Create your own awesome maps

Even on the go

with our free apps for iPhone, iPad and Android

Get Started

Already have an account?
Log In

Methoden & Technieken van Wetenschappelijk Onderzoek A1&2 by Mind Map: Methoden & Technieken  van
Wetenschappelijk  Onderzoek
4.0 stars - 4 reviews range from 0 to 5

Methoden & Technieken van Wetenschappelijk Onderzoek A1&2

General introduction


The Nature of Social Scientific Research, wat?, Academic research on topics, related to questions, relevant to social scientific field, Waarom?, A gap in our knowledge/ understanding of what goes on in society, inspiratiebronnen, society itself, ‘Burning social problem’, Research opportunity, Personal interest, literature/theory, gap, Unsolved issue, Inconsistency, Link to Theory is critical!, otherwise ‘only fact-finding’ => empiricism, Deductief (p10), from theory => Data, Explicit hypothesis to be tested, often quantitative research, Inductief, Depart from data => theory, Generalizable patterns based upon observations, Usually cyclic iterative (wisselwerking theorie en data), often qualitative research, Grand theory vs middle range theory, Grand Theory, Darwin – Survival of the fittest, Limited usefullness => highly abstract, How to research such theory?, Middle Range Theory, Social behaviour-reward theory, Useful for empirical research, Limited domain, To develop research questions, Quantitative vs qualitative research, Quantitative research, measurement of social variables, Numerical & statistical data, Deduction: theory testing, Epistemological: positivism - realism, Ontology: Objectivism, Viewpoint of researcher, Qualitative research, Understanding of social meaning given by actors, Words, texts & stories, Induction: theory creation, Epistemological: interpretivism, Ontology: Constructionism, Viewpoint of participants, Epistemology – kennisleer, How do you gather knowledge?, 1. positivism, “What you see, is what you get”, No interpretation, Knowledge through observation, Wetmatigheid, Quantitative approach, Example: “Measure focus on detail amongst gay men”, 2. (Critical) realism, “You can see what there is, but study is required”, Certain level of interpretation – mechanisms behind it, knowledge through understanding, Quantitative approach, Example: “Gay men – risk behaviour”, 3. Interpretivism, The subjective meaning of social action, High level of Interpretation, Symbolic interactionism (in culture), Qualitative approach, Example: The Rainbow Flag, Ontology – wereldbeeld, How is the world constructed?, Objectivism, Quantitative approach, Reality is pre-given – little revision possible, “World is an external fact beyond our influence”, Essentialist in nature, Example: American fast food culture, Constructionism, Qualitative approach, World is not a given fact” – non-essentialist, Continuously constructed & reconstructed, People can change the world by act & speech, Example: UK fast food culture => kerrie, Values – Practical considerations, Values: Personal beliefs or the feelings of researcher, 1. Pretention of objectivity, value free, objective, 2. Self-reflective, exhibit reflexivity, underdog position, 3. Conscious partiality, preconceptions always there, example feminism, Practical considerations, Research question –> research design, Prior literature available?, Topic, deviant activities or sensitive issues =>qualitative research, Cost/funding, Time


Research Designs, 1. What is a Research Design?, Based upon Research Question, “A framework to guide data collection & analysis”, Differs from:, A research strategy (H1), Qualitative, Quantitative, A research method (H17-22), Participatory Observation, Questionnaire, Discourse Analysis, etc., 2. Five research designs:, A. Experimental design, In Search of causality, Is there a causal relationship between X and Y?, X – independent variable, Y – dependent variable, If X changes, is there significant change in Y?, how?, 1. Manipulate X - all other variables constant, 2. Random assignment of subjects, To Experimental group, To Control group, 3.Control group, important! danger of no control group:, Due to events instead of manipulation (history), Test effect or real impact (sensitivity for testing), Reliability of the test (instability of testing), Mortality/attrition, Maturation, Often practical restrictions to laboratory experiment. Alternatives:, Quasi experiment, If random distribution over an experimental or control group is not ethical or impossible, Example: Physical punishment in teaching – Exam results, Field experiment, Ecological validity issue, Natural environment, Natural experiment, No direct intervention, Less often used, B. Cross-sectional design, ‘Study of multiple variables among various subpopulations’, Looking for patterns of association, Direction of relationship difficult to verify – no causality, instead:, Correlation between variables, Strong/weak, Positive/negative, Coherence, Example: Survey on ‘Mister Perfect’, Set of variables: physical features, educational level, empathy, ect., Women all over the world, Advantages, Addresses also variables not to be manipulated, Example: Gender/Age + Medication (aids treatment), Still random in its selection, Disadvantages, Survey – questionnaires => Ecological Validity, niet geheel de waarheid invullen, Only one point in time – Registration, 3rd variable involved & overlooked?, C. Longitudinal design, Study of the same sample on various occasions in time, Focus on Social change over time, 2 types of longitudinal design:, Cohort study, A shared characteristic, Example: Date of birth, Aging effects, Panel study, National level, Random selection – different people, Difference in ages, advantage, Evolution in time, Disadvantage, Often little planning – when T2, T3, …Tn, Attrition, sterfte, emigratie, Panel conditioning, besef van onderzoek, Note. Panel study often not LTD because often randomly, Actually repeated cross-sectional design, D. Case study design, Detailed & intensive analysis of one case, A specific person, event, organization or community, Qualitative research, The case is the focus, Types of case, 1. Critical – well developed theory/hypothesis, 2. Unique, extreme – Example: CEO dating, 3. Revelatory – e.g. Dating through FB, 4. Exemplifying – e.g. Dating in office, Advantages, Detailed focus, Elucidate unique features, Disadvantage, Limited external validity, Difficult to extrapolate to different situations, Often not the goal of the design, rather understanding, E. Comparative design, The same method to compare two or more meaningfully contrasting cases, Qualitative or quantitative, Often cross-cultural comparison, advantage, Some level of comparison, Enables theory-building, Beyond cultural specificity, Disadvantage, Translating research instruments, Finding comparable samples, Learn strengths/weaknesses of each of the five designs!, Evaluate according to 6 criteria, 3. Criteria for Social Research, 1. Reliability, Are the measurements consistent?’, Every time the same result, Betrouwbaarheid, Example: IQ test, 2. Replication, ‘Can the study be easily replicated?’, The procedures should be clearly enough explained, herhaalbaarheid, 3. validity, a. Measure validity, ‘Do you really measure the key concepts?’, Each variable should refer to the actual concept, Appropriate tool to measure?, Example: IQ test – intelligence, b. Internal validity, ‘Is there a true causal relationship between X and Y?’, No other possible explanations?, Benefit of a labo-experiment – control of environment, Example: Dance Moves – Partner choice, c. External validity, Can the results be generalized beyond the study?, Generalizeerbaar, how?, Representative sample (steekproef), Random selection, Example: Mister Perfect survey, d. Ecological validity, ‘Do findings apply to everyday settings?’, Test-environment vs natural environment, Importance of naturalness, Choice between labo-experiment vs field experiment, Example: Labo vs Music Club, These criteria primarily in quantitative research


Research Question, Planning a Research Project (P. 71), 1. Define research area, 2. Look at specific gap, 3. Various research questions, 4. Selection of key research question, Formulating Research Questions, Research questions help to guide:, the literature search, the data collection, the analysis of data, the writing of data, Avoid side tracks, to remain focused, Criteria for Suitable Research Questions, clearly formulated: What, who, where?, Should be researchable, connection with theory and research, sub-questions logically linked to each other, original contribution, Neither too broad nor too narrow


Literature Review, 1. Why literature review?, “Not to simply reinvent the wheel”, Affirming credibility as researcher, Knowledgeable, Not simply reproduce but interpret, Critical attitude, Revise and refine research questions, Develop argument about the relevance of your research & where it leads, More specifically:, What is already known about this area?, What concepts & theories are relevant?, What kind of research methods and research strategies have been employed?, Any controversies, inconsistencies, unanswered questions?, 2. Getting most from your reading (p. 102), 1.Bibliography, Take good note otherwise risk of Plagiarism, = Het overnemen van andermans werk en dat laten doorgaan voor eigen werk, 2. Critical reading skills, Not all articles- only those to prove your argument, 2 risks:, 1. Avoid gaps:Don’t forget critical literature, 2. Avoid Faulty articles: Carefulness – Quality check, 3. Continue to read, 4. Develop a story, 1. Constructing intertextual coherence, Synthesized coherence, unusual combination of different types of literature, Progressive coherence, Logical combination of types of literature, Built upon former theories, 2. Problematizing the situation, Incomplete, Inadequate, Incommensurate, 3. Types of literature review, 1. Systematic Review, All relevant research carried out on the topic, Exhaustive – Accumulation, Progress of study on a specific topic, nadelen, Bureaucratization of literature, Not suitable for all qualitative studies, Example: parenting – 488.888 hits, 2 types:, Meta-analysis, Summary of quantitative tests, See if a variable has effect, Repetition of studies, Correct certain sampling & non-sampling errors, Danger: “file drawer problem”, Medical tests. Example: HIV-aids research, Meta-ethnography, Systematic!, Interpretative synthesis of qualitative research, No generalizations but translations of studies, Comparative and synthesis, Example: ‘All ethnographic work on motherhood in 21st century’, 2. Narrative Review, Only the most contributing ones, Interpretative researchers, Generation of knowledge: inductive, Focus on aligned topics as well, Less explicit criteria, Research question revisable, Identify key debates + linking them, nadelen, Lack of thoroughness, Tendentious, 4. Practical tips for literature review (p. 95-101), 5. Referencing your work (p.101-107)

