Evidence Synthesis and Literature
Literature provides warrants for claims in research. It does this in multiple ways, including through providing theoretical foundations (theory, models, frameworks, etc.) to help us frame our work, or/and more granular insights (hypothesis x is supported more strongly than y, method a was developed by…, x provide a case study of b, etc.). Disciplines vary in their practices relating to citation, in ways that reflect how they build knowledge (Hyland, 1999).
"Insert citation here".
It is quite common to see people use placeholders in text such as “add a reference”. Sometimes that’s because the researcher knows a piece of work or theory but doesn’t have it to hand - that’s probably ok. Be cautious though if you find you are writing your argument and then only looking for literature that is confirmatory, this is bad practice and may lead to writing that doesn’t reflect available evidence in the literature.
Researchers or/and other stakeholders such as policy makers or practitioners often want to bring together evidence that may be of different types to provide an overview or synthesis of what that evidence says, or provides a model or theory grounded in a broad literature. Evidence synthesis methods provide ways of understanding the state of evidence on an issue, to support research and practice that is informed by, and fosters development of, that evidence.
Evidence synthesis methods
Evidence synthesis methods provide ways of understanding the state of evidence on an issue, to support research and practice that is informed by, and fosters development of, that evidence.
Considerations in selecting an evidence synthesis approach
Different approaches to evidence synthesis serve different purposes, and evidence synthesis artefacts should be evaluated with respect to those purposes, rather than any pre-conceived hierarchy of methods. In some cases, a literature review that builds an argument towards the topic of a project is more appropriate than a comprehensive systematic review; in others vice-versa.
Identifying the appropriate type of review involves considering the need or aims, with respect to possible approaches given any constraints (e.g., regarding resourcing).
block-beta
block:all
columns 1
block:rows
columns 3
NI<["Needs identification: What will the evidence synthesis be used for?\\nWhat requirements does that imply?"]>(right)
OA<["Outputs assessment: Consider possible synthesis outputs with respect to key factors (rigour, relevance, readability, and resourcing)"]>(right)
OS<["Output selection: Select the desired synthesis output"]>(right)
end
block:body
columns 3
EN(["Recognised need for evidence"])
block:cycle
columns 4
space:2 RI(["Rigour and transparency"])
space:5
space:2 RES(["Resources \\n(e.g., personnel & time)"])
space
space RE(["Relevance \\n(e.g., focus for purpose)"]) space REA(["Readability \\n(e.g., length, technical focus)"])
RE--> RI
RI-->RE
RE-->REA
REA-->RE
REA-->RI
RI-->REA
RES-->RE
RE-->RES
RES-->RI
RI-->RES
RES-->REA
REA-->RES
end
ES(["Evidence synthesis output identified"])
EN --> cycle
cycle --> ES
end
ref["Evidence synthesis identification process, Knight 2024 CC-By. Based on Figure 1 Wickremasinghe et al.’s ‘Process for matching information needs with an evidence synthesis output’ (Wickremasinghe et al., 2016, p. 535) article published under a CC-By license."]
end
style RI fill:#ffcc80,stroke:#333,stroke-width:2px
style RE fill:#ffcc80,stroke:#333,stroke-width:2px
style RES fill:#ffcc80,stroke:#333,stroke-width:2px
style REA fill:#ffcc80,stroke:#333,stroke-width:2px
style EN fill:#ffab91,stroke:#333,stroke-width:2px
style ES fill:#ffab91,stroke:#333,stroke-width:2px
style NI fill:#ffe082,stroke:#333,stroke-width:2px
style OA fill:#ffe082,stroke:#333,stroke-width:2px
style OS fill:#ffe082,stroke:#333,stroke-width:2px
classDef nopad padding:0px
class ref nopad
classDef circle fill:#ffcc80,stroke:#333,stroke-width:2px
classDef text fill:#ffe082,stroke:#333,stroke-width:2px
What’s a synthesis or review involve?
The flowchart below outlines a broad approach to evidence synthesis, although there are variations to this model (and more nuance across models than the flowchart displays).
