Model-building Approach Research Articles

Feasibility of delivering parent-implemented NDBI interventions in low-resource regions: a pilot randomized controlled study

BackgroundThis implementation feasibility study was conducted to determine whether an evidence-based parent-implemented distance-learning intervention model for young children at high likelihood of having ASD could be implemented at fidelity by Part C community providers and by parents in low-resource communities.MethodsThe study used a community-academic partnership model to adapt an evidence-based intervention tested in the current pilot trial involving randomization by agency in four states and enrollment of 35 coaches and 34 parent-family dyads. After baseline data were gathered, providers in the experimental group received 12–15 h of training while control providers received six webinars on early development. Providers delivered 6 months of intervention with children-families, concluding with data collection. Regression analyses were used to model outcomes of the coach behaviors, the parent fidelity ratings, and child outcomes.ResultsA block design model-building approach was used to test the null model followed by the inclusion of group as a predictor, and finally the inclusion of the planned covariates. Model fit was examined using changes in R2 and F-statistic. As hypothesized, results demonstrated significant gains in (1) experimental provider fidelity of coaching implementation compared to the control group; and (2) experimental parent fidelity of implementation compared to the control group. There were no significant differences between groups on child developmental scores.ConclusionsEven though the experimental parent group averaged less than 30 min of intervention weekly with providers in the 6 months, both providers and parents demonstrated statistically significant gains on the fidelity of implementation scores with moderate effect sizes compared to control groups. Since child changes in parent-mediated models are dependent upon the parents’ ability to deliver the intervention, and since parent delivery is dependent upon providers who are coaching the parents, these results demonstrated that two of these three links of the chain were positively affected by the experimental implementation model. However, a lack of significant differences in child group gains suggests that further work is needed on this model. Factors to consider include the amount of contact with the provider, the amount of practice children experience, the amount of contact both providers and parents spend on training materials, and motivational strategies for parents, among others.Trial registrationRegistry of Efficacy and Effectiveness Studies: #4360, registered 1xx, October, 2020 – Retrospectively registered, https://sreereg.icpsr.umich.edu/sreereg/

Synergies Between Low Carbon Technologies in a Large-Scale MV/LV Distribution System

Power distribution systems are experiencing large deployment of behind-the-meter distributed generation and storage along with electrified transport and heat, requiring fundamental changes in planning and operation. There are increasing efforts to directly model real large-scale networks and conduct detailed analysis beyond conventional practices. Utilizing optimization methods to leverage distributed generation to the benefit of the distribution system and all customers is the ideal. However, progress towards adoption of such controls by system operators are impeded by concerns over privacy, regulatory and market issues in many jurisdictions. While these issues are being addressed, there may be opportunity to exploit the synergistic effect of mixing low carbon technologies (LCT). This paper presents a large-scale medium voltage (MV)-low voltage (LV) integrated system model building, scenario determination and analysis approach for distribution systems. Data with different formats from several databases are used in model building. In collaboration with the national distribution system operator (DSO) in Ireland (ESB Networks), a pilot study is conducted for a rural system in the Southwest of Ireland, highlighting the challenges of directly modelling real distribution systems and investigating the potential synergies between multiple low carbon technologies.

High-Resolution Velocity Model Building Based on Common-Source Migration Images and Convolutional Neural Networks

Building a reliable velocity model plays a vital role in seismic imaging and quantitative reservoir description. However, the current data-driven deep-learning-based velocity model building (VMB) approaches directly reconstruct the velocity model of the subsurface from prestack seismic recordings, which are very sensitive to the noise and amplitude mismatch in the data domain. In this letter, we propose a novel VMB approach based on common-source migration image gathers (CSMIGs) and convolutional neural networks (CNNs). The proposed CNN architecture uses the CSMIGs reconstructed by the reverse time migration approach and migration velocity model as the input data. It aims to capture the nonlinear relationship between the amplitude and phase information of CSMIGs and the optimal subsurface reflectivity model. Trained with realistic subsurface models, it can determine that the VMB approach is a computationally efficient solution for a high-resolution velocity reconstruction. In addition, the proposed approach has a better reconstruction performance, antinoise ability, and can be generalized much more easily than the data-driven VMB approach.

Assessing Differential Effects of Somatic Amplification to Positive Affect in Midlife and Late Adulthood-A Regression Mixture Approach.

