Monte Carlo confidence intervals for the indirect effect with missing data.

Missing data is a common occurrence in mediation analysis. As a result, the methods used to construct confidence intervals around the indirect effect should consider missing data. Previous research has demonstrated that, for the indirect effect in data with complete cases, the Monte Carlo method performs as well as nonparametric bootstrap confidence intervals (see MacKinnon et al., Multivariate Behavioral Research, 39(1), 99-128, 2004; Preacher & Selig, Communication Methods and Measures, 6(2), 77-98, 2012; Tofighi & MacKinnon, Structural Equation Modeling: A Multidisciplinary Journal, 23(2), 194-205, 2015). In this manuscript, we propose a simple, fast, and accurate two-step approach for generating confidence intervals for the indirect effect, in the presence of missing data, based on the Monte Carlo method. In the first step, an appropriate method, for example, full-information maximum likelihood or multiple imputation, is used to estimate the parameters and their corresponding sampling variance-covariance matrix in a mediation model. In the second step, the sampling distribution of the indirect effect is simulated using estimates from the first step. A confidence interval is constructed from the resulting sampling distribution. A simulation study with various conditions is presented. Implications of the results for applied research are discussed.

FINDOUT: Using Either SPSS Commands or Graphical User Interface to Identify Influential Cases in Structural Equation Modeling in AMOS.

The results in a structural equation modeling (SEM) analysis can be influenced by just a few observations, called influential cases. Tools have been developed for users of R to identify them. However, similar tools are not available for AMOS, which is also a popular SEM software package. We introduce the FINDOUT toolset, a group of SPSS extension commands, and an AMOS plugin, to identify influential cases and examine how these cases influence the results. The SPSS commands can be used either as syntax commands or as custom dialogs from pull-down menus, and the AMOS plugin can be run from AMOS pull-down menu. We believe these tools can help researchers to examine the robustness of their findings to influential cases.

betaDelta and betaSandwich: Confidence Intervals for Standardized Regression Coefficients in R.

The multivariate delta method was used by Yuan and Chan to estimate standard errors and confidence intervals for standardized regression coefficients. Jones and Waller extended the earlier work to situations where data are nonnormal by utilizing Browne's asymptotic distribution-free (ADF) theory. Furthermore, Dudgeon developed standard errors and confidence intervals, employing heteroskedasticity-consistent (HC) estimators, that are robust to nonnormality with better performance in smaller sample sizes compared to Jones and Waller's ADF technique. Despite these advancements, empirical research has been slow to adopt these methodologies. This can be a result of the dearth of user-friendly software programs to put these techniques to use. We present the betaDelta and the betaSandwich packages in the R statistical software environment in this manuscript. Both the normal-theory approach and the ADF approach put forth by Yuan and Chan and Jones and Waller are implemented by the betaDelta package. The HC approach proposed by Dudgeon is implemented by the betaSandwich package. The use of the packages is demonstrated with an empirical example. We think the packages will enable applied researchers to accurately assess the sampling variability of standardized regression coefficients.

manymome: An R package for computing the indirect effects, conditional effects, and conditional indirect effects, standardized or unstandardized, and their bootstrap confidence intervals, in many (though not all) models.

Mediation, moderation, and moderated mediation are common in behavioral research models. Several tools are available for estimating indirect effects, conditional effects, and conditional indirect effects and forming their confidence intervals. However, there are no simple-to-use tools that can appropriately form the bootstrapping confidence interval for standardized conditional indirect effects. Moreover, some tools are restricted to a limited type of models. We developed an R package, manymome, which can be used to estimate and form confidence intervals for indirect effects, conditional effects, and conditional indirect effects, standardized or not, using a two-step approach: model parameters are estimated either by structural equation modeling using lavaan or by a set of linear regression models using lm, and then the coefficients are used to compute the requested effects and form confidence intervals. It can be used when there are missing data if the model is fitted by structural equation modeling. There are only a few limitations on some aspects of a model, and no inherent limitations on the number of predictors, the number of independent variables, or the number of moderators and mediators. The goal is to have a tool that allows researchers to focus on model fitting first and worry about estimating the effects later. The use of the model is illustrated using a few numerical examples, and the limitations of the package are discussed.

