Coherence,contradiction, and the development of school science identities

This study demonstrates the potential for collaborative research among participants in local settings to effect positive change in urban settings characterized by diversity. It describes an interpretive case study of a racially, ethnically, and socioeconomically diverse eighth grade science classroom in an urban magnet school in order to explore why some of the students did not achieve at high levels and identify with school science although they were both interested in and knowledgeable about science. The results of this study indicated that structural issues such as the school's selection process, the discourses perpetuated by teachers, administrators, and peers regarding “who belongs” at the school, and negative stereotype threat posed obstacles for students by highlighting rather than mitigating the inequalities in students' educational backgrounds. We explore how a methodology based on the use of cogenerative dialogues provided some guidance to teachers wishing to alter structures in their classrooms to be more conducive to all of their students developing identities associated with school science. Based on the data analysis, we also argue that a perspective on classrooms as communities of practice in which learning is socially situated rather than as forums for competitive displays, and a view of students as valued contributors rather than as recipients of knowledge, could address some of the obstacles. Recommendations include a reduced emphasis on standardized tasks and hierarchies, soliciting unique student contributions, and encouraging learning through peripheral participation, thereby enabling students to earn social capital in the classroom. © 2010 Wiley Periodicals, Inc. J Res Sci Teach 47: 1209–1228, 2010     

Child sexual abuse and persistence of risky sexual behaviors and negative sexual outcomes over adulthood: Findings from a birth cohort

ObjectivesTo determine the impact of child sexual abuse (CSA) on adult sexual behaviors and outcomes over three age periods.MethodsA longitudinal study of a birth cohort born in Dunedin, New Zealand in 1972/1973 was used. Information on CSA was sought at age 26, and on sexual behaviors and outcomes at ages 21, 26, and 32. Comparisons were over the whole period from age 18 to 32, then for the three age periods from age 18 to 21, 21 to 26, and 26 to 32, adjusting for measures of family environment.ResultsOverall, 465 women and 471 men (91.9% of the surviving cohort) answered questions about CSA. Contact CSA was reported by 30.3% of women and 9.1% of men. For abused women, significantly increased rates were observed for number of sexual partners, unhappy pregnancies, abortion, and sexually transmitted infections from age 18 to 21; with rates approaching those of nonabused over time. Conversely, for abused men rates were not significantly elevated in the youngest age period, but were for number of partners from age 26 to 32 and acquisition of herpes simplex virus type 2 from age 21 to 32.ConclusionsGender and age are critical when considering the effect of CSA. While the profound early impact of CSA demonstrated for women appears to lessen with age, abused men appear to carry increased risks into adulthood.Practice implicationsCSA is common and should be considered when young women present with unwanted conceptions or seek multiple terminations, and when men continue to have high risk sexual behavior into adulthood. Furthermore, if CSA is disclosed, sexual risks in adulthood need to be considered.     

Bayes not Bust! Why Simplicity is no Problem for Bayesians

The advent of formal definitions of the simplicity of a theoryhas important implications for model selection. But what isthe best way to define simplicity? Forster and Sober ([1994])advocate the use of Akaike's Information Criterion (AIC), anon-Bayesian formalisation of the notion of simplicity. Thisforms an important part of their wider attack on Bayesianismin the philosophy of science. We defend a Bayesian alternative:the simplicity of a theory is to be characterised in terms ofWallace's Minimum Message Length (MML). We show that AIC isinadequate for many statistical problems where MML performswell. Whereas MML is always defined, AIC can be undefined. WhereasMML is not known ever to be statistically inconsistent, AICcan be. Even when defined and consistent, AIC performs worsethan MML on small sample sizes. MML is statistically invariantunder 1-to-1 re-parametrisation, thus avoiding a common criticismof Bayesian approaches. We also show that MML provides answersto many of Forster's objections to Bayesianism. Hence an importantpart of the attack on Bayesianism fails.

1. Introduction
2. TheCurve Fitting Problem

   2.1 Curves and families of curves

   2.2 Noise

   2.3 Themethod of Maximum Likelihood

   2.4 ML and over-fitting

3. Akaike's Information Criterion(AIC)
4. The Predictive Accuracy Framework
5. The Minimum MessageLength (MML) Principle

   5.1 The Strict MML estimator

   5.2 Anexample: Thebinomial distribution

   5.3 Properties ofthe SMML estimator

   5.3.1  Bayesianism

   5.3.2  Languageinvariance

   5.3.3Generality

   5.3.4  Consistencyand efficiency

   5.4 Similarity to false oracles

   5.5 Approximationsto SMML

6. Criticisms of AIC

   6.1 Problems with ML

   6.1.1  Smallsample biasin a Gaussian distribution

   6.1.2  Thevon Misescircular and von Mises—Fisherspherical distributions

   6.1.3  The Neyman–Scottproblem

   6.1.4  Neyman–Scott,predictive accuracyandminimum expected KL distance

   6.2 Otherproblems with AIC

   6.2.1  Univariate polynomial regression

   6.2.2  Autoregressiveeconometric time series

   6.2.3  Multivariatesecond-orderpolynomial modelselection

   6.2.4  Gapor no gap:a clustering-like problem forAIC

   6.3 Conclusionsfrom the comparison of MML and AIC

7. Meeting Forster's objectionsto Bayesianism

   7.1 The sub-family problem

   7.2 Theproblem of approximation,or, which framework forstatistics?

8. Conclusion

1. Details of the derivation of the Strict MMLestimator
2. MML, AIC and the Gap vs. No Gap Problem

   B.1 Expectedsize of the largest gap

   B.2 Performanceof AIC on thegap vs. no gap problem

   B.3 Performanceof MML in thegap vs. no gap problem