DAAF-Contribution-Community/daaf / .claude/skills/literature-review
literature-review
VerifiedMethodology for literature reviews in DAAF, organized by three levels the researcher picks at the start: quick scan (abstracts and a short memo of pointers), standard review (full reads of key papers, page-checked claims, a stated weight per paper, calibrated wording) and formal or systematic review (researcher-led protocol, second screener, checklist guides for RoB 2, ROBINS-I V2, ROBINS-E, WWC, AMSTAR 2, JBI and CASP, GRADE-style certainty). Covers choosing a review type and level, screening, per-paper notes, citation snowballing, synthesis without pooling or vote counting, labeled descriptive summaries, the certainty phrase table (certainty-language.yaml) and the gap register. Use when planning, screening, appraising, synthesizing or reporting a literature review, or wording what a body of evidence shows. Not for running searches (scholarly-search), reading PDFs (pdf-reading), causal verbs for one's own results (data-scientist), or lay audiences (science-communication).
| File | Kind | Size |
|---|---|---|
| SKILL.md | skill md | 26.9 KB |
| references/certainty-language.yaml | config | 12.3 KB |
| references/certainty-of-evidence.md | markdown | 15.3 KB |
| references/formal-review.md | markdown | 10.3 KB |
| references/gaps-and-saturation.md | markdown | 8.2 KB |
| references/guide-descriptive.md | markdown | 5.7 KB |
| references/guide-education-wwc.md | markdown | 5.3 KB |
| references/guide-nonrandomized.md | markdown | 11.4 KB |
| references/guide-qualitative.md | markdown | 6.0 KB |
| references/guide-randomized.md | markdown | 8.2 KB |
| references/guide-reviews.md | markdown | 5.0 KB |
| references/guide-theory.md | markdown | 2.9 KB |
| references/lit-corpus.md | markdown | 67.2 KB |
| references/lit-verify-report.md | markdown | 62.3 KB |
| references/paper-notes.md | markdown | 10.7 KB |
| references/quick-scan.md | markdown | 13.7 KB |
| references/review-types-and-levels.md | markdown | 17.8 KB |
| references/screening.md | markdown | 7.9 KB |
| references/standard-review.md | markdown | 18.5 KB |
| references/synthesis-methods.md | markdown | 11.9 KB |
references/gaps-and-saturation.md
View on GitHub# Gaps and Saturation
What the review found missing, how sure we are that it is missing, and when to
stop: the gap register, evidence of absence, the saturation table and counts
per round, and the follow-up plan the researcher approves between rounds.
## Contents
- [The gap register](#the-gap-register)
- [Gap types and modes](#gap-types-and-modes)
- [Evidence of absence](#evidence-of-absence)
- [Gap confidence](#gap-confidence)
- [Could DAAF data answer it?](#could-daaf-data-answer-it)
- [Problematization](#problematization)
- [Counts per round](#counts-per-round)
- [Saturation table](#saturation-table)
- [The follow-up plan](#the-follow-up-plan)
- [Sources](#sources)
## The gap register
One entry per gap:
| Field | Content |
|---|---|
| `gap_id` | `G01`, `G02`, … |
| `statement` | The gap in one sentence |
| `miles_type` | `evidence`, `knowledge`, `practical`, `methodological`, `empirical`, `theoretical`, `population`, or `unclassified` (the default when left out; a researcher's judgment at the formal level, optional at the quick and standard levels, § Gap types and modes) |
| `sa_mode` | `confusion`, `neglect`, `application_spotting`, `problematization`, or `unclassified` (as `miles_type`) |
| evidence of absence | The exact searches that returned nothing, or whose results were all excluded (next sections) |
| `supporting_claim_ids` | Claims that point to the gap (authors' stated open questions, contradictory findings) |
| `confidence` | `high`, `moderate`, `low`: confidence that the gap is real, not a finding's certainty |
| `data_answerable` | Whether data DAAF can reach could answer it, with a note |
`lit_corpus gaps record` fills the evidence of absence from the search
MANIFEST rows themselves, so it is never typed: a gap names the rows'
sha256 and kind (`zero_results` or `none_eligible`), and the tool refuses a
row whose search returned results, or whose results were not all excluded at
screening (`lit-corpus.md` § Subcommands).