This study aims to provide an empirical demonstration of a novel method, regression mixture model, by examining differential effects of somatic amplification to positive affect and identifying the predictors that contribute to the differential effects. Data derived from the second wave of Midlife in the United States. The analytic sample consisted of 1,766 adults aged from 33 to 84 years. Regression mixture models were fitted using Mplus 7.4, and a two-step model-building approach was adopted. Three latent groups were identified consisting of a maladaptive (32.1%), a vulnerable (62.5%), and a resilient (5.4%) group. Six covariates (i.e., age, education level, positive relations with others, purpose in life, depressive symptoms, and physical health) significantly predicted the latent class membership in the regression mixture model. The study demonstrated the regression mixture model to be a flexible and efficient statistical tool in assessing individual differences in response to adversity and identifying resilience factors, which contributes to aging research.

Self-validated ensemble models for design of experiments

One of the possible objectives when designing experiments is to build or formulate a model for predicting future observations. When the primary objective is prediction, some typical approaches in the planning phase are to use well-established small-sample experimental designs in the design phase (e.g., Definitive Screening Designs) and to construct predictive models using widely used model selection algorithms such as LASSO. These design and analytic strategies, however, do not guarantee high prediction performance, partly due to the small sample sizes that prevent partitioning the data into training and validation sets, a strategy that is commonly used in machine learning models to improve out-of-sample prediction. In this work, we propose a novel framework for building high-performance predictive models from experimental data that capitalizes on the advantage of having both training and validation sets. However, instead of partitioning the data into two mutually exclusive subsets, we propose a weighting scheme based on the fractional random weight bootstrap that emulates data partitioning by assigning anti-correlated training and validation weights to each observation. The proposed methodology, called Self-Validated Ensemble Modeling (SVEM), proceeds in the spirit of bagging so that it iterates through bootstraps of anti-correlated weights and fitted models, with the final SVEM model being the average of the bootstrapped models. We investigate the performance of the SVEM algorithm with several model-building approaches such as stepwise regression, Lasso, and the Dantzig selector. Finally, through simulation and case studies, we show that SVEM generally generates models with better prediction performance in comparison to one-shot model selection approaches.

Establishing a job-crafting and intrapreneurial link: Towards new approaches to innovative work

Orientation: This research study focuses on establishing a link between job crafting and landmark studies on intrapreneurship.Research purpose: The purpose of this study was to provide a theoretical overview of intrapreneurship, intrapreneurial orientation and job crafting, and to explore theoretical linkages between these areas of enquiry.Motivation for the study: There is currently a dearth of research studies that explore the link between job crafting and intrapreneurial behaviours in existing organisations in the form of intrapreneurial orientation.Research design, approach and method: The study is presented as a conceptual paper in the form of a qualitative, theoretical study, employing a model-building approach. A deductive research approach is followed, and a narrative review methodology is employed.Main findings: The findings of this study from a literature search acknowledge the contributions of job crafting and intrapreneurial research within the management sciences, and we remain cognisant of the organisational implications of each, which have, to date, focused on the organisation, rather than the individual. With this in mind, we suggest that job crafting and intrapreneurial behaviours are empirically researched to validate the recommendations made.Practical/managerial implications: This study will help to establish the type of job-crafting interventions and job-crafting strategies needed to promote intrapreneurial behaviours in practice.Contribution/value-add: This study provides noteworthy insights, which include the suggestion that employees with a forward-looking disposition will engage in job crafting, with a focus on intrapreneurial behaviour. Furthermore, the study fills a void left in the current body of knowledge.

Food-Seeking Behaviors and Food Insecurity Risk During the Coronavirus Disease 2019 Pandemic

ObjectiveFood insecurity risk increases among disaster-struck individuals. The authors employed the social determinants of health framework to (1) describe the characteristics and food-seeking behaviors of individuals coping with the coronavirus disease 2019 pandemic and (2) evaluate the relationship between these factors and food insecurity risk. DesignA cross-sectional Qualtrics survey was administered May 14–June 8, 2020. ParticipantsAdults living in New York were recruited online (n = 410). Main Outcome MeasureFood insecurity risk. AnalysisLogistic regression analyses were conducted using a model-building approach. ResultsA total of 38.5% of the sample was considered food insecure after the coronavirus disease 2019 outbreak. The final model revealed that not knowing where to find help to acquire food, reporting that more food assistance program benefits would be helpful, being an essential worker, having general anxiety, and being a college student were risk factors for food insecurity regardless of demographic characteristics. Conclusions and ImplicationsWith more individuals experiencing food insecurity for the first time, there is a need for enhanced outreach and support. The findings complement emerging research on food insecurity risk during and after the pandemic and can help to inform food assistance programs and policies.

