Multilevel Factor Research Articles

The Factorial Structure of Stigma and Its Targets

Abstract. We aimed to determine (1) the attributes of multiple stigmatized populations, (2) whether Kurzban and Leary’s (2001) functional typology of stigma emerges and identifies the dimensions upon which each stigma type differs, and (3) the emotional responses toward emergent stigma types. Participants ( N = 2,674) were assigned to 1/52 stigma target conditions and their attitudes surveyed. Data were analyzed by multilevel factor analysis with stigma targets at Level 2. There were five within-factors at Level 1 (social competence, interpersonal access, social inclusion, immorality, perceived permanence) and three between-factors at Level 2 (interpersonal stigmatization, morality-based stigmatization, chronic stigmatization). Interpersonal stigmatization was predicted by disgust and pity, and morality-based stigmatization was predicted by disgust and [-]happiness, with no predictors for chronic stigmatization. These results support a functional typology of stigma.

Empirical research methods department: Mission, learnings, and future plans

Empirical research methods department: Mission, learnings, and future plans

The goods in everyday love: Positivity resonance builds prosociality.

The positivity resonance theory of coexperienced positive affect (Fredrickson, 2016) identifies the emotion of love as a collective state. This state, termed positive resonance, is defined by the presence of three key features: shared positive affect, caring nonverbal synchrony, and biological synchrony. The current study examined whether a modest behavioral intervention focused on increasing social connectedness could increase study participants' perceptions of day-to-day positivity resonance with corollary impacts on their tendencies for prosociality and self-centeredness. Adults (N = 416, M age = 33.8) were randomized to one of four study conditions: either of two variants of the social connectedness intervention or either of two control groups. Positivity resonance, prosociality, and self-centeredness were measured nightly for 35 consecutive days. Dynamic multilevel factor models of nightly reports showed significant growth in positivity resonance, relative to a passive control group, for the two intervention groups and higher mean levels of prosociality for one of them. In addition, significant dose-response relations were evident (both between persons and within persons), linking positivity resonance to both prosociality and self-centeredness. The within-persons effect for prosociality (but not self-centeredness) was significantly stronger for those randomized to the intervention groups, relative to both passive and active control groups. Taken together, findings suggest that the affective quality of people's day-to-day social encounters may have implications for community flourishing. Discussion centers on theoretical and practical implications as well as directions for future research. (PsycInfo Database Record (c) 2022 APA, all rights reserved).

Estimation of a dynamic multi-level factor model with possible long-range dependence

A dynamic multi-level factor model with possible stochastic time trends is proposed. In the model, long-range dependence and short memory dynamics are allowed in global and local common factors as well as model innovations. Estimation of global and local common factors is performed on the prewhitened series, for which the prewhitening parameter is estimated semiparametrically from the cross-sectional and local average of the observable series. Employing canonical correlation analysis and a sequential least-squares algorithm on the prewhitened series, the resulting multi-level factor estimates have centered asymptotic normal distributions under certain rate conditions depending on the bandwidth and cross-section size. Asymptotic results for common components are also established. The selection of the number of global and local factors is discussed. The methodology is shown to lead to good small-sample performance via Monte Carlo simulations. The method is then applied to the Nord Pool electricity market for the analysis of price comovements among different regions within the power grid. The global factor is identified to be the system price, and fractional cointegration relationships are found between local prices and the system price, motivating a long-run equilibrium relationship. Two forecasting exercises are then discussed.

Testing the Performance of Level-Specific Fit Evaluation in MCFA Models With Different Factor Structures Across Levels.

