A meta-analytic approach to growth curve analysis

A meta-analytic approach to growth curve analysis is described and illustrated by applying it to the evaluation of the Arizona Pilot Project, an experimental project for financing the treatment of the severely mentally ill. In this approach to longitudinal data analysis, each individual subject for which repeated measures are obtained is initially treated as a separate case study for analysis. This approach has at least two distinct advantages. First, it does not assume a balanced design (equal numbers of repeated observations) across all subjects; to accommodate a variable number of observations for each subject, individual growth curve parameters are differentially weighted by the number of repeated measures on which they are based. Second, it does not assume homogeneity of treatment effects (equal slopes) across all subjects. Individual differences in growth curve parameters representing potentially unequal developmental rates through time are explicitly modeled. A meta-analytic approach to growth curve analysis may be the optimal analytical strategy for longitudinal studies where either (1) a balanced design is not feasible or (2) an assumption of homogeneity of treatment effects across all individuals is theoretically indefensible. In our evaluation of the Arizona Pilot Project, individual growth curve parameters were obtained for each of the 13 rationally derived subscales of the New York Functional Assessment Survey, over time, by linear regression analysis. The slopes, intercepts, and residuals obtained for each individual were then subjected to meta-analytic causal modeling. Using factor analytic models and then general linear models for the latent constructs, the growth curve parameters of all individuals were systematically related to each other via common factors and predicted based on hypothesized exogenous causal factors. The same two highly correlated common factors were found for all three growth curve parameters analyzed, a general psychological factor and a general functional factor. The factor patterns were found to be nearly identical across the separate analyses of individual intercepts, slopes, and residuals. Direct effects on the unique factors of each subscale of the New York Functional Assessment Survey were tested for each growth curve parameter by including the common factors as hierarchically prior predictors in the structural model for each of the indicator variables, thus statistically controlling for any indirect effect produced on the indicator through the common factors. The exogenous predictors modeled were theoretically specified orthogonal contrasts for Method of Payment (comparing Arizona Pilot Project treatment or "capitation" to traditional or "fee-for-service" care as a control), Treatment Administration Site (comparing various locations within treatment or control groups), Pretreatment Assessment (comparing general functional level at intake as assigned by an Outside Assessment Team), and various interactions among these main effects. The intercepts, representing the initial status of individual subjects on both the two common factors and the 13 unique factors of the subscales of the New York Functional Assessment Survey, were found to vary significantly across many of the various different treatment conditions, treatment administration sites, and pretreatment functional levels. This indicated a severe threat to the validity of the originally intended design of the Arizona Pilot Project as a randomized experiment. When the systematic variations were statistically controlled by including intercepts as hierarchically prior predictors in the structural models for slopes, recasting the experiment as a nonequivalent groups design, the effects of the intercepts on the slopes were found to be both statistically significant and substantial in magnitude. (ABSTRACT TRUNCATED)     

Reconsidering the Use of Autoregressive Latent Trajectory (ALT) Models

The simultaneous estimation of autoregressive (simplex) structures and latent trajectories, so called ALT (autoregressive latent trajectory) models, is becoming an increasingly popular approach to the analysis of change. Although historically autoregressive (AR) and latent growth curve (LGC) models have been developed quite independently from each other, the underlying pattern of change is often highly similar. In this article it is shown that their integration rests on the strong assumption that neither the AR part nor the LGC part contains any misspecification. In practice, however, this assumption is often violated due to nonlinearity in the LGC part. As a consequence, the autoregressive (simplex) process incorrectly accounts for part of this nonlinearity, thus rendering any substantive interpretation of parameter estimates virtually impossible. Accordingly, researchers are advised to exercise extreme caution when using ALT models in practice. All arguments are illustrated by empirical data on skill acquisition, and a simulation study is provided to investigate the conditions and consequences of mistaking nonlinear growth curve patterns as autoregressive processes.     

