Using Bayesian graphical models to model biases in observational studies and to combine multiple sources of data: application to low birth weight and water disinfection by-products

Summary.  Data in the social, behavioural and health sciences frequently come from observational studies instead of controlled experiments. In addition to random errors, observational data typically contain additional sources of uncertainty such as missing values, unmeasured confounders and selection biases. Also, the research question is often different from that which a particular source of data was designed to answer, and so not all relevant variables are measured. As a result, multiple sources of data are often necessary to identify the biases and to inform about different aspects of the research question. Bayesian graphical models provide a coherent way to connect a series of local submodels, based on different data sets, into a global unified analysis. We present a unified modelling framework that will account for multiple biases simultaneously and give more accurate parameter estimates than standard approaches. We illustrate our approach by analysing data from a study of water disinfection by-products and adverse birth outcomes in the UK.     

Ethnic segregation in context: Social discrimination among native Dutch pupils and their ethnic minority classmates

Social discrimination, defined as the relative preference for intra-ethnic over inter-ethnic relationships, was studied in pupils’ networks in Dutch secondary school classes. While native Dutch pupils (ethnic majority members) mainly named fellow majority members, ethnic minority members reported ties with as many majority as minority members. Considering the ethnic composition of the classroom, however, majority members discriminated less than minorities. No strong effect of classroom ethnic composition on social discrimination was found, whereas neighborhood composition was shown to clearly influence social discrimination: ethnic minority members were more, and majority members less inclined to discriminate in neighborhoods with more ethnic presence.     

Impact of low fertility and early age at sterilisation on women’s formal education and skill development in South India

This paper examines the impact of low fertility and early age at sterilisation on women’s formal education and skill development in South India. Multilevel ordered-logit modelling of pseudo-cohort data re-organised from the three rounds of National Family Health Survey, and thematic analysis of qualitative data collected from Tamil Nadu and Kerala states showed no evidence of women’s resumption of formal education or uptake of skill development training in the post-sterilisation and post-childcare period. While resuming formal education in the post-sterilisation and post-childcare period is harder to achieve for various individual, household, community and policy reasons, there is greater preparedness and support for women to undertake skill development training. As low fertility and early age at sterilisation are widely regarded as the emerging reproductive norm in India, post-sterilisation and -childcare women will be a significant population group both in number and in proportional terms. No government policies or programs have so far recognised this group. India’s new government should consider targeted skill development programs for post-sterilisation and -childcare women appropriate to their social, economic and educational levels. An important contribution of the family planning program, particularly female sterilisation, for the economic and social development of the family and the wider society will otherwise be lost.     

Signal extraction goodness-of-fit diagnostic tests under model parameter uncertainty: Formulations and empirical evaluation

We present a time-domain goodness-of-fit (gof) diagnostic test that is based on signal-extraction variances for nonstationary time series. This diagnostic test extends the time-domain gof statistic of Maravall (2003 Maravall, A. (2003). A class of diagnostics in the ARIMA-model-based decomposition of a time series. Memorandum, Bank of Spain. Available at http://www.bde.es/servicio/software/tramo/diagnosticsamb.pdf [Google Scholar]) by taking into account the effects of model parameter uncertainty, utilizing theoretical results of McElroy and Holan (2009 McElroy, T., Holan, S. (2009). A local spectral approach for assessing time series model misspeci?cation. Journal of Multivariate Analysis 100:604–621.[Crossref], [Web of Science ®] , [Google Scholar]). We demonstrate that omitting this correction results in a severely undersized statistic. Adequate size and power are obtained in Monte Carlo studies for fairly short time series (10 to 15 years of monthly data). Our Monte Carlo studies of finite sample size and power consider different combinations of both signal and noise components using seasonal, trend, and irregular component models obtained via canonical decomposition. Details of the implementation appropriate for SARIMA models are given. We apply the gof diagnostic test statistics to several U.S. Census Bureau time series. The results generally corroborate the output of the automatic model selection procedure of the X-12-ARIMA software, which in contrast to our diagnostic test statistic does not involve hypothesis testing. We conclude that these diagnostic test statistics are a useful supplementary model-checking tool for practitioners engaged in the task of model-based seasonal adjustment.     

Gendered experiences of physical restraint on locked wards for women

Physical restraint is used in inpatient services for people with intellectual disabilities as a way of holding a person to avoid injury. This article uses data from an ethnographic study in a locked unit in the north of England to explore women’s experiences of physical restraint using a feminist disability studies analysis. Data consists of field notes as well as interviews with 16 of the women who had experienced restraint, and 10 staff who worked with them. The women gave insights into the gendered phenomenon of restraint in light of their past experiences of violence. The authors argue that restraint is used with women to encourage passivity at times when more relational and therapeutic methods could be used. The article offers recommendations for alternative strategies that services can encourage.     