A Monte Carlo study was conducted to compare the performance of a level-specific (LS) fit evaluation with that of a simultaneous (SI) fit evaluation in multilevel confirmatory factor analysis (MCFA) models. We extended previous studies by examining their performance under MCFA models with different factor structures across levels. In addition, various design factors and interaction effects between intraclass correlation (ICC) and misspecification type (MT) on their performance were considered. The simulation results demonstrate that the LS outperformed the SI in detecting model misspecification at the between-group level even in the MCFA model with different factor structures across levels. Especially, the performance of LS fit indices depended on the ICC, group size (GS), or MT. More specifically, the results are as follows. First, the performance of root mean square error of approximation (RMSEA) was more promising in detecting misspecified between-level models as GS or ICC increased. Second, the effect of ICC on the performance of comparative fit index (CFI) or Tucker-Lewis index (TLI) depended on the MT. Third, the performance of standardized root mean squared residual (SRMR) improved as ICC increased and this pattern was more clear in structure misspecification than in measurement misspecification. Finally, the summary and implications of the results are discussed.

The evolution and inter-sectoral interaction mechanism of China's national medical alliance: An analysis based on complex systems theory.

This work investigates the performance and inter-sectoral interaction mechanism of China's largest vertically integrated care network, the national medical alliance (NMA). The data collected derive from the China Health Statistics Bulletin and the China Health Statistical Yearbook for the period 2009-2018. The data include 64 observation indicators for five medical sectors in the NMA, namely, tertiary hospitals (THS), secondary hospitals (SHS), community health centres (CHCS), township hospitals (TsHS) and professional public health institutions (PPHIS). This research combines complex systems theory with a multilevel structural dynamic factor model, and yields two main results. First, although the trend for the NMA's global factor is increasing, the evolutionary paths for sectoral factors differ substantially. Among the sectoral factors, the sectoral factor of THS continued to decline, and neither the sectoral factor of CHCS nor the sectoral factor of TsHS has significantly improved. Then, the interaction mechanism between the various NMA sectors is investigated. While a close relationship has been formed between THS and CHCS and between SHS and CHCS, there remains no close two-way relationship between either THS and TsHS or THS and SHS. Thus, going forward, to reach the policy expectations, China's NMA implementation must consider the interaction between different constituent sectors.

Determining Reliability of Daily Measures: An Illustration with Data on Teacher Stress

ABSTRACT Intensive longitudinal data is increasingly used to study state-like processes such as changes in daily stress. Measures aimed at collecting such data require the same level of scrutiny regarding scale reliability as traditional questionnaires. The most prevalent methods used to assess reliability of intensive longitudinal measures are based on the generalizability theory or a multilevel factor analytic approach. However, the application of recent improvements made for the factor analytic approach may not be readily applicable for all researchers. Therefore, this article illustrates a five-step approach for determining reliability of daily data, which is one type of intensive longitudinal data. First, we show how the proposed reliability equations are applied. Next, we illustrate how these equations are used as part of our five-step approach with empirical data, originating from a study investigating changes in daily stress of secondary school teachers. The results are a within-level (ωw), between-level (ωb) reliability score. Mplus syntax for these examples is included and discussed. As such, this paper anticipates on the need for comprehensive guides for the analysis of daily data.

Construction of evaluation system and evaluation model of urban environmental engineering geological problems —— a case study of Jinan City

In order to further study the problems of urban engineering geological environment, taking Jinan as an example, the environmental engineering geological problems are divided into four aspects: primary environmental engineering geological problems, secondary environmental engineering geological problems, the influence of human activities and the restriction of spring water on engineering construction. A multi-level, multi-objective and multi factor evaluation index system suitable for environmental engineering geology is established. The mathematical model is established by using analytic hierarchy process and fuzzy comprehensive evaluation method to evaluate the primary environmental engineering geological problems, secondary environmental engineering geological problems, the impact of human activities and the restrictive effect of spring water on engineering construction. The mathematical model is established by using Nemero index method to comprehensively evaluate the environmental engineering geological problems in Jinan, and the suitability zoning of engineering construction is carried out.

Can Ecological Momentary Assessments Be Used as Daily Markers of Personality Traits?

