Introduction to Social Research Methodology

Conducting a literature review

Ben Stanley

Department of Social Sciences, SWPS University

November 17, 2026

Course outline

Course outline: session 5 (today) Course outline: 15 stations. 1 — Nature and purpose of social research (done); 2 — Research goals and research questions (done); 3 — Hypotheses and research design (done); 4 — Research ethics (done); 5 — Conducting a literature review (today); 6 — Secondary data analysis (to come); 7 — Qualitative approaches (to come); 8 — Quantitative approaches and measurement (to come); 9 — Mixed methods research (to come); 10 — Sampling and participant recruitment (to come); 11 — Questionnaire design and survey fieldwork (to come); 12 — Interviews and focus groups (to come); 13 — Quantitative data analysis (to come); 14 — Qualitative data analysis (to come); 15 — Reporting research; course review (to come). 1 Nature and purpose of social research 2 Research goals and research questions 3 Hypotheses and research design 4 Research ethics 5 Conducting a literature review TODAY 6 Secondary data analysis 7 Qualitative approaches 8 Quantitative approaches and measurement 9 Mixed methods research 10 Sampling and participant recruitment 11 Questionnaire design and survey fieldwork 12 Interviews and focus groups 13 Quantitative data analysis 14 Qualitative data analysis 15 Reporting research; course review

Today’s lecture

  • Why review the literature — what a review is, what it does for your project, and where it sits in the research process
  • Types of review — narrative and systematic reviews, and how to prepare for either
  • Searching the literature — databases, keywords, Boolean operators, and refining a search string
  • Evaluating and reading sources — judging credibility, and reading efficiently rather than exhaustively
  • Organising, synthesising and citing — turning a pile of sources into an argument, without plagiarising anyone
  • By the end, you should be able to plan and begin a defensible review of what is already known about a management problem

Why review the literature?

You are not the first

You are not the first: building on the literature On the left, the literature: decades of published research on employee turnover, customer satisfaction, engagement and leadership; turnover alone has been studied intensively since at least the 1970s. An arrow marked draw on leads to your project, whose base is what is known, built from their concepts, measures and findings. On top of it sits what I add: you distinguish your work from theirs. Here is what is known, and here is what I add: the basic shape of a contribution. If you ignore the literature, you reinvent wheels at best and repeat known mistakes at worst. At Meridian: before designing a single survey question, the team should know what forty years of turnover research has already established. THE LITERATURE Decades of published research Employee turnover Customer satisfaction Engagement Leadership turnover alone: studied intensively since at least the 1970s YOUR PROJECT draw on What is known built from their concepts, measures and findings — the raw material of your study What I add distinguish your work from theirs “Here is what is known, and here is what I add”: the basic shape of a contribution IF YOU IGNORE IT You reinvent wheels at best — and repeat known mistakes at worst. AT MERIDIAN Before designing a single survey question, the team should know what forty years of turnover research has already established.

What is a literature review?

What a literature review is At the top, the definition: a literature review is a critical examination of existing research on the phenomena of interest, and of the relevant theoretical ideas. Two panels follow. In the natural sciences the review sets out an agreed body of knowledge, drawn as a wall of bricks: the established base on which the researcher builds a new study. In the social sciences there are competing approaches and competing accounts: an economist, a psychologist and a sociologist each point at the same question, why do employees quit, and agreement is the exception. So a social-science review conveys the sense of the contest: what we know, meaning what has been found so far, and how we know it, meaning what research approaches have been taken. It is an argument about the state of knowledge, not a list of everything ever written. A LITERATURE REVIEW a critical examination of existing research on the phenomena of interest, and of the relevant theoretical ideas Natural sciences an agreed body of knowledge your study The review sets out the established base; the researcher builds on it. Social sciences competing approaches and competing accounts Why do employees quit? Economist Psychologist Sociologist agreement is the exception So a social-science review conveys the sense of the contest: WHAT WE KNOW what has been found so far HOW WE KNOW IT what research approaches have been taken ✓ an argument about the state of knowledge ✗ not a list of everything ever written

What the review does for your project

What the review does for your project Following Creswell and Creswell, the literature review is drawn as a bridge between a topic, on the left bank, and a researchable question, on the right. Four pillars hold it up. 1, Establishes context: how the topic has been studied, and how understanding has developed over time. 2, Avoids reinventing the wheel: you learn what is settled, and spend effort where it matters. 3, Strengthens your methodology: you inherit tested measures, designs, and warnings about what fails. 4, Identifies trends, debates and gaps: the gap you find is where your own research question lives. At Meridian: if research consistently shows that pay explains less turnover than supervision quality, Meridian's study, and its budget, should be designed accordingly. Creswell & Creswell: the review is the bridge between a topic and a researchable question A topic A researchable question The literature review 1 Establishes context how the topic has been studied, and how understanding has developed over time 2 Avoids reinventing the wheel you learn what is settled, and spend effort where it matters 3 Strengthens your methodology you inherit tested measures, designs, and warnings about what fails 4 Identifies trends, debates and gaps the gap you find is where your own research question lives AT MERIDIAN If research consistently shows that pay explains less turnover than supervision quality, Meridian's study — and its budget — should be designed accordingly.

