Alcohol use trajectories and the ubiquitous cat's cradle: Cause for concern?

[Correction Notice: An erratum for this article was reported in Vol 120(2) of Journal of Abnormal Psychology (see record 2011-05801-001). Due to formatting problems, some of the data and graphic information in the original Figure 7 are not correctly displayed. Although not all data in the original Figure 7 are visible, nothing displayed in that figure was in error.] In recent years, trajectory approaches to characterizing individual differences in the onset and course of substance involvement have gained popularity. Previous studies have sometimes reported 4 prototypic courses: (a) a consistently “low” group, (b) an “increase” group, (c) a “decrease” group, and (d) a consistently “high” group. Although not always recovered, these trajectories are often found, despite these studies varying in the ages of the samples studied and the duration of the observation periods employed. Here, the authors examined the consistency with which these longitudinal patterns of heavy drinking were recovered in a series of latent class growth analyses that systematically varied the age of the sample at baseline, the duration of observation, and the number and frequency of measurement occasions. Data were drawn from a 4-year, 8-wave panel study of college student drinking (N = 3,720). Despite some variability across analyses, there was a strong tendency for these prototypes to emerge regardless of the participants' age at baseline and the duration of observation. These findings highlight potential problems with commonly employed trajectory-based approaches and the need to not over-reify these constructs. (PsycINFO Database Record (c) 2011 APA, all rights reserved)     

Exact and approximate algorithms for part-machine clustering based on a relationship between interval graphs and Robinson matrices

An important manufacturing cell formation problem requires permutations of the rows (parts) and columns (machines) of a part-machine incidence matrix such that the reordered matrix exhibits a block-diagonal form. Numerous objective criteria and algorithms have been proposed for this problem. In this paper, a new perspective is offered that is based on the relationship between the consecutive ones property associated with interval graphs and Robinson structure within symmetric matrices. This perspective enables the cell formation problem to be decomposed into two permutation subproblems (one for rows and one for columns) that can be solved optimally using dynamic programming or a branch-and-bound algorithm for matrices of nontrivial size. A simulated annealing heuristic is offered for larger problem instances. Results pertaining to the application of the proposed methods for a number of problems from the literature are presented.     

Local Optima in K-Means Clustering: What You Don't Know May Hurt You.

The popular K-means clustering method, as implemented in 3 commercial software packages (SPSS, SYSTAT, and SAS), generally provides solutions that are only locally optimal for a given set of data. Because none of these commercial implementations offer a reasonable mechanism to begin the K-means method at alternative starting points, separate routines were written within the MATLAB (Math-Works, 1999) environment that can be initialized randomly (these routines are provided at the end of the online version of this article in the PsycARTICLES database). Through the analysis of 2 empirical data sets and 810 simulated data sets, it is shown that the results provided by commercial packages are most likely locally optimal. These results suggest the need for some strategy to study the local optima problem for a specific data set or to identify methods for finding "good" starting values that might lead to the best solutions possible. (PsycINFO Database Record (c) 2010 APA, all rights reserved)     

Stability and change in patterns of concerns related to eating, weight, and shape in young adult women: A latent transition analysis.

Although college women are known to be at high risk for eating-related problems, relatively little is known about how various aspects of concerns related to eating, weight, and shape are patterned syndromally in this population. Moreover, the extent to which various patterns represent stable conditions or transitory states during this dynamic period of development is unclear. The present study used latent class and latent transition analysis (LCA/LTA) to derive syndromes of concerns related to eating, weight, and shape and movement across these syndromes in a sample of 1,498 women ascertained as first-time freshmen and studied over 4 years. LCA identified 5 classes characterized by (a) no obvious pathological eating-related concerns (prevalence: 28%–34%); (b) a high likelihood of limiting attempts (prevalence: 29%–34%); (c) a high likelihood of overeating and binge eating (prevalence: 14%–18%); (d) a high likelihood of limiting attempts and overeating or binge eating (prevalence: 14%–17%); and (e) pervasive bulimiclike concerns (prevalence: 6%–7%). Membership in each latent class tended to be stable over time. When movement occurred, it tended to be to a less severe class. These findings indicate that there are distinct, prevalent, and relatively stable forms of eating-related concerns in college women. (PsycINFO Database Record (c) 2010 APA, all rights reserved)     

K-means clustering and mixture model clustering: Reply to McLachlan (2011) and Vermunt (2011).