A new paradigm for hypothesis testing in medicine,with examination of the Neyman Pearson condition

In the past, hypothesis testing in medicine has employed the paradigm of the repeatable experiment. In statistical hypothesis testing, an unbiased sample is drawn from a larger source population, and a calculated statistic is compared to a preassigned critical region, on the assumption that the comparison could be repeated an indefinite number of times. However, repeated experiments often cannot be performed on human beings, due to ethical or economic constraints. We describe a new paradigm for hypothesis testing which uses only rearrangements of data present within the observed data set. The token swap test, based on this new paradigm, is applied to three data sets from cardiovascular pathology, and computational experiments suggest that the token swap test satisfies the Neyman Pearson condition.     

如何描述发展趋势的差异：潜变量混合增长模型

在追踪研究中,研究者不仅关心某一特质随时间的发展趋势,而且关注个体之间发展趋势的差异及其存在差异的原因。在总体发展同质的情形下,多层线性模型和潜变量增长曲线模型为解决这一问题提供了切实有效的方法。但是如果所研究的总体本身不同质,就需要一种能够描述总体中不同质子总体的不同发展特点的方法。该文简要介绍了一种能够描述不同群体不同发展趋势特征的统计模型——潜变量混合增长模型,并通过一个实际例子介绍了这一方法的应用过程,同时说明了潜变量混合增长模型与多层线性模型和潜变量增长曲线模型之间的关系     

基于增长模型的非随机缺失数据处理:选择模型和极大似然方法

对含有非随机缺失数据的潜变量增长模型,为了考察基于不同假设的缺失数据处理方法:极大似然(ML)方法与DiggleKenward选择模型的优劣,通过Monte Carlo模拟研究,比较两种方法对模型中增长参数估计精度及其标准误估计的差异,并考虑样本量、非随机缺失比例和随机缺失比例的影响。结果表明,符合前提假设的Diggle-Kenward选择模型的参数估计精度普遍高于ML方法;对于标准误估计值,ML方法存在一定程度的低估,得到的置信区间覆盖比率也明显低于Diggle-Kenward选择模型。     

Latent Growth Models with Floors,Ceilings, and Random Knots

In longitudinal/developmental studies, individual growth trajectories are sometimes bounded by a floor at the beginning of the observation period and/or a ceiling toward the end of the observation period (or vice versa), resulting in inherently nonlinear growth patterns. If the trajectories between the floor and ceiling are approximately linear, such longitudinal growth patterns can be described with a linear piecewise (spline) model in which segments join at knots. In these scenarios, it may be of specific interest for researchers to examine the timing when transition occurs, and in some occasions also to examine the levels of the floors and/or ceilings if they are not known and fixed. In the current study, we propose a reparameterized piecewise latent growth curve model so that a direct estimation of the random knots (and, if needed, a direct estimation of random floors and ceilings) is possible. We derive the model reparameterization using a 4-step structured latent curve modeling approach. We provide two illustrative examples to demonstrate how the proposed reparameterized models can be fitted to longitudinal growth data using the popular SEM software Mplus and we supply the full coding for applied researchers’ reference.     

Optimization-Based Model Fitting for Latent Class and Latent Profile Analyses

Statisticians typically estimate the parameters of latent class and latent profile models using the Expectation-Maximization algorithm. This paper proposes an alternative two-stage approach to model fitting. The first stage uses the modified k-means and hierarchical clustering algorithms to identify the latent classes that best satisfy the conditional independence assumption underlying the latent variable model. The second stage then uses mixture modeling treating the class membership as known. The proposed approach is theoretically justifiable, directly checks the conditional independence assumption, and converges much faster than the full likelihood approach when analyzing high-dimensional data. This paper also develops a new classification rule based on latent variable models. The proposed classification procedure reduces the dimensionality of measured data and explicitly recognizes the heterogeneous nature of the complex disease, which makes it perfect for analyzing high-throughput genomic data. Simulation studies and real data analysis demonstrate the advantages of the proposed method.     