(Qualitative) research methods:


The Nature of Qualitative Research, 1. What is qualitative research?, Def: A research strategy that emphasizes on words rather than quantification in the collection & analysis of data”, Inductive in nature, Constructionism, Interpretivism, Different research methods:, Ethnography / participatory observation, Qualitative interviewing, in-depth, semi-structured, unstructured, Focus groups, Discourse / conversation analysis, Documentary analysis, 2. Main steps in qualitative research? (p. 370), 3. Theory – Data & Concepts, Inductive in nature, Theory-building, Different types of concepts:, 1. Definitive concepts, Fixed indicators, Key features of phenomena, Straitjacket on the social world, Quantitative, Example: IQ, 2. Sensitizing concepts, Loose concept as a sense of guidance, Varied forms of phenomena, Changes throughout research, Too broadly => no guideline (love), Too narrow => little innovation (kissing), Qualitative, Example: Terrorism, 4. Sampling, 1. Probability sampling, Population is too large – Selection, Random, Quantitative research, except when:, 1. Importance to generalize / replication, 2. Research question doesn’t require specific group, 2. Purposive sampling:, Selection in function of the research question, No random approach – purposive, Theory-Oriented, 3 subtypes:, 1. Snowball sampling, Starts with one entry-point, Sample gradually grows, 2. Theoretical sampling, Driven by emerging theory, Theoretical saturation: Sufficient information per relevant category, 3. Convenience sampling, 5. Criteria for qualitative research, Degree of realism => 3 popular schools, 1. Adapting quantitative criteria, 2. Mid-position, 3. Alternative criteria – ‘Trustworthiness’, A. Credibility (Internal validity), Credible research, Results are ‘acceptable’ for researchers & others, Acceptable understanding of social meaning, Example: “All Indians are terrorists”, 2 Methods:, 1. Respondents’ Validation, 2. Triangulation, B. Transferability (External validity), No random sampling – how to generalize?, Thick description of your case, C. Dependability (Reliability), Keep records of all phases of research, Auditing process, Example: “I have spent 3 months in the field, did research in 5 villages. I have talked to 350 people often by means of semi-structured interviews, yet occasionally also unstructured, in-depth interviews”, D. Confirmability (Objectivity), Shown to have acted in good faith, No personal values guide the research, + Authenticity, Actively involve respondents in your research, Action Research, + Relevance, For the ongoing academic debate, Relevance for policy, 6. Key features qualitative research, 1. seeing through the eyes of those studied, Taking the role of the other, Understanding the meanings people give their worlds, Unexpected findings, 2. Description and emphasis on context, Detailed account of the social setting, ‘Thick descriptions’ of what is going on, 3. Emphasis on social process, How patterns of events unfold over time, Social worlds characterized by change and flux, 4. Flexibility and limited structure, No ‘prior contamination’ by rigid schedules, Sensitizing concepts, 5. Concepts and theory emerge from the data, 7. qualitative research critiques, 1. Too subjective => researcher decides what to focus on, 2. Difficult to replicate => unstructured format, 3. Problems of generalization => samples not ‘representative’ of all cases, 4. Lack of transparency => often unclear what researcher actually did, 8.Differences in relation to quantitative research, See page 393, 9.Similarities in relation to quantitative research, Both about data reduction, Seek to answer research questions, Relating data analysis to research literature, Concerned with variation, Avoid deliberate distortion


Ethnography & Participatory Observation, 1. What is ethnography, Take part in people’s everyday life in a particular social setting for an extensive period of time, Interest in the way people live & how they give meaning to that life, Focus on behaviour, culture, rituals, norms, etc., Participatory observation, informal interview, ‘hanging around’, Writing a detailed account of a particular setting, Example: ‘Gang leader for a day’ - Venkatesh, Note. Ethnography is both a research method & the written product of research, 2. How to get access?, A. overt vs. covert role as ethnographer, Overt role, what?, Participants aware of researcher’s intentions, Research goal is immediately disclosed, Why?, practical considerations, Less anxiety - No part to play, Ability to take notes, Ethics, Challenges?, Have to negotiate access – in closed settings, Often “strategic planning, hard work and dumb luck”, Importance of Entry points:, Gatekeepers:, Define your level of access, Need their support/approval, Sponsors:, Often your buddy throughout your research, Prepared to give information, Covert role, What?, Particiants are unaware of the researcher’s identity, Disguised identity, or vague about real intentions, Why?, Little negotiations to access, No reactivity, Disadvantages?, Practical difficulties in taking notes, Cannot use other methods, Anxiety about ‘blowing your cover’, Ethical problems:, Deception, Informed consent, Ethnography is mostly overt => Practical considerations & ethics, Throughout research: Frequently shift in overt & covert roles, B. open/public vs. closed settings, Open/public settings, Public spaces such as streets, neighborhoods, In principle little negotiation required to access the place, Possible hostility & distrust from members, Key informants:, Sponsor/gatekeeper, Danger of undue reliance, Solicited <=> unsolicited accounts, Closed settings, Private Places such as police corps, prisons, schools, Require substantial amount of negotiation to access, Thrust-building critical, Key role of gatekeepers, Restrictions to ability to get around, 3 types of collection:, 1. Through personal network - sponsor, 2. Research bargain – gatekeeper, 3. ‘Hanging around’, How to keep ensuring access?, 1. Entering the setting, 2. Gaining access to individuals, Credibility as ethnographer:, To gain trust of participants, Emphasize your knowledge & understanding of the setting, Pass tests of competence/insider knowledge, Participate/ take part in group activities, Role of key informants, Alliances, Backup in times of need, 3. Level of participation, 1. Complete observer => No interaction at all, 2. Observer-as-participant => Limited level of involvement, 3. Participant-as-observer => Participate in an overt role – others are aware of your role as researcher, 4. Participant => Covert role – others are UNAWARE of your research intentions, Always evaluate your own position!!, 4. Purposive sampling, In function of the research question, Theory-Oriented, 3 subtypes:, 1. Convenience sampling, 2. Snowball sampling, Starts with one entry-point, Sample gradually grows, 3. Theoretical sampling:, Driven by emerging theory, Discover variation in categories & then densify these categories, Ongoing process to generate theory, See page 416, Not merely sampling of people, but also of time units & places, Theoretical saturation:, When categories are dense enough, no more new data per category is being found, 5. Taking field notes, Write down notes as soon as possible after events, Be vivid and clear – detailed descriptions, Danger of dictaphone or taking notes, Three types of field notes:, 1. Mental notes, 2. Jotted notes, 3. Full field notes, 6. When & how to end?, 1. Practical considerations, End of sabbatical leave, Funding runs out, Personal/family commitments, Deadlines, 2. Emotional pressure, situation too risky, 3. Situation seizes to exist, 4. Reached the theoretical saturation point, 7. The rise of visual ethnography, 2 types of materials:, Extant visual materials, Items that already exist, Example: personal photographs, Research-driven visual materials, Created at the request of the researcher, Goal: aide memoires, data sources, or prompts for discussion, Research position:, 1. Realist position: images directly represent situation, 2. Reflexive position: researcher’s influence on what is portrayed & how images are interpreted


Interviewing in Qualitative Research, 1. What is qualitative interviewing?, Less structured/standardized then Structured interview, Take the participant’s viewpoint, Encourage ‘rambling’ off the topic, Rich, detailed answers, Aim to understand rather than to generalize, 2. Types of qualitative interviewing, Unstructured interview, Few, loosely defined topics, Open-ended questions, Conversational style, Semi-structured interview, List of specific topics to cover: interview guide, Flexible question order & phrasing, Life history interview, Subject looks back to their entire life, How they understand their own life, Unravel how life events unfolded, Types:, naturalistic, researched, reflexive, Oral history interview, Subject reflects on specific events in the past, Testimonies of ordinary people, Risk of memory lapses & distortions, 3. How to do qualitative interviewing, Preparing an interview guide:, 1. List of topics to be covered, 2. ‘What do I need to know’?, To answer research questions, 3. Logical but flexible order, 4. ‘Face sheet’ information:, Name, age, gender, position, 5. Quiet & private setting, 6. Good quality recording machine, Kinds of questions:, 1. Introducing: “Could you perhaps tell me a bit more about…?”, 2. Follow-up: “What do you mean...?”, 3. Probing: “You said earlier...”, 4. Specifying: “What happened next?”, 5. Direct – indirect: “Do you...?” <=> “What do most people think about…?”, 6. Structuring: “Let’s move on to…”, 7. Silence, 8. Interpreting: “Do you mean that…?”, 9. Vignette question, Question about hypothetical but realistic situation, Hypothetical questions, “Say that you....what would you do...?”, Sensitive topics, What makes a good interviewer?, Knowledgeable, Structuring, Sensitive/empathic, Talks neither too much nor too little, Open, Remembering, Aware of own position, 4. Recording & transcription, Why?, ‘WHAT but also HOW they said it’, No need for note-taking, Listening & interpreting, Less limitations of memory & intuitive glossing, Detailed & accurate record, Public scrutiny, Keep in mind!! 1 hour interview = 5-6 hours transcription, Selective transcription saves time, 5. Purposive Sampling, Select interviewees who are relevant to your research questions, Snowball sampling => Useful when no sampling frame, Theoretical sampling, Focus on emerging theory rather than statistical adequacy of sample, Theoretical saturation, Cannot predict sample size in advance!!, 6. Benefits of qualitative interviewing, Qualitative Interviewing, Resistance to observation, Reflect on past events/life course, Ethically defensible, Fewer reactive effects, Less intrusive, Longitudinal research, Greater coverage, Specific focus, Participatory Observation, Seeing through others’ eyes, Native language, Less taken-for granted ideas, Access to deviant or hidden activities, Sensitivity to context of action, Flexibility in encountering the unexpected, See pages 465-466