McLachlan (2011) and Vermunt (2011) each provided thoughtful replies to our original article (Steinley & Brusco, 2011). This response serves to incorporate some of their comments while simultaneously clarifying our position. We argue that greater caution against overparamaterization must be taken when assuming that clusters are highly elliptical in nature. Specifically, users of mixture model clustering techniques should be wary of overreliance on fit indices, and the importance of cross-validation is highlighted. Additionally, we note that K-means clustering is part of a larger family of discrete partitioning algorithms, many of which are designed to solve problems identical to those for which mixture modeling approaches are often touted. (PsycINFO Database Record (c) 2011 APA, all rights reserved)     

Correction to Sher et al. (2011).

Reports an error in "Alcohol use trajectories and the ubiquitous cat's cradle: Cause for concern" by Kenneth J. Sher, Kristina M. Jackson and Douglas Steinley (Journal of Abnormal Psychology, np). Due to formatting problems, some of the data and graphic information in the original Figure 7 are not correctly displayed. Although not all data in the original Figure 7 are visible, nothing displayed in that figure was in error. (The following abstract of the original article appeared in record 2011-02783-001.) In recent years, trajectory approaches to characterizing individual differences in the onset and course of substance involvement have gained popularity. Previous studies have sometimes reported 4 prototypic courses: (a) a consistently “low” group, (b) an “increase” group, (c) a “decrease” group, and (d) a consistently “high” group. Although not always recovered, these trajectories are often found, despite these studies varying in the ages of the samples studied and the duration of the observation periods employed. Here, the authors examined the consistency with which these longitudinal patterns of heavy drinking were recovered in a series of latent class growth analyses that systematically varied the age of the sample at baseline, the duration of observation, and the number and frequency of measurement occasions. Data were drawn from a 4-year, 8-wave panel study of college student drinking (N = 3,720). Despite some variability across analyses, there was a strong tendency for these prototypes to emerge regardless of the participants' age at baseline and the duration of observation. These findings highlight potential problems with commonly employed trajectory-based approaches and the need to not over-reify these constructs. (PsycINFO Database Record (c) 2011 APA, all rights reserved)     

Clustering, seriation, and subset extraction of confusion data.

The study of confusion data is a well established practice in psychology. Although many types of analytical approaches for confusion data are available, among the most common methods are the extraction of 1 or more subsets of stimuli, the partitioning of the complete stimulus set into distinct groups, and the ordering of the stimulus set. Although standard commercial software packages can sometimes facilitate these types of analyses, they are not guaranteed to produce optimal solutions. The authors present a MATLAB *.m file for preprocessing confusion matrices, which includes fitting of the similarity-choice model. Two additional MATLAB programs are available for optimally clustering stimuli on the basis of confusion data. The authors also developed programs for optimally ordering stimuli and extracting subsets of stimuli using information from confusion matrices. Together, these programs provide several pragmatic alternatives for the applied researcher when analyzing confusion data. Although the programs are described within the context of confusion data, they are also amenable to other types of proximity data. (PsycINFO Database Record (c) 2010 APA, all rights reserved)     

Profiling local optima in K-means clustering: Developing a diagnostic technique.

Using the cluster generation procedure proposed by D. Steinley and R. Henson (2005), the author investigated the performance of K-means clustering under the following scenarios: (a) different probabilities of cluster overlap; (b) different types of cluster overlap; (c) varying samples sizes, clusters, and dimensions; (d) different multivariate distributions of clusters; and (e) various multidimensional data structures. The results are evaluated in terms of the Hubert-Arabie adjusted Rand index, and several observations concerning the performance of K-means clustering are made. Finally, the article concludes with the proposal of a diagnostic technique indicating when the partitioning given by a K-means cluster analysis can be trusted. By combining the information from several observable characteristics of the data (number of clusters, number of variables, sample size, etc.) with the prevalence of unique local optima in several thousand implementations of the K-means algorithm, the author provides a method capable of guiding key data-analysis decisions. (PsycINFO Database Record (c) 2010 APA, all rights reserved)