LGM模型中缺失数据处理方法的比较:ML方法与Diggle-Kenward选择模型

追踪研究中缺失数据十分常见。本文通过Monte Carlo模拟研究,考察基于不同前提假设的Diggle-Kenward选择模型和ML方法对增长参数估计精度的差异,并考虑样本量、缺失比例、目标变量分布形态以及不同缺失机制的影响。结果表明:(1)缺失机制对基于MAR的ML方法有较大的影响,在MNAR缺失机制下,基于MAR的ML方法对LGM模型中截距均值和斜率均值的估计不具有稳健性。(2)DiggleKenward选择模型更容易受到目标变量分布偏态程度的影响,样本量与偏态程度存在交互作用,样本量较大时,偏态程度的影响会减弱。而ML方法仅在MNAR机制下轻微受到偏态程度的影响。     

Comparative longitudinal structural analyses of the growth and decline of multiple intellectual abilities over the life span.

Latent growth curve techniques and longitudinal data are used to examine predictions from the theory of fluid and crystallized intelligence (Gf-Gc theory; J. L. Horn & R. B. Cattell, 1966, 1967). The data examined are from a sample (N approximately 1,200) measured on the Woodcock-Johnson Psycho-Educational Battery-Revised (WJ-R). The longitudinal structural equation models used are based on latent growth models of age using two-occasion "accelerated" data (e.g., J. J. McArdle & R. Q. Bell, 2000; J. J. McArdle & R. W. Woodcock, 1997). Nonlinear mixed-effects growth models based on a dual exponential rate yield a reasonable fit to all life span cognitive data. These results suggest that most broad cognitive functions fit a generalized curve that rises and falls. Novel multilevel models directly comparing growth curves show that broad fluid reasoning (Gf) and acculturated crystallized knowledge (Gc) have different growth patterns. In all comparisons, any model of cognitive age changes with only a single g factor yields an overly simplistic view of growth and change over age.     

Investigating Ceiling Effects in Longitudinal Data Analysis

Score limitation at the top of a scale is commonly termed “ceiling effect.” Ceiling effects can lead to serious artifactual parameter estimates in most data analysis. This study examines the consequences of ceiling effects in longitudinal data analysis and investigates several methods of dealing with ceiling effects through Monte Carlo simulations and empirical data analyses. Data were simulated based on a latent growth curve model with T = 5 occasions. The proportion of the ceiling data [10%–40%] was manipulated by using different thresholds, and estimated parameters were examined for R = 500 replications. The results showed that ceiling effects led to incorrect model selection and biased parameter estimation (shape of the curve and magnitude of the changes) when regular growth curve models were applied. The Tobit growth curve model, instead, performed very well in dealing with ceiling effects in longitudinal data analysis. The Tobit growth curve model was then applied in an empirical cognitive aging study and the results were discussed.     

On growth curves and mixture models

The multilevel model of change and the latent growth model are flexible means to describe all sorts of population heterogeneity with respect to growth and development, including the presence of sub‐populations. The growth mixture model is a natural extension of these models. It comes at hand when information about sub‐populations is missing and researchers nevertheless want to retrieve developmental trajectories from sub‐populations. We argue that researchers have to make rather strong assumptions about the sub‐populations or latent trajectory classes in order to retrieve existing population differences. A simulated example is discussed, showing that a sample of repeated measures drawn from two sub‐populations easily leads to the mistaken inference of three sub‐populations, when assumptions are not met. The merits of methodological advises on this issue are discussed. It is concluded that growth mixture models should be used with understanding, and offer no free way to growth patterns in unknown sub‐populations. Copyright © 2006 John Wiley & Sons, Ltd.     

Latent Growth Curve and Repeated Measures ANOVA Contrasts: What the Models are Telling You

The relationship between the latent growth curve and repeated measures ANOVA models is often misunderstood. Although a number of investigators have looked into the similarities and differences among these models, a cursory reading of the literature can give the impression that they are very different models. Here we show that each model represents a set of contrasts on the occasion means. We demonstrate that the fixed effects parameters of the estimated basis vector latent growth curve model are merely a transformation of the repeated measures ANOVA fixed effects parameters. We further show that differences in fit in models that estimate the same means structure can be due to the different error covariance structures implied by the model. We show these relationships both algebraically and through using data from a simulation.     

