Cervical cancer

Several statistical methods are available for the analysis of responses with ordinal categories or continuous distributions for the respective visits in longitudinal studies. This paper discusses an alternative nonparametric strategy for studies with more than two groups through Mann-Whitney rank measures of association for all pairs of groups. The proposed method is based on U-statistic theory, and it applies a linear or linear logistic model to the Mann-Whitney estimators for the probabilities of better response for each group relative to each of the others. In addition, the ways of adjusting for covariables and managing stratification factors are explained. Analysis of parallel dose-response relationships for two treatments is illustrated for the proposed method with data from a multicenter study with repeated measurements. A nonparametric estimator for relative potency is provided from the method.     

Pupillographic assessment of sleepiness in sleep-deprived healthy subjects

Medical research frequently involves the statistical comparison of >2 groups, often using data obtained through the application of complex experimental designs. Fortunately, inferential statistical methodologies exist to address these situations. Analysis of variance (ANOVA) in its many forms is used to simultaneously test the equality of all groups in a study. One-way (with 1 independent variable), 2-way (with 2 independent variables), and repeated-measures (patients serve as their own controls) ANOVAs are forms of this technique. Each form has been developed to analyze data from a specific experimental design. Analysis of covariance (ANCOVA) allows the researcher to control for confounding variables that may influence the response of the dependent variable. Finally, multivariate analysis of variance (MANOVA) evaluates the simultaneous responses of multiple dependent variables to > or = 1 independent variable. Whereas ANOVA is the correct alternative to statistically inappropriate multiple t-tests, MANOVA is the correct alternative to statistically inappropriate multiple univariate ANOVA calculations. Use of each of these statistical methods requires an appropriate experimental design and data meeting a number of assumptions. When used properly, each of these methods provides a powerful statistical analysis technique.     

Propensity score methods for bias reduction in the comparison of a treatment to a non-randomized control group

In observational studies, investigators have no control over the treatment assignment. The treated and non-treated (that is, control) groups may have large differences on their observed covariates, and these differences can lead to biased estimates of treatment effects. Even traditional covariance analysis adjustments may be inadequate to eliminate this bias. The propensity score, defined as the conditional probability of being treated given the covariates, can be used to balance the covariates in the two groups, and therefore reduce this bias. In order to estimate the propensity score, one must model the distribution of the treatment indicator variable given the observed covariates. Once estimated the propensity score can be used to reduce bias through matching, stratification (subclassification), regression adjustment, or some combination of all three. In this tutorial we discuss the uses of propensity score methods for bias reduction, give references to the literature and illustrate the uses through applied examples.     

The use of covariance functions and random regressions for genetic evaluation of milk production based on test day records

In the analysis of test day records for dairy cattle, covariance functions allow a continuous change of variances and covariances of test day yields on different lactation days. The equivalence between covariance functions as an infinite dimensional extension of multivariate models and random regression models is shown in this paper. A canonical transformation procedure is proposed for random regression models in large-scale genetic evaluations. Two methods were used to estimate covariance function coefficients for first parity test day yields of Holsteins: 1) a two-step procedure fitting covariance functions to matrices with estimated genetic and residual covariances between predetermined periods of lactation and 2) REML directly from data with a random regression model. The first method gave more reliable estimates, particularly for the periphery of the trajectory. The goodness of fit of a random regression model based on covariables describing the shape of the lactation curve was nearly the same as random regression on Legendre polynomials. In the latter model, two and three regression coefficients were sufficient to fit the covariance structure for additive genetic and permanent environment, respectively. The eigenfunction pattern revealed the possibility of selection for persistency. Covariance functions can be usefully implemented in large-scale test day models by means of random regressions.     

Statistical adjustments and uncontrolled studies.

Many evaluations of social interventions are based on uncontrolled assignments of individuals to treatment groups. Statistical adjustments are often used to compensate for naturally occurring differences between groups. There is much confusion and controversy about the adequacy of these statistical methods. A variety of interrelated problems have been identified, including measurement error, unequal growth rates across groups, and regression artifacts. It is shown that these problems can all be subsumed under a general conceptual framework, as particular examples of model misspecification. This perspective is helpful in revealing clearly the nature of the problems posed by lack of experimental control. The important case of linear adjustment (analysis of covariance) is given special attention. An expression is derived for the proportion of bias remaining after adjustment, in terms of easily interpretable parameters. Implications of these results for research and evaluation design are considered. (41 ref) (PsycINFO Database Record (c) 2010 APA, all rights reserved)     

Child care history and kindergarten adjustment.