**This link provides excellent guidance re: developing search strategies: https://ait.libguides.com/c.php?g=689530&p=4936115** Writing your questions
flowchart TD
consider["`Consider the aims of the synthesis and approach to adopt based on this, develop your aims`"] --> select["`Select an evidence synthesis approach (or, narrow options to review in light of the next step), develop the rationale for selecting this approach`"]
select --> collate["`Check requrements for the review and collate documentation needed; consider if you and your collaborators have the right expertise (and address gaps)`"]
collate --> pilot["`Identify key concepts and terms (including in previous syntheses on this topic), select databases and pilot searches; ensure the queries are syntactically valid, and return data addressing your need (e.g., by peer-checking)`"]
pilot --> collate
pilot --> execute["`Execute the searches (or/and purposive sampling), recording results returned as appropriate. Deduplicate results`"]
pilot --> grey["`Consider whether grey literature or sources that are not included in your target indexes should also be searched`"]:::extra
pilot --> register["`If required of your synthesis type, register a protocol`"]:::extra
execute --> prisma["`If required for your review type, you should ensure you record data for a PRISMA flow diagram or modification`"]:::extra
execute --> screen["`Screen the title and abstracts of items retrieved, for inclusion (recording this process, and dual-screening as appropriate)`"]
screen --> full["`Retrieve the full texts of items included post-screening.`"]
full --> review["`Conduct full-text review of the items for inclusion (recording this process, and dual-reviewing as appropriate)`"]
review --> appraise["`If required for your type of review, appraise items retrieved using a standard model`"]:::extra
review --> chase["`Conduct forward/backward citation chasing where appropriate and consider if there are implications for updating search strategy`"]:::extra
review --> extract["`Extract information/data from items using an identified method explaining how the extraction was conducted`"]
extract --> synthesis["`Conduct analysis of data using an identified approach to produce a synthesis e.g., of charts and tables alongside narrative summary`"]
synthesis <--> validate["`If appropriate, validate synthesised findings or recommendations e.g., with an expert panel or end user participation`"]:::extra
validate <--> synthesis
synthesis --> article["`Compile required documentation and synthesis into outputs for publication (e.g., a research article)`"]
article --> communicate["`Consider communication of the output to relevant stakeholders, including key implications for policy/practice/research`"]
classDef extra fill:#ffe082,stroke:#333,stroke-width:2px
Stakeholder involvement
One way that evidence synthesis approaches vary is the degree to which stakeholders are engaged in the process of identifying the need, working with the data, and developing the synthesis, as well as validating any end-product artefacts.
For example the below visual depictions indicate ways that syntheses might be iterative and involve abduction - explanatory reasoning based on best evidence - and stakeholder contextualisation through the use of narratives or scenarios that connect to practice.


What approaches are good for what?
There’s an excellent decision aid to help you think about what approach to adopt at: Right Review
This resource is also useful in providing detail over a range of approaches: Library Guides: Which review is that? A guide to review types: Decision Tool
Are there tools to help?
There are also various software tools designed to help in various aspects of the synthesis process. Including those in the table below. There are some other excellent open source (and often code based) resources available: Developing collaborations and technology for evidence synthesis
| tool | url | cost (where relevant, 2 users * 6 months) | pros | cons | in/out |
|---|---|---|---|---|---|
| RevMan | https://subscribe.cochrane.org/info/pricing | 330 (12 months) | Targets meta analysis q | ||
| EPPI Reviewer | https://eppi.ioe.ac.uk/cms/Default.aspx?tabid=2937 | 650 | Feature rich for review types https://eppi.ioe.ac.uk/cms/Portals/35/ER Web Presentation 12 June 2020.pdf | cost | |
| ReLiS | https://relis.iro.umontreal.ca/user/user_help | 0 | open source | Experimental tool Requires SQL-ish queries | |
| distillersr | https://www.distillersr.com/pricing/student-pricing | 1,500 | cost | ||
| ASReview | https://asreview.nl/ | 0 | open source use of active learning to semi-automate screening | Mostly for screening, and then reviews involving labelling (rather than extracting) data I think. Somewhat experimental. | |
| Covidence | https://www.covidence.org/pricing/ | 440 (12 months; but possibly free with <500 records?) | https://support.covidence.org/help | ||
| rayyan | https://www.rayyan.ai/pricing/ | 120 | https://help.rayyan.ai/hc/en-us/articles/22697630697617-Getting-Started-with-Rayyan-A-Quick-Start-Guide |
Some examples of less conventional syntheses
This is largely from my own work (so, caveats), but may be of interest to people considering how they draw on evidence synthesis and literature to make claims.
