Selecting the Optimal Transformation of a Continuous Covariate in Cox's Regression: Implications for Hypothesis Testing

ABSTRACT

Background: Many exposures in epidemiological studies have nonlinear effects and the problem is to choose an appropriate functional relationship between such exposures and the outcome. One common approach is to investigate several parametric transformations of the covariate of interest, and to select a posteriori the function that fits the data the best. However, such approach may result in an inflated Type I error. Methods: Through a simulation study, we generated data from Cox's models with different transformations of a single continuous covariate. We investigated the Type I error rate and the power of the likelihood ratio test (LRT) corresponding to three different procedures that considered the same set of parametric dose-response functions. The first unconditional approach did not involve any model selection, while the second conditional approach was based on a posteriori selection of the parametric function. The proposed third approach was similar to the second except that it used a corrected critical value for the LRT to ensure a correct Type I error. Results: The Type I error rate of the second approach was two times higher than the nominal size. For simple monotone dose-response, the corrected test had similar power as the unconditional approach, while for non monotone, dose-response, it had a higher power. A real-life application that focused on the effect of body mass index on the risk of coronary heart disease death, illustrated the advantage of the proposed approach. Conclusion: Our results confirm that a posteriori selecting the functional form of the dose-response induces a Type I error inflation. The corrected procedure, which can be applied in a wide range of situations, may provide a good trade-off between Type I error and power.     

Adoptive Families in Distress

In this study, features of 22 adoptive families were examined in the light of some possible explanations of the reported over-representation of adoptive families among child psychiatry referrals. Themes which emerged included problems with bonding, motivations for adoption and psychiatric illness of parents. Implications for adoption workers are discussed.¹     

Returns to scale to family size - who gains from high fertility?

Summary In this paper the hypothesis put forward by J. C. Caldwell in a number of recent articles, including one in this journal is tested, that there are social and economic increasing returns to scale to family size. Using two village samples of household data collected in Bangladesh in 1977, it is shown that there is a statistically significant inverse correlation between adjusted consumption of rice per head and family size. This result holds true when age and size of landholding are taken into account. From this, it is concluded that the hypothesis is not empirically supported by the data from Bangladesh. This result is partially explained by the extreme poverty of the region and the fact that women are exploited in Bangladesh society.     

Fostering Integrity in Postgraduate Research: An Evidence-Based Policy and Support Framework

Postgraduate research students have a unique position in the debate on integrity in research as students and novice researchers. To assess how far policies for integrity in postgraduate research meet the needs of students as “research trainees,” we reviewed online policies for integrity in postgraduate research at nine particular Australian universities against the Australian Code for Responsible Conduct of Research (the Code) and the five core elements of exemplary academic integrity policy identified by Bretag et al. (2011 Bretag, T., Mahmud, S., Wallace, M., Walker, R., Green, M., East, J., James, C., McGowan, U. and Partridge, L. 2011. Core elements of exemplary academic integrity policy in Australian higher education. International Journal for Educational Integrity, 7: 3–12.  [Google Scholar]), i.e., access, approach, responsibility, detail, and support. We found inconsistency with the Code in the definition of research misconduct and a lack of adequate detail and support. Based on our analysis, previous research, and the literature, we propose a framework for policy and support for postgraduate research that encompasses a consistent and educative approach to integrity maintained across the university at all levels of scholarship and for all stakeholders.     

Intersectional oppression: A reflexive dialogue between Muslim academics and their experiences of Islamophobia and exclusion in UK Higher Education

Islamophobia has seen a disturbing rise in British and international contexts; however, we see limited attention being afforded to this in the context of Higher Education (HE), particularly the impacts of this through the lens of staff members. HE continues to frame itself as a post-racial, secular, progressive, and inclusive space, yet is perpetually opposed due to its surreptitious and violent incidences of racism, classism, sexism and ableism. We discuss here, using liberatory-based and autoethnographic methodologies, Muslim staff members' experiences and insights of Islamophobia in UK HE institutions (HEIs)—a severely underrepresented group and largely invisible issue in the academy. We highlight the need to expose, challenge, question, and destabilise the continuous victimisation of Muslims within these spaces and beyond. We unpick issues related to intersectionality and different levels of exclusion by exploring the sense of belonging felt by Muslim staff; specific challenges and episodes of Islamophobia faced; and ways the sector can dismantle Islamophobia and begin developing practices to genuinely support anti-Muslim, anti-racist inclusion and social justice.