Bereavement in twin relationships: an exploration of themes from a study of twinship.

The primary foci of twin research have been on twins as a biological phenomenon, on the management of medical complications of twin pregnancy and birth and on the study of individual differences and development by psychologists and geneticists for whom twins provide a powerful research tool. The twin relationship - twinship - has received less research attention, including the aspect of bereavement which may occur at any stage of life. This paper presents some findings from a qualitative study that used grounded theory methodology to explore the nature of twinship especially issues relevant to psychological counselling with twins. Fourteen participants were interviewed in-depth about their understanding and experience of twinship. Participants were selected who had a rich knowledge and experience of twinship and of therapeutic work. Segments of transcribed interviews were constantly compared to describe and elaborate frequently repeated concepts in the data. Two of the main themes of discernment, the core category of the theory of twinship, are described. These are 'defining boundaries' and 'managing ambiguity'. Aspects of these as experienced by twins are discussed in the light of current psychological theories about bereavement.     

The comparative constitution of twinship: strategies and paradoxes.

In both traditional and modern societies, twinship, as an unusual mode of reproduction, involves difficulties for social systems in maintaining consistent classification systems. It is proposed that the most prevalent response to twinship involves various 'strategies of normalisation' to defuse and contain the potential disruption. This proposition is illustrated and analysed in relation to ethnographic maternal drawn mainly (but not exclusively) from African communities in the twentieth century. Following a discussion of twin infanticide as the most extreme of the normalising strategies, the article concludes by identifying a number of paradoxes in the social construction of twinship.     

Preadolescent development: case studies in twins.

The authors studied eight sets of healthy twins, ranging in physical maturity from prepubertal to late pubertal, and their parents, to assess psychological changes in early adolescence. A phase-specific psychosocial regression correlates with the biological onset of puberty. We present preliminary findings related to twinship, parental reactions, and longitudinal development, based on detailed case studies. Parents experienced increased conflict, detachment, and pride in response to their twins'' psychosocial development at puberty. The more advanced child in a pair tended to lead in all spheres of adjustment-school success, heterosexual interest, peer friendships, and independent behavior. Although the more advanced child began to break away from his twin, in general the twin relationships remained close. Differences in personality traits and rate of psychosocial development within a twinship tended to remain consistent or to become accentuated. They were occasionally narrowed but rarely reversed from early childhood through early adolescence. Some of the differences were fostered by dissimilar patterns of identifying with the parents.     

Estimating marginal proportions and intraclass correlations with clustered binary data

A logistic regression with random effects model is commonly applied to analyze clustered binary data, and every cluster is assumed to have a different proportion of success. However, it could be of interest to obtain the proportion of success over clusters (i.e. the marginal proportion of success). Furthermore, the degree of correlation among data of the same cluster (intraclass correlation) is also a relevant concept to assess, but when using logistic regression with random effects it is not possible to get an analytical expression of the estimators for marginal proportion and intraclass correlation. In our paper, we assess and compare approaches using different kinds of approximations: based on the logistic‐normal mixed effects model (LN), linear mixed model (LMM), and generalized estimating equations (GEE). The comparisons are completed by using two real data examples and a simulation study. The results show the performance of the approaches strongly depends on the magnitude of the marginal proportion, the intraclass correlation, and the sample size. In general, the reliability of the approaches get worsen with low marginal proportion and large intraclass correlation. LMM and GEE approaches arises as reliable approaches when the sample size is large.     

Comparison of bioaerosol sampling methods in barns housing swine.

The air in livestock buildings contains bioaerosol levels that are sufficiently high to cause adverse health effects in animals and workers. These bioaerosols are complex mixtures of live and dead microorganisms and their products as well as other aeroallergens. The effectiveness of sampling methods used for quantifying the very high concentrations of microorganisms in these environments has not been well studied. To facilitate an accurate assessment of respiratory hazards from viable organisms in agricultural environments, three bioaerosol sampling methods were investigated: the Andersen microbial sampler method (AMS), the all-glass impinger method (AGI), and the Nuclepore filtration-elution method (NFE). These methods were studied in a parallel fashion in 24 swine confinement buildings. Measurements were taken in two seasons with three types of culture media in duplicate to assess total bacteria, gram-negative enteric bacteria, and total fungi. Methods were analyzed for the proportion of samples yielding data within the limits of detection, intraclass reliability, and correlation between methods. For sampling viable bacteria, the AMS had a poor data yield because of overloading and demonstrated weak correlation with the AGI. Conversely, the AGI and NFE gave sufficient numbers of valid data points (90%), yielded high intraclass reliabilities (alpha greater than or equal to 0.92), and were highly correlated with each other (r = 0.86). The AGI and the NFE were suitable methods for assessing bacteria in this environment, but the AMS was not. The AMS was the only method that consistently recovered enteric bacteria (73% data yield). For sampling fungi, the AGI and AMS both yielded sufficient data and all three methods demonstrated high intraclass reliability.(ABSTRACT TRUNCATED AT 250 WORDS)     

