Introduction to Policy Processes Dan Laitsch • 1
Overview (Class meeting 5) Sign in Agenda PBL break out, final project polishing Centre Jobs Review last class Stats PBL planning (presentations) Policy Conclusions [Lunch] Action research Course review Evaluation PBL and dismiss • 2
Centre Jobs Program Assistant (CSELP) Identify, organize, and provide an overview of electronic education policy resources in Canada, including Federal and provincial government resources; think tanks, policy centres, professional organizations, and NGOs; judicial decisions and resources; research resources and data repositories; and news and information sources. Graduate Student Editor (IJEPL) Assist with review of articles; responsible for article layout and posting.
Class : Review Cohort break outs Mid term assessment results Significance and t-tests Policy and unifying content Action research • 4
Part IV:Significantly DifferentUsing Inferential Statistics Chapter 12 Two Groups Too Many? Try Analysis of Variance (ANOVA)
What you learned in Chapter 12 • What Analysis of Variance (ANOVA) is and when it is appropriate to use • How to compute the F statistic • How to interpret the F statistic
Analysis of Variance (ANOVA) • Used when more than two group means are being tested simultaneously • Group means differ from one another on a particular score / variable • Example: DV = GRE Scores & IV = Ethnicity • Test statistic = F test • R.A. Fisher, creator
Path to Wisdom & Knowledge • How do I know if ANOVA is the right test?
Different Flavors of ANOVA • ANOVA examines the variance between groups and the variances within groups • These variances are compared against each other • Similar to t Test. ANOVA has more than two groups • Single factor (or one way) ANOVA • Used to study the effects of 2 or more treatment variables • One-way ANOVA for repeated measures • Used when subjects subjected to repeated measures.
More Complicated ANOVA • Factorial Design • More than one treatment/factor examined • Multiple Independent Variables • One Dependent Variable • Example – 3x2 factorial design
Computing the F Statistic • Rationale…want the within group variance to be small and the between group variance large in order to find significance.
Hypotheses • Null hypothesis • Research hypothesis
Omnibus Test • F test is an “omnibus test” and only tells you that a difference exist • Must conduct follow-up t tests to find out where the difference is… • BUT…Type I error increases with every follow-up test / possible comparison made
Glossary Terms to Know • Analysis of variance • Simple ANOVA • One-way ANOVA • Factorial design • Omnibus test • Post Hoc comparisons
Part IV: Significantly Different Chapter 14 Cousins or Just Good Friends? Testing Relationships Using the Correlation Coefficient
What you will learn in Chapter 14 • How to test the significance of the correlation coefficient • The interpretation of the correlation coefficient • The distinction between significance and meaningfulness (Again!)
The Correlation Coefficient • Remember…correlations examine the relationship between variables they do not attempt to determine causation • Examine the “strength” of the relationship • Range -1 to +1 • Direct relationships • Positive correlations • Indirect relationships • Negative correlations
Computing the Test Statistic • Use the Pearson formula
So How Do I Interpret… • r(27) = .393, p < .05? • r is the test statistic • 27 is the degrees of freedom • .393 is the obtained value • p < .05 is the probability • Critical value (Table B4) for r(27) is .3494
Causes and Associations (Again!) • Just because two variables are related has no bearing on whether there is a causal relationship. • Example: • Quality marriage does not ensure a quality parent-child relationship • Two variables may be correlated because they share something in common…but just because there is an “association” does not mean there is “causation.”
Significance Versus Meaningfulness(Again, Again!!) • Even if a correlation is significant, it doesn’t mean that the amount of variance accounted for is meaningful. • Example • Correlation of .393 • Squaring .393 shows that the variance accounted for .154 or 15.4% • 84.6% remains unexplained!!! • “What you see is not always what you get.”
Policy (conclusions) • Analysis • Frameworks • Organize • Structure • Cannot explain • Theories • Models • Theme: Science, research as a framework • Frame-->theory-->model
Conclusions • Common pool resource theory • Governance from the common pool • Agenda setting and policy adoption • Advocacy coalitions • Policy networks • Punctuated equalibrium • Incrementalism • Major chance • Rationality and the role of the individual • Asimov and Seldon • Micro-policy and the role of the institutions
Conclusions • Strengthening policy theory • Building logical coherence • Seeking causality • Empirically falsifiable • Defined scope • Useful (presents more than obvious outcomes) • Developing field (mostly descriptive) • From qualitative to testable
Conclusions • Next steps • Clarify and specify (ability to be proven wrong) • Broad in scope • Defines the causal process • Develop a coherent model of the individual • Resolve internal inconsistencies • Develop a research program • Respect and use multiple theories when appropriate