Russo ioe_nov12

Click here to load reader

download Russo ioe_nov12

of 42

  • date post

    03-Dec-2014
  • Category

    Documents

  • view

    1.563
  • download

    1

Embed Size (px)

description

 

Transcript of Russo ioe_nov12

  • 1. Interactions between the individual and the group:reflections from multilevel modellingin educational research Federica Russo Center Leo Apostel, VrijeUniversiteitBrussel& Centre for Reasoning, University of Kent

2. OverviewPhilosophy of education and empirical research Reverse the question: does empirical research look into philosophy?Multilevel models in educational research Definition and examples The need for an accompanying substantive theoryA substantive theory for multilevel Recent work by Little and Yilokoski Main features 2 3. PHILOSOPHY OF EDUCATION ANDEMPIRICAL RESEARCH3 4. Does PhilEd pay enough attention to empirical research?PhillipsHyslop-Margison&NaseemJ Phil Ed 39(4), (2005) Phl Ed Archive (2007) PhilEd hasnt paid enoughand serious attention to Straw man: wrong selectionempirical researchof critics It is possible to study Counterexamples: PhilEdnormative processes does pay attention toempirically empirical research Mutual benefit of PhilEd Problem of empiricaland empirical research to generalisabilitylook into real cases 4 5. PhilSci and PhilEduPhillips (2005, p.582): The marked change in doing philosophy of science came about when it was realised that there was much to gain by taking scientific research seriously, rather than discussing an artefact of the philosophers imagination. [] The present essay is making a call for a parallel revolution in philosophical discussions of educational research, a revolution that entails taking examples of educational research seriously. [] Processes that humans engage in, in the real world, whether normative or cultural or psychological (or all three at once) can be studiedand probably ought to be studiedempirically, but they also need to be assessed in terms of the values (and if relevant the conception of education) that they embody.5 6. Turn the question on its headDoes empirical research look sufficientlyinto philosophy of education? (Or, for the matter, into philosophy?) 7. EMPIRICAL RESEARCH 7 8. Modelling in the social sciencesCausal relations in social contextsMarital problems migrationMaternal education child survivalStress + physical health + self-rated healthTwo approachesQualitative: smaller and focused samplesQuantitative: statistical analyses of large data 8 9. A crash courseMULTILEVEL MODELS9 10. Why multilevel?An example of quantitative methods used inempirical research in education Going quantitative, the new panacea for evidenceBut is it really panacea?It models hierarchical structures Typical of social (and education) contextsA sounding board for the question: does empirical research look into philosophy?10 11. Multilevel modelsA special type of statistical model used in causal analysis to model hierarchical structures: Individuals / family / local population / national population Firms / regional market / national market / global market Pupils / classes / school / school systemsNo a priori reason to choose the level of analysisActually, good reasons to study the interactions betweenthe levels 11 12. Traditional approachesHolism the system as a whole determines the behaviour of the parts in a fundamental way; the properties of a given system cannot be reduced to the mere sum of its componentsIndividualism social phenomena and behaviours have to be explained by appealing to individual decisions and actions, without invoking any factor transcending them 12 13. The statistical counterpartsAggregate-level models explain aggregate-level outcomes through aggregate-level variablesIndividual-level models explain individual-level outcomes by individual-level explanatory variables13 14. Types of variablesIndividual: measure individual characteristics, takevalues of each of the lower units in the sample. e.g. income of each individual in the sampleAggregate: summary of the characteristics ofindividuals composing the group e.g.: mean income of state residents 14 15. DangersAtomistic fallacy wrongly infer a relation between units at a higher level of analysis from units at a lower level of analysisEcological fallacy draw inferences about relations between individual level variables based on the group level data 15 16. Robinson: illiteracy and immigration1930 census in the US, for each of 48 states + district ofColumbiaIndividual correlation: descriptive properties of individuals Positive correlation: immigrants more illiterate than native citizensEcological correlation: descriptive properties of groups Negative correlation: correlation between being foreign-born and illiterate magnified and in the reversed directionExplanation: immigrants tend to settle down in states wherenative population is more literate 16 17. Courgeau: Farmers migration in NorwayData from the Norwegian population registry (since 1964) andfrom two national censuses (1970 and 1980)Aggregate model