DIY bootstrapping: Getting the nonparametric bootstrap confidence interval in SPSS for any statistics or function of statistics (when this bootstrapping is appropriate).

Researchers can generate bootstrap confidence intervals for some statistics in SPSS using the BOOTSTRAP command. However, this command can only be applied to selected procedures, and only to selected statistics in these procedures. We developed an extension command and prepared some sample syntax files based on existing approaches from the Internet to illustrate how researchers can (a) generate a large number of nonparametric bootstrap samples, (b) do desired analysis on all these samples, and (c) form the bootstrap confidence intervals for selected statistics using the OMS commands. We developed these tools to help researchers apply nonparametric bootstrapping to any statistics for which this method is appropriate, including statistics derived from other statistics, such as standardized effect size measures computed from the t test results. We also discussed how researchers can extend the tools for other statistics and scenarios they encounter.

The influence of nonnormality from primary studies on the standardized mean difference in meta-analysis.

In this study we investigated the influence of data nonnormality in the primary studies on meta-analysis of the standardized mean difference (SMD) for a two-independent-group design. The bias, mean squared error, and confidence interval coverage probability of the mean effect sizes under different types of population distributions were compared. Also, the performance of the Q test was examined. The results showed that oppositely skewed distributions (i.e., distributions skewed in different directions) showed poor performance for point and interval estimates of mean effect sizes in meta-analysis, especially when the tails were pointing toward each other. The previously found adverse impacts due to nonnormality in primary studies do not disappear when primary studies with nonnormal data are meta-analyzed, even when the average sample size and number of studies are large. The results also showed that, when the tails were pointing toward each other, the Type I error rates of the Q test were inflated. We suggest that the impact of violating the assumption of normality should not be ignored in meta-analysis.

Meta-analyzing dependent correlations with correction for artifacts that multiplicatively attenuate the true correlation.

Previous procedures for meta-analyzing dependent correlations have been found to overestimate or underestimate the true variation in effect sizes. Samplewise-adjusted procedures have been shown to perform better than simple within-study means when meta-analyzing dependent correlations. However, such procedures cannot be applied when correction for artifacts such as unreliability is desired. In the present study, we extended the procedures to correct for attenuation due to artifacts when meta-analyzing dependent correlations. Monte Carlo simulation was conducted in order to examine conditions with various degrees of dependence, degrees of heterogeneity, sample sizes, and numbers of studies, among other factors. The previous procedures, including the samplewise-adjusted procedures without correction, yielded biased point estimates and confidence intervals with low coverage probabilities of the population mean correlation and degree of heterogeneity. More importantly, the bias and undercoverage of the confidence interval increased with the mean sample size and number of studies in many conditions. The new samplewise-adjusted procedures with correction for attenuation yielded negligible biases when estimating the mean population correlation, even in the presence of dependent correlations. Given that the need for correction for attenuation due to artifacts is becoming more recognized in meta-analysis, our findings highlight the importance of such considerations when meta-analyzing dependent correlations. Conditions under which these procedures can be further improved are also discussed.

Meta-analyzing dependent correlations: an SPSS macro and an R script.

The presence of dependent correlation is a common problem in meta-analysis. Cheung and Chan (2004, 2008) have shown that samplewise-adjusted procedures perform better than the more commonly adopted simple within-sample mean procedures. However, samplewise-adjusted procedures have rarely been applied in meta-analytic reviews, probably due to the lack of suitable ready-to-use programs. In this article, we compare the samplewise-adjusted procedures with existing procedures to handle dependent effect sizes, and present the samplewise-adjusted procedures in a way that will make them more accessible to researchers conducting meta-analysis. We also introduce two tools, an SPSS macro and an R script, that researchers can apply to their meta-analyses; these tools are compatible with existing meta-analysis software packages.

Personal spiritual values and quality of life: evidence from Chinese college students.

