Standard errors of fit indices using residuals in structural equation modeling

The asymptotic standard errors of the correlation residuals and Bentler's standardized residuals in covariance structures are derived based on the asymptotic covariance matrix of raw covariance residuals. Using these results, approximations of the asymptotic standard errors of the root mean square residuals for unstandardized or standardized residuals are derived by the delta method. Further, in mean structures, approximations of the asymptotic standard errors of residuals, standardized residuals and their summary statistics are derived in a similar manner. Simulations are carried out, which show that the asymptotic standard errors of the various types of residuals and the root mean square residuals in covariance, correlation and mean structures are close to actual ones.The author is indebted to the reviewers for their comments and suggestions which have led to an improvement of this work.     

Skew and internal consistency

The effects of skew on the standardized item alpha were examined with Monte Carlo techniques. Alphas computed from normal variables were compared with alphas from lognormal variables, ranks, and skewed versus normal Likert-type variables. The extent and direction of skew were varied, as was the size of the population interitem correlation (rho), the number of items, and the number of categories for Likert-type variables. Because the average interitem correlation affects alpha and skew affects the average interitem correlation, the effect of skew on the average interitem correlation also was examined. Results indicated that skew decreased the average interitem correlation and produced small decreases in alpha that were largest when skew was large, rho was small, items were skewed in opposite directions, and there were fewer items.     

On the distribution of the maximum likelihood estimator of Cronbach's alpha

The asymptotic normal distribution of the maximum likelihood estimator of Cronbach's alpha (under normality) is derived for the case when no assumptions are made about the covariances among items. The asymptotic distribution is also considered for the special case of compound symmetry and compared to the exact distribution.The authors would like to thank Willem J. Heiser, an associate editor and the reviewers for valuable and helpful comments to improve the quality of this work.     

A taxonomy of item response models

A number of models for categorical item response data have been proposed in recent years. The models appear to be quite different. However, they may usefully be organized as members of only three distinct classes, within which the models are distinguished only by assumptions and constraints on their parameters. “Difference models” are appropriate for ordered responses, “divide-by-total” models may be used for either ordered or nominal responses, and “left-side added” models are used for multiple-choice responses with guessing. The details of the taxonomy and the models are described in this paper. The present study was supported in part by two postdoctoral fellowships awarded to Lynne Steinberg: an Educational Testing Service Postdoctoral Fellowship at ETS, Princeton, NJ and an NIMH Individual National Research Service Award at Stanford University, Stanford, CA. Helpful comments by the editor and three anonymous reviewers are gratefully acknowledged.     

Response to Ted Peters’ “Models of God”

In “Models of God,” Ted Peters discusses a methodology for formulating and evaluating models of God, surveys nine models, and proposes one that he entitles “Eschatological Panentheism.” This paper provides critical comments on Peters’ methodological claims, taxonomy of models of God, and specific proposal. This paper has been delivered during APA Pacific 2007 Mini-Conference on Models of God.Both Peters’ “Models of God” and these comments were presented at the “Models of God” mini-conference at the Pacific Division Meetings of the American Philosophical Association in April of 2007.     

Speaking truth to power: Empowerment ideology as social intervention and policy

The popularity, and subsequent ambiguity, in the use of the term “empowerment” has created an even greater need for reassessment in the applied context than in the theory and research literatures. This paper outlines some of the areas of community, organizational, and societal level social intervention and policy ostensibly based on the concept of empowerment. These include neighborhood voluntary associations (for environmental protection, community crime prevention, etc.), self-help groups, competence-building primary prevention, organizational management, health care and educational reforms, and national and international community service and community development policies. Issues in applying social research to community organizations and to legislative and administrative policy making are reviewed. Ten recommendations are offered, including the value of a dialectical analysis, for helping researchers and policy makers/administrators make more effective use of empowerment theory and research. Humpty Dumpty: “When I use a word, it means just what I choose it to mean neither more nor less.” King of Hearts: “If there's no meaning in it, that saves a world of trouble, you know, as we needn't try to find any.”—Lewis Carroll Portions of this paper were first presented in the program “Empowerment Theory, Research and policy” at the Biennial Conference on Community Research and Action, Williamsburg, Virginia, June 18, 1993. The author thanks Barbara B. Brown, Jo Ann Lippe, Ken Maton and his students, David V. Perkins, Marc A. Zimmerman, and the anonymous reviewers for their very helpful comments on earlier drafts.     

Effects of Parental Soothing Behavior on Stress Levels of 2–8 Year Old Children During Voiding Cystourethrograms by Phase of Procedure

Using the example of a voiding cystourethrogram (VCUG), a painful radiological procedure, this study investigated whether parental soothing behavior (reassuring comments like “it’s almost over” or “You’re O.K.” and soothing by “sh, sh”) in one phase of the procedure influenced the child’s distress in the following phase. The sample was comprised of 68 2–8 year-old children and the accompanying parent(s). Child and parental behavior during the VCUG was coded using a standardized rating scale (CAMPIS-R). Parental reassurance during the anticipatory phase significantly increased the child’s distress of the following phase, while parental “sh, sh” significantly reduced it. Both parental behaviors showed no significant effect on the child’s distress of the following phase when applied during the procedure itself. Results underline the importance of differentiating between anticipatory and procedural phases of the VCUG. Counselling methods for parents on more appropriate strategies to assist their children during procedural phases of the VCUG are necessary.     