This study hypothesized that select ecological momentary assessment (EMA) survey items are sensitive to day-to-day fluctuations in personality traits Extraversion (E) and Neuroticism (N). As part of the Einstein Aging Study, 312 older adults (Mage=76.96 years, SD=4.85 years, range=70-90 years) completed up to 5 EMA surveys per day for 16 days and a Big Five trait personality measure. Parallel two-factor multilevel confirmatory factor analyses were conducted for E (Daily-E; Trait-E) and N (Daily-N; Trait-N). The E model showed good fit (CFI=.95; TLI=.94; RMSEA=.02) and a significant correlation of .20 between Daily-E and Trait-E factors. The N model showed poor fit (CFI=.68; TLI=.61; RMSEA=.06). Results suggest EMA items can be used as daily markers of Extraversion, yet results are unclear for Neuroticism due to poor model fit. Daily markers of Extraversion can be used to detect fluctuations in personality traits across days that may predict long-term personality change.

Cultural Variation in Aggressive Behavior: A Cross-Cultural Comparison of Students’ Exposure to Bullying Across 32 Countries

Introduction. The prevalence rates of bullying vary significantly across countries and continents. Specifically, UNESCO estimates that the prevalence rates vary from 22.8% (Central America) to 48.2% (Sub-Saharan Africa). Recently, these differences among countries and regions have been attributed to culture- and country-level variables. Thus, the first purpose of this study is to examine the comparability of bullying in schools across countries. Secondly, a cross-cultural comparison of the latent mean scores of bullying is implemented. Method. The data of 286,481 adolescent students (M=15.78, SD=0.29) from 32 countries were analyzed using multilevel confirmatory factor analyses (MLCFA) and multigroup factor alignment. Results. Results indicated that the meaning of bullying is equivalent within and between cultures. However, cross-cultural differences in bullying are apparent. East Asian countries have the lowest latent means of bullying, while Southeast Asian countries have the highest means. Anglo-Saxon, Eastern European, Mediterranean, South American, and Middle East countries displayed rather higher scores. Discussion and Conclusion. These findings underscore the existence of cross-cultural differential responding in bullying measures. Further, the implicit role of culture as an important variable that determines the rates of bullying is underscored.

Mirror-Bridge Phase-Shift Modulation With Low Common-Mode Noise for Single-Phase CHB PFC

For ac/dc power conversion applications like server power supply, telecom power supply, and on-board charger of an electric vehicle, the rising electricity demand has presented challenges toward high efficiency and high power density. The cascaded H-bridge (CHB) based multilevel power factor correction (PFC) converter has much higher power density than that of traditional boost PFC converter. However, the CHB-PFC converter has multiple floating dc outputs, which leads to a common mode (CM) electromagnetic interference (EMI) issue due to the parasitic capacitance of cascaded dc/dc stages and layouts and distributed noise sources existing in the ac/dc system. Thus, a large EMI filter is necessary for the PFC system that sharply degrades the system's power density. In order to quantify the CM noise caused by the floating outputs and cascaded parasitic capacitance, a unified equivalent CM circuit model is derived in this article, which decouples the effects among the multiple CM noise sources in the CHB converter. Based on the derived circuit model, a mirror-bridge phase-shift modulation is proposed to cancel the CM noise within the CHB converter itself. A single-phase 2-kW CHB PFC prototype has been built up to verify the analysis, and 40-dB CM noise reduction compared with conventional unipolar modulation is achieved by the proposed modulation scheme.

Instructional Practices, Students’ Self-Efficacy and Math Achievement: a Multi-level Factor Score Path Analysis

Instructional Practices, Students’ Self-Efficacy and Math Achievement: a Multi-level Factor Score Path Analysis

Development and initial validation of a scale to measure momentary self-concept clarity

ABSTRACT Several studies have suggested that momentary self-concept clarity (SCC) levels are important for emotion regulation and self-control processes, but these studies have used unvalidated measures of momentary SCC. Here, we report on the development and preliminary validation of a brief self-report scale, the Momentary Self-Concept Clarity Scale (M-SCCS). One hundred and twenty-two adults completed momentary SCC items 6-7 times per day for two weeks. Multilevel factor analyses suggested the M-SCCS has good factorial validity. The scale also showed excellent between-person reliability, fair within-person reliability, and patterns of criterion relations that resemble other self-report measures of SCC. There was little measurement reactivity over time. The M-SCCS may be useful for future experience-sampling studies aimed at uncovering short-term self-regulatory processes involving SCC.