Focus groups, 1. What is a focus group?, Form of a group interview, Several participants & moderator, Discuss a specific issue, Study the interaction between group members, Are opinions expressed & modified through the group discussions?, 2. Use of focus groups, constructionist world view, The way people collectively construct & organize knowledge, to understand why people hold certain views, A wide range of views, Media and cultural studies, Audience reception, Example: “How do prostitutes see R&B clips?”, 3. Conducting focus groups, Recording & transcription:, To study not only what people say, but who is saying it, How was the topic discussed?, Processes of collective meanings, Nuances in language, 1 hour of interview = 8 hours of transcription, Reminder!! Can be difficult to distinguish the voices, Number of focus groups:, Average 5-10 focus groups per study, Theoretical saturation point, Socio-demographic characteristics:, Use stratifying criteria (age, gender, etc.), Large number of groups => Diverse range of viewpoints, Reminder!! more groups means more volume of data & adds to complexity of analysis, Size of focus groups:, Average 6-10 members per group, Over-recruit in anticipation of ‘no-shows’, Smaller sized groups when:, The topic is sensitive or controversial, Each person will have plenty to say, Personal, detailed accounts, Larger sized groups when:, You want to hear numerous, brief suggestions, 4. Level of moderator involvement, Be unobtrusive & non-directive, Questions:, Small number of general questions to stimulate the discussion, Intervene if discussion wanders off /long silence, Interesting points not picked up by the participants, Beginning:, Thank people for coming, Introduce yourself & the project, Outline format & procedure, Ethical issues, Collect demographic information, Name cards, Ending:, Thank people for participating, Explain what will happen to the data, Arrange any further meetings, 5. Role of group interaction, Selection of participants:, Usually on the basis of a shared experience or characteristic, Socio-demographic factors – stratifying criteria, Look for systematic variation between groups, Strangers versus ‘natural groups’:, Which group makes it easier to discuss the topic?, Natural groups => Often granted assumptions, Group interaction:, Important but often overlooked, 1. Complementary interactions:, Consensus emerges, Agreement between viewpoints, Each participant builds on the previous remarks, 2. Argumentative interactions:, Participants challenge each other, Opinions are revised & modified, Makes people account for their views, 6. Limitations of focus groups, Researcher has less control over proceedings, Data are more difficult to manage, Produce large volume of data, Need to examine both themes in what people say & patterns of interaction, Difficult to organize/risk of no-shows, Transcription is time consuming, Groupthink’/ social desirability, Potential to cause discomfort, Sensitive: Private lives, Power aspect: Highly hierarchical


Language in Qualitative Research, 1. Functions of language in social research:, 1. Language as a resource of information, Medium through which social research can be conducted, 2. Language as a topic in itself, Conversation analysis (CA), Discourse analysis (DA), 2. Conversation analysis (CA), What?, Study of LANGUAGE IN CONVERSATIONS, Better understanding of people’s reasoning, DETAILED ANALYSIS OF TALK in everyday settings, Features?, Focus on speech: ‘How is dialogue being created?’, The ‘HERE-AND-NOW’ IN DIALOGUE is pivotal, No extrapolation beyond the dialogue itself, Use of detailed TRANSCRIPTS of ‘natural’ conversation, Key assumptions:, 1. Talk is STRUCTURED, Tacit rules, Recurring patterns, 2. Talk is CONTEXTUAL, Has to be analyzed in that context, 3. Analysis is INDUCTIVE in nature, Departs from the conservation itself, Basic analytic tools:, 1. Notational symbols, We:ll - prolonged sound, .hh - intake of breath, (0.8) - silence for 0.8 seconds, 2. Turn-taking, Shared codes, Smooth transactions: A ends… // B starts, 3. Adjacency pairs, Question & Answer, Invitation & Response, 4. Preference organization, Acceptance is preferred over refusal, Dispreferred response needs to be justified => use of accounts (smoes), 5. Accounts, Justifies action by reference to common values, Prevents damage to the relationship, 6. Repair mechanisms, Restore interaction to ‘normal’, When A stops & B doesn’t start, A resumes, When A hasn’t stopped & B starts, A stops, 3. Discourse analysis (DA), What?, Study of all forms of linguistic communication, Discourse is a frame of how we perceive/define the social world, Context !, Key features:, Constructionist worldview, Not value-free frame, People use of a particular discourse to accomplish something, Action-oriented – Power structures, Need of critical analysis, Not just codification, Skeptical reading of intended meanings, 2 main goals:, 1. Uncovering interpretative repertoires, Which style of speech – references, The way discourse is construed, Example: Bush’s speech – off stage/on stage, 2. How is factual knowledge being produced?, Use of Rhetorical devices:, Quantification rhetoric, Use of variation, Role of detail to strengthen the argument, Accounts to argue these devices, Critical discourse analysis, Emphasizes the power aspect language carries, Related to ideology & socio-cultural change, Foucault – Disciplining through discourse, Example: American schools – Textbooks, 4. Differences & Similarities, Both focus on language & context, DA more flexible than CA in looking beyond immediate context of talk, DA recognizes the wider socio-cultural & political context, Epistemological: CA almost positivist in nature while DA departs from a more interpretivist stance


Documents as Sources of Data, 1. Documents as topic of research, use of extant documents!, Already exist in the social world, Have not been produced for the purpose of social research, Have been preserved & available for analysis, Unobtrusive & non-reactive method, 4 Criteria for evaluation:, 1. Authenticity, Genuine goal of the author, Is the origin clear?, 2. Credibility, What is not displayed – silence?, Free from error or distortion?, 3. Representativeness, Is the evidence typical of its kind?, 4. Meaning, Is it clear and comprehensible?, 2. Type of documents, 1. Personal Documents, Diaries, letters and autobiographies, Unsolicited personal documents, Keep socio-historical context in mind, Blurred distinction between autobiographies & biographies, Visual objects, 3 types of home photograph, 1. Idealization, 2. Natural portrayal, 3. Demystification, Role of context to discern underlying reality, Representativeness?, Different interpretations of images, 2. Official documents, 1. from the state:, Acts of parliament/laws/proceedings, Officials reports and judgments, 2. from private sources:, Company reports, meeting minutes, memos, 3. Mass-media outputs, Newspapers, magazines, radio, TV, film, Often contradictions – different styles in reporting =>Reflected values of editors/journalists, 4. Virtual documents, Official documents published on the internet, Email listings, Facebook, Twitter, blogs, etc., Private email communication, Evaluate each of them according to the 4 criteria, 3. The world as a text, Reading world as a ‘text’, Wider than only documents, Visual images, artifacts, landscapes, performances…, Readers/audiences, Active or passive?, Audiences may interpret different meanings to those intended by author, Caution: Social researcher often adds another layer of interpretation – 2nd/3rd layer of interpretation, 4. Interpreting documents, 3 types of studying documents:, 1. Qualitative content analysis, Study of themes & implicit meanings in the text, Code & retrieve, 2. Semiotics, Analysis of symbols in everyday life, What are the signs/symbols?, Two types of meaning, 1. Denotative meaning, Explicit meaning, 2. Connotative meaning, Given meaning at the individual level, Polysemic nature of signs, 3. Hermeneutics, Interpretivism, Discern meaning of texts from the perspective of the author, Socio-historical context crucial, Active audience – plurality of interpretations

