Academic skills 2 Lecture 1 World views: • • • • Post-positivism o Verify theories, use of variables, deductive o ME ☺ o Quantitative Constructivism o Social construction of meaning o Theory generation, inductive o Qualitative Transformative o Transform the world o Political/activism, change-oriented o Qualitative Pragmatism o Problem-centered, pluralistic o Real-world practice oriented o Mixed methods Quantitative: • • Causality o X = cause, Y = effect o X = independent, Y = dependent o X = predictor/antecedent, Y = outcome/criterion/response Distribution of participants o 1st: sample o 2nd: assignment to conditions ▪ Random assignment = classic experiment ▪ Based on existing group = quasi-experiment Academic skills 2 Lecture 2 Qualitative • • • • • Key features: • Natural setting • Researcher as key instrument • Multiple sources of data • Participants meaning (instead of meaning from literature/researcher) • Emergent design (plan can always change) • Reflexivity (on researcher’s personal background etc.) • Holistic (complexity etc.) Data collection • Observations • Interviews • Focus groups • Documents/texts “Qualitative lenses” • Conversation analysis ▪ Interaction as series of observable actions ▪ Only analyze what is observable/made relevant by participants • Narrative analysis ▪ Importance of narratives in our lives ▪ Building blocks of narratives as analytical framework • Critical discourse analysis ▪ Discourses in competition with one another ▪ Importance of power • Ethnography ▪ Importance of spending a lot of time with a community Steps of qualitative analysis 1) Organize/prepare data 2) Browse through all the data 3) Start coding (Predefined/partly predefined/open) 4) Generate description (identifying themes/countertrends, select labels) 5) Representing description and themes Qualitative validity procedures o Triangulation ▪ Comparing different data sources for justification of the themes o Member checking o Rich, thick description o Think about bias (self-reflection) o Present negative/discrepant information also o Spend prolonged time in the field o Use peer debriefing Academic skills 2 Lecture 3 Types of Variables • X o Independent >> Manipulated in an experiment o Predictor/antecedent >> Cannot be manipulated • • • • Y o Dependent >> Used in experiment like independent o Outcome/criterion/response >> Used in survey just like predictor/antecedent Intervening or Mediating variables o Transmit the effect of an independent on a dependent variable. ▪ E.g. measuring the amount of polyphenols in the wine, when asking people to drink red wine. (polyphenols cause health benefits) Moderating variables o Predictor variables that affect the direction/strength of the relationship between independent and dependent variables. ▪ E.g. are the effects of wine consumption larger for males than females Confounding variables o Another cause for the possible outcome ▪ E.g. the effects do not come from the wine, but from the cheese everyone ate with the wine. Selecting a Sample 1) Set the population 2) Choose a sampling frame 3) Select units from frame o Power analysis: can help estimate a target sample size ▪ Estimated effect size/correlation ▪ Required significance level (false positive, type 1 error) ▪ Desired power (false negative, type 2 error) 4) Approach units Internal validity/reliability • • • Improving construct validity o Operationalization of complex concepts ▪ Definition based on literature ▪ Sub dimensions Improving reliability (internal consistency) o Use of multiple items that measure the same construct Cronbach’s alpha o Internal consistency of the scale o Possible values: 0 – 1.0 o Optimal values: 0.7 – 0.9 Academic skills 2 Lecture 4 Quantitative data analysis • • Measurement levels o Ratio ▪ There is an absolute zero o Interval ▪ Distance between values is meaningful ▪ Distance can be expressed in numbers o Ordinal ▪ Values can be ordered ▪ Difference can’t be expressed in numbers o Nominal ▪ Words, not numbers Measures of central tendency o Mean ▪ Average o Median ▪ Middle number o Mode ▪ Most common number Academic skills 2 Lecture 6 Types of validity • • • • Construct validity o Do you measure what you intent to measure? Internal validity o Assessing causal relations External validity o Can you generalize the results? Statistical conclusion validity o Type 1 or type 2 error Academic skills 2 Creswell & Creswell – Ch. 1 Research designs • • • Quantitative o Experimental ▪ applied behavioural analysis/single-subject analysis o Non experimental ▪ causal-comparative research, correlational design o Longitudinal Qualitative o Narrative research (humanities) ▪ Studies the lives of individuals/asks stories about their lives o Phenomenology (philosophy/psychology) ▪ Describes experiences of individuals about a phenomenon o Grounded theory (sociology) ▪ Derives a general theory of a process, action or interaction grounded in the views of participants. o Ethnography (anthropology/sociology) ▪ Studies a cultural group over a long time o Case study ▪ Develops an in-depth analysis of a case, collecting detailed information Mixed methods o Convergent ▪ Merging of quantitative/qualitative data. Collected at same time and both integrated in the interpretation of the results. o Explanatory sequential ▪ First quantitative research, then explaining results with qualitative o Exploratory sequential ▪ First qualitative research, then quantitative o Complex designs with embedded core designs ▪ Combination of previous designs to make complex mixed method Academic skills 2 Creswell & Creswell – Ch. 3 Quantitative theory use • • • • • Theory in quantitative research o Interrelated set of constructs/variables formed into propositions/hypotheses, that specify the relationship among variables Theoretical rationale o Specifying how and why variables and relational statements are interrelated Developing a theory 1) Combine independent, mediating and dependent variables into questions >> these provide information about the type and magnitude of their relationship 2) Form this information into a hypothesis 3) Researchers test hypothesis