Secondary Analysis & Meta Analysis
- 1. EDUCATIONAL RESEARCH BASED ON METHOD- SECONDARY ANALYSIS & META ANALYSIS JOSHINI JOSE FIRST SEMESTER ,M.Ed SPS,MGU SPSMPCO3-INTRODUCTION TO EDUCATIONAL RESEARCH AND STATISTICS
- 2. DESCRIPTIVE RESEARCH Describes & interprets what it is: conditions or relationships that exist, opinions held, processes going on ,effects evident or trends that are developing Primarily concerned with present Deals with relationship between variables, hypothesis testing & development of generalizations, principles or theories that has a universal validity
- 3. Either correlational research or causal comparative research or neither describing existing conditions Unlike experimental research no manipulation of variable, Appropriate for behavioural science
- 4. UNIQUE CHARACTERISTICS Involves hypothesis formulation & testing Employ logical methods of inductive – deductive reasoning to arrive at generalization Employ methods of randomization Accurate description of variables and procedures helpful for other researchers Methods are non-experimental
- 5. METHOD-1.SECONDARY ANALYSIS Reanalyzing data gathered by a previous investigator But may have a different hypothesis, different experimental designs or different methods of statistical analysis Subject and the data are same Purpose and method of analysis different
- 6. CHARACTERISTICS OF SECONDARY DATA Insight into total situation Helpful in hypothesis formulation Supplementary information Cost, time &effort saving Helpful in testing the hypothesis
- 7. FACTORS TO BE CONSIDERED AVAILABILITY RELEVANCE ADEQUACY ACCURACY RELIABILITY/CONSISTENCY
- 8. DISADVANTAGES Quality of research Not specific Incomplete information outdated
- 9. ADVANTAGES O SECONDARY ANALYSIS The new investigator bring objectivity or a fresh point to the investigation Better questions will be raised or hypothesis to be tested It brings greater expertise to the area of investigation and greater skill in experimental design and statistical analysis
- 10. Saves time & money Useful for students of research methodology by enabling them to use real data, rather than simulated or inferior data for exercise purpose Played an important part in educational research
- 11. 2.META-ANALYSIS It is a quantitative approach for systematically combining results of previous research to arrive at conclusions about the body of research A study collects data from individual subjects (such as 100 subjects=100 “data points”) Meta-analysis collects data from individual studies(such as 100 studies=100 “data points”)
- 12. GLASS(1976) coined the term “META ANALYSIS” It refers to the analysis of analysis It summarizes data from several individual studies that concern a specific research question
- 13. Meta analysis is considered as the statistical analysis of a large amount of analyzed results from individual studies for the purpose of integrating the findings It starts with selecting studies with similar variables & population samples followed by- identifying & coding study characteristics & finally statistically analyzing & reporting the assessed findings of the meta analysis study
- 14. FUNCTION OF META- ANALYSIS Identify heterogeneity in effects among multiple studies and, where appropriate, provide summary measure Increase statistical power & precision Develop, refine & test hypothesis Reduce subjectivity of study comparisons by using systematic &explicit comparison procedure
- 15. Identify data gap in the knowledge base & suggest direction for future research Calculate sample size for future studies
- 16. STEPS IN META ANALYSIS 1.Define hypothesis 2.Locate the studies 3.Input data 4.Calculate effect size 5.Analyze variables
- 17. DEFINE HYPOTHESIS There must be a well defined statement of the relationship between the variables under investigation So that the researcher can carefully define the inclusion & exclusion criteria when locating potential studies
- 18. LOCATE THE STUDIES Meta Analysis should adequately summarize the existing literature, therefore a good literature search is critical to retrieve every relevant study
- 19. INPUT DATA Empirical findings are gathered from primary studies & input into statistical database.
- 20. CALCULATE EFFECT SIZE The overall effect is calculated by converting all statistics to a common metric, making adjustments as necessary to correct different issues
- 21. ANALYSE VARIABLES Analyze the variables if heterogeneity exist. Some researchers insist upon analyzing moderating variables as appropriate E.g. :A study on coping behaviour of patients with cancer
- 22. THANK YOU