Qualitative Data analysis


Qualitative Data Analysis, 1. What is qualitative data analysis?, After collection of data, time for analysis:, “A large corpus of unstructured textual material”, From field notes, interview transcripts, documents, ‘Attractive nuisance’:, 1. Richness of documents, 2. Difficulty of finding analytical paths, ‘Analytic interruptus’:, Make sure to carry out a real analysis, Typical to qualitative data analysis:, No well-established, widely accepted set of rules, Different strategies to analyze:, 1. Analytic induction, 2. Grounded theory, 3. Narrative analysis (Thematic analysis, 4. Secondary analysis of qualitative data, 2. Analytic induction, What?, Rigorous analysis of data, with the aim to find an universal explanation of phenomena, Steps, 1. Rough definition of research question, 2. Hypothetical explanation, 3. Data analysis => Examination of the cases, 4. If any deviant case is found => redefine or even reformulate hypothesis, 5. Continue until all cases fit the hypothesis, Key problems:, 1. Each time have to reanalyze & reorganize data, 2. Often ‘conditions that are SUFFICIENT rather NECESSARY conditions’, Circumstances explained, not the reason?, Example: Link illegal drug trade – real estate/prostitution => really explanatory?, 3. No guidelines to know when sufficient cases investigated, and hypothesis is valid, See pages 539-541, 3. Grounded theory, Main features:, Theory is derived from the data which is systematically gathered & analyzed, iterative process, Repetitive interplay between data collection & analysis/theory building, Theory is grounded in the data, A STRATEGY to analyze data, not really a theory, Often the GENERATION OF CONCEPTS instead of a theory, Tools of grounded theory:, 1. theoretical sampling, 2. coding, Defining of themes in your data: codes, Shorthand devices to label, separate, compile & organize data => categories of information, Different types:, 1. Open concepts => in the initial stage, 2. Axial concepts => links between initial concepts, 3. Focused concepts => refine, discern the core category, 3. constant comparison between concepts & data, 4. theoretical saturation, Outcomes of grounded theory:, Concepts => produced by open coding, Categories (of information), Core categorization of your data, Further development of the concepts, Properties => Typical features of a key category, Hypotheses, Theory, Explanation of relationship between concepts, Substantive theory, Lower level of abstraction, “Theory in a particular empirical stance”, Example: Hash trade, formal theory, Higher level of abstraction, “Applicable in a wider range of areas”, Example: Drug trade, See page 544, Criticisms on grounded theory:, Too inductive => Suspend relevant theories/concepts till late in the process, Funding => requires clear aims, theories & research questions, Time consuming, Not necessarily production of a theory, Confusing use of terms ‘concepts’ & ‘categories’, Fragments data: Loss of context & narrative flow, 4. Basic operations in qualitative data analysis, 1. How to code?, Read through field notes/transcripts more than once => make notes the second time, Review the index of codes you made, Consider theoretical ideas related to data & connections between codes, Multiple coding of data items: 2/3 codes, Use of Memos (basis for properties), Reminder of what terms mean, Encourages reflection, 2. Turning data into fragments:, Cut & paste, After coding, start to retrieve – put fragments together, Not just a mechanical task of data management, Helps to generate ideas and build theory, See pages 551-552, 3. Problems with coding:, 1. Loss of context => Extracts of data, 2. Loss of narrative flow => Fragmentation of data, 5. Narrative analysis, Key features:, Focus on stories people employ to refer to their lives and life world, Significance of the context, to link events, Narrative flow:, 1. Narrative approach during the collection, During interviews discover their life history, 2. Narrative approach during the analysis, Clear biographical intention, Thematic analysis:, Seek for recurring themes in data, Keep the language of the research participant, (Matrix-based method to order & synthesize data), Repetitions Local expressions Metaphors Transitions Similarities/differences Linguistic connectors Missing data Theory-related material, 6. Secondary analysis of qualitative data, Analysis of the data already analyzed, Compare own interpretation with original reading, More in-depth analysis, Disadvantages:, 1. lack ‘insider’ understanding of social setting, 2. participants may not have given consent for secondary analysis

Kwantitatieve onderzoeks methoden


The Nature of Quantitative Research, 1. Means steps in quantitative research, See pages 140-141, 1. THEORY:, 2. Hypothesis:, 3. RESEARCH DESIGN:, 4. CONCEPTS – OPERATIONALISATION, 5. SELECTION RESEARCH SITES, 6. SAMPLING, 7. ADMINISTER INSTRUMENTS, 8. PROCESS DATA, 9. ANALYSIS, 10. FINDINGS/CONCLUSION, 2. Concepts, WHAT ARE CONCEPTS?, 1. The BUILDING BLOCKS of a particular theory, 2. The labels we give to phenomena we detect in the social world, Broad categorization – FOCUS UPON COMMON FEATURES, CERTAIN DEGREE OF ABSTRACTION, WHY TO MEASURE?, 1. Delineate FINE DIFFERENCES between people/cases, Do not have to focus on extremes only, 2. CONSISTENT YARDSTICK for addressing differences, Consistent over time/researcher, 3. Degree of CORRELATION between concepts, HOW TO MEASURE?, INDICATOR, operational definition of a concept, LESS DIRECTLY QUANTIFIABLE THAN MEASURES, CONCEPT: ‘problematic drinking’, MEASURE: age, personal income, etc., INDICATOR: “memory loss”, DIRECT VS. INDIRECT INDICATORS, DIRECT: “hangover”, INDIRECT: “getting into a fight”, MULTIPLE-INDICATOR MEASURE, Often concept has MORE THAN ONE DIMENSION, VARIOUS INDICATORS TO HAVE A REAL IDEA, Example: ‘Problematic drinking’ Weekly frequency Time frame (throughout the year Consequences afterwards (hangover - missing class), EACH TO BE CODED INTO QUANTITIES, Frequency categories, Never, less than monthly, monthly, weekly, every day, LIKERT SCALE: STUDY OF ATTITUDE !, SET OF STATEMENTS – LEVEL OF AGREEMENT, TO MEASURE ATTITUDES, Interrelated statements, 5-POINT / 7-POINT SCALE TO ANSWER, BENEFITS, Less chance of misinterpretation, More than ONE DIMENSION, Allows FINER DISTINCTIONS, 3. Reliability & validity, RELIABILITY OF YOUR INSTRUMENT, 1. STABILITY OVER TIME, Test-retest method of your instrument, T1 = T2, Note: AVOID REACTIVITY/ONLY REAL CHANGE, 2. INTERNAL RELIABILITY, SPLIT-HALF METHOD, 10 statements – 5/5 => ‘same level of problematic drinking’, CRONBACH’S ALPHA – all possible split-halfs !!, If 0.70 correlation => satisfactory level, 3. INTER-OBSERVER CONSISTENCY, Same findings between different researchers, VALIDITY - FOCUS HERE ON MEASURE VALIDITY:, “Do we actually measure the concept?”, FACE VALIDITY – Are results valid AT FIRST SIGHT?, CONCURRENT VALIDITY – if x up & y up than also x down y down, PREDICTIVE VALIDITY – in time, CONSTRUCT VALIDITY – theory/concept, CONVERGENT VALIDITY – same result through other measures, 6 criteria ( chapter 2.3), VALIDITY PRESUPPOSES RELIABILITY!, 4. Mean concerns of quantitative researcher, 1. MEASUREMENT, Can a concept be quantified?, Example: ‘7/10 binge drinking (BD), when problematic?’, Comparisons between measures, Example: Questionnaire – Likert Scale, Changes in a variable over time, Example: Redefinition of BD, unknown concept before, 2. CAUSALITY, Explain, sometimes even predict, Causal relationships between IV (independent variable) and DV (dependent variable), VARIABLE: “A CHARACTERISTIC THAT VARIES”, Example: Type of friends – level of BG, Correlation various indicators , if direction unclear, 3. GENERALIZATION, Can the results be applied to individuals BEYOND THE SAMPLE?, Aim TO GENERALIZE TO TARGET POPULATION, Requires REPRESENTATIVE SAMPLE, 1. Random selection, 2. Probability sampling (H 7), 4. REPLICATION, Detailed description of procedures allows other researchers to replicate study, Low incidence of published replications, 5. Criticisms of quantitative research, 1. FAILURE TO DISTINGUISH BETWEEN OBJECTS IN THE NATURAL WORLD & SOCIAL PHENOMENA, 2. OVERLY FOCUS ON PRECISION AND ACCURACY, Presumed connection between concepts & measures, REALITY ‘MEASUREMENT BY FIAT’, 3. LACK OF ECOLOGICAL VALIDITY, Reliance on instruments and measurements, Little relevance to participants' everyday lives, 4. STATIC VIEW OF SOCIAL LIFE, Flexible relationship possible, IGNORES PROCESSES OF HUMAN INTERPRETATION, 6. Exceptions, Discrepancy between ideal type & actual practice of social research, 1. REVERSE OPERATIONISM, Measurements can sometimes lead to inductive theorising, FACTOR ANALYSIS, Is employed in relation to multiple-indicator measures (like likert scales) to determine wether groups or indicators tend to bunch together to form distinct clusters, referred to as factors., goal: reduce number of variables, INDUCTIVE, 2. RELIABILITY & VALIDITY TESTING, Often little details –> time consuming, MEASUREMENT BY FIAT/ Cronbach’s alpha, 3. SAMPLING, often non-probability samples instead of probality samples