## Gap types and modes
Classifying a gap by type and mode is formal-level work. At the quick and
standard levels it is optional: the researcher is not asked to classify gaps,
and the statement, the evidence of absence and the confidence carry the gap.
Leave both fields out of the `lit_corpus gaps record` line; the tool stores
them as `unclassified` (1.1.0, gap-words-v1). The tool does not know the
review's level, so at the formal level the protocol, not the tool, requires
a type and a mode for every gap. A code the agent does give must be one of
the values below (anything else is refused).
**Codes stay in the machine fields.** In anything the researcher reads (the
memo, the synthesis, a checkpoint message), describe a gap in plain words, for
example "G01: a population not yet studied (how sure we are the gap is real:
low)", never "G01 (population, neglect; confidence low)". The tool's own
`output/gap_register_v{N}.md` does the same: each heading gives the gap's
plain words from the tables below, or "not classified", and the evidence
lines say "the search returned nothing" or "every result was excluded at
screening"; the codes stay in the `gap_register` table.
**Miles 2017's seven gap types** (read from a snippet only, so the plain words
are DAAF's paraphrase of the type names, not Miles's definitions):
| Code | Plain words for the researcher |
|---|---|
| `evidence` | studies point different ways |
| `knowledge` | the question has not been studied |
| `practical` | practice and research disagree (Miles: practical-knowledge conflict) |
| `methodological` | a method that could answer it has not been used |
| `empirical` | a claim has not been tested with data |
| `theoretical` | no theory explains it |
| `population` | a population not yet studied |
**Sandberg & Alvesson 2011's gap-spotting modes:**
| Code | Mode | Plain words |
|---|---|---|
| `confusion` | Confusion spotting: competing explanations exist | the explanations disagree |
| `neglect` | Neglect spotting: an area is overlooked, under-researched, or lacks empirical support | an overlooked area |
| `application_spotting` | Application spotting: existing work could be extended or complemented | existing work could be extended |
| `problematization` | Combined or disruptive modes (critical confrontation, a new idea), including problematization (below) | an assumption worth questioning |
## Evidence of absence
A search that found nothing is evidence about the index searched, not proof
that no literature exists (`scholarly-search` § Literature-review workflow).
State absence the way DAAF states any negative claim: quote the probe.
- **Zero results:** the exact query, the backends that answered (not the ones
that failed), the date and the result count, from the MANIFEST row.
- **None eligible:** a search that returned papers, every one of which was
excluded at screening; name the exclusion reasons.
- Say what was not searched (paywalled databases, Google Scholar, languages not
covered; `scholarly-search` § Known gaps).
Levac et al. 2010 warn that gaps identified without assessing evidence quality
"may yield false conclusions". A gap where weak studies exist ("only
correlational evidence") is different from a gap where none exist; say which.
## Gap confidence
DAAF's convention (no published scale was found for gap confidence):
| Confidence | Use when |
|---|---|
| high | Several phrasings and sources returned nothing relevant, snowballing from the key papers found nothing, and the key papers' authors name the gap as open |
| moderate | Searches returned nothing relevant, but coverage has a known hole (a missing database, a language) |
| low | One phrasing returned nothing, or the gap rests on the reader's impression |
## Could DAAF data answer it?
Mark `data_answerable` when a dataset DAAF can reach (for example the Education
Data Portal sources) could address the gap, with a one-line note naming it. The
orchestrator can offer a Full Pipeline analysis for such a gap; whether to
pursue it is the researcher's call.
## Problematization
An optional pass that challenges the assumptions behind a literature rather
than filling gaps in it (Alvesson & Sandberg 2011). Assumption types: "in-house,
root metaphor, paradigm, ideology, and field". Use it when the researcher asks
for it; label its output as an argument, not a finding.
