We need to change how we compute RMSEA for nested model comparisons in structural equation modeling.

Comparison of nested models is common in applications of structural equation modeling (SEM). When two models are nested, model comparison can be done via a chi-square difference test or by comparing indices of approximate fit. The advantage of fit indices is that they permit some amount of misspecification in the additional constraints imposed on the model, which is a more realistic scenario. The most popular index of approximate fit is the root mean square error of approximation (RMSEA). In this article, we argue that the dominant way of comparing RMSEA values for two nested models, which is simply taking their difference, is problematic and will often mask misfit, particularly in model comparisons with large initial degrees of freedom. We instead advocate computing the RMSEA associated with the chi-square difference test, which we call RMSEAD. We are not the first to propose this index, and we review numerous methodological articles that have suggested it. Nonetheless, these articles appear to have had little impact on actual practice. The modification of current practice that we call for may be particularly needed in the context of measurement invariance assessment. We illustrate the difference between the current approach and our advocated approach on three examples, where two involve multiple-group and longitudinal measurement invariance assessment and the third involves comparisons of models with different numbers of factors. We conclude with a discussion of recommendations and future research directions. (PsycInfo Database Record (c) 2023 APA, all rights reserved).

Correcting bias in extreme groups design using a missing data approach.

Extreme groups design (EGD) refers to the use of a screening variable to inform further data collection, such that only participants with the lowest and highest scores are recruited in subsequent stages of the study. It is an effective way to improve the power of a study under a limited budget, but produces biased standardized estimates. We demonstrate that the bias in EGD results from its inherent missing at random mechanism, which can be corrected using modern missing data techniques such as full information maximum likelihood (FIML). Further, we provide a tutorial on computing correlations in EGD data with FIML using R. (PsycInfo Database Record (c) 2023 APA, all rights reserved).

Estimating classification consistency of screening measures and quantifying the impact of measurement bias.

Screening measures are used in psychology and medicine to identify respondents who are high or low on a construct. Based on the screening, the evaluator assigns respondents to classes corresponding to different courses of action: Make a diagnosis versus reject a diagnosis; provide services versus withhold services; or conduct further assessment versus conclude the assessment process. When measures are used to classify individuals, it is important that the decisions be consistent and equitable across groups. Ideally, if respondents completed the screening measure repeatedly in quick succession, they would be consistently assigned into the same class each time. In addition, the consistency of the classification should be unrelated to the respondents' background characteristics, such as sex, race, or ethnicity (i.e., the measure is free of measurement bias). Reporting estimates of classification consistency is a common practice in educational testing, but there has been limited application of these estimates to screening in psychology and medicine. In this article, we present two procedures based on item response theory that are used (a) to estimate the classification consistency of a screening measure and (b) to evaluate how classification consistency is impacted by measurement bias across respondent groups. We provide R functions to conduct the procedures, illustrate the procedures with real data, and use Monte Carlo simulations to guide their appropriate use. Finally, we discuss how estimates of classification consistency can help assessment specialists make more informed decisions on the use of a screening measure with protected groups (e.g., groups defined by gender, race, or ethnicity). (PsycInfo Database Record (c) 2021 APA, all rights reserved).

Factors associated with health behaviors in middle childhood.

Health behaviors of school-aged children are the precursors of adolescents' health-risk behaviors. The study reported here is the first wave of a cohort-sequential longitudinal study of health-risk behaviors in youth. Using a youth resilience framework, researchers collected data from 1,934 youth in Grades 4-6. Statistically significant differences in health behaviors were found for gender, ethnicity, and grade level, which accounted for only 2% of the variance. Risk and protective factors accounted for an additional 18% of variance in health behaviors. With the federal mandate for schools to establish wellness policies, school nurses are ideally situated to influence school environments to support children's healthy behaviors.

Cumulative concussion exposure in rugby players: neurocognitive and symptomatic outcomes.

A total of 111 rugby players underwent comprehensive testing to determine the impact of self-reported concussion exposure. Reliable estimates of concussion exposure were associated with an increase in postconcussion symptoms (PCS), but not diminished neurocognitive functioning. Importantly, the effects of concussion exposure on PCS varied as a function of player status. More specifically, extent of concussion exposure was associated with increased memory complaints and overall PCS endorsements in a dose-dependent manner for retired and older recreational players, but not for those who were younger and playing at more competitive levels. Future work should systematically evaluate the constituent participant factors that may influence differential concussion outcomes.

Improvement of rural children's asthma self-management by lay health educators.

