Mastering Qualitative Research: Key Concepts and Sampling

School

McMaster University**We aren't endorsed by this school

Course

HLTHAGE 2A03

Subject

Sociology

Date

Dec 10, 2024

Pages

Uploaded by SuperHumanRedPandaPerson1140

2A03 EXAM REVIEW WEEK 7.1 – QUALITATIVE RESEARCH-Max Weber oPrecursor to symbolic interactionism oMacro sociologist oVerstehen ; understanding Necessary to consider subjective meaning and understanding-Anti positivism oReality as subjective, shifting, emergent -Allows us to oSee through perspective of those being study oDescribe/emphasize social context oUnderstand social processesoFlexibility o--> rich descriptive analysis-INDUCTIVE APPROACH-Steps oGenerate general research question oSelect relevant site/subjects oCollect data oInterpret dataoConceptual and theoretical work (sensitizing concepts)Focus research questionsCollect further data oFindings/conclusions -Sensitizing concepts oTo guide the research oDerive concepts from participnts perspective using language and expressions -Qualitative research methods oEthnography (participant observation)oFocus groups oLanguage based approach Conversation and discourse analysis Critical discourse analysis oPhenomenology Examines lived experiences Explores meanings, structures and essences of a phenomena MODULE 7.2-Judging quality of research oReliability Consistency and dependability of research measures

oValidity Refers to integrity of conclusions in research oQualitative Credibility Parallels internal validity Are findings believable Depends on oMethods oCredibility of researcher Transferability Parallels external validity Do findings apply to other people and context Can findings be generalized across social settings Dependability Parallels reliability Are findings likely to be consistent over time Refers to systematic process followed oAuthenticity Consciousness of ones own perspectiveoReflexivity Understanding how ones own experiences and background affect actions Confirmability Parallels objectivity Researcher as the research instrument oView world as socioculturally constructed-Comparing and contrasting types of research oNumbers vs words oPOV of researcher and participants oResearcher involvement oTesting vs developing theories and concepts

oStructured vs unstructured approach oGeneralizable knowledge vs contextual understanding oHard reliable vs rich, deep data oMacro vs micro level analysis oTrends/patterns vs meanings oArtificial vs natural settings -Drawbacks oToo subjective oDifficult to replicateoRarely generalizableoNot transparent WEEK 8 – SAMPLING STRATEGIES -What is sampling oProcess of choosing a sample from population oInvolves selecting individuals/units of research oProbability sampling QuantitativeSamples drawn such that every case can be chosen Chooses representative sample oNon probability Qualitative Sample chosen based on subjective judgement of researcher -Key terms oPopulation Name for larger general group from which sample is drawn oSampling element Unit of analysis/case in a population Person, group, organization etc. Specify sampling unity, geographical location and temporal boundaries oTarget population Specific pool of cases they want to study oSampling frame List of cases in a population or best approximation of all elements in population Must accurately represent target population oSampling ratioRatio of sample size to population size Ex: 150 ppl from 50 000 pop150/50k = 0.003 = 0.3%oParameter Characteristic of a population

Determined when all elements in population are measured Exl percentage of older adults in Hamilton who smoke oStatistic Qualitiative research tool used to estimate parameters Numerical estimate of population parameter computed from sample oInference Conclusion based on evidence and reasoning -Sampling errors oStatistical error resulting from non representative sample oOccurs when there is a discrepancy bw sample and population oRandomization minimizes this oDecreases as sample size increases -3 sources of bias in sampling oNot randomized samples oInaccurate selection of sample oSubjects do not participate -Types of sampling in order from simplest to most complex oSimple random Chosen completely at random All have = likelihood oSystematic sample Type of random probability sampling Samples members at a fixed periodic interval Ex: population size / desired sample size = sampling ratio Problem ; periodicity Occurs when cases in sampling frame are arranged in systematic order oStratified random sample Researcher divides population into stratas then randomly choose from each Tends to produce more representative results if stratum are accurate oMulti-stage cluster sampling Used when Large populationsNo adequate sampling frames Geographical dispersion Involves 2 or more stages Selecting clusters, groups, elements Selecting subunits in clusters Issues Technical complications (ex different size clusters)-Sample size oLarge samples TEND to be more representative