Psychometric analysis of the Stunkard-Messick Eating Questionnaire (SMEQ) and Comparison with the dutch Eating Behavior Questionnaire (DEBQ)

The factor structure of the Stunkard-Messick Eating Questionnaire (SMEQ) differs from that of the Dutch Eating Behavior Questionnaire (DEBQ). Using students and members of a weight watchers club, we factor-analyzed responses to the SMEQ and produced a solution which was different from that obtained by Stunkard and Messick, but which had some similarities to the factor structure of the DEBQ. “External” taste items and “internal” hunger items loaded on the same factor, a finding inconsistent with externality theory but consistent with the common sense notion of appetite strength. Although our finding suggests factor instability of the SMEQ, the stability of the DEBQ appears to be the consequence of a restricted range of DEBQ items compared with SMEQ items. Financial support for this research was given by Lilly Industries.     

A note on using alpha and stratified alpha to estimate the reliability of a test composed of item parcels

Several authors have suggested that prior to conducting a confirmatory factor analysis it may be useful to group items into a smaller number of item ‘parcels’ or ‘testlets’. The present paper mathematically shows that coefficient alpha based on these parcel scores will only exceed alpha based on the entire set of items if W, the ratio of the average covariance of items between parcels to the average covariance of items within parcels, is greater than unity. If W is less than unity, however, and errors of measurement are uncorrelated, then stratified alpha will be a better lower bound to the reliability of a measure than the other two coefficients. Stratified alpha are also equal to the true reliability of a test when items within parcels are essentially tau‐equivalent if one assumes that errors of measurement are not correlated.     

Psychometric analysis of the Stunkard-Messick Eating Questionnaire (SMEQ) and Comparison with the dutch Eating Behavior Questionnaire (DEBQ)

The factor structure of the Stunkard-Messick Eating Questionnaire (SMEQ) differs from that of the Dutch Eating Behavior Questionnaire (DEBQ). Using students and members of a weight watchers club, we factor-analyzed responses to the SMEQ and produced a solution which was different from that obtained by Stunkard and Messick, but which had some similarities to the factor structure of the DEBQ. “External” taste items and “internal” hunger items loaded on the same factor, a finding inconsistent with externality theory but consistent with the common sense notion of appetite strength. Although our finding suggests factor instability of the SMEQ, the stability of the DEBQ appears to be the consequence of a restricted range of DEBQ items compared with SMEQ items. Financial support for this research was given by Lilly Industries.     

Interference resolution in major depression

In two experiments, we investigated individual differences in the ability to resolve interference in participants diagnosed with major depressive disorder (MDD). Participants were administered the “Ignore/Suppress” task, a short-term memory task composed of two steps. In Step 1 (“ignore”), participants were instructed to memorize a set of stimuli while ignoring simultaneously presented irrelevant material. In Step 2 (“suppress”), participants were instructed to forget a subset of the previously memorized material. The ability to resolve interference was indexed by response latencies on two recognition tasks in which participants decided whether a probe was a member of the target set. In Step 1, we compared response latencies to probes from the to-be-ignored list with response latencies to nonrecently presented items. In Step 2, we compared response latencies to probes from the to-be-suppressed list with response latencies to nonrecently presented items. The results indicate that, compared with control participants, depressed participants exhibited increased interference in the “suppress” but not in the “ignore” step of the task, when the stimuli were negative words. No group differences were obtained when we presented letters instead of emotional words. These findings indicate that depression is associated with difficulty in removing irrelevant negative material from short-term memory.     

Differential incidence of procrastination between blue and white-collar workers

The present study compared chronic procrastination tendencies claimed by “whitecollar” working adults (n = 141) with rates reported by Harriott and Ferrari ( 1996) of “blue-collar” working adults (n - 211). Participants completed reliable and valid measures of everyday procrastination tendencies (decisional, arousal, and avoidant), and socio-demographic items at one of several public forums. Results showed no significant difference between classes of working adults on age, gender, marital status, or the number of children, although “white-collar” workers claimed higher levels of education than “blue-collar” workers. “White-collar” workers also reported significantly higher scores on all three forms of chronic procrastination than “blue-collar” workers. It seems that professional employees report procrastination more frequently than unskilled workers. Further research is needed to explore the causes and consequences associated with differences in chronic procrastination by occupational group.     