Evaluating the relative efficiency among robust estimation methods for multilevel factor analysis with categorical data

Multilevel measurement models are more frequently applied to help answer questions when data arise from hierarchically structured multivariate data. In this simulation study of multilevel factor models, we evaluated the relative efficiency among three estimation methods: robust maximum likelihood, unweighted least squares, and weighted least squares. We found that weighted least squares yielded more or equally efficient parameter estimates under all sample size conditions for all model parameters. The relative efficiency of standard errors was less straightforward where maximum likelihood was more efficient for loadings and residual variances, but weighted least squares was more efficient for the factor covariance matrices. Finally, we give recommendations for estimating multilevel confirmatory factor analysis models and directions for future research.

Validation of the Child Behavior Rating Scale (CBRS) using multilevel factor analysis.

Behavioral regulation supports children to control their cognitive and emotional skills and participate fully in classroom learning and interactions. Teacher ratings are frequently used to collect data but are highly susceptible to the teacher's response bias, meaning much child level variance is attributable to the teacher level. Multilevel modeling techniques are commonly used to assess the psychometric properties of measures at different levels by disaggregating within- and between-teacher variation in student behaviors. In this study, we attempt to identify the factor structure at each level using teacher ratings on the Child Behavior Rating Scale (CBRS). The sample of 1,375 preschool children (49.2% males) was drawn from a longitudinal study in Singapore. The teacher ratings of these children were collected at three time points (average age at Time 1 was 58.5 months, Time 2 was 69.8 months, and Time 3 was 80.9 months). Using multilevel exploratory and confirmatory factor analyses, a three-factor model at the within level and a two-factor model at the between level best fit the data. At the within level, the three factors were identified as basic interpersonal skills, advanced interpersonal skills, and learning-related social skills. The three factors showed good internal consistency and differential correlates. The two factors at the between level may be related to teacher and classroom characteristics. This study provides empirical support for (a) differential factor structures of CBRS at the within and between levels; and (b) two distinct interpersonal skills. Future studies should aim at exploring the sources of between level variability. (PsycInfo Database Record (c) 2021 APA, all rights reserved).

Psychometric analysis and validity of the Daily Sexual Minority Stressors Scale among young adult same-sex female couples.

Sexual minority women experience greater health disparities relative to heterosexual women, which is thought to be due to unique stressors related to their sexual identities. Daily diary or momentary assessments may provide a more nuanced approach to understanding how sexual minority stressors relate to health behaviors than cross-sectional studies provide. To date, there is no validated measure to examine daily sexual minority stressors. A recent pilot study developed a brief (8-item) measure for assessing sexual minority stressors in self-identified lesbian or mostly lesbian women (Heron et al., 2018). Although an optimal number of items were generated to best capture the daily experiences of lesbian women, psychometric examination and validation of this new measure is necessary. Using multilevel confirmatory factor analysis among a fully scaled sample of sexual minority women, the present study established that the Daily Sexual Minority Stressors Scale has good model fit as a unidimensional measure (i.e., one factor at each level of analysis). Intraclass correlations indicate the majority of variation (57%) is within person. Additionally, we established convergent and discriminant validity using similar measures (single-item assessment, general stressors, negative affect, history of discrimination, and heterosexism). Finally, criterion validity was supported. At the daily level, experiencing daily sexual minority stressors was associated with a significantly greater likelihood of drinking alcohol that day. Experiencing more daily sexual minority stressors during the study period was significantly associated with a history of harassment and discrimination, victimization, isolation, vigilance, and also with acceptance concerns, difficult processes, and internalized homonegativity. (PsycInfo Database Record (c) 2021 APA, all rights reserved).

Canonical correlation-based model selection for the multilevel factors

We develop a novel approach based on the canonical correlation analysis to identify the number of the global factors in the multilevel factor model. We propose the two consistent selection criteria, the canonical correlations difference (CCD) and the modified canonical correlations (MCC). Via Monte Carlo simulations, we show that CCD and MCC select the number of global factors correctly even in small samples, and they are robust to the presence of serially correlated and weakly cross-sectionally correlated idiosyncratic errors as well as the correlated local factors. Finally, we demonstrate the utility of our approach with an application to the multilevel asset pricing model for the stock return data in 12 industries in the U.S.

