Likelihood‐based and marginal inference methods for recurrent event data with covariate measurement error

Recurrent event data arise commonly in medical and public health studies. The analysis of such data has received extensive research attention and various methods have been developed in the literature. Depending on the focus of scientific interest, the methods may be broadly classified as intensity‐based counting process methods, mean function‐based estimating equation methods, and the analysis of times to events or times between events. These methods and models cover a wide variety of practical applications. However, there is a critical assumption underlying those methods–variables need to be correctly measured. Unfortunately, this assumption is frequently violated in practice. It is quite common that some covariates are subject to measurement error. It is well known that covariate measurement error can substantially distort inference results if it is not properly taken into account. In the literature, there has been extensive research concerning measurement error problems in various settings. However, with recurrent events, there is little discussion on this topic. It is the objective of this paper to address this important issue. In this paper, we develop inferential methods which account for measurement error in covariates for models with multiplicative intensity functions or rate functions. Both likelihood‐based inference and robust inference based on estimating equations are discussed. The Canadian Journal of Statistics 40: 530–549; 2012 © 2012 Statistical Society of Canada     

The KIND Program for Adolescent Family and Dating Violence

Adolescent relational violence has significant and pervasive impacts for young people's current and future relationships. It is known that there is a cross-over between adolescent family violence and other offending behaviour, meaning there is a need for youth justice systems to recognise and respond to these forms of violence. This paper outlines a pilot intervention, the KIND Program, run in Youth Justice South Australia, and aimed at intervening with adolescent family and dating violence. KIND aims to provide tailored systemic intervention to young people and their support network, reducing violence while increasing connection and strength within relationships. The pilot of KIND ran for seven months in 2017, and had eight families participating in the program. Participation in, feedback from, and impact of the program suggest that participation was worthwhile and useful for young people and their families. The KIND Program would benefit from a more extensive pilot period to explore the outcomes of this form of intervention.     

The Relationship Between Distance from Gambling Venues and Gambling Participation and Problem Gambling Among U.S. Adults

In this article we examine the relationship between extent of gambling for U.S. adults and the distance from their residence to the nearest casino or track. We employ data from a telephone survey of U.S. adults conducted in 2011–2013. The chances that the respondents gambled in the past year, were frequent gamblers, or were problem gamblers were greater if they lived close to a casino. The chances that the respondents gambled in the past year or were frequent gamblers were greater if they lived close to a horse or dog track. The effects of closeness to a casino on the likelihood of past-year gambling, frequent gambling, and problem gambling, as well as the effect of closeness to a track on past-year gambling, extended to about 30 miles from the respondent’s home. In addition, the concentration of casinos within 30 miles of the respondent’s home was positively related to the respondents’ chance of being a frequent or problem gambler. If a respondent had no casinos within 30 miles, he or she had a 2.7 % chance of being a problem gambler; if one casino, a 3.9 % chance; if six or more, a 6.2 % chance. The authors estimate that at least part of this effect is causal.     

Establishing turf grass increases soil greenhouse gas emissions in peri-urban environments

Urbanization is becoming increasingly important in terms of climate change and ecosystem functionality worldwide. We are only beginning to understand how the processes of urbanization influence ecosystem dynamics and how peri-urban environments contribute to climate change. Brisbane in South East Queensland (SEQ) currently has the most extensive urban sprawl of all Australian cities. This leads to substantial land use changes in urban and peri-urban environments and the subsequent gaseous emissions from soils are to date neglected for IPCC climate change estimations. This research examines how land use change effects methane (CH₄) and nitrous oxide (N₂O) fluxes from peri-urban soils and consequently influences the Global Warming Potential (GWP) of rural ecosystems in agricultural use undergoing urbanization. Therefore, manual and fully automated static chamber measurements determined soil gas fluxes over a full year and an intensive sampling campaign of 80 days after land use change. Turf grass, as the major peri-urban land cover, increased the GWP by 415 kg CO₂-e ha^?1 over the first 80 days after conversion from a well-established pasture. This results principally from increased daily average N₂O emissions of 0.5 g N₂O ha^?1 d^?1 from the pasture to 18.3 g N₂O ha^?1 d^?1 from the turf grass due to fertilizer application during conversion. Compared to the native dry sclerophyll eucalypt forest, turf grass establishment increases the GWP by another 30 kg CO₂-e ha^?1. The results presented in this study clearly indicate the substantial impact of urbanization on soil-atmosphere gas exchange in form of non-CO₂ greenhouse gas emissions particularly after turf grass establishment.     