Where the review sits in the research process

Where the review sits in the research process A bar across the top: the literature review is an early stage of the research process, after the topic and before the question and design are final, but not a self-contained one. Arrows marked feeds run from it down to four later stages along the research process. Concepts and theories: the ideas that frame the study come from the literature; at Meridian, job satisfaction, organisational commitment, embeddedness. Research questions: sharpened against what is already known: you cannot claim to fill a gap until you have mapped the terrain. Design and measurement: borrowed and adapted from previous studies. Interpretation: findings mean little until compared with what others found; at Meridian, 40% of leavers cite scheduling: typical of retail, or anomalous?. Underneath: the review is a living document; you keep returning to the literature throughout the project, not a chapter you finish and forget. THE LITERATURE REVIEW an early stage — after the topic, before the question and design are final — but not a self-contained one the research process feeds Concepts and theories the ideas that frame the study come from the literature AT MERIDIAN job satisfaction, organisational commitment, embeddedness feeds Research questions sharpened against what is already known: you cannot claim to fill a gap until you have mapped the terrain feeds Design and measurement borrowed and adapted from previous studies feeds Interpretation findings mean little until compared with what others found AT MERIDIAN 40% of leavers cite scheduling: typical of retail, or anomalous? A LIVING DOCUMENT You keep returning to the literature throughout the project — not a chapter you finish and forget.

A review is not secondary data analysis

A literature review is not secondary data analysis Across the top, another researcher's study: data collected, such as survey responses and records; their analysis, run on their data; and published findings and arguments. Secondary data analysis taps the start of that chain: it reuses other researchers' data, and you run your own analysis on datasets someone else collected, going from their data to your analysis to new findings, without new fieldwork. A literature review taps the end: it engages with other researchers' findings and arguments, their published conclusions, and tells you what is known, and how. Keep the two shelves separate: this session is about the literature; the next session is about data, where to find existing datasets and how to judge whether they can answer your question. Another researcher's study Data collected survey responses, records Their analysis run on their data Published findings and arguments SECONDARY DATA ANALYSIS reuses other researchers' data: you run your own analysis on datasets someone else collected their data your analysis new findings new findings without new fieldwork LITERATURE REVIEW engages with other researchers' findings and arguments: their published conclusions their conclusions what is known, and how reviewing tells you what is known Keep the two shelves separate: this session, literature — next session, data where to find existing datasets, and how to judge whether they can answer your question

Types of review

Narrative reviews

Narrative reviews: strengths and weaknesses The traditional narrative review in three steps: the reviewer reads widely, selects what seems important, and constructs a critical narrative of the field. Selection and interpretation rest on the reviewer's judgement; there is no formal protocol. Strengths: it is flexible and can follow the argument wherever it leads, and it suits broad, theoretical or loosely defined topics; most textbook chapters and thesis literature reviews are narrative reviews. An illustration of cherry-picking: a field of sources, some fitting the reviewer's view and some contradicting it; a dashed red loop around the cited ones takes in mostly those that fit. A reviewer who thinks pay drives turnover cites the studies that agree. Weaknesses: cherry-picking, since consciously or not reviewers favour sources that fit their view; and it is hard to check, because there is no record of what was searched, so a reader cannot tell what was left out. 1 The reviewer reads widely 2 selects what seems important 3 constructs a critical narrative of the field Selection and interpretation rest on the reviewer's judgement — there is no formal protocol. STRENGTHS + Flexible: it can follow the argument wherever it leads + Suits broad, theoretical or loosely defined topics Most textbook chapters and thesis literature reviews are narrative reviews. Cherry-picking fits the view contradicts it cited A reviewer who thinks pay drives turnover cites the studies that agree. WEAKNESSES − Cherry-picking: consciously or not, reviewers favour sources that fit their view − Hard to check: no record of what was searched, so a reader cannot tell what was left out

Systematic literature reviews

Systematic literature reviews: the protocol and its aim A bar marks the protocol, decided before searching begins; an arrow leads from it to searching begins. Under the bar, the protocol's four key features. 1, Defined research question: the question the review sets out to answer — not “turnover in general”. 2, Stated search strategy: which databases, which search strings, which dates. 3, Inclusion and exclusion criteria: rules for what counts, applied consistently to everything found. 4, Documented screening: how many sources were found, how many kept — and why. Beside them, two researchers, A and B, follow the same protocol and arrive at the same set of documents. The aim is replicability: another researcher following your protocol should retrieve substantially the same literature. The systematic review originated in medicine and is now standard in management research, where the evidence base is large and mixed in quality. THE PROTOCOL — DECIDED BEFORE SEARCHING BEGINS searching begins 1 Defined research question the question the review sets out to answer — not “turnover in general” 2 Stated search strategy which databases, which search strings, which dates 3 Inclusion and exclusion criteria rules for what counts, applied consistently to everything found 4 Documented screening how many sources were found, how many kept — and why Same protocol, same literature A B THE AIM: REPLICABILITY Another researcher following your protocol should retrieve substantially the same literature. Originated in medicine; now standard in management research, where the evidence base is large and mixed in quality.