Sampling, 1. What is sampling?, Sample: the segment of the population (=universe of units -> not necessarily people, but also nations, cities, firms, etc) that is selected for investigation., OFTEN POPULATION IS TOO LARGE =>Need to take A SAMPLE, Preferably the sample has to be representative in order to generalize back to the population, PROBABILITY SAMPLING, SAMPLING OF PEOPLE, TIME/DATE, MEDIA, 2. Different steps of sampling, 1. POPULATION:, UvA - 27 000 students, 2. SAMPLE:, Selection of the population for research – 900 students, 3. SAMPLING FRAME:, List of all units – See administration, 4. REPRESENTATIVE SAMPLE:, A sample that reflects the population accurately, Random/probability sampling, Example: “Only those in class = not representative”, 5. SAMPLE BIAS:, Distortion in the representativeness of the sample, 1. Non-probability vs. random sampling, 2. Sampling frame is inaccurate – wrong list, 3. Non-response – “Those studying won’t answer”, some members of the sampling frame stand little or no change of being selected for inclusion in the sample, 6. DANGER OF SAMPLING ERROR, OVERREPRESENTATION, Risk high in non-probability sampling, Example: “Those who attend class”, PREFERENCE IN QUANTITATIVE RESEARCH FOR PROBABILITY SAMPLING!, Each unit has A KNOWN CHANCE OF SELECTION, RANDOM SELECTION, 3. Types of probability sampling, 1. SIMPLE RANDOM SAMPLING, Every unit has an EQUAL PROBABILITY of selection, SAMPLING FRACTION: n/N, n = sample size, N = population size, List all units & number them consecutively, Use ‘random numbers table’ to select the units, 2. SYSTEMATIC RANDOM SAMPLING, Select units DIRECTLY FROM SAMPLING FRAME, A RANDOM STARTING POINT, choose every nth unit, Ensure sampling frame has no inherent ordering, Example: every 30th name, 3. STRATIFIED RANDOM SAMPLING, 1. Stratify population ALONG SPECIFIC SOCIAL-DEMOGRAPHIC CRITERIA, 2. Randomly SELECTION WITHIN EACH CATEGORY, Benefits:, PROPORTIONATELY REPRESENTATIVE, Detailed information:, “32 % of UvA students has engaged into cantus drinking “, “Variation exits within the UvA student population. 14% of the ASW students is familiar with CD, while among chemistry students that number rises up to 45%”, Disadvantage:, REQUIRES A DETAILED SAMPLING FRAME, Great deal of work, 4. MULTI-STAGE CLUSTER SAMPLING, Useful to study GEOGRAPHICALLY DISPERSED populations, Different STAGES OF SAMPLING, 1. DIVIDE POPULATION IN GROUPS (= CLUSTERS), + 1’. SAMPLE A FEW CLUSTERS, 2. DIVIDE THESE CLUSTERS IN SUB-CLUSTERS, + 2’. SAMPLE A FEW SUB-CLUSTERS, 3. SAMPLE RESPONDENTS IN THESE SUB-CLUSTERS, Stages uit responsiecollege, 1. SAMPLE REGIONS (N, W, E, S) 2. SAMPLE UNIVERSITIES WITHIN SAMPLED REGIONS 3. SAMPLE RESPONDENTS IN THESE SAMPLED UNIVERSITIES, Example: “Cantus-drinking amongst Dutch students”, 1. Group universities according to region, 2. Randomly sample two regions, 3. Randomly sample five universities per region, 4. Sample students from these universities, 4. Qualities/Benefits of probability sampling, Benefit: Allows generalizations about the population, Limits: Not beyond the population, 5.Sample size, ABSOLUTE SIZE MATTERS MORE THAN RELATIVE SIZE, THE LARGER THE SAMPLE = THE MORE PRECISE, SAMPLING ERROR decreases, Larger sample => time consuming/costly, Saturation, RESPONSE RATE, RESPONSE RATE = % OF SAMPLE WHO AGREE TO PARTICIPATE (or % who provide usable data), NON-RESPONSE (Reason why, is important to know!), HETEROGENEITY OF THE POPULATION, THE MORE DIVERSIFIED the population is, THE LARGER THE SAMPLE has to be, TYPE OF ANALYSIS TO BE CARRIED OUT, Some techniques require larger sample, Example: “Cantus-drinking amongst students of the age between 12 & 25”, 6. Types of non-probability sampling, 1. CONVENIENCE SAMPLING, the most easily accessible individuals, useful when piloting a research instrument, may be a chance to collect data that is too good to miss, 2. SNOWBALL SAMPLING, researcher makes initial contact with a small group, these informants lead you to others in their network, useful for qualitative studies of deviant groups, 3. QUOTA SAMPLING, OFTEN USED IN MARKET RESEARCH & OPINION POLLS, Cheap, quick & easy to manage, PROPORTIONATELY REPRESENTATIVE, Socio-demographic categories – STRATA, But NON-RANDOM SAMPLING, Interviewers SELECT PEOPLE TO FIT THEIR QUOTA for each category, SAMPLE BIASED:, THE FRIENDLY AND ACCESSIBLE: Students/women with young children, UNDERREPRESENTATION OF LESS ACCESSIBLE GROUPS: Males in mid thirties, 7. Error in survey research, Sampling error, Error due to difference between sample & population, Inadequate sampling frame/non-response, Makes it difficult to generalize findings, Data collection error, Implementation of research instruments, e.g. poor question wording in surveys, Data processing error, Faulty management of data, e.g. coding errors


Structured Interviewing, 1. What is a structured interview?, Key method in quantitative research, STANDARDIZED INTERVIEW SCHEDULE, Each interviewee is given the SAME QUESTIONS, + IN THE SAME WAY, + IN THE SAME ORDER, CLOSED + PRE-CODED OR FIXED CHOICE QUESTIONS, DIFFERENCE IN RESPONSES ONLY TO BE ATTRIBUTED “TO TRUE VARIATION”, Avoid Interviewer variability, 2. Conducting structured interviews, CLOSED QUESTIONS, PRE-CODED, No coding/categorization by interviewer, Steps, 1. KNOW YOUR WAY AROUND THE SCHEDULE, To avoid awkward questions, mistakes in reading, 2. INTRODUCE THE RESEARCH, Spoken or written RATIONALE, IDENTIFY YOURSELF, your employer, purposes of research and procedure of interview, ETHICAL ISSUES:, Anonymity, confidentiality, right to withdraw, Opportunity for interviewee TO ASK QUESTIONS, 3. BUILDING RAPPORT, Can be difficult if limited time & little opportunity for discussion (closed questions), Respondent – prepared to invest time and effort, 4. ASKING QUESTIONS, Stick to the schedule!!, Even SMALL VARIATIONS in WORDING CAN AFFECT RESPONSES, 5. RECORDING ANSWERS, Write exact words used by interviewee, or use FIXED CHOICE questions, 6. CLEAR INSTRUCTIONS, Certain questions not relevant to every interviewee, FILTER QUESTIONS help interviewer navigate the schedule, Example: 7. ‘Have you ever been in a committed, romantic relationship?’ (If no, jump to question 20), 7. QUESTION ORDER, Every interviewee – questions in the same order, GENERAL QUESTIONS before specific questions, Earlier questions AFFECT SALIENCE OF LATER ONES, First questions directly related to the topic, Embarrassing or SENSITIVE QUESTIONS AT THE END, 8.PROBING (DIEPER GRAVEN), Respondent does NOT UNDERSTAND QUESTION/INSUFFICIENT ANSWER, 1. Non-directive probes: “Mhmm”, “Can you say a bit more about that?”, 2. REPEAT FIXED CHOICE ALTERNATIVES, 9. PROMPTING (GEHEUGENSTEUN), Interviewer suggests possible answers, SHOW CARDS, 10. LEAVING THE INTERVIEW, Thank the interviewee, Debriefing should be minimal, 11. TRAINING AND SUPERVISION, If researcher hires interviewer(s), Ensure that interviewers know the schedule and follow standardized procedures, EVALUATE: Examine completed forms, tape record a sample of interviews, call-backs to respondents, 3. How to avoid measurement errors?, 1. AVOID ERROR DUE TO INTERVIEWER VARIABILITY, See page194, Differences in responses are due to ‘true variation’, not inconsistencies in the conduct of interviews, Error reduced by STANDARDIZATION, Reduces INTRA-INTERVIEWER & INTER-INTERVIEWER VARIABILITY, 2. AVOID ERROR DUE TO DATA PROCESSING, CLOSED ENDED, PRE-CODED OR FIXED CHOICE QUESTIONS (Limited choice of possible answers), INTERVIEWER DOES NOT INTERPRET responses before recording (In contrast to open questions), CODING FRAME REDUCES VARIABILITY IN CODING PROCEDURE, What? Rules for assigning answers to categories, Example: Male 0, Female 1, Unknown 9, LIMITS INTRA-CODER & INTER-CODER VARIABILITY, 3. INTERVIEW CONTEXT, COMPUTER-ASSISTED PERSONAL INTERVIEWING (CAPI) & TELEPHONE INTERVIEWING (CATI), More efficient filtering of questions, Immediate data entry, BY TELEPHONE?, voordelen, Quicker and cheaper – no travel required, Easier to monitor/evaluate, Reduces interviewer effect – no non-verbal cues, PROBLEMS:, No telephone, ex-directory, Limited time and rapport?, Cannot respond to non-verbal signs of confusion, Less satisfying experience for interviewee, 4. Problems with structured interviewing, CHARACTERISTICS OF INTERVIEWERS, Gender, age, ethnicity, class (rapport), Can evoke socially desirable responses, RESPONSE SETS, People may respond in consistent but irrelevant ways, 1. Acquiescence – agreeing or disagreeing to all questions, 2. Social desirability, THE PROBLEM OF MEANING, Interpretivist critique, Interviewer & respondents MAY GIVE DIFFERENT MEANINGS to discussed concepts