Parents gave histories of 589 children just before kindergarten. Children were later assessed with teacher, peer, and observer measures of social adjustment in school. Children with higher day-care amounts in each of 3 eras (0–4, and 4–5 yrs) scored higher on the composite negative adjustment and lower on positive adjustment (however, they also scored lower on teacher-rated internalizing problems). Day care predicted even after statistical control for measures representing alternative explanations, such as family stress and socioeconomic status, accounting for 2.7% of variance in negative adjustment and 2.9% of positive adjustment. Interactions between day care and other variables did not add to predictions of the molar adjustment composites. Extensive infancy care did not in itself predict adjustment, according to planned contrasts that controlled for total amount of day care received across the 3 eras of the child's life. (PsycINFO Database Record (c) 2010 APA, all rights reserved)     

Group sequential monitoring of distribution-free analyses of repeated measures

In many clinical trials the principal analysis consists of a 1 degree of freedom test based on an aggregate summary statistic for a set of repeated measures. Various methods have been proposed for the marginal analysis of such repeated measures that entail estimates of a measure of treatment group difference (the treatment effect) at each of K repeated measures and a consistent estimate of the covariance matrix, where asymptotically these estimates are normally distributed. One can then obtain an overall large sample 1-d.f. test of group differences, such as by taking the average of these K estimates. These methods include the Wei-Lachin family of multivariate rank tests and a corresponding multivariate analysis using the Mann-Whitney difference estimator as a measure of treatment group differences. Other methods, such as O'Brien's non-parametric test, are based on a single summary score for each patient, such as the within-patient mean value. These, and other such methods, allow for some observations to be missing at random. Herein I employ sequential data augmentation to conduct group sequential analyses using a 1 degree of freedom test from a multivariate Mann-Whitney analysis and for the O'Brien rank test. Su and Lachin used this method to perform group sequential analyses of a vector of Hodges-Lehmann estimators. By augmentating the data from the sequential looks in a single analysis, one obtains an estimate of the covariance of the estimates at each look, from which one obtains an estimate of the correlations among the sequential 1-d.f. test statistics. I describe a simple secant algorithm to determine the group sequential boundaries based on recursive integration of the standard multivariate normal distribution with the estimated correlation matrix. Although the boundary obtains readily using the method of Slud and Wei, the more flexible method of Lan and DeMets may be preferred. The true information fraction at each look, needed to apply the spending function method of Lan and DeMets, however, is unknown. Thus, I also describe the use of a surrogate measure of information.     

Therapists as Fixed Versus Random Effects-Some Statistical and Conceptual Issues: A Comment on Siemer and Joormann (2003).

The authors disagree with M. Siemer and J. Joormann's (see record 2003-10163-009) assertion that therapist should be a fixed effect in psychotherapy treatment outcome studies. If treatment is properly standardized, therapist effects can be examined in preliminary tests and the therapist term deleted from analyses if such differences approach zero. If therapist effects are anticipated and either cannot be minimized through standardization or are specifically of interest because of the nature of the research question, the study has to be planned with adequate statistical power for including therapist as a random term. Simulation studies conducted by Siemer and Joormann confounded bias due to small sample size and inconsistent estimates. (PsycINFO Database Record (c) 2010 APA, all rights reserved)     

Planned and post hoc methods for multiple-sample McNemar (1947) tests with missing data.

Pre-post experimental designs with dichotomous dependent variables are encountered frequently in behavioral research. If there is only one group, the McNemar (1947) test can be used to test the hypothesis of no change in parameters. The McNemar test has been extended to cover multiple groups. The problem with these tests is that complete data must be available for all subjects. If post- or pretest data are missing, subjects must be discarded. Ekbohm (1982) provided a solution for the one-group model that has good statistical properties. We extend this method to the multiple-group case. We describe and illustrate procedures for planned and post hoc contrasts. (PsycINFO Database Record (c) 2010 APA, all rights reserved)     

Regression models: calculating the confidence interval of effects in the presence of interactions

The main goal of regression analysis (multiple, logistic, Cox) is to assess the relationship of one or more exposure variables to a response variable, in the presence of confounding and interaction. The confidence interval for the regression coefficient of the exposure variable, obtained through the use of a computer statistical package, quantify these relationships for models without interaction. Relationships between variables that present interactions are represented by two or more terms, and the corresponding confidence intervals can be calculated 'manually' from the covariance matrix. This paper suggests an easy procedure for obtaining confidence intervals from any statistical package. This procedure is applicable for modifying variables which are continuous as well as categorical.     