and individual model show opposite results: Aggregate: regions with more farmers are those with higher rates of migrations; Individual: in a same region migration rates are lower for farmers than for non-farmersReconciliation: multilevel model aggregate characteristics (e.g. the percentage of farmers) explain individual behaviour (e.g. migrants behaviour)17 18. Types of models - summaryIndividual: explain individual-level outcomes by individual- level explanatory variables e.g.: explain the individual probability of migrating through the individual characteristics of being/not being farmerAggregate: explain aggregate-level outcomes throughexplanatory aggregate-level variables e.g.: explain the percentage of migrants in a region through the percentage of people in the population having a certain occupational status (e.g. being a farmer)Multilevel: make claims across the levels, from the aggregate- level to the individual-level and vice-versa e.g.: explain the individual probability to migrate for non-farmers through the percentage of farmers in the same region18 19. Grouping in multilevelUnits grouped at different levels, a-contextual languageGrouping may be more or less randomOnce the grouping is done, differentiation: group and its member influence and are influenced by the group membership19 20. Statistical modelling of hierarchiesYij 0 j 1 j x ij 2 z j ijresponse variable at theindividual level explanatory variable at the individual levelexplanatory variable at the group leveli: index for the individualsj: index for the groupthese vary depending on the group Errors are independent at each level and between levels20 21. Goldstein:Multilevel in educational researchStudy school effectiveness, examination results, All quantifiable aspects of educationStatistical advantages of multilevelEfficient estimates of regression coefficientCorrect standard errors, confidenceintervals, significance tests for the clustersEnables measuring differences between clustershttp://www.math.helsinki.fi/msm/banocoss/Goldstein_course.pdf21 22. Hierarchies in educational researchSimple hierarchy:Pupil / class / school / neighbourhood /Cross-classified structurePupil ethnicity // school neighbourhood // 22 23. Goldstein et al: examination results and school differencesInner London schools Response variable: examination results Explanatory variables: standardised London reading tests, verbal reasoning category, gender, school gender (mixed, boys, girls), school religious denomination (State, Church of England, Roman Catholic, other)Results: Small effect of school gender; Roman Catholic slightly better; girls better than boys; large differences for different verbal reasoning categories.Differences between schools in examination results depend on intake achievement and curriculum subject consideredNo single dimension in which schools differ23 24. Driessen: School composition and primary school achievementDutch primary schools Response variable: language and math proficiency Explanatory variables: parental ethnicity and education, pupils sex and age, school composition, ethnic diversityResults: Quite strong effect of school composition on language, weak on math; all children, independently of background, perform worse in schools with high ethnic diversityQuestion about distribution policy and other measures 24 25. [] despite their usefulness, models formultilevel analysis cannot be a universalpanacea.[]They are notsubstitutes for wellgrounded substantive theories []Multilevel models are tools to be used withcare and understanding. Goldstein, Multilevel statistical models, http://www.bristol.ac.uk/cmm/team/hg/multbook1995.pdf 25 26. WHAT SUBSTANTIVE THEORYFOR MULTILEVEL?26 27. Modelling and explainingWhat does a multilevel model model? Relations between different levels in a hierarchical structureWhat does a multilevel model explain? How group behaviour influences individual behaviour (but not vice-versa)Statistically, multilevel achieves bothBut the substantive theory is still wanting27 28. What needs the substantive theorySchool religious denominationWhat social practices, norms, values are involved?School composition and ethnicityHow do these influence peer relation among pupils?What is the extra information that we need and that statistics does not give us? 28 29. LEVELS IN A SUBSTANTIVE THEORY 29 30. Levels, beyond statisticsDan Little Levels of the social:Ontology what social entities?Explanation reduction?Causation causal powers?Inquiry what level?Description what level requirements?Generalisation recurrence of types? Avoid analogies with natural sciences, dont reify social phenomena 30 31. Levels, beyond the received viewsMethodological individualism Holism Social facts must be Social entities andreducible to facts about structures have primacy andindividualsare independent There is no higher level Individuals are influencedwithout lower levelby social facts, but do not influence them E.g.: Austrian schooleconomics, some political E.g.: sociologists in thescientists Durkheim tradition 31 32. M