative needs.
5. Conclusion
This systematic review provides the first comprehensive synthesis of Ripple Effects Mapping (REM) applications, highlighting a distinct increase in REM publications since the publication of Chazdon et al. (2017) ‘A Field Guide to Ripple Effects Mapping’. Most applications have been conducted in the USA and are closely associated with the ‘Extension’ profession or medical, health and wellbeing fields. Despite this growth, reporting practices vary considerably: while some studies provided detailed appendices outlining all four stages of REM - Appreciative Inquiry, Participatory Approach, Interactive group interviewing and reflection, and Radiant Thinking, others offer minimal methodological information. This inconsistency raises questions about whether REM was fully applied or whether if only rippling terminology was used.
This review has reported on the advantages and disadvantages of applying REM, identified adaptions to the traditional REM approach, and proposes four pathways for future development 1) continuing REM as an evaluation method, 2) advancing REM as a co-production approach, 3) conceptualising rippling as a framework to guide research activities and 4) integrating REM with other approaches. These directions position REM to evolve beyond its established role in evaluation towards broader research applications.
We acknowledge some limitations to this systematic review. The review was limited to SCOPUS with a criterion of only including English language journals, peer-reviewed journal publications, excluding book chapters and conference proceedings. We also chose to exclude grey literature, such as evaluation reports, organisational documents and unpublished studies, because these sources often lack standardised peer-review processes and may vary widely in methodological rigour, making systematic comparison challenging. However, given the co-production nature of REM and its frequent use in community-based projects, it is likely that additional examples exist in grey literature. Excluding these sources may mean that our review underrepresents practical applications of REM outside academic contexts, and future reviews could address this gap by incorporating grey literature to provide a more comprehensive picture.
5.1. Recommendations on Practicalities and Reporting of REM
Given the wide range of reporting on REM applications, we recommend the following as core information that should be included when reporting REM applications (either in the manuscript or in appendix documents): the number of REM sessions held, the number of participants attending, the outputs being examined and included in analysis (for example is it just the REM map? Or are transcripts being analysed), the analytical approach, especially of the map, details of the Appreciative Inquiry process, whether participants were provided with an opportunity to comment on the final map, and the level of co-production. We encourage readers who are looking to apply REM to refer to Nobles, Wheeler, et al. (2022) and Creaser et al. (2024) for practical considerations on applying REM and Washburn et al. (2020) for recommendations on notetaking during REM sessions. We also encourage Extension professionals and educators to explain the terminology they use to enable those outside of the discipline to gain a further understanding of the work they conduct. Nevertheless, we draw the conclusion that REM is a strong method for capturing a wider range of anticipated and unanticipated outcomes from interventions of diverse types that go beyond typical reporting metrics. It is also a method that meaningfully engages the intended beneficiaries or participants in capturing the quality of the outcomes, not just the quantity. This review has identified its tried and tested role in evaluation, but also demonstrates potential as a research tool for understanding participants, groups and activities beyond traditional evaluation.
Supplemental Material
Supplemental Material - A Systematic Review of the Applications of Ripple Effects Mapping
Supplemental Material for A Systematic Review of the Applications of Ripple Effects Mapping by Llinos Brown, Kate Pangbourne in International Journal of Qualitative Methods.
Footnotes
Acknowledgments
We would like to thank our colleagues at the Institute for Transport Studies and EDRC for their support during the online co-working sessions, which provided valuable time and space to focus on writing this review.
Ethical Considerations
This study did not require ethics approval, as it did not contain human or animal trials, and the studies reviewed had acquired various permissions from various ethical boards.
Funding
The authors disclosed receipt of the following financial support for the research, authorship, and/or publication of this article: Energy Demand Research Centre (EDRC), supported by the Engineering and Physical Sciences Research Council and the Economic and Social Research Council (grant number EP/Y010078/1).
Declaration of Conflicting Interests
The authors declared no potential conflicts of interest with respect to the research, authorship, and/or publication of this article.
Supplemental Material
Supplemental material for this article is available online.
References
Supplementary Material
Please find the following supplemental material available below.
For Open Access articles published under a Creative Commons License, all supplemental material carries the same license as the article it is associated with.
For non-Open Access articles published, all supplemental material carries a non-exclusive license, and permission requests for re-use of supplemental material or any part of supplemental material shall be sent directly to the copyright owner as specified in the copyright notice associated with the article.