A quasi-experimental approach to assessing treatment effects in the nonequivalent control group design.

Evaluates 4 statistical tests of treatment effect for the nonequivalent control group design. This design consists of pre- and posttreatment measures of a dependent variable with biased assignments to treatment groups. The biased assignment creates a treatment-pretest confounding for which different statistical techniques adjust. The different statistical tests discussed are the analysis of covariance, analysis of covariance with reliability correction, raw change score analysis, and standardized change score analysis. If assignment to treatment groups is based on the pretest score (a very infrequent event), analysis of covariance is the appropriate mode of analysis. Selection based on the pretest true scores necessitates a reliability correction procedure. Selection based on stable group differences and selection that occurs midway between the pre- and posttest necessitates change score analysis. (35 ref) (PsycINFO Database Record (c) 2010 APA, all rights reserved)     

Measurement of psychological disorders using cognitive diagnosis models.

Cognitive diagnosis models are constrained (multiple classification) latent class models that characterize the relationship of questionnaire responses to a set of dichotomous latent variables. Having emanated from educational measurement, several aspects of such models seem well suited to use in psychological assessment and diagnosis. This article presents the development of a new cognitive diagnosis model for use in psychological assessment--the DINO (deterministic input; noisy "or" gate) model--which, as an illustrative example, is applied to evaluate and diagnose pathological gamblers. As part of this example, a demonstration of the estimates obtained by cognitive diagnosis models is provided. Such estimates include the probability an individual meets each of a set of dichotomous Diagnostic and Statistical Manual of Mental Disorders (text revision [DSM-IV-TR]; American Psychiatric Association, 2000) criteria, resulting in an estimate of the probability an individual meets the DSM-IV-TR definition for being a pathological gambler. Furthermore, a demonstration of how the hypothesized underlying factors contributing to pathological gambling can be measured with the DINO model is presented, through use of a covariance structure model for the tetrachoric correlation matrix of the dichotomous latent variables representing DSM-IV-TR criteria. (PsycINFO Database Record (c) 2010 APA, all rights reserved)     

Interactions for dichotomous variables in repeated measures designs.

Describes a method for analyzing interactions in a repeated measures design for a dichotomous dependent variable. The method is based upon the Cochran Q model but is restricted to the case in which a planned analysis is adopted. The model is illustrated for ordered and unordered independent variables. (PsycINFO Database Record (c) 2010 APA, all rights reserved)     

Analysis of dichotomous variables in repeated measures experiments.

Provides an alternative to the method of L. A. Marascuilo and R. Serlin (1977) for the analysis of repeated measures experiments with dichotomous outcomes. Weighted linear regression methods are used to provide tests analogous to ANOVA. Tests of main effects and interactions are presented. (14 ref) (PsycINFO Database Record (c) 2010 APA, all rights reserved)     

Ordinal invariant measures for individual and group changes in ordered categorical data

Subjective judgements of complex variables are commonly recorded as ordered categorical data. The rank-invariant properties of such data are well known, and there are various statistical approaches to the analysis and modelling of ordinal data. This paper focuses on the non-additive property of ordered categorical data in the analysis of change. A rank-invariant non-parametric method of analysis is presented that is valid regardless of the number of response categories. The unique feature of this method is the augmented ranking approach that is related to the joint distribution of paired observations. This approach makes it possible to measure separately the individual order-preserved categorical changes, which are attributable to the group change, and the individual categorical changes that are not consistent with the pattern of group change. The method is applied to analysis of change in a three-point scale and in a visual analogue scale of continuous ordinal responses.     

Detecting differential item functioning with confirmatory factor analysis and item response theory: Toward a unified strategy.