Values are guiding principles in our life. While some studies found spiritual values to be "healthier," Sagiv and Schwartz (Eur J Soc Psychol 30:177-198, 2000) showed that people holding non-spiritual values were higher on affective well-being. We examined the predictive power of these two types of values with a longitudinal data set collected from Chinese students mainly in Hong Kong. Structural equation modeling revealed that spiritual values (as well as family income) positively predicted quality of life a year later. Non-spiritual, self-enhancement values, did not show any association. Results suggest that developing spiritual values may promote well-being through enabling individuals to find meaning and purpose in life.

Improving an old way to measure moderation effect in standardized units.

Moderation effects in multiple regression, tested usually by the inclusion of a product term, are frequently investigated in health psychology. However, several issues in presenting the moderation effects in standardized units and their associated confidence intervals are commonly observed. While an old method had been proposed to standardize variables in moderated regression before fitting a moderated regression model, this method was rarely used due to inconvenience and even when used, the confidence intervals derived were biased. Here, we attempt to solve these two problems by providing a tool to conveniently conduct standardization in moderated regression without the step of standardizing the variables beforehand and to accurately form the nonparametric bootstrapping confidence intervals for this standardized measure of moderation effects. Health psychology researchers are now equipped with a tool that can be used to report and interpret standardized moderation effects correctly. (PsycInfo Database Record (c) 2022 APA, all rights reserved).

Development of an Indigenous Inventory GMAB (Gambling Motives, Attitudes and Behaviors) for Chinese gamblers: an exploratory study.

Scale development in the extant gambling literature has been dominated by pathological gamblers, but the non- or sub-clinical gamblers have been overlooked. Moreover, most scales are predominantly based on Western samples; only a few of the scales have Chinese versions validated with Chinese samples. A rarely explored niche still exists for the development of an indigenous scale for Chinese gamblers. The current exploratory study made the first step towards such a direction by identifying factors through the construction of an indigenous Gambling Motives, Attitudes and Behavior (GMAB) scale for Chinese gamblers. Preliminary items were generated primarily from focus group discussions. The items were administered through a telephone survey in which 791 randomly sampled gamblers participated. Exploratory factor analyses revealed (a) five dimensions of gambling motives, namely, self-worth, monetary gains, sensation seeking, boredom alleviation, and learning; (b) four dimensions of gambling attitudes, namely, luck and fate, attitudes toward negative consequences in gambling, techniques, and superstition; and (c) six dimensions of behavior, namely, impaired control in gambling, gambling involvement, arousal reaction, superstitious behavior, controlled gambling and casino exploration. Implications of the interplay among these factors and future research directions were discussed.

A longitudinal investigation of the bidirectional relationship of sleep quality with emotional stability and social cynicism in a large community sample.

OBJECTIVE: Dispositional characteristics like emotional stability and social cynicism have been consistently associated with negative affect, which is a known predictor and outcome of poor sleep quality. This study hypothesized a bidirectional relationship of sleep quality with emotional stability and social cynicism over a five-year period. METHODS: Participants were 7,181 Chinese people, who completed two waves of online surveys from a larger panel study. Questionnaires on Big Five personality traits, social cynicism, and sleep quality were administered twice at a five-year interval. RESULTS: Cross-lagged analysis revealed a significant bidirectional relationship between emotional stability and sleep quality over five years as hypothesized. However, there was no association between sleep quality and social cynicism in either direction. CONCLUSION: Our study provides consistent evidence of a bidirectional relationship between sleep quality and emotional stability in a five-year time-frame among a younger population (mean age = 24.86). Given the critical role of emotional stability in various areas of functioning, the findings highlight the importance of sleep health education in young adults, who are going through a critical period of personality development.

SEM-Based Methods to Form Confidence Intervals for Indirect Effect: Still Applicable Given Nonnormality, Under Certain Conditions.

