Non-significant p-values? Strategies to understand and better determine the importance of effects and interactions in logistic regression.

In the context of generalized linear models (GLMs), interactions are automatically induced on the natural scale of the data. The conventional approach to measuring effects in GLMs based on significance testing (e.g. the Wald test or using deviance to assess model fit) is not always appropriate. The objective of this paper is to demonstrate the limitations of these conventional approaches and to explore alternative strategies for determining the importance of effects. The paper compares four approaches to determining the importance of effects in the GLM using 1) the Wald statistic, 2) change in deviance (model fitting criteria), 3) Bayesian GLM using vaguely informative priors and 4) Bayesian Model Averaging analysis. The main points in this paper are illustrated using an example study, which examines the risk factors for cyber abuse victimization, and are further examined using a simulation study. Analysis of our example dataset shows that, in terms of a logistic GLM, the conventional methods using the Wald test and the change in deviance can produce results that are difficult to interpret; Bayesian analysis of GLM is a suitable alternative, which is enhanced with prior knowledge about the direction of the effects; and Bayesian Model Averaging (BMA) is especially suited for new areas of research, particularly in the absence of theory. We recommend that social scientists consider including BMA in their standard toolbox for analysis of GLMs.

Bayesian modeling predicts age and sex are not required for accurate stature estimation from femoral length.

Despite the recognized flaws in applying traditional stature estimation equations such as those of Trotter and Gleser (1952) [5] to a contemporary population, there are currently no available alternatives for stature estimation in Australia that address these limitations. Post mortem computed tomography (PMCT) DICOM scans of the left and right femora were acquired from 76 Australian deceased individuals aged 17-76 years for metric analysis. Femoral bicondylar length, femoral epicondylar breadth and anterior-posterior (AP) diameter, medial-lateral (ML) diameter, circumference and cortical area at the femoral midshaft were measured on three-dimensional (3D) models to build statistical models for estimating stature. In addition, Australian individuals aged 16-63 years (n=111) were measured in standing and supine positions to aid in the adjustment of supine stature of deceased individuals utilized in this study to standing stature. The results of this preliminary evaluation strongly indicate that the optimal model for estimating stature includes bicondylar femoral length and epicondylar breadth, that the effect of sex as an independent variable is very low, and there is limited practical benefit in including age in the estimation of stature. Our study indicates that the Australian population sampled represents a small yet significant shift in stature from the original Trotter and Gleser sample. Additionally, in the case of fragmentary remains, it was found that epicondylar breadth and AP diameter had the highest probability of accurate stature estimation in the absence of bicondylar femoral length. As stature forms a significant component of a biological profile and therefore aids in the personal identification of human remains, it is important that forensic anthropologists utilize the most accurate methodologies available. Stature estimation of Australian individuals is therefore achieved with higher accuracy through utilizing the femoral equations proposed in this study.

Apophyseal Ossification of the Iliac Crest in Forensic Age Estimation: Computed Tomography Standards for Modern Australian Subadults.

This study contrasts the ontogeny of the iliac crest apophysis using conventional radiography and multislice computed tomography (MSCT), providing probabilistic information for age estimation of modern Australian subadults. Retrospective abdominopelvic MSCT data acquired from 524 Australian individuals aged 7-25 and surveillance radiographs of adolescent idiopathic scoliosis patients included in the Paediatric Spine Research Group Progression Study (n = 531) were assessed. Ossification scoring of pseudo-radiographs and three-dimensional (3D) volume-rendered reconstructions using Risser (1958) quantitative descriptors indicate discrepancies in age estimates, stage allocation, and conflicting morphological progression. To mitigate visualization limitations associated with two-dimensional radiographs, we provide and validate a modified 3D-MSCT scoring tier of ossification, demonstrating complete fusion between 17.3-19.2 and 17.1-20.1 years in males and females. Legal demarcation for doli incapax presumption and age of majority (18 years) can be achieved using probability estimates from a fitted cumulative probit model for apophyseal fusion using the recalibrated standards.