## Counts per round
Each round reports simple counts (no flow diagram; PRISMA 2020 counts are built
by hand at the formal level, `review-types-and-levels.md` § Reporting standards
by level):
| Count | From |
|---|---|
| Records found, per source, and unique works after merging versions | The search log (MANIFEST rows) |
| Records from citation searching (`cites`, `refs`, `similar`) | The search log, kept apart from database searches |
| Screened; included; excluded, by reason code | The screening log (`screening.md`) |
| Read in full, by reading depth (`FULL_TEXT_VERIFIED`, `FULL_TEXT_PARTIAL`) | Evidence cards and their checks |
| Abstract-only | The screening log and cards |
| Cited in the latest deliverable | The deliverable's citations |
| Second-screener disagreements | The screening log |
`lit_corpus counts` produces this table (`output/counts_v{N}.json` and `.md`;
`--round R` for one round, `--deliverable PATH` for the cited count).
## Saturation table
Per round, list what was new: concepts or themes, findings, gaps. When rounds
stop adding concepts, the review is "nearing completion" (Webster & Watson
2002). The table is the evidence the researcher uses to apply the stopping rule
(`review-types-and-levels.md` § Stopping rules).
| Round | Searches run | New papers included | New concepts or themes | New findings | New gaps |
|---|---|---|---|---|---|
| 1 | … | … | … | … | … |
## The follow-up plan
At the end of each round, propose the next one for the researcher to approve:
recommended searches and snowball targets, with the expected yield, the reading
load and the cost (search costs from `scholarly-search` § Costs and exit codes;
reading costs per paper). Name what stopping now would leave open.
## Sources
Miles 2017 (snippet only); Sandberg & Alvesson 2011, Organization; Alvesson &
Sandberg 2011, Academy of Management Review; Levac et al. 2010; Webster &
Watson 2002; Müller-Bloch & Kranz 2015. Read for DAAF's 2026-10-02 methodology
survey.
Related skills
| Skill | Stars | Updated | Works with | Status | Install |
|---|---|---|---|---|---|
| canvas-designanthropics/skillsCreate beautiful visual art in .png and .pdf documents using design philosophy. You should use this skill when the user asks to create a poster, piece of art, design, or other static piece. Create original visual designs, never copying existing artists' work to avoid copyright violations. | Stars: 198k | Updated | Claude CodeCodex | Verified | Install |
| docxanthropics/skillsUse this skill whenever the user wants to create, read, edit, or manipulate Word documents (.docx files) or Word templates (.dotx files). Triggers include: any mention of 'Word doc', 'word document', '.docx', '.dotx', or requests to produce professional documents with formatting like tables of contents, headings, page numbers, or letterheads. Also use when extracting or reorganizing content from .docx or .dotx files, inserting or replacing images in documents, performing find-and-replace in Word files, working with tracked changes or comments, or converting content into a polished Word document. If the user asks for a 'report', 'memo', 'letter', 'template', or similar deliverable as a Word or .docx file, use this skill. Do NOT use for PDFs, spreadsheets, Google Docs, or general coding tasks unrelated to document generation. | Stars: 198k | Updated | Claude CodeCodex | Verified | Install |
| pdfanthropics/skillsUse this skill whenever the user wants to do anything with PDF files. This includes reading or extracting text/tables from PDFs, combining or merging multiple PDFs into one, splitting PDFs apart, rotating pages, adding watermarks, creating new PDFs, filling PDF forms, encrypting/decrypting PDFs, extracting images, and OCR on scanned PDFs to make them searchable. If the user mentions a .pdf file or asks to produce one, use this skill. | Stars: 198k | Updated | Claude CodeCodex | Verified | Install |
| pptxanthropics/skillsUse this skill any time a .pptx or .potx file is involved in any way — as input, output, or both. This includes: creating slide decks, pitch decks, or presentations; reading, parsing, or extracting text from any .pptx or .potx file (even if the extracted content will be used elsewhere, like in an email or summary); editing, modifying, or updating existing presentations; combining or splitting slide files; working with templates (.potx), layouts, speaker notes, or comments. Trigger whenever the user mentions "deck," "slides," "presentation," or references a .pptx or .potx filename, regardless of what they plan to do with the content afterward. If a .pptx or .potx file needs to be opened, created, or touched, use this skill. | Stars: 198k | Updated | Claude CodeCodex | Verified | Install |