Diagnostics of Robust Growth Curve Modeling Using Student's t Distribution

Growth curve models with different types of distributions of random effects and of intraindividual measurement errors for robust analysis are compared. After demonstrating the influence of distribution specification on parameter estimation, 3 methods for diagnosing the distributions for both random effects and intraindividual measurement errors are proposed and evaluated. The methods include (a) distribution checking based on individual growth curve analysis; (b) distribution comparison based on Deviance Information Criterion, and (c) post hoc checking of degrees of freedom estimates for t distributions. The performance of the methods is compared through simulation studies. When the sample size is reasonably large, the method of post hoc checking of degrees of freedom estimates works best. A web interface is developed to ease the use of the 3 methods. Application of the 3 methods is illustrated through growth curve analysis of mathematical ability development using data on the Peabody Individual Achievement Test Mathematics assessment from the National Longitudinal Survey of Youth 1997 Cohort (Bureau of Labor Statistics, U.S. Department of Labor, 2005).     

The normative development of child and adolescent problem behavior

The aim of this study was to identify normative developmental trajectories of parent-reported problems assessed with the Child Behavior Checklist (CBCL; T. M. Achenbach, 1991) in a representative sample of 2,076 children aged 4 to 18 years from the general population. The trajectories were determined by multilevel growth curve analyses on the CBCL syndromes in a Longitudinal multiple birth-cohort sample that was assessed 5 times with 2-year intervals. Most syndromes showed a linear increase or decrease with age or a curvilinear trajectory, except for thought problems. Trajectories for most syndromes differed for boys versus girls, except those for withdrawn, social problems, and thought problems. These normative developmental trajectories provide information against which developmental deviance in childhood and adolescence can be detected.     

First-Order Learning Models With the GDINA: Estimation With the EM Algorithm and Applications

In learning environments, understanding the longitudinal path of learning is one of the main goals. Cognitive diagnostic models (CDMs) for measurement combined with a transition model for mastery may be beneficial for providing fine-grained information about students’ knowledge profiles over time. An efficient algorithm to estimate model parameters would augment the practicality of this combination. In this study, the Expectation–Maximization (EM) algorithm is presented for the estimation of student learning trajectories with the GDINA (generalized deterministic inputs, noisy, “and” gate) and some of its submodels for the measurement component, and a first-order Markov model for learning transitions is implemented. A simulation study is conducted to investigate the efficiency of the algorithm in estimation accuracy of student and model parameters under several factors—sample size, number of attributes, number of time points in a test, and complexity of the measurement model. Attribute- and vector-level agreement rates as well as the root mean square error rates of the model parameters are investigated. In addition, the computer run times for converging are recorded. The result shows that for a majority of the conditions, the accuracy rates of the parameters are quite promising in conjunction with relatively short computation times. Only for the conditions with relatively low sample sizes and high numbers of attributes, the computation time increases with a reduction parameter recovery rate. An application using spatial reasoning data is given. Based on the Bayesian information criterion (BIC), the model fit analysis shows that the DINA (deterministic inputs, noisy, “and” gate) model is preferable to the GDINA with these data.     

Generalized graded unfolding model with structural equation for subject parameters

The generalized graded unfolding model (GGUM) is capable of analyzing polytomous scored, unfolding data such as agree‐disagree responses to attitude statements. In the present study, we proposed a GGUM with structural equation for subject parameters, which enabled us to evaluate the relation between subject parameters and covariates and/or latent variables simultaneously, in order to avoid the influence of attenuation. Additionally, an algorithm for parameter estimation is newly implemented via the Markov Chain Monte Carlo (MCMC) method, based on Bayesian statistics. In the simulation, we compared the accuracy of estimates of regression coefficients between the proposed model and a conventional method using a GGUM (where regression coefficients are estimated using estimates of θ). As a result, the proposed model performed much better than the conventional method in terms of bias and root mean squared errors of estimates of regression coefficients. The study concluded by verifying the efficacy of the proposed model, using an actual data example of attitude measurement.     