Uncovering causes of stagnating product sales of a healthy snack: A system dynamics group model building project in a food processing company

Product life cycle (PLC) analysis studies patterns of sales over time and the reasons behind those patterns. Such analysis can be challenging because of the interconnected effect of technical product characteristics, marketing mix elements, and the environment (e.g. competition, consumers). Moreover, employees from different functions have different backgrounds and perceptions about the origins of patterns, which constrains mutual understanding and hampers decision-making. The group model building (GMB) approach can be an effective method to support teams in solving complex dynamic problems, such as a PLC analysis of product sales. This study uses a GMB approach and presents a system dynamics model, which supported cross-functional collaboration in analysing causes of stagnating product sales in the PLC of a healthy snack product in a food processing company. Through multiple GMB sessions with the company team, a system dynamics model was developed, and several strategies were analysed. The model has proven useful in explaining the reasons behind the sales stagnation and in studying possible interventions. Moreover, GMB was successful in increasing participants’ insight into the causes of the problem, improving communication, and creating a shared vision about the problem.

Development and Evaluation of a Physiologically Based Pharmacokinetic Drug-Disease Model of Propranolol for Suggesting Model Informed Dosing in Liver Cirrhosis Patients.

AimThe study was aimed to understand the underlying causes for the differences in propranolol pharmacokinetics (PK) between healthy and cirrhosis populations by using a systematic whole-body physiologically based pharmacokinetic (PBPK) model-building approach for suggesting model informed propranolol dosing in liver cirrhosis patients with different stages of disease severity.MethodsA whole-body PBPK model was developed by using population simulator PK-Sim® by using reported physicochemical and clinical data for propranolol in healthy and liver cirrhosis populations. The model evaluation was done by visual verification and comparison of PK parameters using their observed/predicted ratios (Robs/pred).ResultsThe developed model has effectively described the disposition of propranolol after intravenous and oral application in healthy and liver cirrhosis populations. All the model predictions were comparable to the observed clinical data and the Robs/pred for all the PK parameters were within a 2-fold range. A significant increase in plasma concentration of propranolol and decrease in drug clearance was observed in progressive stages of liver cirrhosis. The developed model after evaluation with the reported clinical PK data was used for suggesting model informed propranolol dosing in different stages of liver cirrhosis based on systemic unbound drug concentration.ConclusionThe developed PBPK model has successfully described propranolol PK in healthy and cirrhosis populations after IV and oral administration. The evaluated PBPK propranolol-cirrhosis model can have many implications in predicting propranolol dosing in liver cirrhosis patients with different stages of disease severity.

Predictors of Chemotherapy Safe-Handling Precautions and Knowledge Among a Sample of Jordanian Oncology Nurses: A Model-Building Approach.

There are adverse side effects associated with handling chemotherapy; however, using safe-handling precautions can minimize or prevent these potential effects. Despite availability of international guidelines for chemotherapy handling, adherence to precautions is below expectations. This study examined knowledge of safe-handling precautions among a sample of oncology nurses in Jordon. A cross-sectional study was employed that included a convenience sample of 153 oncology nurses. Oncology nurses from two hospitals completed the Chemotherapy Handling Questionnaire. Descriptive analysis, Spearman rank correlation coefficients, and regression analyses were performed to determine the predictors of precaution use when handling hazardous drugs among participants. We observed that age, number of patients for whom the worker administered chemotherapy per day, the number of patients receiving chemotherapy per day in the participant's work unit, nurses' knowledge about safe-handling precautions, perceived risk, perceived barriers, self-efficacy, organization influence/workplace safety climate, conflict of interest, and interpersonal influences were predictors of use of safe-handling precautions (adjusted R2 = .66, p < .001). Several predictors for using safe-handling precautions were identified. Clinically, chemotherapy handling procedures should be evaluated frequently to identify barriers to safe practices and to improve worker safety.

Exploring AdaBoost and Random Forests machine learning approaches for infrared pathology on unbalanced data sets.