BACKGROUND: The purpose of the present analysis is to examine changes in rural children's asthma self-management after they received lay health educator (LHE)-delivered classes. METHODS: Elementary schools were randomly assigned to the treatment or attention-control condition and their participating students received either asthma education or general health promotion education, respectively. The triethnic sample was composed of 183 children (46% Hispanic, 29.5% non-Hispanic white, 22% African American, and 2.6% other categories) who had a mean age of 8.78 years (SD = 1.24). The time frame from baseline to postintervention was 12 weeks. RESULTS: Repeated measures analysis of variance found main effects in changes in scores for children's asthma knowledge, asthma self-management, self-efficacy for managing asthma symptoms, and metered dose inhaler (MDI) technique and significant group interaction effects for the treatment intervention on the measures of children's asthma knowledge, asthma self-management, and MDI technique. CONCLUSIONS: The delivery of an asthma health education intervention by trained LHEs to school-aged children was an effective means for improving children's knowledge and skills in asthma self-management.

Expressive writing as a presurgical stress management intervention for breast cancer patients.

This study evaluated whether expressive writing (EW) was an effective stress management intervention for breast cancer patients. Women were recruited at the end of neoadjuvant chemotherapy and assigned to write about their cancer experience (EW group; n = 24) or neutral topics (neutral writing [NW] group; n = 25). Women were asked to write for 20 minutes a day for a total of four writing sessions that were completed over a 7-day period. Participants were reassessed approximately 3 days before and 2 weeks after surgery. The intervention did not significantly decrease women's distress, perceived stress, sleep disturbance, or pain. There was some evidence that the EW group used more sleep medication at the presurgical assessment than the NW group. Social constraints moderated the effect of the intervention. Among women with high social constraints, the EW group reported lower average daily pain than the NW group. Among women with low social constraints, the EW group reported higher average daily pain than the NW group. EW was not broadly effective as a stress management intervention for women with breast cancer. These data do not support the use of EW as a presurgical mind-body complementary medicine program for this population.

Robust step-down tests for multivariate independent group designs.

A composite step-down procedure, in which a set of step-down tests are summarized collectively with Fisher's combination statistic, was considered to test for multivariate mean equality in two-group designs. An approximate degrees of freedom (ADF) composite procedure based on trimmed/Winsorized estimators and a non-pooled estimate of error variance is proposed, and compared to a composite procedure based on trimmed/Winsorized estimators and a pooled estimate of error variance. The step-down procedures were also compared to Hotelling's T (2) and Johansen's ADF global procedure based on trimmed estimators in a simulation study. Type I error rates of the pooled step-down procedure were sensitive to covariance heterogeneity in unbalanced designs; error rates were similar to those of Hotelling's T (2) across all of the investigated conditions. Type I error rates of the ADF composite step-down procedure were insensitive to covariance heterogeneity and less sensitive to the number of dependent variables when sample size was small than error rates of Johansen's test. The ADF composite step-down procedure is recommended for testing hypotheses of mean equality in two-group designs except when the data are sampled from populations with different degrees of multivariate skewness.

Outcomes of a brief sexual health intervention for homeless youth.

Homeless youth face various health challenges. The effectiveness of a short intervention to promote sexual health in 572 homeless 16-23-year-olds (M = 19.467+1.89) was conducted using a quasi-experimental repeated measures design. Data collected at three time points (pre-intervention, immediately post-intervention and follow-up) via laptop computers were analyzed using multivariate general linear mixed models. A significant condition by time interaction was found for self-reported AIDS/STD knowledge; intervention participants had higher scores at first post-test. Females scored significantly higher on cognitive and behavioral outcomes while males reported significantly more sexual risk-taking behaviors. Findings support gender-specific interventions.

Patient assessment of tests to detect cervical cancer.

OBJECTIVES: This study was undertaken to understand how women view characteristics of tests for cervical dysplasia, because these characteristics can affect patient decision-making about screening and follow-up. METHODS: We recruited women who participated in a clinical trial of optical spectroscopy for the diagnosis of cervical dysplasia and used conjoint analysis to assess the women's preferences concerning test attributes. One group of women had a history of an abnormal Papanicolaou smear (diagnostic sample), while the other group did not (screening sample). Participants rated pairs of test scenarios that varied on characteristics such as test sensitivity and painfulness. Based on their responses, the relative importance of test sensitivity, specificity, timing of results feedback and treatment, and pain were calculated, and a cluster analysis was done to identify subgroups of participants with different preference patterns. RESULTS: In the overall sample, sensitivity was the most important attribute, followed by timing, specificity, and pain. Cluster analysis revealed four distinct groups who placed varying importance on each characteristic. The participants in the cluster for which pain was the most important attribute were more likely to be diagnostic patients, non-white, and have low education levels. They also reported more anxiety and pain during the examination than participants in other clusters. CONCLUSIONS: To continue to reduce morbidity and mortality from cervical cancer, developers of new testing procedures should take into account test attributes such as these, which may affect adherence to screening and diagnostic follow-up to further minimize morbidity and mortality from cervical cancer.