Self-completion Questionnaires, 1. What is a self-completion questionnaire (SCQ)?, -NO INTERVIEWER present -Respondent writes answers on form -Returned to researcher or deposited for collection -Postal questionnaires -Distributed in person/by email/online, 2. Advantages/Disadvantages of SCQ, TYPICAL OF SCQ:, LESS NUMBER of questions, Simpler interview schedule, FEW OPEN-ENDED QUESTIONS, ADVANTAGES OF SCQ:, 1. Cheaper & quicker to administer (To widely dispersed populations), 2. No INTER- OR INTRA-INTERVIEWER EFFECT, 3. Fill in at their own time, 4. On sensitive issues, DISADVANTAGES OF SCQ:, No PROBING OR PROMPTING, Only very salient questions => RESPONDENT FATIGUE, No complex questions, Can see the whole questionnaire before answering => QUESTION ORDER EFFECT, WHO IS ANSWERING it?, Little context, People with limited literacy skills – error, LOWER RESPONSE RATE, 3. Response-rate & how to improve?, Relatively low => Risk of sample bias, 60-70% = acceptable, Strategies to improve response rates:, Monetary incentive Stamped addressed envelope Covering letter Reminders Clear instructions Attractive layout, 4. How to design a SCQ?, 1. UNCLUTTERED LAYOUT, 2. CLEAR PRESENTATION, 3. VERTICAL ALIGNMENT OF FIXED CHOICE ANSWERS, Distinguishes questions from answers, Respondent less likely to make a mistake, Easier to pre-code, Exception: Likert scale, 4. CLEAR INSTRUCTIONS, How to indicate choice of answer, Can they select more than one answer?, 5. KEEP QUESTIONS AND ANSWERS TOGETHER, DON’T SPREAD A QUESTION OVER TWO PAGES, Put answers alongside each corresponding question, 5. Diaries as a form of SCQ, RESEARCHER-DRIVEN DIARIES, Alternative to structured observation in quantitative research, STRUCTURED DIARY, TIME-USE DIARY, Record amount of time spent on different/certain activities, CLEAR INSTRUCTIONS HOW TO COMPLETE DIARY, How often/level of detail?, Give an example of a diary entry, Checklist of items, events or behaviors to include in each entry, SHOW BLOCKS OF TIME IN COLUMNS, Advantages, precise estimation of time spent on activities (valid, reliable data), Chronological order of events, Personal or sensitive issues, Disadvantages, Cost of producing diaries and monitoring completion, boredom, fatigue and attrition, failure to record details, selective inclusion of events


Asking Questions, 1. Open vs closed questions, Open questions, Advantages, - Respondents answer in their own terms - Allow new, unexpected responses - Exploratory -> generate fixed answer questions, Disadvantages, - Time-consuming for interviewer & respondent - Difficult to code - More effort required from respondent - Interviewer variation in recording answers, Closed questions, Advantages, - Quicker & easier to complete: BETTER RESPONSE RATE + LESS MISSING DATA - Easy to process data: PRE-CODED - Easy to compare answers: INTER-CODER RELIABILITY, Disadvantages, - Restrictive range of answers: No spontaneity - Difficult to make fixed choice answers exhaustive -Respondents may interpret questions differently, coding, PRE-CODING: Closed answers in surveys, POST-CODING: Use A CODING FRAME, 1. Categorize unstructured material, 2. Assign a number/code to each category, GENERAL GUIDELINES:, Categories must not overlap, Cover all possible answers, Consistent over time/ between coders, 2. Types of questions, - Personal factual questions - Factual questions about others - Informant factual questions - Attitudes (houding) - Beliefs (overtuigingen) - Normative standards & values - Lay knowledge (vragen over kennis: ken je...?) - for example: see sheets, 3. Designing questions: Rules of thumb, 1. REMEMBER YOUR RESEARCH QUESTIONS, 2. DECIDE EXACTLY WHAT YOU WANT TO FIND OUT, “Do you own a car?” – possession, “Do you drive a car?” - use, 3. IMAGINE YOURSELF AS A RESPONDENT, Please look into the commonly made mistakes p.240-245!, Specifiekere regels, 1. Vermijd ambigue termen in vragen -> niet 'hoe vaak', 2. Vermijd lange vragen, 3. " 2 ledige vragen, 4. " te algemene vragen, 5. geen 'leading questions' -> niet sturen, 6. geen dubbele vragen, 7. geen negaties/ontkenningen, 8. vermijd technische termen, 9. zorg dat vraag/antwoord opties goed afgesteld zijn, 10. Zijn de antwoord opties gebalanceerd?, 11. vermijd gebrek aan visuele ondersteuning, 12. Geen idee? -> probleem luiheid (soms moet het erbij), 4. Vignette questions, Present a scenario, Ask them how they would respond or what they think the characters should do, Useful for sensitive topics, 5. Piloting and pre-testing questions, CHECK THAT THE RESEARCH INSTRUMENT WORKS, - Gain practice at using interview schedule - Does each question flow smoothly on to the next? - Identify vague or confusing questions - Remove any questions that received uniform responses, OPEN QUESTIONS CAN GENERATE FIXED CHOICE ANSWERS TO INCLUDE IN THE FINAL STUDY, PILOT RESPONDENTS SHOULD NOT BE IN FINAL SAMPLE!, 6. Using existing questions, Questions have already been piloted, Known properties of reliability & validity, Helps you to draw comparisons with other studies, ‘QUESTION BANKS’ Repositories of questions used in previous surveys


Structured Observation, 1. What is structured observation?, Research method to SYSTEMATICALLY OBSERVE respondents’ BEHAVIOUR, EXPLICIT RULES for observation/coding, Rigid OBSERVATION SCHEDULE, FIXED SET OF CATEGORIES TO MEASURE PEOPLE’S ACTIVITY, Aggregate & compare behaviour IN THE SAMPLE, Popular in cross-sectional research design, 2. Advantages of structured observation, Observation overcomes typical problems of social survey:, 1. MEANINGS OF QUESTIONS less an issue 2. OMISSION OF KEY TERMS can be avoided 3. Don’t rely on MEMORIES OF BEHAVIOUR 4. SOCIAL DESIRABILITY effect less 5. NO INTERVIEWER BIAS 6. NO GAP BETWEEN STATED & ACTUAL BEHAVIOUR, 3. Observation schedule/observation strategies?, A. OBSERVATION SCHEDULE:, Must have clear focus – easy to use, DEFINE SPECIFIES CATEGORIES OF BEHAVIOUR + TO ALLOCATE BEHAVIOUR TO A CATEGORY, CATEGORIES MUST BE INCLUSIVE (COVER ALL OPTIONS) + MUTUALLY EXCLUSIVE, INTER-OBSERVER CONSISTENCY, Example: Dark room activities-Recorded every three seconds, B. OBSERVATION STRATEGIES:, 1. RECORD INCIDENTS – EVENTS AND INTERVENTIONS, Example: Every entry in dark room, 2. OBSERVE FOR SHORT PERIODS OF TIME – REPEATEDLY, Example: Between 1-1.15 am, between 3-4 am, etc., 3. OBSERVE FOR LONG PERIODS OF TIME – CONTINUOUSLY, Example: One full night, TIME SAMPLING: Record whatever is happening every x minutes, 4. Sampling in structured observation, 1. SAMPLING PEOPLE: Random sample of individuals to observe, 2. SAMPLING TIME PERIODS Observe same individual(s), at randomly selected times, CANNOT ALWAYS USE PROBABILITY SAMPLING => No sampling frame for public encounters, Example: gay men in a bar/doormen, LIMITED EXTERNAL VALIDITY, If conducted over a short span of time, 5. Reliability & validity, RELIABILITY, INTER-OBSERVER CONSISTENCY/INTRA-OBSERVER CONSISTENCY, MEASUREMENT VALIDITY, Does schedule measure underlying concept?, Implementation of schedule, Reactive effect, Participants’ awareness of being studied can create change in their behaviour - Self-consciousness - Role selection: playing out a role - Researcher’s presence alters the social setting - Response sets, 6. Field stimulation, Researcher directly intervenes in a setting & observes the results, CONTRIVED OBSERVATION, Quasi-experimental design, 7. Criticisms of structured observation, - IMPOSING AN IRRELEVANT FRAMEWORK on the social setting - NEGLECTS THE MEANINGS/INTENTIONS BEHIND BEHAVIOUR - Ignores the SOCIAL CONTEXT OF BEHAVIOUR - GENERATES FRAGMENTED DATA – difficult to see the wider picture, BUT, - MORE ACCURATE THAN INTERVIEWS & QUESTIONNAIRES => See what people really do, not what they say they do - USEFUL TOGETHER WITH OTHER METHODS => Study behaviour, attitudes and social context => Often unstructured observation - pilot