oSmaller the population --> bigger the sample ratio most be for accurate sample Populations under 1000 = 30% sample sizes oCommon sample sizes are: 100, 400, 900, 1600, 2500Each size increase cuts the sampling error by 1/2, then 1/3, then 1/4, and then 1/5 respectivelyThe biggest change occurs between 100 and 400oOften sample size is dictated by financial concernsoAdditional Issues with Sample SizeNon-responseThe response rate is the percentage of the sample that participates in the studyIs there is some particular issue common to the non-responders that brings them to differ in some important way from those who participateHeterogeneity ( diversity) of the PopulationGenerally, the greater the heterogeneity of the population on the characteristics of interest, the larger the sample size should beKind of AnalysisThe sample size needed may vary depending on what sort of analysis will be doneIf small groups in the population are to be compared to larger groups, it may be necessary to oversample the smaller group in order to make meaningful comparisonsCertain statistical procedures, such as some multivariate analyses,require large sample sizes to work properly-Non probability sampling oConvenience sampling Sampling units (elements) chosen on basis of availability not representativeness Ex: stand outside a store Problems Cannot have confidence in generalizability of findings Useful for pilot studies, testing research instruments, measuring reliability etc.oQuota sampling Non probability version of stratified sampling Population is stratified on basis of population variants (ex race, age etc.)Samples are drawn from stratum on basis of convenience Strengths Cheaper, easier Can be done quick Good for pilots, exploratory, commercial market research, opinionpolling

Weaknesses Not representative Eligibility judgements can be wrong oPurposive sampling Requires researcher to use their judgement Useful when Researcher is very knowledgable on population Goal is to get info on atypical members of society oSnowball sampling Identifying one member of population and then asking them to identify others Useful for studying members of underground population who want to keep identities secret oTheoretical sampling When researcher simultaneously collects, analyzes and decides what datato find next Iterative process from emerging theory Going back and forth between data collection and theory Theoretical saturation No new relevant data emerging The categories are well developed in terms of their properties andvariations Relationships among categories are well established/validated WEEK 9 – RESEARCH PROCESS-Steps in process oMotivation oResearch question oLiterature review oChoose general orientation Epistemology, ontology, inductive/deductive oChoose research design Quantitative or qualitative types oEthical considerations Obtain clearance oGather data oAnalyze data oFindings oConclusions -Elements of research proposal oQuestion

o2 thru 6 oSampling -Research questions oInteresting, clear, researchable, connected to established theory and research oLogically linked to one another oCan make original contribution to existing knowledge on topic -Why oGuides literature search oLimits project scope oGuides data collection and analysis -Literature review oIdentify what is already known oRelevant concepts and theoriesoAppropriate research methods oInconsistencies in findings and unanswered questions WEEK 10.1– SURVEY RESEARCH --Key terms oInterview schedule Name of survey

oProbe Follow up question to clarify/elaborate on concepts -Appropriate categories of questions oBehaviour oAttitudes/beliefs/opinionsoCharacteristics oExpectations oSelf classification oKnowledge -Types of surveys oSelf administered Pros Cheap Respondent freedom No interview apprehension ConsRespondent controlled Lack of oversight oInterview survey Person administered Pros Easy for respondent Survey can be adapted to respondent Clarification can be offered Instant feedback from respondent Cons Slow Costly Margin for human error Apprehension on interviewed oCATI Computer assisted telephone interview Used in academic commercial and govt researchers Random digit dialingoOnline survey Pros Speed, reach Cheap Real time capture Can include videos/graphics etc. Cons Technical issues

--Guidelines for survey interviewing oAppearance Researcher should dress similarly to intervieweeoDemeanour Friendly, pleasant, relaxedoFamiliarity with questionnaire Unfamiliarity places burden on respondent Takes too much time Must know specifications about how to respond to questions oFollow wording exactly oRecord responses exactly Important to also include behavioural responses oProbing oLikert scale Often solicits need for probes oCoordination and control -Secondary research oData collected and processed by one researcher and reanalysed for different purpose oAdvantages Cheaper FasterMay allow you to benefit from work of professionals Question of validity

-Characteristics of good research questionsoClear, unambiguous Must also be specific oMake sense given context of the survey Must fit within purpose of study Keep respondent expectations in mind oInformation must be accessible to respondents Social salience -Questions to avoid oDouble barelled Ask more than 1 thing at a time, do not allow for true answers oJargon and slang Has to do with accessibility oConfusing questions Too many negatives oLeading or loaded questions -