The estimation of heritability of some developmental characters in natural populations ofArabidopsis thaliana (L.)Heynh

The investigation concerns to what extent the phenotypic variability of some characters connected with the developmental rate inArabidopsis thaliana (L.)Heynh. is of genetic nature. For this reason, in a variable natural populationSt? the values of coefficients of heritability were estimated by means of offspring-parent regression coefficients and of intraclass correlation coefficients for the three following characters: “number of days to appearance of the flower primordia”, “number of rosette leaves” and “number of rosette leaves per day”. The regression coefficients were calculated (1) by using simple progeny means, (2) the maternal data were repeated with each individual offspring value, and (3) the weighted regression coefficients were used. It has been found that about a half of phenotypic differences in developmental characters studied is determined genetically. The question is discussed whether such heritability estimates can be influenced by dominance or interactions, as it results from considerations about the population structure inArabidopsis. There are no substantial differences between the methods used, and in the given case, the estimation of heritability with the aid of the intraclass correlation coefficient appears to be very advantageous.     

Comparison of methods for calculating the heritability of adult field resistance to yellow rust and grain yield in spring wheat

Several methods are available for estimating heritability in disomic species, including parent-offspring regression, realized heritability, intraclass correlations of recombinant inbred lines, and diallel-cross analysis. Estimates were obtained by these various methods for a set of eight bread wheat (Triticum aestivum) lines adapted to the East African highlands, which had been intercrossed and selfed in a half-diallel arrangement to give F₁, F₂ and F₃ generations, and F₆ recombinant inbred lines. Significant genetic variation existed among parents and crosses for both grain yield and yellow rust resistance in all generations. Based on the heritability calculated from the analysis of F₆ recombinant inbred lines, analysis of the F₂ diallel crosses was recommended for determining the heritability of both characters in early segregating generations. The results also suggest that a form of tandem selection may be effective in developing locally adapted germplasm which combines high grain yield with yellow rust resistance. Received: 15 February 1999 / Accepted: 11 March 1999     

Control charting instrumental analyses

Control chart methodology has important applications in the analytical chemistry laboratory. This paper explores application of statistical process control (SPC) techniques to chemical measuring systems and highlights problems when data are correlated. Instrumental methods of chemical analysis typically generate correlated quality-control (QC) sample data owing to the effect of calibration. That is, QC sample measurements stemming from the same calibration curve will tend to be more closely related than measurements derived from different calibration curves. Data correlation will bias the estimated sample variance and essentially invalidate the control limits if not accounted for. A way of calculating statistically valid control limits in the presence of intraclass correlation is examined and an illustrative example given.     

Proportional Hazards Regression for the Analysis of Clustered Survival Data from Case–Cohort Studies

Summary Case–cohort sampling is a commonly used and efficient method for studying large cohorts. Most existing methods of analysis for case–cohort data have concerned the analysis of univariate failure time data. However, clustered failure time data are commonly encountered in public health studies. For example, patients treated at the same center are unlikely to be independent. In this article, we consider methods based on estimating equations for case–cohort designs for clustered failure time data. We assume a marginal hazards model, with a common baseline hazard and common regression coefficient across clusters. The proposed estimators of the regression parameter and cumulative baseline hazard are shown to be consistent and asymptotically normal, and consistent estimators of the asymptotic covariance matrices are derived. The regression parameter estimator is easily computed using any standard Cox regression software that allows for offset terms. The proposed estimators are investigated in simulation studies, and demonstrated empirically to have increased efficiency relative to some existing methods. The proposed methods are applied to a study of mortality among Canadian dialysis patients.     

