Classifying Speech Sonority Functional Data using a Projected Kolmogorov-Smirnov Approach

This paper addresses a linguistically motivated question of classification of functional data, namely the statistical classification of languages according to their rhythmic features. This is an important open problem in phonology. The analysis is based on the information provided by the sonority, which is an index of local regularity of the speech signal. Our main tool is the projected Kolmogorov-Smirnov test. This is a new goodness of fit test for functional data. The result obtained supports the linguistic conjecture of the existence of three rhythmic classes.     

LYING ABOUT WHAT YOU KNOW OR ABOUT WHAT YOU DO?

We compare communication about private information to communication about actions in a one‐shot 2‐person public good game with private information. The informed player, who knows the exact return from contributing and whose contribution is unobserved, can send a message about the return or her contribution. Theoretically, messages can elicit the uninformed player's contribution, and allow the informed player to free‐ride. The exact language used is not expected to matter. Experimentally, however, we find that free‐ride depends on the language: the informed player free‐rides less—and thereby lies less frequently—when she talks about her contribution than when she talks about the return. Further experimental evidence indicates that it is the promise component in messages about the contribution that leads to less free‐ride and less lying.     

Bridging 1990 and 2000 census race data: Fractional assignment of multiracial populations

In contrast to previous censuses, Census 2000 permitted individuals to mark more than one race. Because the new race tables include both single-race and mixed-race categories, measuring change during the 1990s requires some method of bridging between the two data sets.To accomplish this bridging, we first identified biracial populations as of 1990 through the race and ancestry responses of individuals in the PUMS file. With race responses assumed to represent a person's primary race identity, we then determined the percentage of each biracial group that preferred each race as the primary identity. The same percentages can be used to assign biracial persons from Census 2000 into single-race categories. We also provide fractional assignment percentages for selected states and for the larger specific nationality groups of mixed-race Asians.Comparison of our 1990 estimates of the numbers in leading biracial groups with those reported in Census 2000 suggests that our fractional assignment values are reasonable for biracial groups other than those involving American Indians and Alaska Natives. For the latter biracial groups and for all groups representing three or more races, we recommend equal fractional assignment into the appropriate single-race categories.     

The school experience of children in residential care: A multiple case study

Children in residential care are the out‐of‐home population group with the lowest rate of access to postsecondary education. Research aimed at understanding their situation within the school context suggests that besides learning progress and outcomes, subjective experiences should be taken into account. Qualitative research based on a multiple‐case study was designed and carried out in Spain with the aim of deepening the understanding of the individual school experiences of three children in residential care. Each case was analysed from the perspective of the child, the educators and the teachers. Semistructured, one‐to‐one interviews were carried out, and children's files were consulted. The results pointed to emotional distress and accumulated educational delay as common traits of the three children in the study. In shaping each particular school experience, the key themes identified included (1) individual characteristics, (2) the children's academic aspirations and strategies, (3) relationships with peers and teachers, (4) response experienced in school, (5) the complex structure of the residential centre, (6) the sense of belonging to the residential centre, (7) team work and (8) political action and investment in inclusive education.     

On multivariate confidence regions and simultaneous confidence limits for ratios

Convenient general linear model computational procedures are presented for constructing multivariate confidence regions and simultaneous confidence limits for ratios of linear combinations of the parameters. The practical consequence is that a single general linear model computer program, capable of validating the underlying model and estimating the parameters, can (after slight modification) also construct the confidence regions, and even determine their precise analytic form (ellipsoid, hyperboloid, etc.). The text is deliberately factual while the appendices extend and help clarify earlier work by Henry Scheffe. As an example, a confidence ellipse and simultaneous confidence limits are constructed for several relative potencies in a classical multiple parallel line bioassay.     

Three models of collaborative child protection: what is their influence on short stays in foster care?

The paper articulates three models of collaboration between child protection services and law enforcement agencies on investigations of child maltreatment, and examines the relative impact of each model on child removals and short‐term foster care placements. An extensive review of the research literature is used to derive lessons for jurisdictions using the different models. Promising strategies from several jurisdictions are described as they relate to collaborative investigations, child removal and short‐term placements. Finally, we provide recommendations to improve child protection investigations and reduce the use of short‐term foster care placements.     

Incorporating support constraints into nonparametric estimators of densities

Let f_n ^? (x) be the usual Parzen-Rosenblatt kernel estimator of the pdf f of a random variable X based on a sample X₁,…,X_n from X.In many practical applications,it is knownt hat X>c and/or X<d for given constants c and d.Additionally, one might know the values of(c)and/or f(d).“mirrorimage”and“tieddown”modifications of f_n ^?incorporate this additional information into an estimator f_n which has support [c,d].This estimatoris interpreted in a manner which allows one to use most of the known convergence properties of kernel estimates in studying the behavior of f_n.     

A bartlett correction factor for a likelihood ratio statistic in multivariate bioassay

A more accurate Bartlett correction factor is proposed for a likelihood ratio statistic in multivariate bioassay.