Narrative vs systematic: a comparison

Narrative and systematic reviews compared A comparison in five rows, the narrative review in one column and the systematic review in the other. Selection: narrative, reviewer's judgement; systematic, explicit criteria, set in advance. Search: narrative, undocumented, flexible; systematic, documented strategy, stated databases. Replicability: narrative, low; systematic, high — the protocol can be repeated. Bias risk: narrative, cherry-picking; systematic, reduced, though not eliminated. Best for: narrative, broad or theoretical terrain; systematic, focused, answerable questions. Examples: how should we think about the employment relationship is narrative-review territory; does flexible scheduling reduce turnover in retail is systematic-review territory. For your projects, borrow the discipline: a fully systematic review is beyond a semester's work, but three habits cost about an hour: state your keywords, name your databases, and write down your criteria for keeping or discarding sources. Narrative review Systematic review Selection Search Replicability Bias risk Best for Reviewer's judgement Undocumented, flexible Low Cherry-picking Broad or theoretical terrain “How should we think about the employment relationship?” Explicit criteria, set in advance Documented strategy, stated databases High — the protocol can be repeated Reduced, though not eliminated Focused, answerable questions “Does flexible scheduling reduce turnover in retail?” FOR YOUR PROJECTS: BORROW THE DISCIPLINE A fully systematic review is beyond a semester's work — but three habits cost about an hour: state your keywords name your databases write down your criteria for keeping or discarding sources

Preparing for the review

Preparing for the review: three decisions that make a protocol Before searching, three decisions, in writing. 1, define your question or focus: not turnover, which will drown you in sources, but why frontline retail employees quit, a manageable, searchable object. 2, establish inclusion and exclusion criteria, for example peer-reviewed sources published since 2010, empirical studies of retail or service-sector employees, and English-language publications; adjust them deliberately, and record the change. 3, decide on sources and time frames: which databases you will search and how far back you will go; ten years is a common default in management, but a field with a pivotal older literature may need more. Together the three decisions are your protocol: it turns searching from wandering into method, and it is what your methods section will describe. Before searching: three decisions, in writing 1 Define your question or focus NOT “turnover” will drown you in sources BUT “why frontline retail employees quit” a manageable, searchable object 2 Establish inclusion and exclusion criteria for example: peer-reviewed sources published since 2010 empirical studies of retail or service-sector employees English-language publications Adjust them deliberately, and record the change. 3 Decide on sources and time frames which databases you will search how far back you will go Ten years is a common default in management; a field with a pivotal older literature may need more. = YOUR PROTOCOL It turns searching from wandering into method — and it is what your methods section will describe.

Searching the literature

From concepts to keywords

From concepts to keywords List the key concepts in your research question, then synonyms and related terms for each. Three concept headers with their synonyms underneath. Employee turnover: staff turnover, attrition, quit rates, retention, intention to leave. Retail workforce: frontline employees, service sector, sales staff, store employees. Causes: antecedents, determinants, predictors, drivers. Under causes, a note on the move from everyday English to academic English: authors write antecedents of turnover intention, not causes of turnover. Below, where keywords live, each with a small article whose keyword is highlighted. In the abstract: a decent signal of relevance, and more hits, so a broad first pass. In the title: far fewer hits, but almost every hit is relevant, so a precision pass. List the key concepts in your research question — then synonyms and related terms for each Employee turnover Retail workforce Causes staff turnover attrition quit rates retention intention to leave frontline employees service sector sales staff store employees antecedents determinants predictors drivers everyday English → academic English: authors write “antecedents of turnover intention”, not “causes of turnover” Where keywords live IN THE ABSTRACT a decent signal of relevance — and more hits broad first pass IN THE TITLE far fewer hits — but almost every hit is relevant precision pass