Toddlers Involve Their Caregiver to Help Another Person in Need

Three experiments with 18‐ to 35‐month‐old children (n = 169) studied toddler–caregiver interactions when being confronted with another person in need. In particular, we explored whether toddlers would request their caregiver to help a needy other when they are not able to help themselves. Children observed another person who needed help to accomplish a task, but were either not able to provide help as the object was out of reach (Experiment 1) or because an obstacle prevented children from interacting with the other person (Experiments 2, 2b). The experiments revealed the same developmental trend with 2.5‐year‐olds selectively involving their caregiver to help the needy other. The results are interpreted in terms of toddlers’ motivation to see others helped and with respect to their developing ability to actively involve others to regulate their emotions.     

Opening the Book

Open access book publishers often face different challenges than their counterparts in journal publishing. In particular, the funding models and discovery solutions that work for journal articles may not always function as well for books. Charles Watkinson of the University of Michigan Press answered questions about recent activities in these two areas. University of Michigan Press has participated in Knowledge Unlatched, a platform that allows a large network of libraries to financially support the open access publication of books from several publishers. In order to enhance discovery of these titles and learn more about their use, the press has also begun participating in the new Open Access eBooks on JSTOR program.     

Islamophobia and the Problematization of Mosques: A Critical Exploration of Hate Crimes and the Symbolic Function of “Old” and “New” Mosques in the United Kingdom

Most scholarly studies have tended to focus on the building of new and proposed mosques, and in particular how they are sites of conflict and contestation symbolic of wider “problems” associated with Muslims and Islam in the United Kingdom. This study focuses on an overlooked aspect within this, the extent to which attacks on mosques that are neither new nor proposed perform a similar symbolic function. Presenting new empirical evidence from research undertaken with ten mosques across the United Kingdom that had been targeted for attack, we begin by exploring the existing literature on the problematization of mosques using the lens of critical Islamophobia studies to do so. Setting out what is known about attacks on mosques in the British setting, empirical findings from the research are used to illustrate the type and manifestation of attacks experienced, going on to consider the drivers and catalysts for them. Exploring the similarities and differences between the conflict and contestation associated with new mosques and the attacks on mosques that are not new, this study concludes that some resonance exists in the symbolic function mosques continue to serve in the community. In conclusion, the significant resonance between Islamophobically motivated attacks against mosques with those against the individuals is considered.     

A Blueprint for Full Collective Flood Risk Estimation: Demonstration for European River Flooding

Floods are a natural hazard evolving in space and time according to meteorological and river basin dynamics, so that a single flood event can affect different regions over the event duration. This physical mechanism introduces spatio‐temporal relationships between flood records and losses at different locations over a given time window that should be taken into account for an effective assessment of the collective flood risk. However, since extreme floods are rare events, the limited number of historical records usually prevents a reliable frequency analysis. To overcome this limit, we move from the analysis of extreme events to the modeling of continuous stream flow records preserving spatio‐temporal correlation structures of the entire process, and making a more efficient use of the information provided by continuous flow records. The approach is based on the dynamic copula framework, which allows for splitting the modeling of spatio‐temporal properties by coupling suitable time series models accounting for temporal dynamics, and multivariate distributions describing spatial dependence. The model is applied to 490 stream flow sequences recorded across 10 of the largest river basins in central and eastern Europe (Danube, Rhine, Elbe, Oder, Waser, Meuse, Rhone, Seine, Loire, and Garonne). Using available proxy data to quantify local flood exposure and vulnerability, we show that the temporal dependence exerts a key role in reproducing interannual persistence, and thus magnitude and frequency of annual proxy flood losses aggregated at a basin‐wide scale, while copulas allow the preservation of the spatial dependence of losses at weekly and annual time scales.     

Asking about the future: methodological insights from energy biographies

Temporality is fundamental to qualitative longitudinal (QLL) research, inherent in the design of returning to participants over time, often to explore moments of change. Previous research has indicated that talking about the future can be difficult, yet there has been insufficient discussion of methodological developments to address these challenges. This paper presents insights from the Energy Biographies project, which has taken a QLL and multimodal approach to investigating how everyday energy use can be understood in relation to biographical pasts and imagined futures. In particular, we detail innovative techniques developed within the project (e.g. SMS photograph activities) to elicit data on anticipated futures, in ways that engender thinking about participants’ own biographical futures and wider societal changes. We conclude by considering some of the significant benefits and challenges such techniques present. These methodological insights have a wider relevance beyond the substantive topic for those interested in eliciting data about futures in qualitative research.