over and over > a theory emerges 4) Someone gives it a name Levels of theories o Micro-level ▪ Explanations about small slices of time, space or numbers of people o Macro-level ▪ Explanations about larger aggregates o Meso-level ▪ Link micro- and macro-levels Forms of theory o Interconnected hypotheses ▪ E.g. the higher this, the greater that o If-then logic ▪ If {independent variable}, then {dependent variable} o Visual models ▪ Visual picture of the variables (e.g. arrows that show relations) Qualitative theory use • • Theory in qualitative research o As a broad explanation for behavior and attitudes. o As a theoretical lens or perspective. o As the end point/end goal. Forms of theory o Propositional generalization ▪ Summary of interpretations and claims o Naturalistic generalization ▪ Own personal experiences of researcher o Pattern theory ▪ Explanation that develop during naturalistic or qualitative research Academic skills 2 • Qualitative theoretical perspectives o Feminist perspective ▪ View women’s diverse situations and the institutions that frame those situations as problematic o Racialized discourses ▪ Raise important questions about the control and production of knowledge about people/communities of color o Critical theory ▪ Concerned with empowering human beings to transcend the constraints placed on them by race, gender or class o Queer theory ▪ Focuses on queer community. Transfers the voices and experiences of those who are suppressed o Disability inquiry ▪ Understanding this population’s sociocultural perspective, allowing them to take control over their lives Mixed methods theory use • Two frameworks o Social science framework ▪ Presented as a literature review, conceptual model or theory ▪ Places the theory at the beginning as a priori framework ▪ Names the theory followed by how it mixes quan and qual components and discusses studies that used the theory ▪ Includes a diagram of the theory ▪ Should be framework for quan and qual data collection efforts ▪ Review of the theory at the end o Participatory-social justice framework ▪ Have the effect of involving participants collaboratively in the research ▪ Especially applicable to the study of community health issues and marginalized groups ▪ Transformative approach ▪ Important that people are treated respectfully through gathering and communicating data and the reporting of the results. ▪ Ends the study by advocating for change to help the population/issues Academic skills 2 Creswell & Creswell – Ch. 7 Quantitative research question • • • • Three approaches o Experiment/group comparisons o Survey o Descriptive study Use of questions OR hypotheses Two forms of hypotheses o Null: no differences o Alternative: difference Two models for questions/hypotheses o Descriptive: describing something o Inferential: drawing inferences from a sample to a population Mixed methods research question • • Should have at least 3 RQs! o Qualitative question o Quantitative question o Mixed method question: combining/integrating the other two ▪ Hybrid/integrated question Three forms of question: o Conveying the methods/procedures in the study o Conveying the content of the study o Combining method and content as hybrid question Academic skills 2 Creswell & Creswell – Ch. 8 Research design • • Survey design o Quantitative description of trends, attitudes and opinions of a population by studying a sample o Descriptive questions, questions about relationships between variables or questions about predictive relationships between variables over time o Rationale description: ▪ Identify the purpose ▪ Indicate why survey is best approach ▪ Indicate whether data will be collected cross-sectional or longitudinal ▪ Specify the form of data collection and explain why Experimental design o Manipulates one or more variables in order to evaluate how this manipulation impacts an outcome o Between-subject design ▪ Compares two or more groups o Within-group design ▪ Studies only one group o Mixed design: ▪ Studies including both between-subject and within-group designs Sampling • • • Sampling design o Single stage o Multistage/clustering Stratification o Proportions in the sample reflect proportions in the population Response bias with surveys o When nonresponses would have changed the results o Wave analysis ▪ Examine survey every week to determine if average responses change Academic skills 2 Creswell & Creswell – Ch. 10 Mixed method designs • • • • • • • • Convergent design o Single-phase data collection. o Collecting quan and qual data using same variables/concepts o Three phases: 1) Analysing qual data by coding/making broad themes 2) Analysing quan data statistically 3) Integrating both databases: mixed methods analysis o Ways of merging qual and quan: • Side-by-side comparison • Present findings after each other, compare in discussion • Data transformation • Take qual themes/codes and count to form quan measures • Joint display • Merge both data in table or graph and make interpretation of the display Explanatory sequential design o Two-phase data collection. o Collect quan data, analyse results, explain results with qual data o Both samples should be the same people Exploratory sequential design o Two-phase data collection o Collect qual data and analyse to explore, then build a feature to be tested, then quan data and analysis o Both samples should be from same population, but different people Complex designs o Intersecting mixed methods in other designs Mixed methods experimental (intervention) design o Collecting quan and qual data and integrating info within an experiment or intervention trial Case study design o Integrates one or more core designs into a case study • Deductive: establish the cases through quan and qual data • Inductive: collect and analyse both data and form cases and make comparisons among the cases Participatory-social justice design o Integrates one or more core designs into a participatory/social justice study Evaluation design o Integrates one or more core designs into the steps of an evaluation procedure. (e.g. evaluating the success of an intervention or policy