Mode effects in the center for epidemiologic studies depression (CES-D) scale: personal digital assistant vs. paper and pencil administration.

As interest grows in creating computerized versions of established paper-and-pencil (P&P) questionnaires, it becomes increasingly important to explore whether changing the administration modes of questionnaires affects participants' responses. This study investigated whether mode effects exist when administering the Center for Epidemiologic Studies Depression (CES-D) scale by a personal digital assistant (PDA) versus the classic P&P mode. The Differential Functioning of Items and Tests (DFIT) procedure identified mode effects on the overall test and individual items. A mixed-effects regression model summarized the mode effects in terms of CES-D scores, and identified interactions with covariates. When the P&P questionnaire was administered first, scores were higher on average (2.4-2.8 points) than those of the other administrations (PDA second, PDA first, and P&P second), and all 20 questionnaire items exhibited a statistically significant mode effect. Highly educated people and younger people demonstrated a smaller difference in scores between the two modes. The mode-by-order effect influenced the interpretation of CES-D scores, especially when screening for depression using the established cut-off scores. These results underscore the importance of evaluating the cross-mode equivalence of psychosocial instruments before administering them in non-established modes.

Attachment and sexual health behaviors in homeless youth.

PURPOSE: To explore relationships and group differences in attachment-related constructs (social connectedness and social support) and sexual health behaviors in a sample of homeless youth. DESIGN AND METHODS: An exploratory design, this study analyzed baseline data from an ongoing intervention study. Survey data were collected by audio, computer-assisted self-interview (A-CASI) from a nonprobability sample of homeless youth (n = 176). RESULTS: Social connectedness was positively correlated with period of time homeless, while social support was positively related to sexual self-concept and to self-efficacy and intention to use condoms. PRACTICE IMPLICATIONS: The street group that homeless youth identify with may be a potential resource for peer-mentoring and skill-building regarding healthy sexual behavior.

A Monte Carlo study of the impact of missing data and differential item functioning on theta estimates from two polytomous Rasch family models.

This paper examines the impact of differential item functioning (DIF), missing item values, and different methods for handling missing item values on theta estimates with data simulated from the partial credit model and Andrich's rating scale model. Both Rasch family models are commonly used when obtaining an estimate of a respondent's attitude. The degree of missing data, DIF magnitude, and the percentage of DIF items were varied in MCAR data conditions in which the focal group was 10% of the total population. Four methods for handling missing data were compared: complete-case analysis, mean substitution, hot-decking, and multiple imputation. Bias, RMSE, means, and standard errors of the theta estimates for the focal group were adversely affected by the amount and magnitude of DIF items. RMSE and fidelity coefficients for both the reference and focal group were adversely impacted by the amount of missing data. While all methods of handling missing data performed fairly similarly, multiple imputation and hot-decking showed slightly better performance.

Reaching and treating Spanish-speaking smokers through the National Cancer Institute's Cancer Information Service. A randomized controlled trial.

Although the prevalence of smoking is lower among Hispanics than among the general population, smoking still levies a heavy public health burden on this underserved group. The current study, Adiós al Fumar (Goodbye to Smoking), was designed to increase the reach of the Spanish-language smoking cessation counseling service provided by the National Cancer Institute's Cancer Information Service (CIS) and to evaluate the efficacy of a culturally sensitive, proactive, behavioral treatment program among Spanish-speaking smokers. Adiós was a 2-group randomized clinical trial evaluating a telephone-based smoking cessation intervention. Spanish-speaking smokers (N = 297) were randomized to receive either standard counseling or enhanced counseling (EC). Paid media was used to increase the reach of the Spanish-language smoking cessation services offered by the CIS. The Adiós sample was of very low socioeconomic status (SES), and more than 90% were immigrants. Calls to the CIS requesting smoking cessation help in Spanish increased from 0.39 calls to 17.8 calls per month. The unadjusted effect of EC only approached significance (OR = 2.4, P = .077), but became significant after controlling for demographic and tobacco-related variables (OR = 3.8, P = .048). Adiós al Fumar demonstrated that it is possible to reach, retain, and deliver an adequate dose of treatment to a very low SES population that has traditionally been viewed as difficult to reach and hard to follow. Moreover, the findings suggest that a proactive, telephone-counseling program, based on the Treating Tobacco Use and Dependence Clinical Practice Guideline and adapted to be culturally appropriate for Hispanics, is effective. Cancer 2007. (c) 2006 American Cancer Society.