Boolean operators

Boolean operators as Venn diagrams Five panels. AND narrows, because both terms must appear: two overlapping circles, turnover and retail, with only their overlap shaded; turnover AND retail. OR broadens, because either term may appear: turnover and attrition with both circles shaded; turnover OR attrition OR quit rates. NOT, written as a minus sign in Google Scholar, excludes a term: the turnover circle is shaded except where it overlaps financial; turnover minus financial removes asset-turnover finance papers. Quotation marks search an exact phrase: quoting employee turnover finds high employee turnover in retail, but not employee on page 2 and turnover on page 9. Parentheses group the logic: in open bracket employee turnover OR attrition close bracket AND retail, the shaded area is the part of either the employee turnover circle or the attrition circle that lies inside the retail circle: either name for the phenomenon, but definitely in retail. AND narrows: both terms must appear turnover retail turnover AND retail OR broadens: either term may appear turnover attrition turnover OR attrition OR "quit rates" NOT ( − ) excludes a term turnover financial turnover -financial removes asset-turnover finance papers “Quotation marks” search an exact phrase "employee turnover" ✓ …high employee turnover in retail… ✗ “employee” on page 2, “turnover” on page 9 ( Parentheses ) group the logic ("employee turnover" OR attrition) AND retail either name for the phenomenon — but definitely in retail employee turnover attrition retail

Building a search string

Building a search string from concept blocks Running example: the Meridian team wants literature on why retail employees quit. Three dashed blocks, one per concept. Employee turnover: quote employee turnover, OR, quote staff attrition. Retail: retail. Causes: causes OR antecedents. The blocks are joined by AND: OR inside a block, AND between blocks. Assembled, the string reads: open bracket, quote employee turnover, OR quote staff attrition, close bracket, AND retail AND, open bracket, causes OR antecedents, close bracket. This is the skeleton of every competent search string. RUNNING EXAMPLE The Meridian team wants literature on why retail employees quit. concept: employee turnover "employee turnover" OR "staff attrition" concept: retail retail concept: causes causes OR antecedents AND AND OR inside a block — AND between blocks ("employee turnover" OR "staff attrition") AND retail AND (causes OR antecedents) the skeleton of every competent search string

Building a search string: three attempts

Building a search string in three attempts Three attempts at a search string for Meridian's question. Attempt 1: employee turnover. Verdict: millions of hits, hopelessly broad — but informative: the term is far too broad on its own. Attempt 2: "employee turnover" AND retail, changed by + phrase in quotes, + AND a second concept. Verdict: better, but swamped by US supermarket studies, mixing descriptions with explanations. Attempt 3: ("employee turnover" OR "staff attrition") AND retail AND (causes OR antecedents), changed by + OR synonyms, + AND a third concept. Verdict: focused, on studies that explain quitting, not merely count it. No first search string is right; the skill is in the refinement: read the results critically, and adjust. 1 employee turnover MILLIONS OF HITS hopelessly broad — but informative: the term is far too broad on its own 2 "employee turnover" AND retail + phrase in quotes, + AND a second concept BETTER but swamped by US supermarket studies, mixing descriptions with explanations 3 ("employee turnover" OR "staff attrition") AND retail AND (causes OR antecedents) + OR synonyms, + AND a third concept FOCUSED on studies that explain quitting, not merely count it No first search string is right — the skill is in the refinement. Read the results critically, and adjust.

Citation searching

Citation searching, backwards and forwards A good source you have found sits in the middle of a time line running from older work on the left to newer work on the right. Backwards, or snowballing: arrows run from the source to three older documents; read its reference list, because the studies it builds on are probably relevant to you too. Forwards: arrows run from three newer documents back to the source; Google Scholar's cited by finds the studies that later built on it, and shows its impact on the field. Shortcuts: review articles, and editors' introductions to collections and special issues, where someone has already surveyed the terrain for you. A handful of good recent articles, plus citation searching in both directions, can map a literature faster than keyword searching alone. a good source you have found ← older newer → the studies it builds on BACKWARDS — “SNOWBALLING” Read its reference list: the studies it builds on are probably relevant to you too. the studies that later built on it FORWARDS — “CITED BY” Google Scholar's “cited by” finds the studies that later built on it — and shows its impact on the field. SHORTCUTS Review articles, and editors' introductions to collections and special issues: someone has already surveyed the terrain for you. A handful of good recent articles, plus citation searching both ways, can map a literature faster than keywords alone.

Evaluating and reading sources

Not all sources are equal

A rough hierarchy of credibility for academic sources Five tiers, from most to least credible for academic purposes. 1, Peer-reviewed journal articles: vetted by independent experts before publication. 2, Scholarly books and chapters: reviewed by editors and academic publishers, less formally. 3, Grey literature: consultancy, industry and government reports: no academic review. 4, Quality press: journalistic standards, but written for news value, not cumulative knowledge. 5, Blogs, corporate content, Wikipedia: no systematic quality control at all. An arrow runs down beside the tiers: descending, the burden of proof grows; the lower the source, the harder you interrogate it before relying on it. Wikipedia: orient yourself and mine its references, but never cite it. Underneath, in a dashed box: the hierarchy is a starting point, not a verdict; weak journal articles exist, and so do excellent reports. 1 Peer-reviewed journal articles vetted by independent experts before publication 2 Scholarly books and chapters reviewed by editors and academic publishers, less formally 3 Grey literature consultancy, industry and government reports: no academic review 4 Quality press journalistic standards, but written for news value, not cumulative knowledge 5 Blogs, corporate content, Wikipedia no systematic quality control at all Descending, the burden of proof grows The lower the source, the harder you interrogate it before relying on it. Wikipedia: orient yourself and mine its references — never cite it. A starting point, not a verdict: weak journal articles exist — and so do excellent reports.