Checking the Assumptions of Rasch's Model for Speed Tests

Two new tests for a model for the response times on pure speed tests by Rasch (1960) are proposed. The model is based on the assumption that the test response times are approximately gamma distributed, with known index parameters and unknown rate parameters. The rate parameters are decomposed in a subject ability parameter and a test difficulty parameter. By treating the ability as a gamma distributed random variable, maximum marginal likelihood (MML) estimators for the test difficulty parameters and the parameters of the ability distribution are easily derived. Also the model tests proposed here pertain to the framework of MML. Two tests or modification indices are proposed. The first one is focused on the assumption of local stochastic independence, the second one on the assumption of the test characteristic functions. The tests are based on Lagrange multiplier statistics, and can therefore be computed using the parameter estimates under the null model. Therefore, model violations for all items and pairs of items can be assessed as a by-product of one single estimation run. Power studies and applications to real data are included as numerical examples.     

Random effects and extended generalized partial credit models

In this paper it is shown that under the random effects generalized partial credit model for the measurement of a single latent variable by a set of polytomously scored items, the joint marginal probability distribution of the item scores has a closed-form expression in terms of item category location parameters, parameters that characterize the distribution of the latent variable in the subpopulation of examinees with a zero score on all items, and item-scaling parameters. Due to this closed-form expression, all parameters of the random effects generalized partial credit model can be estimated using marginal maximum likelihood estimation without assuming a particular distribution of the latent variable in the population of examinees and without using numerical integration. Also due to this closed-form expression, new special cases of the random effects generalized partial credit model can be identified. In addition to these new special cases, a slightly more general model than the random effects generalized partial credit model is presented. This slightly more general model is called the extended generalized partial credit model. Attention is paid to maximum likelihood estimation of the parameters of the extended generalized partial credit model and to assessing the goodness of fit of the model using generalized likelihood ratio tests. Attention is also paid to person parameter estimation under the random effects generalized partial credit model. It is shown that expected a posteriori estimates can be obtained for all possible score patterns. A simulation study is carried out to show the usefulness of the proposed models compared to the standard models that assume normality of the latent variable in the population of examinees. In an empirical example, some of the procedures proposed are demonstrated.     

Selecting a linear mixed model for longitudinal data: repeated measures analysis of variance, covariance pattern model, and growth curve approaches

With increasing popularity, growth curve modeling is more and more often considered as the 1st choice for analyzing longitudinal data. Although the growth curve approach is often a good choice, other modeling strategies may more directly answer questions of interest. It is common to see researchers fit growth curve models without considering alterative modeling strategies. In this article we compare 3 approaches for analyzing longitudinal data: repeated measures analysis of variance, covariance pattern models, and growth curve models. As all are members of the general linear mixed model family, they represent somewhat different assumptions about the way individuals change. These assumptions result in different patterns of covariation among the residuals around the fixed effects. In this article, we first indicate the kinds of data that are appropriately modeled by each and use real data examples to demonstrate possible problems associated with the blanket selection of the growth curve model. We then present a simulation that indicates the utility of Akaike information criterion and Bayesian information criterion in the selection of a proper residual covariance structure. The results cast doubt on the popular practice of automatically using growth curve modeling for longitudinal data without comparing the fit of different models. Finally, we provide some practical advice for assessing mean changes in the presence of correlated data.     

Latent growth curve analysis with dichotomous items: Comparing four approaches

A Monte Carlo study was used to compare four approaches to growth curve analysis of subjects assessed repeatedly with the same set of dichotomous items: A two‐step procedure first estimating latent trait measures using MULTILOG and then using a hierarchical linear model to examine the changing trajectories with the estimated abilities as the outcome variable; a structural equation model using modified weighted least squares (WLSMV) estimation; and two approaches in the framework of multilevel item response models, including a hierarchical generalized linear model using Laplace estimation, and Bayesian analysis using Markov chain Monte Carlo (MCMC). These four methods have similar power in detecting the average linear slope across time. MCMC and Laplace estimates perform relatively better on the bias of the average linear slope and corresponding standard error, as well as the item location parameters. For the variance of the random intercept, and the covariance between the random intercept and slope, all estimates are biased in most conditions. For the random slope variance, only Laplace estimates are unbiased when there are eight time points.