Predicting pickle harvests using a parametric feedforward neural network

Feedforward networks have demonstrated their ability to model non-linear data. Despite this success, their use as a statistical analysis tool has been limited by the persistent assumption that these networks can only be implemented as non-parametric models. In fact, a feedforward network can be used for parametric modeling, with the result that many of the common parametric testing procedures can be applied to the nonlinear network. In this paper, a feedforward network for predicting the biological growth rate of pickles is developed. Using this network, the parametric nature of the network is demonstrated. Once trained, the network model is tested using standard parametric methods. In order to facilitate this testing, it is first necessary to develop a method for calculating the degrees of freedom for the neural network, and the residual covariance matrix. It is shown that the degrees of freedom is determined by the number of parameters that actually contribute to an output. With this information, the covariance matrix can be created by adapting the error matrix. Using these results, the trained network is tested using a simple F-statistic.     

The Analysis of Designed Experiments and Longitudinal Data by Using Smoothing Splines

In designed experiments and in particular longitudinal studies, the aim may be to assess the effect of a quantitative variable such as time on treatment effects. Modelling treatment effects can be complex in the presence of other sources of variation. Three examples are presented to illustrate an approach to analysis in such cases. The first example is a longitudinal experiment on the growth of cows under a factorial treatment structure where serial correlation and variance heterogeneity complicate the analysis. The second example involves the calibration of optical density and the concentration of a protein DNase in the presence of sampling variation and variance heterogeneity. The final example is a multienvironment agricultural field experiment in which a yield–seeding rate relationship is required for several varieties of lupins. Spatial variation within environments, heterogeneity between environments and variation between varieties all need to be incorporated in the analysis. In this paper, the cubic smoothing spline is used in conjunction with fixed and random effects, random coefficients and variance modelling to provide simultaneous modelling of trends and covariance structure. The key result that allows coherent and flexible empirical model building in complex situations is the linear mixed model representation of the cubic smoothing spline. An extension is proposed in which trend is partitioned into smooth and non-smooth components. Estimation and inference, the analysis of the three examples and a discussion of extensions and unresolved issues are also presented.     

Emptying the Nest: Older Men in the United States, 1880–2000

Between 1880 and 2000, the percentage of married men 60 and older living only with their wives in empty nest households rose from 19 percent to 78 percent. Data drawn from the US census show that more than half of this transformation occurred in the 30‐year period from 1940 to 1970, bookended by moderate increases between 1880 and 1940 and very modest increases after 1970. Two literatures have presented demographic, cultural, and economic explanations for the decline in elderly co‐residence with their children, but none adequately accounts for a sharp change in the mid‐twentieth century. Both aggregate comparisons and multivariate analysis of factors influencing the living arrangements of elderly men suggest that economic advances for all age groups in the critical 30‐year period, along with trends in fertility and immigration, best explain the three‐stage shift that made the empty nest the dominant household form for older men by the beginning of the twenty‐first century.     

Family support in late life: A review of the literature on aging,disability, and family caregiving

ABSTRACT

For older adults and people with disabilities in the United States, family caregiving is an important part of remaining at home and in the community. As care recipients and caregivers age, family dynamics change, and the health, social, and financial impacts of this (largely unremunerated) work have implications for individuals, families, and social policy. In this review, the authors map the literature across multiple fields related to disability and aging to understand caregiving in late life, what it means to be an older caregiver and/or to care for older people. The authors summarize the findings of 97 articles to address the care, services, and supports family caregivers provide for older adults; negative and positive impacts for caregivers serving in this role; supports that family members use or need; and societal impact of family caregiving. Much of the literature describes the work family caregivers provide and negative impacts of caregiving. Less attention is devoted to caregiving benefits, supports used by family caregivers, and societal impacts. The authors conclude with an agenda for future research that attends to the need for research that includes: more diverse samples, new types of caregivers, longitudinal data, qualitative data and analysis, and comparative research.     

Using process evaluation to determine effectiveness of participatory ergonomics training interventions in construction

The construction industry continues to experience high rates of musculoskeletal injuries despite the widespread promotion of ergonomic solutions. Participatory ergonomics (PE) has been suggested as one approach to engage workers and employers for reducing physical exposures from work tasks but a systematic review of participatory ergonomics programs showed inconclusive results.. A process evaluation is used to monitor and document the implementation of a program and can aid in understanding the relationship between the program elements and the program outcomes. The purpose of this project is to describe a proposed process evaluation for use in a participatory ergonomic training program in construction workers and to evaluate its utility in a demonstration project among floor layers.     

An organizational model to distinguish between and integrate research and evaluation activities in a theory based evaluation

Theory-based evaluation (TBE) is an evaluation method that shows how a program will work under certain conditions and has been supported as a viable, evidence-based option in cases where randomized trials or high-quality quasi-experiments are not feasible. Despite the model's widely accepted theoretical appeal there are few examples of its well-implemented use, probably due to time and money limitations necessary for planning and a confusion over the definitions between research and evaluation functions and roles. In this paper, we describe the development of a theory-based evaluation design in a Math and Science Partnership (MSP) research project funded by the U.S. National Science Foundation (NSF). Through this work we developed an organizational model distinguishing between and integrating evaluation and research functions, explicating personnel roles and responsibilities, and highlighting connections between research and evaluation work. Although the research and evaluation components operated on independent budgeting, staffing, and implementation activities, we were able to combine datasets across activities to allow us to assess the integrity of the program theory, not just the hypothesized connections within it. This model has since been used for proposal development and has been invaluable as it creates a research and evaluation plan that is seamless from the beginning.     

Confidence in preferences

Indeterminate preferences have long been a tricky subject for choice theory. One reason for which preferences may be less than fully determinate is the lack of confidence in one’s preferences. In this paper, a representation of confidence in preferences is proposed. It is used to develop and axiomatise an account of the role of confidence in choice which rests on the following intuition: the more important the decision to be taken, the more confidence is required in the preferences needed to take it. This theory provides a natural account of when an agent should defer a decision; namely, when the importance of the decision exceeds his confidence in the relevant preferences. Possible applications of the notion of confidence in preferences to social choice are briefly explored.     

A Longitudinal Study of the Developmental Trajectories of Parental Attachment and Career Maturity of South Korean Adolescents

This study tested the developmental trajectories of career maturity (CM) and parental attachment (PA), the longitudinal influence of both, and gender as a moderator. Findings showed developmental progressions in adolescents’ PA and CM over 4 years. The change in PA was positively related to the developmental change in CM. For gender, there was a statistically significant difference in the initial status of CM and PA but no significant difference in developmental change. This study verified the longitudinal influence of PA on CM and showed that the cross‐sectional effect of predictors on CM needs to be distinguished from the longitudinal effect in future studies.     

In the Eye of the Beholder: Exploring Variations in the Perceptions of Pornography

While many studies have explored the correlates and outcomes associated with pornography use, the measurement of such use has relied heavily on self-definitions of pornography from research participants, and little is known regarding how individuals define pornography for themselves. Using a diverse sample of 2,089 individuals sampled from the online MTurk Web site, participants were asked to rate whether they felt 20 different examples of sexual media were pornography. Differences by gender, religious attendance, marital status, and pornography use patterns were explored. Results suggested significant variation across the sample in how specific examples of sexual media were perceived. In addition, response profiles significantly differed by gender, religious attendance, marital status, and use of pornography. Implications for the measurement of pornography are discussed.