What peer review does — and does not — guarantee

What peer review does, and does not, guarantee Top: a manuscript goes to independent experts, usually anonymised, who scrutinise its theory, methods and conclusions; the outcome is accept, revise or reject. Weak work is filtered out or revised, which is why journal articles anchor a review. Middle: peer review is a quality floor, not a truth guarantee. Reviewers cannot re-run the study: They read the manuscript; honest errors, and occasionally fraud, pass through. Journals vary enormously: From rigorous to barely reviewed: predatory journals publish almost anything for a fee. Which journal always matters. Findings can be superseded: Before leaning on a study, check whether later work confirmed or challenged it. Bottom, citation counts: on the plus side, influence is evidence of scrutiny, since heavily cited work has been read, used and tested; on the minus side, new work has not had time to be cited, and some work is cited only to be criticised. Citations measure attention, not agreement. manuscript Independent experts usually anonymised; they scrutinise its theory, methods and conclusions accept revise reject Weak work is filtered out or revised: this is why journal articles anchor a review. A quality floor, not a truth guarantee Reviewers cannot re-run the study They read the manuscript; honest errors, and occasionally fraud, pass through. Journals vary enormously From rigorous to barely reviewed: predatory journals publish almost anything for a fee. Which journal always matters. Findings can be superseded Before leaning on a study, check whether later work confirmed or challenged it. CITATION COUNTS + influence is evidence of scrutiny: heavily cited work has been read, used and tested − new work has not had time to be cited; some work is cited only to be criticised Citations measure attention, not agreement.

Grey literature in management research

Grey literature in management research In management, much current evidence appears outside academic journals: consultancy studies, for example from McKinsey or Deloitte; industry-association reports; government statistics. Its value: it is current, since a consultancy can publish in weeks what a journal takes two years to print; it is practical, addressed to managerial questions; and it is often built on data academics cannot access. Its dangers: no peer review, with methods often undisclosed; commercial interest, since a report may exist to sell the consultancy's services; and samples and definitions chosen to flatter the headline. The working rule: grey literature supplies context and currency, such as the state of the market and the scale of the problem; academically reviewed work carries your arguments. Always ask: who produced this, for whom, and what do they gain from its conclusions? Grey literature: in management, much current evidence appears outside academic journals consultancy studies (McKinsey, Deloitte) industry-association reports government statistics ITS VALUE + Current: a consultancy can publish in weeks what a journal takes two years to print + Practical: addressed to managerial questions + Often built on data academics cannot access ITS DANGERS − No peer review: methods often undisclosed − Commercial interest: a report may exist to sell the consultancy's services − Samples and definitions chosen to flatter the headline THE WORKING RULE Grey literature context and currency the market, the scale of the problem Academically reviewed work your arguments ALWAYS ASK Who produced this, for whom — and what do they gain from its conclusions?

Questions to ask of any source

Questions to ask of any source Six questions to ask of each publication you consider keeping. 1, What question or problem? Is it actually close to yours, or merely wearing similar keywords? 2, Which key concepts, and how defined? Definitions bite: one study counts all leavers, another only voluntary quits. 3, Which theories, models and methods? Established frameworks, or an innovative approach? 4, What results and conclusions? Stated precisely, not as you vaguely remember them. 5, How does it relate to the field? Does it confirm, add to, or challenge established knowledge? 6, What strengths and weaknesses? Every study has both; note them while you read. Behind them all, Babbie's advice to read as a sceptical consumer: how do they know that? If the source gives you the means to answer, it is probably worth keeping. 1 What question or problem? Is it actually close to yours, or merely wearing similar keywords? 2 Which key concepts, and how defined? Definitions bite: one study counts all leavers, another only voluntary quits. 3 Which theories, models and methods? Established frameworks, or an innovative approach? 4 What results and conclusions? Stated precisely, not as you vaguely remember them. 5 How does it relate to the field? Does it confirm, add to, or challenge established knowledge? 6 What strengths and weaknesses? Every study has both; note them while you read. BABBIE: READ AS A SCEPTICAL CONSUMER “How do they know that?” If the source gives you the means to answer, it is probably worth keeping.