Comparison of three modes of measuring stress, coping, and humor in school-age children.

Computer-assisted administration of surveys is gaining popularity among many researchers, but the equivalence of this method to more traditional approaches such as using paper and pencil has not been determined for many commonly used questionnaires, particularly among school-age children. This study examined systematic differences in the responses of 4th, 5th, and 6th graders to measures of stress, coping, and humor among three modes of assessment: paper-and-pencil questionnaires, computer-assisted self-interviewing (CASI), or a combination of paper-and-pencil and CASI. Participants were 1,245 ethnically diverse children enrolled in public schools in the central region of the United States. Psychometric and score distribution characteristics were examined using item analyses and analyses of mean and covariance structure as a function of mode of assessment. Differences in response patterns, primarily at the scale score level, were documented on some of the key measures. In general, CASI medians and means were higher and correlations among CASI measures tended to be lower than those obtained with paper-and-pencil and mixed mode assessment, and CASI variances were lower. This study suggests the importance of the continued examination of the impact of mode of questionnaire administration when assessing these and other domains of well-being in school-age children.

Measuring asthma severity: instrument refinement.

The focus of this study is to evaluate a brief parent-report instrument, the Severity of Chronic Asthma (SCA) scale, that conforms to the national guidelines for assessing asthma. Convergent validity was found between the SCA and other measures related to asthma severity including an illness severity scale (How Bad is the Asthma?), asthma management scales for parents and children, and the pediatric quality-of-life scale. The SCA is a multidimensional scale with appropriate evidence of reliability and validity that may be a heuristic and effective measure in both clinical practice and research endeavors.

At-risk drinking in employed men and women.

BACKGROUND: "At-risk" drinking is associated with a variety of negative health and social consequences. However, little is known about the characteristics of at-risk drinkers or of changes in at-risk status over time. PURPOSE: The objective was to examine the correlates of at-risk drinking and the prospective predictors of maintenance or change in at-risk status. METHOD: Participants were 4,322 employed individuals assessed at baseline and 4 years later. At-risk drinking was defined as 2 or more drinks per day for men and 1 or more drinks per day for women. RESULTS: The baseline prevalence of at-risk drinking was 11%. Four percent of baseline not-at-risk individuals transitioned to at-risk drinking at follow-up, and 54% of the baseline at-risk individuals remained at-risk at follow-up. Several demographic-, work-, and tobacco-related variables differentiated at-risk groups and were prospective predictors of change in at-risk drinking status among those individuals who were not at risk at baseline. However, none of the constructs predicted change among at-risk drinkers. CONCLUSION: The data suggest that at-risk drinking is of public health concern. Eleven percent of the participants met criteria for at-risk drinking. Further, at-risk and not-at-risk drinkers differed on numerous characteristics, and their drinking may be influenced by different factors.

Racial and ethnic differences in predictors of smoking cessation.

Racial/ethnic differences in the determinants of smoking cessation could have important treatment implications. The current study examined racial/ethnic differences in smoking cessation, prospective predictors of cessation, and whether the predictive ability of these factors differed by race/ethnicity. Participants were 709 employed adults recruited through the National Rural Electric Co-op Association or through natural gas pipeline corporations. Data were collected in 1990 and 1994. Although race/ethnicity was not predictive of abstinence, Hispanic, African American, and White smokers displayed differential on tobacco-, alcohol-, and work-related variables. These racial/ethnic differences highlight the specific factors that should be considered when providing smoking cessation treatment to specific populations. Limitations are noted.

Depression in pregnant women seeking smoking cessation treatment.

The proportion of DSM-IV-based threshold and subthreshold depressive disorders was assessed, using the Primary Care Evaluation of Mental Disorders and a DSM-IV based assessment of depression history, in a sample of 139 pregnant smokers seeking smoking cessation treatment. Sixty-three percent of the sample met criteria for a current or past depressive disorder. Forty-six percent had a current threshold or subthreshold depressive disorder. Twenty percent met criteria for current major depression, with half of those having major depression with dysthymia. Twenty-one percent of the sample was chronically depressed (i.e., dysthymia or major depression with dysthymia). Women with current depression were less likely to tell their doctors that they smoked. Our results suggest that investigation of the prevalence of depression in community samples of pregnant smokers is warranted, as depression may present a major obstacle in pregnant women's efforts to quit smoking.

Understanding the associations among education, employment characteristics, and smoking.

The current study examined the association between education level and smoking status in a community-based sample of working adults. Participants were enrolled at the University of Texas M.D. Anderson Cancer Center site of a cancer risk behavior reduction intervention delivered at the worksite. There was a strong educational gradient in smoking status. The prevalence of current smoking was almost threefold higher among individuals with