Xã hội học - Xã hội học - Introduction to qualitative data analysis

Introduction to Qualitative Data AnalysisSamuel K. Frimpong (Ph.D.)Outline of presentationWhat is qualitative data?Goal(s) of qualitative data analysisData ReductionContent analysiscodingRaw Data in Qualitative StudiesText/wordsPicture(s)Video(s)Sources: Interview, focus group discussion, observation, Internet etc.Basic Types of Qualitative DataWhat participants describeWhat participants say or write about the topic or phenomenon of interestWhat Participants DescribesAccording to Holliday (2002), “what participants describe” includes the following five types of data.Description of behaviour – what people are seen or heard doing or saying.Description of event – behaviour defined either by people in the setting (e.g. meeting) or by the researcher (e.g. customer service).Description of institution – the way the setting operates in terms of regulations, tacit rules and ritual.Description of appearance – what the setting or people in it look like (e.g. space, buildings, clothing, arrangements of people or objects).Description of research – what people say or do in the interview, focus group etc..What Participants Say or WriteAccording to Ap (2003), “what participants say or write” includes the following four types of data.Account – what people say or write to the researcher in terms of actual words.Talk – what people are heard saying in terms of actual words.Behaviour in setting – what is actually seen happening (e.g. actions and body language).Document – a piece of writing belonging to or pertaining to the setting, that is, any written information provided by people about the topic of interest.What is Qualitative Data Analysis?Qualitative analysis seeks to identify the underlying themes and patterns of the data collectedThe analysis is an art or craft that needs to be learned – no one standard wayResearcher’s Needs for Qualitative Data AnalysisSelf disciplineCreativityAbility to gain insightGoal of Qualitative Data AnalysisStorytellingConceptual frameworkProcess/Flow (e.g., a service blueprint)PropositionsTheoriesProcess of Qualitative Data AnalysisAccording Miles and Huberman (1994), the process simply involves:data reduction;data display; anddrawing conclusions Process of Qualitative Data AnalysisData Reduction: one attempts to identify categories, themes and concepts that emerge from the data.Data displaying refers to the organisation of the categories and themes etc. into some form of ideograph such as a typology, taxonomy, map, matrix or model.Conclusions are drawn by constantly comparing data against other data. This may involve examining the nature, or patterns, or both of the relationships, if any, between the data.Qualitative Data Analysis SpiralContent AnalysisHolsti (1968, p. 608) defines content analysis as “any technique for making inferences by systematically and objectively identifying special characteristics of messages”.Four Types of Content AnalysisAccording to Jennings (2001)• summation – data are reduced into categories that integrate and generalise the themes of the document;explanation – text is primarily explained by the content of the document;structuration – data are ordered to a pre-determined set of categories or by an order determined by the text; objective hermeneutics – objective and subjective analysis of data is made to explain the nature of the textual interactions.Steps of Content Analysis ProcessWhat are Codes?According to Miles & Huberman (1994) codes are:“ tags or labels for assigning units of meaning to the descriptive or inferential information compiled during a study” (p. 56).Three types:descriptive - this names the section of data being analysed;interpretive - which makes inferences about the data;pattern based – which uses interpretive codes to identify themes, processes and relationships.Coding DataGenerate a coding frameClean your list of raw codesGenerate a coding list with each code (concept) clearly defined and exclusively different.Phases of CodingAccording to Strauss (1987), there are Three Phases:open coding - first phase or preliminary coding which occurs during data collection. Re-occurring words, themes or concepts are searched and recorded;axial coding – involves a search for the relationship between open codes; andselective coding – the final phase of coding which may progress from axial coding and involves an examination of some codes over and above others.Open CodingAssign a code word or phrase that accurately describes the meaning of the text segmentLine-by-line coding is done first in theoretical researchMore general coding involving larger segments of text is adequate for practical research (action research)Coding between the lines- Identifying key EventsAnother Example of Open CodingAxial CodingClassify the codesThe process of looking for categories that cut across all data setsAfter this type of coding, you have identified your themesYou can’t classify something as a theme unless it cuts across the preponderance of the dataCoding ApproachesInductive coding:Researchers “listen” to the dataThe understanding emerges from the contextThe assigning of a code is derived from the data itself without reference to any pre-conceived notions or ideas.Deductive coding:Based on theories and previous literature.Adapt to the context of the studyUsually used with the inductive coding.Coding ApproachesReading / Data immersionRead for contentExplore the data by reading through all of your information to obtain a general sense of the informationAre you obtaining the types of information you intended to collectIdentify emergent concepts and develop tentative explanationsNote (new / surprising) topics that need to be explored in further fieldworkDeveloping hypotheses, questioning and verificationExtract meaning from the dataDo the categories developed make sense?What pieces of information contradict my emerging ideas?What pieces of information are missing or underdeveloped?What other opinions should be taken into account?How do my own biases influence the data collection and analysis process?Data ReductionDistill the information to make visible the most essential concepts and relationshipsGet an overall sense of the dataDistinguish primary/main and secondary/sub- themesSeparate essential from non-essential dataDefine relationships between conceptsUse visual devices – e.g. matrices, diagramsData Display and InterpretationUse visual devices – e.g. matrices, diagramsInterpretationidentifying the core meaning of the data, remaining faithful to the perspectives of the study participants but with wider social and theoretical relevanceQualitative Data Analysis SoftwareWordstat, SPSS text analysis NvivoAtlas.tietc.