Family studies of tooth size factors in the permanent dentition.

Correlations between dimensions of the permanent teeth in Australian Aboriginals were studied by factor analysis to disclose the main sources of shared variability. Findings indicated that in both males and females most of the common variability in the tooth dimensions could be accounted for by factors representing mesiodistal size of anterior teeth, buccolingual size of anterior teeth, generalized size of the premolars and generalized size of the molars. Factor scores derived from the analysis were used to calculate intraclass correlations among brothers and among sisters. These correlations tended to be higher for the factors contributing most to the common variability indicating that the factors might represent fields under direct genetic control. There was no trend for intraclass correlations among siblings derived from multivariate scores to be consistently higher than those based on observed tooth dimensions. The main advantage to the user of factor analysis is the ability to interpret associations between interrelated variables more objectively than is possible by conventional correlation methods.     

Estimating intraclass correlation for binary data

This paper reviews many different estimators of intraclass correlation that have been proposed for binary data and compares them in an extensive simulation study. Some of the estimators are very specific, while others result from general methods such as pseudo-likelihood and extended quasi-likelihood estimation. The simulation study identifies several useful estimators, one of which does not seem to have been considered previously for binary data. Estimators based on extended quasi-likelihood are found to have a substantial bias in some circumstances.     

Multiple regression methods show great potential for rare variant association tests

The investigation of associations between rare genetic variants and diseases or phenotypes has two goals. Firstly, the identification of which genes or genomic regions are associated, and secondly, discrimination of associated variants from background noise within each region. Over the last few years, many new methods have been developed which associate genomic regions with phenotypes. However, classical methods for high-dimensional data have received little attention. Here we investigate whether several classical statistical methods for high-dimensional data: ridge regression (RR), principal components regression (PCR), partial least squares regression (PLS), a sparse version of PLS (SPLS), and the LASSO are able to detect associations with rare genetic variants. These approaches have been extensively used in statistics to identify the true associations in data sets containing many predictor variables. Using genetic variants identified in three genes that were Sanger sequenced in 1998 individuals, we simulated continuous phenotypes under several different models, and we show that these feature selection and feature extraction methods can substantially outperform several popular methods for rare variant analysis. Furthermore, these approaches can identify which variants are contributing most to the model fit, and therefore both goals of rare variant analysis can be achieved simultaneously with the use of regression regularization methods. These methods are briefly illustrated with an analysis of adiponectin levels and variants in the ADIPOQ gene.     

Quantifying the impact of fixed effects modeling of clusters in multiple imputation for cluster randomized trials

In cluster randomized trials (CRTs), identifiable clusters rather than individuals are randomized to study groups. Resulting data often consist of a small number of clusters with correlated observations within a treatment group. Missing data often present a problem in the analysis of such trials, and multiple imputation (MI) has been used to create complete data sets, enabling subsequent analysis with well-established analysis methods for CRTs. We discuss strategies for accounting for clustering when multiply imputing a missing continuous outcome, focusing on estimation of the variance of group means as used in an adjusted t-test or ANOVA. These analysis procedures are congenial to (can be derived from) a mixed effects imputation model; however, this imputation procedure is not yet available in commercial statistical software. An alternative approach that is readily available and has been used in recent studies is to include fixed effects for cluster, but the impact of using this convenient method has not been studied. We show that under this imputation model the MI variance estimator is positively biased and that smaller intraclass correlations (ICCs) lead to larger overestimation of the MI variance. Analytical expressions for the bias of the variance estimator are derived in the case of data missing completely at random, and cases in which data are missing at random are illustrated through simulation. Finally, various imputation methods are applied to data from the Detroit Middle School Asthma Project, a recent school-based CRT, and differences in inference are compared.     