| Type | Why | e.g., |
|---|---|---|
| Taxonomy development with interview validation | Build model with expert consensus across diverse literature | E.g., - Taxonomy of expert disagreement types grounded in literature and expert interviews https://sjgknight.com/finding-knowledge/2022-08-why-do-experts-disagree |
| Critical Integrative Synthesis (CIS) | Build model with expert consensus across diverse literature | Build a model of processing for conflicting information (Deroover, thesis forthcoming). A model of epistemic justice thinking (Knight, forthcoming) |
| Materials review | Review the materials used in research to (a) consider gaps/approaches adopted, and (b) implications for adoption/adaptation of those materials into practice contexts | (forthcoming) |
| Systematic grey literature scoping review and content analysis | Systematically documented approaches to working with grey literature helps foster transparency and warrants claims around the nature of the available resources | E.g., analysis of: - public submissions to a govt inquiry https://doi.org/10.14742/ajet.8922 - resources available for AI ethics doi.org/10.1007/s43681-023-00416-z - cases in public repositories re: AI incidents https://doi.org/10.21203/rs.3.rs-4844649/v1 - ethics guidelines and their methods of development https://doi.org/10.31219/osf.io/n43d6 |
| Scoping review | Understand scope of approaches and use those in creation of new material via expert panel | Forthcoming Knight, S., Viberg, O., Mavrikis, M., Kovanović, V., Khosravi, H., Ferguson, R., Corrin, L., Thompson, K., Major, L., Lodge, J., Hennessy, S., & Cukurova, M. (2024). Emerging technologies and research ethics: Developing editorial policy using a scoping review and reference panel. PLOS ONE. https://doi.org/10.1371/journal.pone.0309715 |
| Design reviews (e.g., design spaces) | To understand range of design aims and their instantiation | E.g., - |
Other resources
On literature reviews
- Paré G, Kitsiou S. Chapter 9 Methods for Literature Reviews. In: Lau F, Kuziemsky C, editors. Handbook of eHealth Evaluation: An Evidence-based Approach. Victoria (BC): University of Victoria; 2017 Feb 27. Available from: https://www.ncbi.nlm.nih.gov/books/NBK481583/
- Pat Thomson: not all literature ‘reviews’ are the same
On reviews:
- Open guide to conducting reviews: https://systematicreviewmethods.github.io/index.html
- UNISA guide on types of review: Guides: Systematic Reviews: Overview of systematic reviews
- Grant, M.J. & Booth, A. (2009). A typology of reviews: An analysis of 14 review types and associated methodologies. Health Information & Libraries Journal, 26(2), 91-108. https://doi.org/10.1111/j.1471-1842.2009.00848.x
- UK Health Security Agency piece on Systematic Reviews (it gives a nice overview): Systematic Reviews
In EdTech you might be interested in this piece which investigates the types of reviews used in the literature, making some recommendations:
- Buntins, Katja, Svenja Bedenlier, Victoria Marín, Marion Händel, und Melissa Bond. 2023. “Methodische Ansätze Zu Evidenzsynthesen in Der Bildungstechnologie: Eine tertiäre Übersichtsarbeit“. MedienPädagogik: Zeitschrift für Theorie Und Praxis Der Medienbildung 54 Research Syntheses):167-91. https://doi.org/10.21240/mpaed/54/2023.12.20.X
There is also a literature around decolonial approaches, and Indigenous approaches, to evidence synthesis (and of course literature and research more broadly). I lack expertise here, but I found the following two methodological pieces rewarding and important.
- Rogers, J. Towards an Indigenous literature re-view methodology: Aboriginal and Torres Strait Islander boarding school literature. Aust. Educ. Res. (2023). https://doi.org/10.1007/s13384-023-00654-4
- Lowe, K., Tennent, C., Burgess, C., Moodie, N., Vass, G., Guenther, J. (2023). Developing a Systematic Methodology to Explore Research in Indigenous Education. In: Moodie, N., Lowe, K., Dixon, R., Trimmer, K. (eds) Assessing the Evidence in Indigenous Education Research. Postcolonial Studies in Education. Palgrave Macmillan, Cham. https://doi.org/10.1007/978-3-031-14306-9_2
An idea for a dumb decision tool
There is a fantastic decision aid to help you think about what approach to adopt at: Right Review
But it currently misses some pieces. So I was curious about other approaches to building a tool. A future idea is around building some form of interactive tool to select a synthesis approach which would then also build template documentation for you, e.g., using:
- http://www.qnamarkup.org/ / https://github.com/colarusso/QnAMarkup
- https://docassemble.org/
- https://github.com/hungrymedia/interactive-decision-tree
I’ve made some progress (with significant chatGPT help), and built two tools:
- Which review? A dumb decision tool which is a decision tool (see code at https://gist.github.com/sjgknight/14f7cf3b5f6457ac5989666a0fa40749), you select options and it’ll re-rank methods. It is dumb in the sense that the weights and outputs are simple. It could be extended or used in other contexts too.