A SEM-based approach using likelihood-based confidence interval (LBCI) has been proposed to form confidence intervals for unstandardized and standardized indirect effect in mediation models. However, when used with the maximum likelihood estimation, this approach requires that the variables are multivariate normally distributed. This can affect the LBCIs of unstandardized and standardized effect differently. In the present study, the robustness of this approach when the predictor is not normally distributed but the error terms are conditionally normal, which does not violate the distributional assumption of ordinary least squares (OLS) estimation, is compared to four other approaches: nonparametric bootstrapping, two variants of LBCI, LBCI assuming the predictor is fixed (LBCI-Fixed-X) and LBCI based on ADF estimation (LBCI-ADF), and Monte Carlo. A simulation study was conducted using a simple mediation model and a serial mediation model, manipulating the distribution of the predictor. The Monte Carlo method performed worst among the methods. LBCI and LBCI-Fixed-X had suboptimal performance when the distributions had high kurtosis and the population indirect effects were medium to large. In some conditions, the problem was severe even when the sample size was large. LBCI-ADF and nonparametric bootstrapping had coverage probabilities close to the nominal value in nearly all conditions, although the coverage probabilities were still suboptimal for the serial mediation model when the sample size was small with respect to the model. Implications of these findings in the context of this special case of nonnormal data were discussed.

Clinical utility of the Cross-Cultural (Chinese) Personality Assessment Inventory (CPAI-2) in the assessment of substance use disorders among Chinese men.

This study examined the clinical utility of the Cross-Cultural (Chinese) Personality Inventory (CPAI-2) in differentiating the personality characteristics of Chinese men with substance use disorders from other psychiatric patients and normal control participants. The CPAI-2 profile of 121 Chinese men with substance use disorders was contrasted with that of a matched psychiatric comparison group (n = 172) and a normal comparison group (n = 187). Multivariate analyses of variance and logistic regression results supported the utility of the CPAI-2 clinical scales, especially Pathological Dependence, Antisocial Behavior, and Depression, for assessing substance use disorders. The Pathological Dependence scale (cutoff T score of 64) achieved good sensitivity and specificity. Apart from the universal personality traits related to neuroticism, conscientiousness, and agreeableness found in Western studies, the indigenously derived CPAI-2 personality scales, including Family Orientation and Harmony, highlighted deficits in social adjustment and interpersonal relationship as important cultural features in the personality characteristics of these participants. The study provided a cross-cultural extension to research on the relationship between personality and substance use disorders and could assist clinicians in considering culturally relevant treatment approaches.

Meta-analysis of stigma and mental health.

Recent research has emphasized the adverse effects of stigma on minority groups' mental health. Governments and service agencies have put much effort into combating stigma against a variety of conditions. Nevertheless, previous empirical research on the stigma-mental health relationship has yielded inconclusive findings, varying from strong negative to zero correlations. Thus, whether stigma is related significantly to mental health is yet to be confirmed. Using meta-analysis, the associations between stigma and mental health from 49 empirical studies were examined across various stigmatized conditions and mental health indices. Possible moderators were also explored. The mean correlation between stigma and average mental health scores corrected for sampling error, unreliability, and other artifacts was -.28 (N=10,567, k=52). No strong moderators were found, yet meaningful patterns were observed. Implications of the results are discussed.

Sleep and optimism: A longitudinal study of bidirectional causal relationship and its mediating and moderating variables in a Chinese student sample.

While both sleep and optimism have been found to be predictive of well-being, few studies have examined their relationship with each other. Neither do we know much about the mediators and moderators of the relationship. This study investigated (1) the causal relationship between sleep quality and optimism in a college student sample, (2) the role of symptoms of depression, anxiety, and stress as mediators, and (3) how circadian preference might moderate the relationship. Internet survey data were collected from 1,684 full-time university students (67.6% female, mean age = 20.9 years, SD = 2.66) at three time-points, spanning about 19 months. Measures included the Attributional Style Questionnaire, the Pittsburgh Sleep Quality Index, the Composite Scale of Morningness, and the Depression Anxiety Stress Scale-21. Moderate correlations were found among sleep quality, depressive mood, stress symptoms, anxiety symptoms, and optimism. Cross-lagged analyses showed a bidirectional effect between optimism and sleep quality. Moreover, path analyses demonstrated that anxiety and stress symptoms partially mediated the influence of optimism on sleep quality, while depressive mood partially mediated the influence of sleep quality on optimism. In support of our hypothesis, sleep quality affects mood symptoms and optimism differently for different circadian preferences. Poor sleep results in depressive mood and thus pessimism in non-morning persons only. In contrast, the aggregated (direct and indirect) effects of optimism on sleep quality were invariant of circadian preference. Taken together, people who are pessimistic generally have more anxious mood and stress symptoms, which adversely affect sleep while morningness seems to have a specific protective effect countering the potential damage poor sleep has on optimism. In conclusion, optimism and sleep quality were both cause and effect of each other. Depressive mood partially explained the effect of sleep quality on optimism, whereas anxiety and stress symptoms were mechanisms bridging optimism to sleep quality. This was the first study examining the complex relationships among sleep quality, optimism, and mood symptoms altogether longitudinally in a student sample. Implications on prevention and intervention for sleep problems and mood disorders are discussed.