Reliability and accuracy of handheld stopwatches compared with electronic timing in measuring sprint performance

This study assessed reliability of split times obtained by handheld stopwatches (HHSs) compared with electronic timing (ET) during a 200-m sprint. Two HHS timing methods were compared with ET: single-split timers (SST) and multiple-split timers (MST). Twenty-six timers without previous experience were given instruction and completed practice trials until good agreement was achieved between ET and HHS. Trained runners (8 males, 10 females) were timed for each 25-m interval on a standard 200-m course. Repeated-measures analysis of variance and intraclass correlation models were used to determine reliability. A total of 248 split times were analyzed. No significant differences were found between the three timing methods (p > 0.99), and calculated intraclass correlation values were high (0.988). Mean error between SST, MST, and ET (-0.04 +/- 0.24 and -0.05 +/- 0.24 seconds, respectively) indicated faster HHS times, though not significantly. However, absolute errors were considerably larger (0.15 +/- 0.20 and 0.16 +/- 0.19 between SST, MST, and ET, respectively). The HHS-recorded splits were faster than ET in 67.3% of splits and slower in 29.4%. The distribution of errors made the development of a reliable correction factor to convert HHS to ET impossible. It was concluded that on the basis of the small mean error and high intraclass correlations, the use of HHSs may be a viable alternative to ET in collecting group data. However, on the basis of the absolute error between HHS and ET, when high degrees of precision are required, ET should be used, and reliable correction of HHS to ET values is not possible. It was further concluded that HHS times should be reported without attempting correction and interpreted in light of the shortcomings of the HHS method.     

Evaluation of fecal mRNA reproducibility via a marginal transformed mixture modeling approach

Background  

Developing and evaluating new technology that enables researchers to recover gene-expression levels of colonic cells from fecal samples could be key to a non-invasive screening tool for early detection of colon cancer. The current study, to the best of our knowledge, is the first to investigate and report the reproducibility of fecal microarray data. Using the intraclass correlation coefficient (ICC) as a measure of reproducibility and the preliminary analysis of fecal and mucosal data, we assessed the reliability of mixture density estimation and the reproducibility of fecal microarray data. Using Monte Carlo-based methods, we explored whether ICC values should be modeled as a beta-mixture or transformed first and fitted with a normal-mixture. We used outcomes from bootstrapped goodness-of-fit tests to determine which approach is less sensitive toward potential violation of distributional assumptions.     

A comparison of methods for determining compartmental analysis parameters

The traditional method for determining compartmental analysis parameters relies on a visual selection of data points to be used for regression of data from each cellular compartment. This method is appropriate when the compartments are kinetically discrete and are easily discernible. However, where treatment effects on compartment parameters are being evaluated, a more objective method for determining initial parameters is desirable.

Three methods were examined for determining initial isotopic contents and half-times of ⁸⁶Rb elution from cellular compartments using theoretical data with known parameters. Experimental data from roots of Douglas fir (Pseudotsuga menziesii [Mirb.] Franco) and barley (Hordeum vulgare L.) intact seedlings were also used. The three methods were a visually assisted, linear regression on data of semilog plot of isotope elution versus time, a microcomputer-assisted, linear regression on semilog plot where maximization of the square of the correlation coefficient (r²) was the criterion to determine data points needed for each regression and a mainframe computer-assisted, direct nonlinear regression on elution data using a model of the sum of three exponential decay functions. The visual method resulted in the least accurate estimates of compartmental analysis parameters. The microcomputer-assisted and nonlinear regression methods calculated the parameters equally well.

     

Parametric mode regression for bounded responses

We propose new parametric frameworks of regression analysis with the conditional mode of a bounded response as the focal point of interest. Covariate effects estimation and prediction based on the maximum likelihood method under two new classes of regression models are demonstrated. We also develop graphical and numerical diagnostic tools to detect various sources of model misspecification. Predictions based on different central tendency measures inferred using various regression models are compared using synthetic data in simulations. Finally, we conduct regression analysis for data from the Alzheimer's Disease Neuroimaging Initiative to demonstrate practical implementation of the proposed methods. Supporting Information that contain technical details and additional simulation and data analysis results are available online.     

Analyses of gene frequencies

Models of variance components and their intraclass correlational equivalences are developed for genes falling into various categories of subdivisions within a population. Estimable functions are elaborated demonstrating that intraclass correlations can be estimated only relative to that for the least related genes in the informational system. The effects of different types of subdivisions-and of ignoring them-on the parameters are demonstrated. Small sample estimators are formulated for all of the parameters by three different methods, including both a weighted and an unweighted method of analysis of the variation among subpopulations. How estimators change with assumptions about the parameters is illustrated. Various tests of hypotheses are outlined in chi(2) and F-test terminology. Discussed are factors which may affect the correlations and the manner in which their effects are manifest, hopefully in clarification of some of the misconceptions that have arisen in this connection.