- And csv_to_mermaid_3 which is a tool to CSV structured data into a mermaid diagram (see code at https://gist.github.com/sjgknight/31851bf2a3cfa96afdbee24e10259d1a) - it works fine as long as there isn’t a lot of syntax.
I’ve used these drawing on this literature:
- “Schematic of the main differences between the types of literature review” by Brennan, M. L., Arlt, S. P., Belshaw, Z., Buckley, L., Corah, L., Doit, H., Fajt, V. R., Grindlay, D., Moberly, H. K., Morrow, L. D., Stavisky, J., & White, C. (2020). Critically Appraised Topics (CATs) in veterinary medicine: Applying evidence in clinical practice. Frontiers in Veterinary Science, 7, 314. https://doi.org/10.3389/fvets.2020.00314
- Grant, M.J. & Booth, A. (2009). A typology of reviews: An analysis of 14 review types and associated methodologies. Health Information & Libraries Journal, 26(2), 91-108. https://doi.org/10.1111/j.1471-1842.2009.00848.x
To build this table outlining components of evidence synthesis approaches evidence_synthesis.xlsx and the diagram below, to which I’d also like to add the kinds of theory, materials, and stakeholders involved in the process, and the sorts of outputs produced by them.
flowchart TD
subgraph types
direction LR
ES["Approaches to Evidence Synthesis"]
ES --> Q["More configurative, interpretative, or/and hypothesis developing"]
ES --> Co["More confirmatory, consolidatory, or/and hypothesis testing"]
Co --> A[fa:fa-ruler-combined Systematic Reviews]:::highlight-node
Co --> B[fa:fa-calculator Meta-Analysis]:::highlight-node
Co --> C[fa:fa-network-wired Scoping Review]:::highlight-node
Co --> E[fa:fa-hourglass-start Rapid Review]:::highlight-node
Co --> G[fa:fa-archive Umbrella Review]:::highlight-node
Co --> H[fa:fa-theater-masks Bibliometric Analysis]:::highlight-node
Co --> I[far:fa-compass State of the Art Review]:::highlight-node
Co --> J[fa:fa-route Evidence Maps]:::highlight-node
Co --> K[fa:fa-list Annotated Bibliographies]:::highlight-node
Co --> L[fa:fa-scale-unbalanced Critically Appraised Topic]:::highlight-node
Q --> F[fa:fa-gears Realist Evidence Synthesis]:::highlight-node
Q --> M["far:fa-comments Qualitative Evidence Synthesis (QES)"]:::highlight-node
Q --> D[fa:fa-quote-left Narrative Review]:::highlight-node
Q --> N[far:fa-share-from-square Materials Review]:::highlight-node
M --> BF[fa:fa-puzzle-piece Including framework-synthesis methods such as best-fit and Rapid Best Fit]:::highlight-node
Q --> O[fa:fa-map]:::highlight-node
A --> A1["fa:fa-clock 18 months+ fa:fa-person fa:fa-repeat fa:fa-person 2+ personnel"]:::highlight-node
B --> B1["fa:fa-clock 18 months+ fa:fa-person fa:fa-repeat fa:fa-person 2+ personnel"]:::highlight-node
C --> C1["fa:fa-clock 2 months+" fa:fa-person 1+ personnel]:::highlight-node
D --> D1["fa:fa-clock 3 months+ fa:fa-person 1+ personnel"]:::highlight-node
E --> E1["fa:fa-clock 2 months+" fa:fa-person 1+ personnel]:::highlight-node
F --> F1["fa:fa-clock 12 months+"]:::highlight-node
G --> G1["fa:fa-clock 6 months+"]:::highlight-node
H --> H1["fa:fa-clock 6 months+"]:::highlight-node
I --> I1["fa:fa-clock 6 months+"]:::highlight-node
J --> J1["fa:fa-clock 6 months+" fa:fa-person 1+ personnel]:::highlight-node
K --> K1["fa:fa-clock 4 months+" fa:fa-person 1+ personnel]:::highlight-node
L --> L1["fa:fa-clock 4 months+" fa:fa-person 1+ personnel]:::highlight-node
M --> M1["fa:fa-clock 9 months+"]:::highlight-node
N --> N1["fa:fa-clock 3 months+ fa:fa-person 1+ personnel"]:::highlight-node
A1 --> |"Components"| A2["Search: Exhaustive\\nAppraisal: Quality determines inclusion/exclusion\\nSynthesis: Narrative + Tabular\\nAnalysis: Known evidence; Recommendations for practice"]:::highlight-node