In this article, the authors developed a common strategy for identifying differential item functioning (DIF) items that can be implemented in both the mean and covariance structures method (MACS) and item response theory (IRT). They proposed examining the loadings (discrimination) and the intercept (location) parameters simultaneously using the likelihood ratio test with a free-baseline model and Bonferroni corrected critical p values. They compared the relative efficacy of this approach with alternative implementations for various types and amounts of DIF, sample sizes, numbers of response categories, and amounts of impact (latent mean differences). Results indicated that the proposed strategy was considerably more effective than an alternative approach involving a constrained-baseline model. Both MACS and IRT performed similarly well in the majority of experimental conditions. As expected, MACS performed slightly worse in dichotomous conditions but better than IRT in polytomous cases where sample sizes were small. Also, contrary to popular belief, MACS performed well in conditions where DIF was simulated on item thresholds (item means), and its accuracy was not affected by impact. (PsycINFO Database Record (c) 2010 APA, all rights reserved)     

Red cell ferritin, a marker of iron deficiency in hemodialysis patients

Methods for evaluating the hazards associated with noncancer responses with epidemiologic data are considered. The methods for noncancer risk assessment have largely been developed for experimental data, and are not always suitable for the more complex structure of epidemiologic data. In epidemiology, the measurement of the response and the exposure is often either continuous or dichotomous. For a continuous noncancer response modeled with multiple regression, a variety of endpoints may be examined: (1) the concentration associated with absolute or relative decrements in response; (2) a threshold concentration associated with no change in response; and (3) the concentration associated with a particular added risk of impairment. For a dichotomous noncancer response modeled with logistic regression, concentrations associated with specified added/extra risk or with a threshold responses may be estimated. No-observed-effect concentrations may also be estimated for categorizations of exposures for both continuous and dichotomous responses but these may depend on the arbitrary categories chosen. Respiratory function in miners exposed to coal dust is used to illustrate these methods.     

A method for exploring the effects of attrition in randomized experiments with dichotomous outcomes.

Attrition from conditions in randomized experiments is common. Yet it is difficult to assess the possible effects of attrition because the outcome status of the dropouts is usually unknown. This article develops methods to assess those effects in studies with dichotomous outcomes, illustrating the methods with randomized experiments in drug abuse treatment, smoking cessation treatment, and alcoholism treatment. The methods include computing the lowest and highest possible effect sizes that could have been observed, enumerating the percent of possible study outcomes below a given threshold, estimating the probability that an outcome beyond any given threshold would be observed if all participants were measured, and constructing attrition analysis plots showing the effects of attrition under varied assumptions. For the kind of study to which they apply, these methods should replace the treatment of missing participants as failures in an "intent-to-treat" analysis. A user-friendly personal computer program is available to implement all of these analyses. (PsycINFO Database Record (c) 2010 APA, all rights reserved)     

Gender, socioeconomic status, age, and jealousy: Emotional responses to infidelity in a national sample.

The authors used a representative national sample (N = 777) to test the evolutionary hypothesis that men would be more bothered by sexual infidelity and women by emotional infidelity, the Jealousy as a Specific Innate Module (JSIM) effect. Our alternative conceptualization of jealousy suggests that there are distinct emotional components of jealousy that did not evolve differently by gender. The authors looked for effects of age, socioeconomic status (SES), and type of measure (continuous or dichotomous) on jealousy. The authors did not find age or SES effects. Forced-choice items provided support for our alternative view; both genders showed more anger and blame over sexual infidelity but more hurt feelings over emotional infidelity. Continuous measures indicated more emotional response to sexual than emotional infidelity among both genders. (PsycINFO Database Record (c) 2010 APA, all rights reserved)     

Generalized full-information item bifactor analysis.

Full-information item bifactor analysis is an important statistical method in psychological and educational measurement. Current methods are limited to single-group analysis and inflexible in the types of item response models supported. We propose a flexible multiple-group item bifactor analysis framework that supports a variety of multidimensional item response theory models for an arbitrary mixing of dichotomous, ordinal, and nominal items. The extended item bifactor model also enables the estimation of latent variable means and variances when data from more than 1 group are present. Generalized user-defined parameter restrictions are permitted within or across groups. We derive an efficient full-information maximum marginal likelihood estimator. Our estimation method achieves substantial computational savings by extending Gibbons and Hedeker's (1992) bifactor dimension reduction method so that the optimization of the marginal log-likelihood requires only 2-dimensional integration regardless of the dimensionality of the latent variables. We use simulation studies to demonstrate the flexibility and accuracy of the proposed methods. We apply the model to study cross-country differences, including differential item functioning, using data from a large international education survey on mathematics literacy. (PsycINFO Database Record (c) 2011 APA, all rights reserved)