Reading efficiently: abstract-first triage

Reading efficiently: abstract-first triage A funnel. At the top, at Meridian, a Scholar search on retail turnover returns 400 plausible hits. Stage 1, title and abstract: is it actually about your question? One minute. Most sources exit here, and that is the system working, not a shortcut. Stage 2, introduction and conclusion: what was found, and does it matter to you? About five minutes. More sources exit. Stage 3, full text: read closely, with structured notes, only for the survivors. Out of the bottom come perhaps 25 sources worth reading properly, a week's work instead of a semester's. Notes go under standard headings: question, method, sample, findings, limitations, and relevance to your question, so the sources are comparable when you write. a Scholar search on retail turnover: 400 plausible hits 1 · Title and abstract Is it actually about your question? one minute ✗ Most sources exit here: that is the system working, not a shortcut. 2 · Introduction and conclusion What was found — and does it matter to you? about five minutes ✗ More sources exit. 3 · Full text read closely, with structured notes only for the survivors ≈ 25 worth reading properly a week's work instead of a semester's NOTES UNDER STANDARD HEADINGS question method sample findings limitations relevance to your question so the sources are comparable when you write

Organising, synthesising and citing

Organising what you find

Organising what you find: connections between sources A map of sources as dots, from earlier on the left to later on the right, with five kinds of connection added one at a time, each with its guiding question. Trends and patterns: an arrow runs from a cluster of earlier studies to a later one; do approaches or findings shift over time? Themes: two clusters are circled, labelled pay and supervision; which questions and concepts recur across the literature? Debates and contradictions: a red zigzag marked not equal joins two sources; where do sources disagree, and why? Assess their methods and samples; unresolved conflicts are candidate gaps. Pivotal publications: one large dot with lines to several later studies; which studies changed the direction of the field? Gaps: an empty dashed circle, with an arrow to your research question; what is missing, under-studied, or methodologically weak? At Meridian: little is known about turnover drivers in Central European multi-site retail; the review ends, and the study begins, in the same sentence. earlier later trend Trends and patterns Do approaches or findings shift over time? pay supervision Themes Which questions and concepts recur across the literature? ≠ Debates and contradictions Where do sources disagree, and why? Assess their methods and samples; unresolved conflicts are candidate gaps. pivotal Pivotal publications Which studies changed the direction of the field? gap your research question Gaps What is missing, under-studied, or methodologically weak? AT MERIDIAN “Little is known about turnover drivers in Central European multi-site retail.” The review ends, and the study begins, in the same sentence.

Structuring the review

Four ways to structure a literature review Four cards, each with a small schematic. Chronological, a time line with two turning points marked: organised by development over time; best when the field has clear turning points and key debates. Thematic, marked most common: sections on pay, supervision, engagement and labour-market conditions, as in a turnover review; organised by topics and sub-questions; best when distinct aspects need separate treatment. Methodological, a qualitative and a quantitative column: organised by research approach; best when qualitative and quantitative studies reach different conclusions. Theoretical, theory A and theory B both pointing at the same phenomenon: organised by competing frameworks; best when rival theories explain the same phenomenon. Never acceptable: source by source; one paragraph per article, in the order you read them, is not a structure but the absence of one. Chronological turning points ORGANISED BY development over time BEST WHEN the field has clear turning points and key debates Thematic most common pay supervision engagement labour-market conditions e.g. a turnover review ORGANISED BY topics and sub-questions BEST WHEN distinct aspects need separate treatment Methodological qualitative quantitative ORGANISED BY research approach BEST WHEN qualitative and quantitative studies reach different conclusions Theoretical theory A theory B phenomenon ORGANISED BY competing frameworks BEST WHEN rival theories explain the same phenomenon NEVER ACCEPTABLE: SOURCE BY SOURCE One paragraph per article, in the order you read them, is not a structure — it is the absence of one.

Synthesis over summary

Synthesis over summary On the left, the cardinal sin: a sequence of summaries, one paragraph per author: Smith found X; Jones found Y; Brown found Z. The synthesis is left to the reader. On the right, a synthesis, which groups, compares and adjudicates, in five moves: 1, Establish the themes that emerge from the literature. 2, Discuss them — especially conflict and difference in how they are understood. 3, State the agreed knowledge, if any, and the nature of the disagreements. 4, Take a position yourself, from those available in the literature. 5, Conclude with your research question, and how it follows from the gaps. Underneath, what synthesis sounds like, organised around ideas rather than authors: the field agrees that supervision matters, but splits on whether its effect runs through job satisfaction or through perceived fairness. ✗ A sequence of summaries Smith found X. Jones found Y. Brown found Z. one paragraph per author: the synthesis is left to the reader ✓ A synthesis: groups, compares, adjudicates 1 Establish the themes that emerge from the literature 2 Discuss them — especially conflict and difference in how they are understood 3 State the agreed knowledge, if any, and the nature of the disagreements 4 Take a position yourself, from those available in the literature 5 Conclude with your research question, and how it follows from the gaps SYNTHESIS SOUNDS LIKE THIS — IDEAS, NOT AUTHORS “The field agrees that supervision matters, but splits on whether its effect runs through job satisfaction or through perceived fairness.”