Content Analysis, 1. What is content analysis?, "approach to the analysis of documents and texts that seeks to quantify content in tems of predetermined categories in a systematic and replicable manner.", Counting of certain words, people, themes, Different from Qualitative content analysis !! (chapter 21), Also approach to analysis of documents and texts, but:, SYSTEMATIC AND RELIABLE, Quantifies content by means of PREDETERMINED CATEGORIES, 1. Coding schedule 2. Coding scheme, FOCUS ON MANIFEST CONTENT, Example: ‘Age’ in my space profiles News items: ‘face book’/ ‘dangerous’ in new papers, Not necessarily the meaning – latent content (different from Qualitative document analysis), 2. Type of sampling, MEDIA:, Which types of text? - Printed or visual data? Documents?Mass media? If mass media, which kind? - TV, radio, newspapers, magazines… - More than one type? Example: Sunday papers/Sample of 4000 news items, DATES:, - After a historical event or in general - When to start & stop analyzing? - PROBABILITY SAMPLING WHERE POSSIBLE * Example: Every 4th month, 3. What to count?, 1. SIGNIFICANT PERSONS/actors, Protagonists or alternative voices, 2. WORDS, - Frequency of certain words * Example: ‘privacy’/ ‘minors’/ ‘face book’ - Connotations/styles of discourse, 3. SUBJECTS & THEMES, 4. DISPOSITIONS, - Positive/negative *Example: ‘work opportunity & FB’ - Values, bias and ideology, 4. Coding: schedule/manual, see sheet or page 283, Coding schedule: a form onto which all the data relating to an item being coded will be entered., Coding manual: statement of instructions to coders and lists all the dimensions, Coding scheme requires:, - NO OVERLAP IN ‘DIMENSIONS’ - MUTUALLY EXCLUSIVE CATEGORIES - EXHAUSTIVE LIST OF CATEGORIES - UNIT OF ANALYSIS CLEAR (Time frame vs news item) - INTER-CODER & INTRA-CODER RELIABILITY, 5. Advantages vs Disadv of content analysis, Advantages, Replicable method Minimal interpretation Scope for longitudinal analysis No reactive effects – unobtrusive method Flexible - can be applied to various texts Info about populations which are difficult to access, Disadvantages, Depends quality of document Risk of intra-coder and inter-coder variation Latent content Descriptive rather than explanatory A-theoretical? (see also CA)


Secondary Analysis & Official Statistics, 1. What is secondary analysis?, Analysis of data that were collected by others for different purposes = secondary - Other researchers - Institutions of the state / business organizations Primary data = collected by oneself Blurred boundary between primary & secondary data, 2. Benefits of secondary analysis, 1. SAVES COST & TIME, DATA IS ALREADY COLLECTED, 2. HIGH QUALITY DATA, - Rigorous sampling - Large sample size - Experienced researchers, 3. OPPORTUNITY FOR LONGITUDINAL ANALYSIS, - Previous waves of a survey - Track social change over time, 4. SUBGROUP ANALYSIS, - Study of one of the subcategories of sample - Example: Crime rate in particular neighborhood, 5. OPPORTUNITY FOR CROSS-CULTURAL ANALYSIS, Fewer limits of time, cost, language, 6. MORE TIME FOR DATA ANALYSIS, 7. REANALYSIS MAY OFFER NEW INSIGHTS, - Focus on one variable or subgroup - New theories can be applied, 3. Disadvantages of secondary data, 1. LACK OF FAMILIARITY WITH DATA, How was it collected? How was it coded & managed?, 2. COMPLEXITY OF DATA, Volume of data Hierarchical data sets, 3. NO CONTROL OVER DATA QUALITY, Validity & reliability, 4. ABSENCE OF KEY VARIABLES, May be no data on a variable of interest to you Inability to apply new theories - might have yielded significant results, 4. Official statistics, COLLECTED BY STATE AGENCIES, ADVANTAGES COMPARED TO DATA FROM SURVEYS:, - Reduced time and cost - No reactivity - Cross-sectional & longitudinal analysis - Cross-cultural analysis, DISADVANTAGES OF OFFICIAL STATISTICS:, - Only reveal ‘tip of the iceberg’ - Say more about collection procedures, than actual phenomenon, 5. The social construction of crime statistics, CRIME RATE = OFFENCES RECORDED BY POLICE, Contingent on social processes of decision-making, 8 STAGES OF (DE) SELECTION: see sheet, 6. Reliability & validity of official statistics, RELIABILITY - DANGER, - Definitions, categories & allocated resources evolve over time - Reflects priorities of agencies/organizations - Example: Moral panics & police ‘crackdowns’, VALIDITY - DANGER, - ‘Fiddling’ the crime figures - Ecological fallacy: "the error of assuming that inferences about individuals can be made from findings relating to aggregate data."

Mixed Methods

H24 & H25

Beyond the Quantitative/Qualitative Divide -Mixed Methods, 1. Divide clear cut?, In principal:, Quantitative research, - FOCUS ON FACTS, FIGURES - VIEWPOINT OF RESEARCHER 1. DEDUCTIVE: Theory testing - hypothesis 2. POSITIVISM - REALISM 3. OBJECTIVIST WORLDVIEW RESEARCH METHODS: 1. Structured interview 2. SC Questionnaire - Dairy 3. Structured Observation 4. Quantitative Content analysis 5. Secondary data analysis (quan), Qualitative research, - FOCUS ON MEANING, STORIES, TEXTS,... - VIEWPOINT OF PARTICIPANTS 1. INDUCTIVE: Theory production 2. INTERPRETIVISM 3. CONSTRUCTIONIST RESEARCH METHODS: 1. Ethnography – Part Observation 2. Unstructured interview 3. Focus group discussion 4. CA - DA 5. Document analysis 6. Secondary data analysis (qual), FROM STRICT EPISTEMOLOGICAL/ONTOLOGICAL STANCE:, EMBEDDED METHODS ARGUMENT, Research methods inextricably linked to epistemological/ontological viewpoint, Irreconcilable sets of beliefs, No overlap => Mixed methods research is not feasible/desirable, PARADIGM ARGUMENT, Paradigm = set of epistemological beliefs, values & methods that dictate how research should be done, Quantitative & qualitative research reflect different, incommensurable paradigms, Research methods defined by epistemological stance only, Mixed research is not posssible, FROM TECHNICAL STANCE:, - Not necessarily connection between epistemological / ontological viewpoint and research methods - TENDENCIES RATHER THAN ABSOLUTE DETERMINISM - A research method from one strategy is viewed as capable of being pressed into the service of another. - Mixed methods research has benefits, WHAT IS MIXED METHODS RESEARCH?, ‘Integrates quantitative & qualitative research within a single project’, Example: Questionnaire – focus group discussion, 2. Four dimensions considered, Contrast overrated:, 1. BEHAVIOUR VS. MEANING, QUANTITATIVE METHODS STUDY ‘MEANING’ TOO, - Attitude, beliefs, norms - Study of texts (Quantitative Content Analysis) - Example: Likert scale, QUALITATIVE RESEARCHERS STUDY HUMAN BEHAVIOUR, BUT IN THE CONTEXT OF NORMS, VALUES AND CULTURE, "In other words, quantitative and qualitative researchers are typically interested in both what people do and what they think, but go abou the investigation of these areas in different ways.", 2. THEORY TESTING VS. PRODUCTION, -QUANTITATIVE RESEARCH NOT ALWAYS INVOLVE HYPOTHESIS TESTING Example: Exploratory surveys - Link between concepts/variable, but less direction - THEORY PRODUCTION WHILE INTERPRETING DATA, 3. NUMBERS VS. WORDS, QUALITATIVE RESEARCHERS SOMETIMES UNDERTAKE A LIMITED AMOUNT OF QUANTIFICATION (To illustrate the frequency of a certain phenomenon), QUASI-QUANTIFICATION IN QUALITATIVE RESEARCH REPORTS (“most”, “many”, “often”, “some”), 4. ARTIFICIAL VS. NATURAL, QUALITATIVE INTERVIEWS & FOCUS GROUPS ARE NOT NATURALISTIC SETTINGS!, 3. Research each other?, Each strategy can be used to analyze the other, A. QUALITATIVE APPROACH TO QUANTITATIVE RESEARCH, Study ‘Research articles based upon SC questionnaires’, ETHNOSTATISTICS:, Study of the construction, interpretation and display of statistics in quantitative social research, HOW USED AS A ‘RHETORICAL DEVICE’? (Credibility on the findings/authority as researcher), Example: On ‘voting preference’ – UvA lecturers, B. QUANTITATIVE APPROACH TO QUALITATIVE RESEARCH, META-ETHNOGRAPHY: To aggregate findings from various qualitative studies, Problem of selecting ‘representative’ sample of studies in literature review, Inter-coder & intra-coder reliability, Example: Content analysis of ‘village ethnographies on gossip/lies’, 4. Arguments for mixed methods research, 1. TRIANGULATION, - Results of one research method cross-checked against the results of another - Planned or unplanned? - If results are inconsistent? (Treat one set of results as definitive), 2. COMPLEMENTARITY, Dovetailing (vervlechten) the best aspects of two research strategies, 3. FACILITATION, Using one research strategy to aid another, 5. Benefits of mixed methods research, 1. Address different research sub-questions, - Statistical data => patterns - In-depth interview data => processes - Example: ‘Telling lies in junior high’, 2. Instrumental development, Providing hypotheses Aiding measurement: - Informing design of survey questions - Example: Focus group – questionnaire on ‘Sexuality, honesty & elderly’, 3. Increases credibility, - Researchers’ & participants’ perspectives - Diversity of views + Context-sensitive - Both meaning & ‘facts’, 4. Different audiences, Policy makers – academic field, 5. Confirm & discover, - New data or to confirm data - Enhancement of data - If less accessible field - Solving a puzzle - Different scales of a phenomenon, But!:, NOT INHERENTLY SUPERIOR TO MONO-METHOD OR MONO-STRATEGY RESEARCH!!, Success depends on 4 factors:, 1. Well-designed and conducted, 2. Methods appropriate to research questions, 3. Effects of spreading limited resources, 4. Skills and training of researchers