B1 --> |"Components"| B2["Search: Exhaustive\\nAppraisal: Quality determines inclusion/exclusion\\nSynthesis: Graphical & Tabular\\nAnalysis: Numerical effect size"]:::highlight-node
C1 --> |"Components"| C2["Search: Comprehensive within constraints\\nAppraisal: No formal quality assessment\\nSynthesis: Graphical & Tabular\\nAnalysis: Characterizes literature quantity & quality"]:::highlight-node
D1 --> |"Components"| D2["Search: May be comprehensive\\nAppraisal: No formal quality assessment\\nSynthesis: Narrative\\nAnalysis: Conceptual innovation, theory derivation"]:::highlight-node
E1 --> |"Components"| E2["Search: Time-constrained\\nAppraisal: Quality assessment\\nSynthesis: Narrative & Tabular\\nAnalysis: Literature quantity & direction of effect"]:::highlight-node
F1 --> |"Components"| F2["Search: Theory identification\\nAppraisal: Contextual analysis\\nSynthesis: Qualitative & Narrative\\nAnalysis: Mechanism explanation"]:::highlight-node
G1 --> |"Components"| G2["Search: Identifies component reviews\\nAppraisal: Reviews or studies\\nSynthesis: Graphical & Tabular\\nAnalysis: Recommendations for practice & research"]:::highlight-node
H1 --> |"Components"| H2["Search: Citation mapping\\nAppraisal: Network analysis\\nSynthesis: Graphical\\nAnalysis: Research trends and connections"]:::highlight-node
I1 --> |"Components"| I2["Search: Comprehensive of current literature\\nAppraisal: No formal quality assessment\\nSynthesis: Narrative + Tabular\\nAnalysis: State of knowledge & future priorities"]:::highlight-node
J1 --> |"Components"| J2["Search: Broad scoping\\nAppraisal: Evidence mapping\\nSynthesis: Graphical & Tabular\\nAnalysis: Gaps identification"]:::highlight-node
K1 --> |"Components"| K2["Search: Source selection\\nAppraisal: Annotations\\nSynthesis: Summarization\\nAnalysis: Key contributions identification"]:::highlight-node
L1 --> |"Components"| L2["Search: Critical appraisal\\nAppraisal: Relevance assessment\\nSynthesis: Summary of findings\\nAnalysis: Relevance to context"]:::highlight-node
A2 --> A3[fa:fa-users-involved Systematic Reviews typically result in comprehensive guidelines, often in consultation with experts and stakeholders at the synthesis stage. Comprehensive approach, with low risk of bias.]:::highlight-node
B2 --> B3[fa:fa-users-together Meta-Analysis outputs often involve aggregated data interpretations, with stakeholders providing feedback on applicability. Comprehensive approach, with low risk of bias.]:::highlight-node
C2 --> C3[fa:fa-users-discuss Scoping Reviews map evidence with some stakeholder input at initial stages for relevance, not comprehensive, but transparent approach, with some risk of bias.]:::highlight-node
D2 --> D3[fa:fa-users-share Narrative Reviews provide an overview of trends, generally with minimal stakeholder engagement. Non-comprehnsive approach, interpretative.]:::highlight-node
E2 --> E3[fa:fa-users-quick Rapid Reviews focus on quickly informing policy or practice, involving stakeholders to prioritize evidence. Non-comprehensive, but transparent approach, bias limited.]:::highlight-node
F2 --> F3[fa:fa-users-gear Realist Evidence Synthesis engages stakeholders to identify the problem space and theory of change and adapt the review focus iteratively.]:::highlight-node
G2 --> G3[fa:fa-users-consolidate Umbrella Reviews synthesize findings from multiple reviews, often consulting stakeholders to interpret results.]:::highlight-node
H2 --> H3[fa:fa-users-networks Bibliometric Analysis maps research networks and trends, usually with expert consultation.]:::highlight-node
I2 --> I3[fa:fa-users-overview State of the Art Reviews summarize leading ideas with stakeholder engagement mainly in initial scoping.]:::highlight-node