Citing and avoiding plagiarism

Citing sources and avoiding plagiarism Top, a workflow: find a source; record full details at once, meaning author, year, title, outlet and pages; use a reference manager such as Zotero, which is free and formats citations automatically; cite in APA style, the programme's standard. Below, does it need a citation? Quoting their words: cite. Using their ideas or findings: cite, because uncredited it is plagiarism, exactly as copying words is. Paraphrasing, their idea in your own words: cite, because the idea is still theirs. Changing three words in their sentence: not paraphrase but copying with cosmetics, which is plagiarism. Read, but not used in the write-up: no citation needed. This semester, the habit of recording and crediting sources matters more than the format; referencing mechanics come later. Find a source Record full details at once author · year · title · outlet · pages Use a reference manager Zotero: free, formats citations automatically Cite in APA style the programme's standard Does it need a citation? Quoting their words ✓ Cite Using their ideas or findings ✓ Cite — uncredited, it is plagiarism, exactly as copying words is Paraphrasing: their idea in your own words ✓ Cite — the idea is still theirs Changing three words in their sentence ✗ Not paraphrase: copying with cosmetics — plagiarism Read, but not used in the write-up – No citation needed This semester, the habit of recording and crediting sources matters more than the format: referencing mechanics come later.

Conclusion

Conclusion

  • A literature review is critical engagement with what is already known — what we know, and how we know it — not an annotated reading list
  • The systematic review’s discipline — stated question, documented search, explicit criteria — is a standard worth borrowing even for modest projects
  • Searching is a craft of iteration: concepts → keywords → Boolean strings → refinement, with every step recorded
  • Sources must be weighed, not counted: peer review anchors the hierarchy, grey literature adds currency and context but never carries the argument alone
  • Read by triage, organise by theme, write by synthesis, and cite everything you use
  • Meridian’s team now knows what the field says about turnover; in the next session we ask what existing data it can reuse before collecting any of its own
  • Questions and discussion are welcome

Exercise

Today’s exercise: Weighing the evidence

QR code linking to the exercise worksheet

bdstanley.netlify.app/social-research-methodology-5-exercise

Model answers: ranking the sources

Model answer to Task 1: a defensible ranking of the seven sources A defensible ranking, one among several; credit the feature named, not the sequence. 1, source B, journal article (Kowalczyk & Ferreira): peer-reviewed; a three-wave panel of 1,842; actual quitting, not intention; controls for rival explanations. 2, source F, scholarly book chapter (Novak): editorial review; maps four decades of research; 2019 is not current; others' evidence, not new findings. 3, source G, industry-association report (Retail Workforce Barometer): method disclosed (240 member firms); no independent review; a lobbying interest; over-represents firms willing to respond. 4, source A, consultancy white paper (Hartwell & Grey): current; 4,000 employees in 12 markets; sells its own framework; method undisclosed; the 68% is intention, not behaviour. 5, source E, newspaper article (The Business Courier): journalistic standards; anecdote plus authority: chase the ministry statistics, cite those. 6, source C, practitioner blog post (Lem): a hypothesis worth testing; no quality control; unverifiable personal experience. 7, source D, wikipedia entry (“Employee turnover”): a gateway: definitions, references; anyone can edit; flags missing citations: never cite it. Dashed yellow brackets mark where reasonable groups diverge: places 1 and 2, F first is defensible if argued, for its synthesis value; places 3 and 4, G against A is the classic dispute, and either order is fine if the sales motive is acknowledged; places 5 and 6, C against E, either order with reasoning. One defensible ranking among several: credit the feature named, not the sequence 1 B Journal article Kowalczyk & Ferreira + peer-reviewed; a three-wave panel of 1,842; actual quitting, not intention; controls for rival explanations 2 F Scholarly book chapter Novak + editorial review; maps four decades of research − 2019 is not current; others' evidence, not new findings 3 G Industry-association report Retail Workforce Barometer + method disclosed (240 member firms) − no independent review; a lobbying interest; over-represents firms willing to respond 4 A Consultancy white paper Hartwell & Grey + current; 4,000 employees in 12 markets − sells its own framework; method undisclosed; the 68% is intention, not behaviour 5 E Newspaper article The Business Courier + journalistic standards − anecdote plus authority: chase the ministry statistics, cite those 6 C Practitioner blog post Lem + a hypothesis worth testing − no quality control; unverifiable personal experience 7 D Wikipedia entry “Employee turnover” + a gateway: definitions, references − anyone can edit; flags missing citations: never cite it F first is defensible if argued: its synthesis value G vs A, the classic dispute: either order, if the sales motive is acknowledged C vs E: either order, with reasoning