Critical factors for improving social sustainability of urban renewal projects

This study reviews the sustainable urban design concept and identifies critical factors for enhancing social sustainability of urban renewal projects. Through a questionnaire survey carried out in Hong Kong, the opinions of architects, planners, property development managers, and local citizens were sought and evaluated. The results derived from factor analysis indicated that certain design features should be incorporated for achieving social sustainability. “Satisfaction of Welfare Requirements”, “Conservation of Resources & the Surroundings”, “Creation of Harmonious Living Environment”, “Provisions Facilitating Daily Life Operations”, “Form of Development” and “Availability of Open Spaces” were believed to be the significant underlying factors for enhancing social sustainability of local urban renewal projects.

Examining the content validity of the WHOQOL-BREF from respondents’ perspective by quantitative methods

Content validity, the extent to which a measurement reflects the specific intended domain of content, is a basic type of validity for a valid measurement. It was usually examined qualitatively and relied on experts’ subjective judgments, not on respondents’ responses. Therefore, the purpose of this study was to introduce and demonstrate how to use quantitative methods to examine the content validity for a certain measurement from respondents’ viewpoint with quantitative methods. In the current study, content validity of the WHOQOL-BREF was examined with quantitative methods among 102 undergraduate students and 128 community adults. They were asked to rate the appropriateness of each item with the four domains of the WHOQOL-BREF respectively and also asked to sort the items into the four domains. Then, three quantitative methods on examining content validity were applied to analyze the rating and sorting data, including (1) proportion of substantive agreement, (2) substantive validity, and (3) ANOVA approach. These results were used to compare with the original content structure of the WHOQOL-BREF, to see if the original structure is consistent with the structure of subjects’ judgments. The results showed that the content structure gained from appropriateness rating and item sorting was not totally consistent with the original content structure of the WHOQOL-BREF. Among 24 items, 12 items did not have adequate content validity. More discussion on these items and the issue of content validity were further provided.     

Comparative Assessment of Methods for Measuring Consensual Poverty: Sort Card Versus CAPI

Using data from the Wisconsin Longitudinal Study, the purpose of this study is twofold. First, the study identifies coping strategies used by older adults. Second, the study examines the impact of older adults’ chosen coping strategies on mortality reduction. The study focuses specifically on differences in the use of religious and secular coping strategies among this population. The findings suggest that although coping strategies differ between those who self-classify as religious and those who self-classify as nonreligious, for both groups social approaches to coping (e.g., attending church and volunteering) are more likely than individual approaches (e.g., praying or active/passive coping) to reduce the risk of mortality. The most efficacious coping strategies, however, are those matched to characteristics of the individual.     

Missing Data Mechanisms for Analysing Longitudinal Data with Incomplete Observations in Both Responses and Covariates

Missing observations in both responses and covariates arise frequently in longitudinal studies. When missing data are missing not at random, inferences under the likelihood framework often require joint modelling of response and covariate processes, as well as missing data processes associated with incompleteness of responses and covariates. Specification of these four joint distributions is a nontrivial issue from the perspectives of both modelling and computation. To get around this problem, we employ pairwise likelihood formulations, which avoid the specification of third or higher order association structures. In this paper, we consider three specific missing data mechanisms which lead to further simplified pairwise likelihood (SPL) formulations. Under these missing data mechanisms, inference methods based on SPL formulations are developed. The resultant estimators are consistent, and enjoy better robustness and computation convenience. The performance is evaluated empirically though simulation studies. Longitudinal data from the National Population Health Survey and Waterloo Smoking Prevention Project are analysed to illustrate the usage of our methods.     

Sex preferences for children in Ghana: the influence of educational attainment

Education influences aspects of demographic behaviour and outcomes including a child sex preference. Sex preferences of children have been studied in different societies because of its associated social and demographic implications. Using the 2014 Ghana Demographic and Health Survey, we examined the association between educational attainment and sex preferences of children. Findings from the study indicated that there is preference for sons (26.1%) compared to daughters (17.4%). At higher levels of education, there is a higher likelihood for no preference for a sex of a child. Among the characteristics of respondents that influenced sex preferences are: gender, lineage, religion, occupation and desired family size. Acquisition of knowledge through education to some extent alter fertility preferences and hence the need to motivate individuals to attain some level of education.