Ethics & Politics in Social Research, 1. Introduction, ETHIC ISSUES ARISES IN VARIOUS STAGES OF SOCIAL RESEARCH, ‘How should we treat the people on whom we conduct research?’, Are there actions in which we should not engage in order to ENDANGER THEIR LIVES/FUTURE RESEARCH, PARTICULARLY IN RESEARCH METHODS SUCH AS COVERT OBSERVATION, EXPERIMENT, ETC., NOT ONLY IN EXTREME CASES, FOUR ETHICAL STANCES:, see page 116, 1. UNIVERSALISM:, - Absolute rules about (un) acceptable conduct - ethical precepts should never be broken, 2. SITUATION ETHICS:, - CASE-BY-CASE ASSESSMENT - Principled relativism 2 arguments - The end justifies the means - No choice, 3. ETHICAL TRANSGRESSION IS PERVASIVE, - Virtually all research involves some ethically questionable practices - Participants never fully informed about research - 'the researcher must be dishonest to get honest data', 4. ANYTHING GOES (MORE OR LESS), Do whatever is necessary to get people to talk (But do not harm participants), 2. Ethical principles, 1. NO HARM TO PARTICIPANTS, Physical, psychological, emotional distress, Both short-term & long term, Confidentiality of records risk (Ethnographic studies of small towns or groups), Minimize the level of disturbance, 2. INFORMED CONSENT, COVERT OBSERVATION: participants not given choice to refuse, Explain the research as fully as possible & in terms meaningful to participants, Informed decision about whether to participate But not always practicable /desirable to inform all participants If deviant behaviour/closed settings/reactivity, >>But if non-consenting participants are harmed, researcher is more culpable, 3. INVASION OF PRIVACY, - Linked to informed consent - Giving consent = abrogating the right to privacy - Anonymity & confidentiality of data - Careful storage of personal information, 4. DECEPTION, Social research presented different than it is, Participants wrongly or partially informed, Widespread - Researchers usually want to limit participants’ understanding/reactivity, Compromises professional self-interest & reputation of the discipline, 3. The difficulties of ethical decision-making, BLURRED BOUNDARY BETWEEN ETHICAL/UNETHICAL PRACTICES, COMMON PRACTICES:, - The down-sizing the length of the interview - Limits of informed consent - If asked for a formal informed consent => willingness to participate drops, INTERNET-BASED RESEARCH PROVIDES NEW ETHICAL DILEMMAS, Public access vs. perceived privacy Chat rooms – public or private Guidelines: 1. Information is publically archived 2. No password required 3. The material is not sensitive in nature 4. No site policy that forbids the use of the material, 4. Politics in social research, VALUES INFLUENCES EVERY STAGE OF RESEARCH PROCESS, CANNOT CONDUCT SOCIAL RESEARCH IN A MORAL VACUUM ( Impossibility of objective, value-free research), SOCIAL RESEARCHERS OFTEN HAVE TO ‘TAKE SIDES’, FUNDING AGENCIES HAVE AN IMPACT:, Vested interests Decisions about type of research projects: - Calls for tender: encourages proposals on specific topics - Preference for quantitative, policy-oriented research Monitoring written reports and dissemination, GATEKEEPERS, Mediate access to research settings, May influence how the investigation takes place, CONFLICTING VALUES WITHIN THE RESEARCH TEAM, ATTEMPTS TO THWART PUBLICATION & DISSEMINATION OF CONTROVERSIAL FINDINGS, USE OF FINDINGS TO FUEL POLITICAL DEBATES

1e helft H14 /H1 Field


Twee soorten data-analyse, Kwalitatieve Methoden: Theorieën toetsen/ generen door middel van rijke data zoals taal, Kwantitatieve Methoden: Theorieën toetsen door middel van cijfers

het onderzoeksproces, zie sheets

Variabelen, Onafhankelijke Variabele, - De vermoedelijke oorzaak - Een ‘predictorvariabele’ - Een gemanipuleerde variabele (in experimenten), Afhankelijke Variabele, - Het vermoedelijke gevolg - Een ‘uitkomstvariabele’ - Gemeten, niet gemanipuleerd (in experimenten)

Meetniveaus, categorisch, >> Dingen worden verdeeld in aparte categorieen:, Binaire variabele: Er zijn maar twee categorieen, Nominale variabele: Er zijn meer dan twee categorieen, Ordinale variabele: Zelfde als nominaal, en de categorieen hebben een logische volgorde (bv opleidingsniveau), continu, >> Dingen hebben een precieze score:, Intervalvariabele: Gelijke afstanden vertegenwoordigen gelijke verschillen van het gemeten kenmerk, Ratiovariabele: Zelfde als interval, maar ook de verhoudingen zijn betekenisvol (bv leeftijd, lengte)(heeft absoluut nulpunt)

Meetfouten (‘measurement error’), Het verschil tussen de werkelijke waarde die we proberen te meten, en het getal dat we gebruiken om die waarde te representeren., Validiteit: of een instrument wel meet wat het beoogt te meten., Betrouwbaarheid: of een instrument dezelfde resultaten oplevert onder dezelfde condities., > Betrouwbaarheid is een voorwaarde voor validiteit!

Hoe meten we?, Correlationeel onderzoek:, Observeer wat natuurlijkerwijs in de wereld gebeurt (natuurlijke samenhang), zonder direct in te grijpen., Let op voor de invloed van extra (‘confounding’) variabelen., Experimenteel onderzoek:, Een of meer variabelen wordt systematisch gemanipuleerd om het effect ervan te zien oop een uitkomstvariabele., uitspraken doen over oorzaak en gevolg., Methoden van Dataverzameling bij een Experiment, ‘Between-group’/ ‘Between-subject’/ onafhankelijk, Verschillende personen of dingen in verschillende experimentele condities, ‘Within-subject’/ herhaalde-metingen, Dezelfde mensen of dingen in alle experimentele condities - Voordelen: vaak goedkoper en praktischer - Nadelen: vermoeid/verveeld of meer geoefend bij 2e, 3e, etc keer > oplossing: ‘counterbalancing’

Typen van Variatie, Systematische Variatie, Verschillen in prestatie gecreeerd door een specifieke experimentele manipulatie., Onsystematische Variatie (‘error’), Verschillen in prestatie gecreeerd door onbekende factoren. (Leeftijd, geslacht, IQ, tijdstip, motivatie, meetfouten, etc.), >> Randomisering: Minimaliseert onsystematische variatie.

Data analyseren: Frequentieverdeling (oftewel: Histogram), Een grafiek waarin de waarden van de observaties op de horizontale as staan, met een staaf die toont hoe vaak elke waarde voorkomt in de dataset, Scheefheid (‘skewness’): De symmetrie van de verdeling, Gepiektheid (‘kurtosis’): De ‘dikte’ van de staarten van de verdeling, zie sheets, Centrummaat 1: Modus, Modus: De meest voorkomende score, Bimodaal: Heeft twee modes, Multimodaal: Heeft meerdere modes, Centrummaat 2: Mediaan, De middelste score wanneer de scores gerangordend zijn., Centrummaat 3: Gemiddelde, De som van de scores gedeelde door het aantal scores., Spreidingsmaten:, ‘Range’, Spreiding: De kleinste score, afgetrokken van de hoogste, Erg beinvloed door uitbijters, Interkwartiel Range, Kwartielen = De drie waarden die een geordende set data in 4 gelijke delen verdelen. Eerste kwartiel = mediaan van de onderste helft van de data Tweede kwartiel = mediaan Derde kwartiel = mediaan van de bovenste helft van de data, Standaarddeviatie, - Geeft aan hoe zeer scores dicht rondom het gemiddelde liggen, of juist erg verspreid zijn - Ook wel ‘standaardafwijking’ genoemd - Wordt berekend met een formule, Een stapje verder dan werkelijke data:, kansen, Standaardnormaalverdeling, - Normale verdeling, dus symmetrisch klokvormig - Met een gemiddelde van 0 - En een standaarddeviatie van 1 - Scheefheid = 0 en gepiektheid = 0 - Hiervan zijn dus alle kansen op de scores bekend!, Z-scores, Drukt een score uit in termen van hoeveel standaarddeviaties hij van het gemiddelde verwijderd is., (toevalsvariabele of stochast - gemiddelde) / standaardafwijking

hypothesen, Nul-hypothese, H0, Er is geen effect. Bijv. In een beker coca-cola blijven evenveel spermacellen in leven als in een beker water, De alternatieve hypothese, H1, - Ook wel de experimentele hypothese genoemd - Bijv. In een beker coca-cola blijven minder spermacellen in leven als in een beker water - De alternatieve hypothese kan een- of tweezijdig zijn