Model answers: what each source is good for

Model answer to the managerial-context question: what each source is good for Credibility for an academic review and usefulness to a decision-maker are different tests. The review's arguments must rest on B and F, the journal article and the book chapter. Context for the Meridian board: A and G are what a board finds most immediately persuasive. G gives current sector benchmarks, since Meridian's 27% only means something against a sector baseline; A gives multi-market scale and an actionable framing; E gives salience and the wage-competition threat from e-commerce warehouses, relevant to Meridian's own e-commerce expansion; C suggests a cheap practice, stay interviews, worth piloting even before research concludes. Other legitimate uses in the review: E is a signpost, so chase the labour-ministry statistics it cites and cite those instead; C is a hypothesis-generator, since a research team could test whether exit interviews mislead employers; D is a gateway, to orient with its definitions and mine its references, but never to cite. Credibility for an academic review and usefulness to a decision-maker are different tests CARRIES THE REVIEW'S ARGUMENTS B and F the journal article and the book chapter: the review's arguments must rest on them CONTEXT FOR THE MERIDIAN BOARD A and G: what a board finds most immediately persuasive G current sector benchmarks: Meridian's 27% only means something against a sector baseline A multi-market scale and an actionable framing E salience, and the wage-competition threat from e-commerce warehouses — relevant to Meridian's own e-commerce expansion C a cheap practice, stay interviews, worth piloting even before research concludes Other legitimate uses in the review E A signpost Chase the labour-ministry statistics it cites, and cite those instead. C A hypothesis-generator Do exit interviews mislead employers? A research team could go on to test it. D A gateway Orient with its definitions and mine its references — never cite it.

Model answers: a model search log

Model answer to Task 2: a search log in three versions A model search log; any coherent iteration earns credit if each refinement is motivated by inspecting the results, and hit counts vary, so do not mark them against a target. Version 1: employee turnover retail; approximately hundreds of thousands hits. Unusable breadth: the words are matched separately, so results include finance papers on asset turnover and anything mentioning retail in passing. Version 2: "employee turnover" AND retail; approximately tens of thousands hits. Quotation marks force the exact phrase: the single highest-value refinement. On-topic now, but dominated by descriptive and non-European studies, and still missing work that says “attrition” or “quit”. Version 3: ("employee turnover" OR "staff attrition" OR "turnover intention") AND retail AND (antecedents OR causes OR "supervisory support"); approximately a few thousand hits. OR bundles the field's synonyms; the third AND block tilts results from describing turnover to explaining it. The first page is now dominated by studies that would survive abstract triage. Any coherent iteration earns credit, if each refinement is motivated by inspecting the results. Hit counts vary: do not mark them against a target. Search string Approx. hits Commentary 1 employee turnover retail Hundreds of thousands Unusable breadth: the words are matched separately, so results include finance papers on asset turnover and anything mentioning retail in passing. 2 "employee turnover" AND retail Tens of thousands Quotation marks force the exact phrase: the single highest-value refinement. On-topic now, but dominated by descriptive and non-European studies, and still missing work that says “attrition” or “quit”. 3 ("employee turnover" OR "staff attrition" OR "turnover intention") AND retail AND (antecedents OR causes OR "supervisory support") A few thousand OR bundles the field's synonyms; the third AND block tilts results from describing turnover to explaining it. The first page is now dominated by studies that would survive abstract triage.

Model answers: refinements and common faults

Model answer to Task 2: refinements to credit and faults to correct At the top, the rule: OR within a concept, AND between concepts. Correct: open bracket, quote employee turnover, OR attrition, close bracket, AND retail, meaning either name, but in retail. Wrong: turnover OR retail, because OR between concepts broadens disastrously. Further refinements worth crediting: a date filter, for example since 2015, for currency; minus financial or minus asset, to remove residual finance hits; intitle colon turnover, to restrict a term to titles; frontline OR quote service sector, to sharpen the population. Common faults to correct: treating hit counts as the goal, when the test is first-page quality, not smallness; stacking AND terms until nothing survives; writing OR between concepts instead of within them; refining without saying why, when the log's final column is where the learning is. OR WITHIN A CONCEPT, AND BETWEEN CONCEPTS ✓ ("employee turnover" OR attrition) AND retail either name, but in retail ✗ turnover OR retail OR between concepts broadens disastrously FURTHER REFINEMENTS WORTH CREDITING A date filter (e.g. since 2015), for currency -financial or -asset to remove residual finance hits intitle:turnover to restrict a term to titles frontline OR "service sector" to sharpen the population COMMON FAULTS TO CORRECT Treating hit counts as the goal: the test is first-page quality, not smallness Stacking AND terms until nothing survives Writing OR between concepts instead of within them Refining without saying why: the log's final column is where the learning is

Study guide

Full summary of this session, for revision: Conducting a literature review

QR code linking to the session handout

bdstanley.netlify.app/social-research-methodology-5-handout