J2 --> J3[fa:fa-users-map Evidence Maps provide visual summaries of evidence, sometimes with stakeholder feedback on gaps.]:::highlight-node
K2 --> K3[fa:fa-users-annotate Annotated Bibliographies collect key sources, occasionally with stakeholder suggestions for inclusion.]:::highlight-node
L2 --> L3[fa:fa-users-appraise Critically Appraised Topics involve stakeholders in the appraisal process to ensure relevance and impact.]:::highlight-node
M2 --> M3[fa:fa-users-qualitative Qualitative Evidence Synthesis gathers qualitative insights with stakeholder collaboration to validate findings.]:::highlight-node
N2 --> N3[fa:fa-users-materials Materials Review organizes and classifies materials with input from stakeholders on categorization.]:::highlight-node
A3 --> |"Questions"| A4["fa:fa-question-circle What is the effect size of interventions?\\nHow effective is treatment X for condition Y?"]:::highlight-node
B3 --> |"Questions"| B4["fa:fa-question-circle How effective is treatment X for condition Y?\\nWhat is the combined effect size?"]:::highlight-node
C3 --> |"Questions"| C4["fa:fa-question-circle What is the extent of research on topic Z?\\nWhat types of interventions exist for condition A?"]:::highlight-node
D3 --> |"Questions"| D4["fa:fa-question-circle What are the general trends in research on X?\\nHow have views on Y evolved over time?"]:::highlight-node
E3 --> |"Questions"| E4["fa:fa-question-circle What is the current evidence on intervention X?\\nWhat are the immediate implications for policy Y?"]:::highlight-node
F3 --> |"Questions"| F4["fa:fa-question-circle How do interventions work in different contexts?\\nWhat mechanisms explain the success of program X?"]:::highlight-node
G3 --> |"Questions"| G4["fa:fa-question-circle What do existing reviews say about intervention X?\\nWhat are the overarching findings across reviews on topic Y?"]:::highlight-node
H3 --> |"Questions"| H4["fa:fa-question-circle Who is working on what topics?\\nHow are different studies connected?"]:::highlight-node
I3 --> |"Questions"| I4["fa:fa-question-circle What is the current state of research on X?\\nWhat are the leading thoughts and gaps?"]:::highlight-node
J3 --> |"Questions"| J4["fa:fa-question-circle What is the distribution of evidence on topic Y?\\nWhere are the gaps in current research?"]:::highlight-node
K3 --> |"Questions"| K4["fa:fa-question-circle What are the key sources on topic X?\\nWhat do they contribute to the understanding of Y?"]:::highlight-node
L3 --> |"Questions"| L4["fa:fa-question-circle What are the critical findings on topic X?\\nHow relevant are these findings for context Y?"]:::highlight-node
M3 --> |"Questions"| M4["fa:fa-question-circle What are the qualitative insights on X?\\nHow do themes emerge across different studies?"]:::highlight-node
N3 --> |"Questions"| N4["fa:fa-question-circle What materials exist on topic X?\\nHow can these materials be classified and organized?"]:::highlight-node
classDef reviewStyle fill:#f9f,stroke:#333,stroke-width:2px;
classDef characteristicsStyle fill:#ccf,stroke:#333,stroke-width:1px;
class highlight-node fill:#ccf,stroke:#333,stroke-width:1px;
class A,B,C,D,E,F,G,H,I,J,K,L,M,N,BF reviewStyle;
class A1,B1,C1,D1,E1,F1,G1,H1,I1,J1,K1,L1,M1,N1 characteristicsStyle;
class A2,B2,C2,D2,E2,F2,G2,H2,I2,J2,K2,L2,M2,N2 characteristicsStyle;
class A3,B3,C3,D3,E3,F3,G3,H3,I3,J3,K3,L3,M3,N3 characteristicsStyle;
end
types ~~~ meta
subgraph meta
note["fa:fa-copyright Issues may arise with substantive parts"]
grey["far:fa-newspaper consider if grey literature included"]
other["fab:fa-creative-commons fab:fa-creative-commons-by"]
end