The role of crosstalk in dual-task performance: evidence from manipulating response-code overlap

The present study examined the role of crosstalk in dual-task interference using a combination of a nonspeeded visual task and an auditory-manual reaction time (RT) task. The potential for dual-task crosstalk was introduced by presenting in the visual task objects (e.g., a cup with a handle), which “afford” associated responses that were either spatially compatible or incompatible with the response in the RT task. Crucially, the degree of crosstalk was varied by instructing participants either to attend to the left–right orientation of the objects, creating explicit cross-task response-code overlap (“strong crosstalk”), or to attend to object identity (no direct overlap; “weak crosstalk”). The data indicated a relative benefit for cross-task compatible trials, which was much greater with strong crosstalk than with weak crosstalk. Crucially, however, even on compatible trials dual-task performance was substantially worse with strong crosstalk than with weak crosstalk. This overall cost of crosstalk suggests interference of response codes even on compatible dual-task trials. This work was presented in August 2007 at the XVth Conference of the European Society for Cognitive Psychology (ESCoP) in Marseille, France. The author would like to thank Peter A. Frensch and two anonymous reviewers for helpful comments, Julia Klotz and Marion Marksteiner for testing the participants, and Lynn Huestegge for helpful discussions.     

Science fictions: Comment on Godfrey-Smith

This is a comment on Peter Godfrey-Smith’s, “Models and Fictions in Science”. The comments explore problems he raises if we treat model systems as fictions in a naturalized and deflationary framework.     

Early Childhood Externalising Behaviour Problems: Child, Parenting, and Family-related Predictors Over Time

This study examined the dynamic relationships between child, parenting, and family-related predictor variables and early childhood externalising behaviour problems. A community sample of 395 Australian children was followed longitudinally, and assessed at 4 and 6 years with the Child Behavior Checklist, Teacher Report Form, and standard measures of parenting, temperament, and familial adversity. Variables based on the average scores across the two assessments and the change in scores between assessments were utilised as predictors of parent-reported and teacher-reported externalising behaviour problems at age 6. It was hypothesised that both higher average scores and more detrimental changes in scores, would independently predict externalising problems at age 6. Multivariable analyses found that the presence of parent-reported child externalising problems in six-year-olds were predicted by: (i) the presence of parent-reported child externalising problems at age 4, (ii) higher average “teacher-reported child externalising behaviour,” “inflexible temperament,” “non-persistent temperament,” and “over-reactive parenting,” and (iii) an increased “inflexible temperament” score between age 4 and age 6. The presence of teacher-reported child externalising problems at age 6 was predicted by higher average “parent-reported child externalising behaviour,” and “over-reactive parenting.” The results provide further evidence of the adverse impact of continuing high levels of temperament difficulties and over-reactive parenting on externalising behaviour in early childhood. However, contrary to expectations, the contribution of including the dynamic change scores was limited.     

Differential incidence of procrastination between blue and white-collar workers

The present study compared chronic procrastination tendencies claimed by “whitecollar” working adults (n = 141) with rates reported by Harriott and Ferrari ( 1996) of “blue-collar” working adults (n - 211). Participants completed reliable and valid measures of everyday procrastination tendencies (decisional, arousal, and avoidant), and socio-demographic items at one of several public forums. Results showed no significant difference between classes of working adults on age, gender, marital status, or the number of children, although “white-collar” workers claimed higher levels of education than “blue-collar” workers. “White-collar” workers also reported significantly higher scores on all three forms of chronic procrastination than “blue-collar” workers. It seems that professional employees report procrastination more frequently than unskilled workers. Further research is needed to explore the causes and consequences associated with differences in chronic procrastination by occupational group.     

Internal consistency of tests: Analyses old and new

A coefficient derived from communalities of test parts has been proposed as greatest lower bound to Guttman's immediate retest reliability. The communalities have at times been calculated from covariances between itemsets, which tends to underestimate appreciably. When items are experimentally independent, a consistent estimate of the greatest defensible internal-consistency coefficient is obtained by factoring item covariances. In samples of modest size, this analysis capitalizes on chance; an estimate subject to less upward bias is suggested. For estimating alternate-forms reliability, communality-based coefficients are less appropriate than stratified alpha.I thank Edward Haertel for comments and suggestions, and Andrew Comrey for data.     

Sleeping Beauty and Self-location: A Hybrid Model

The Sleeping Beauty problem is test stone for theories about self- locating belief, i.e. theories about how we should reason when data or theories contain indexical information. Opinion on this problem is split between two camps, those who defend the “1/2 view” and those who advocate the “1/3 view”. I argue that both these positions are mistaken. Instead, I propose a new “hybrid” model, which avoids the faults of the standard views while retaining their attractive properties. This model appears to violate Bayesian conditionalization, but I argue that this is not the case. By paying close attention to the details of conditionalization in contexts where indexical information is relevant, we discover that the hybrid model is in fact consistent with Bayesian kinematics. If the proposed model is correct, there are important lessons for the study of self-location, observation selection theory, and anthropic reasoning.