Impact of Corporate Governance on Organisational Performance of Indian Firms

This research study explored the link between corporate governance practices (CGPs) and organisational performance in India, especially in the context of some major CG reforms that have been undertaken since the turn of the twenty-first century. The authors also attempted to understand in-depth the implications of these reforms for the companies. For assessing the link between CG practices and organisational performance, data were collected from a sample of 100 listed companies in India using an adapted version of the Institute of Company Secretaries of India (ICSI)’s questionnaire. Multilevel Factor Analysis (MFA) for scores along 5 CG sub-categories revealed 17 first-level and 4 second-level factors. Regression of organisational performance, measured using Compound Annual Growth Rate (CAGR), against these factors showed that the first-level factor representing corporate social responsibility and sustainability (CSRS) was a significant predictor of organisational performance. This finding is significant while considering the introduction of mandatory CG provisions for corporate social responsibility (CSR), applicable to companies meeting specified turnover and profitability thresholds according to CG regulations in India. The findings of this study open the debate on CG regulation and on mandatory and desirable norms in the Indian context. Eligible Indian companies must focus on the CG practice of investing in CSR initiatives through purpose-led CSRS interventions and their long-term benefits, rather than on viewing it as a mandatory CG provision that induces short-term expenses.

Using the three-pillar model of sustainability to understand lay reactions to climate policy: A multilevel approach

Public support for climate policy is necessary to enact the large-scale changes needed to mitigate climate change. We use the three-pillar model of sustainability as a conceptual framework to examine how individuals evaluate climate policies and how these evaluations predict policy support. We consider individuals’ evaluations of 1) environmental impacts (i.e., perceived policy effectiveness), 2) economic impacts, and 3) social impacts of policies. We use multilevel modeling to disentangle variance in policy support attributable to between-person differences (i.e., aggregated policy support) and within-person variations across policies. First, we fit a multilevel factor model to our dataset (Nobs = 1056) to identify whether the factor structure of the three-pillar model at both the between-person and within-person levels of analysis. The three-pillar model emerged at the within-person level. In contrast, items instead loaded onto two factors at the between-person level: benefits and harms. Thus, we created within- and between-person constructs matching these factors. Results from multilevel regressions suggest that a) individuals who anticipate more benefits and fewer harms of climate policies (as a set) also tend to report greater aggregated support for climate change policies, and b) anticipating environmental benefits, economic impacts and social impacts to be above average for a certain policy (relative to other policies) predicts greater support for that policy. Our work suggests the relevance of differentiating between who is most likely to support climate policies in general (i.e., between-person differences) as well as policy-specific evaluations associated with support for some policies over others (i.e. within-person variation in support).

Classroom Engagement Scale: Validation of a Teacher-Report Assessment Used to Scale in the Kindergarten Report Card of a Large Urban School District

ABSTRACT Research Findings: This study provides psychometric validation of the Classroom Engagement Scale (CES), a 14-item, teacher-report, formative assessment of behavioral engagement. It is currently used for all kindergarten children in a large, urban school district where it is embedded in the report card. Multi-level factor analysis revealed two distinct dimensions of the measure – academic engagement and social engagement – that comport with the educational research literature. Measurement invariance investigations provided initial support for the applicability of the CES across student subgroups and time. Finally, the two dimensions of the CES were found to be differentially predictive of important kindergarten and third-grade outcomes, providing evidence of concurrent and predictive validity. Practice or Policy: This study was designed in response to both the national call for routine monitoring of critical engagement competencies and the lack of validated teacher-report measures of these skills that can be used to inform ongoing instruction. This initial validation is an important step in institutionalizing practical, quality measurement within schools. Because of its strategic placement on the kindergarten report card, the CES provides an opportunity for teachers and families to share information about students’ engagement skills and coordinate confluent supports to foster these critical competencies.