Random-effects models for meta-analytic structural equation modeling: review, issues, and illustrations.

Meta-analytic structural equation modeling (MASEM) combines the techniques of meta-analysis and structural equation modeling for the purpose of synthesizing correlation or covariance matrices and fitting structural equation models on the pooled correlation or covariance matrix. Both fixed-effects and random-effects models can be defined in MASEM. Random-effects models are well known in conventional meta-analysis but are less studied in MASEM. The primary objective of this paper was to address issues related to random-effects models in MASEM. Specifically, we compared two different random-effects models in MASEM-correlation-based MASEM and parameter-based MASEM-and explored their strengths and limitations. Two examples were used to illustrate the similarities and differences between these models. We offered some practical guidelines for choosing between these two models. Future directions for research on random-effects models in MASEM were also discussed. Copyright © 2016 John Wiley & Sons, Ltd.

Bidirectional relationship between sleep and optimism with depressive mood as a mediator: A longitudinal study of Chinese working adults.

OBJECTIVE: Sleep and optimism are important psycho-biological and personality constructs, respectively. However, very little work has examined the causal relationship between them, and none has examined the potential mechanisms operating in the relationship. This study aimed to understand whether sleep quality was a cause or an effect of optimism, and whether depressive mood could explain the relationship. METHOD: Internet survey data were collected from 987 Chinese working adults (63.4% female, 92.4% full-time workers, 27.0% married, 90.2% Hong Kong residents, mean age=32.59 at three time-points, spanning about 19 months). Measures included a Chinese attributional style questionnaire, the Pittsburgh Sleep Quality Index, and the Depression Anxiety Stress Scale. RESULTS: Cross-sectional analyses revealed moderate correlations among sleep quality, depressive mood, and optimism. Cross-lagged analyses showed a bidirectional causality between optimism and sleep. Path analysis demonstrated that depressive mood fully mediated the influence of optimism on sleep quality, and it partially mediated the influence of sleep quality on optimism. CONCLUSION: Optimism improves sleep. Poor sleep makes a pessimist. The effects of sleep quality on optimism could not be fully explained by depressive mood, highlighting the unique role of sleep on optimism. Understanding the mechanisms of the feedback loop of sleep quality, mood, and optimism may provide insights for clinical interventions for individuals presented with mood-related problems.

Indigenous measures of personality assessment in Asian countries: a review.

This article reviews attempts to develop multidimensional personality measures in Asia and their applications in clinical assessment. Indigenous personality assessment measures in India, Korea, Japan, the Philippines, and Taiwan are examined. These early attempts have not yielded a comprehensive personality measure that integrates a theoretical framework and an empirical program of validation. The Chinese Personality Assessment Inventory (CPAI) is cited as an example to illustrate the process of developing an indigenous measure that meets the testing standards of established assessment instruments. On the basis of the research findings from the CPAI, the authors discuss the relevance of indigenous measures in clinical assessment in native cultures as well as in informing mainstream personality assessment.

Dependent effect sizes in meta-analysis: incorporating the degree of interdependence.

In the present article, a commonly used meta-analytic procedure for handling dependent effect sizes from a single sample was examined, and 2 revised procedures that estimate and incorporate the degree of interdependence were proposed. The authors' simulation results reveal that the commonly used procedure that averages the effect sizes from a single sample (denoted as the samplewise procedure) underestimates the degree of heterogeneity. The proposed variations are less biased than the samplewise procedure in estimating the degree of heterogeneity in most of the situations that we examined. Future directions to further improve the procedures for handling dependent effect sizes from a single sample are discussed.