The use of infrared spectroscopy to augment decision-making in histopathology is a promising direction for the diagnosis of many disease types. Hyperspectral images of healthy and diseased tissue, generated by infrared spectroscopy, are used to build chemometric models that can provide objective metrics of disease state. It is important to build robust and stable models to provide confidence to the end user. The data used to develop such models can have a variety of characteristics which can pose problems to many model-building approaches. Here we have compared the performance of two machine learning algorithms - AdaBoost and Random Forests - on a variety of non-uniform data sets. Using samples of breast cancer tissue, we devised a range of training data capable of describing the problem space. Models were constructed from these training sets and their characteristics compared. In terms of separating infrared spectra of cancerous epithelium tissue from normal-associated tissue on the tissue microarray, both AdaBoost and Random Forests algorithms were shown to give excellent classification performance (over 95% accuracy) in this study. AdaBoost models were more robust when datasets with large imbalance were provided. The outcomes of this work are a measure of classification accuracy as a function of training data available, and a clear recommendation for choice of machine learning approach.

Topoisomerase II\u03b2 targets DNA crossovers formed between distant homologous sites to induce chromatin opening

Type II DNA topoisomerases (topo II) flip the spatial positions of two DNA duplexes, called G- and T- segments, by a cleavage-passage-resealing mechanism. In living cells, these DNA segments can be derived from distant sites on the same chromosome. Due to lack of proper methodology, however, no direct evidence has been described so far. The beta isoform of topo II (topo IIβ) is essential for transcriptional regulation of genes expressed in the final stage of neuronal differentiation. Here we devise a genome-wide mapping technique (eTIP-seq) for topo IIβ target sites that can measure the genomic distance between G- and T-segments. It revealed that the enzyme operates in two distinctive modes, termed proximal strand passage (PSP) and distal strand passage (DSP). PSP sites are concentrated around transcription start sites, whereas DSP sites are heavily clustered in small number of hotspots. While PSP represent the conventional topo II targets that remove local torsional stresses, DSP sites have not been described previously. Most remarkably, DSP is driven by the pairing between homologous sequences or repeats located in a large distance. A model-building approach suggested that topo IIβ acts on crossovers to unknot the intertwined DSP sites, leading to chromatin decondensation.

The KK principle and rotational symmetry

The KK principle and rotational symmetry

204 - Social Health and Dementia

Mapping the complexity of factors influencing cognitive functioning in dementia with a special focus on social healthKarin Wolf-Ostermann, Prof. Dr., wolf-ostermann@uni-bremen.de, Health Sciences Bremen, University of Bremen, Germany.Henrik Wiegelmann, M.A., Health Sciences Bremen, University of Bremen, Germany, hwiegelmann@uni-bremen.deImke Seifert, M.A, Health Sciences Bremen, University of Bremen, Germany, imke.seifert@uni-bremen.deDorota Szczesniak, PhD, Department of Psychiatry, Wroclaw Medical University, Poland, dorota.szczesniak@umed.wroc.plMarta Lenart, Department of Psychiatry, Wroclaw Medical University, Poland, marta.lenart@umed.wroc.plMateusz Luc, Department of Psychiatry, Wroclaw Medical University, Poland, mateusz.luc@student.umed.wroc.plMarcin Pawlowski, Department of Psychiatry, Wroclaw Medical University, Poland, marcin.pawlowski@student.umed.wroc.plEtienne Rouwette, Prof. Dr., Radboud University, Nijmegen, The Netherlands, e.rouwette@fm.ru.nlAnsgar Gerhardus, Prof. Dr., Health Sciences Bremen, University of Bremen, Germany, ansgar.gerhardus@uni-bremen.deJoanna Rymaszewska, Prof. Dr., Department of Psychiatry, Wroclaw Medical University, Poland, joanna.rymaszewska@umed.wroc.pl and on behalf of the SHARED ConsortiumIntroduction:Over the past years the scientific discourse on health moved beyond unidimensional approaches, focusing on a more complex bio-psycho-social understanding. Regarding dementia and cognitive health, several studies have shown that various multidimensional factors (risk/protective) and their interactions, as well as existing individual and social resources, contribute to the heterogeneity observed for onset and further course of dementia. This presentation will a) give a systematic overview of causal factors in relation to the onset and progression of cognitive functioning and dementia b) show results of a first generic system dynamics model as Causal Loop Diagram (CLD) visualizing how different factors of the system are interrelated.Methods:A mixed methods approach was used. First, we conducted a systematic literature review on factors influencing cognitive health with an emphasis on social factors. Second, the Group Model Building (GMB) approach was used to further elaborate the knowledge base from the literature review and to start building a Causal Loop Diagram as a first comprehensive system dynamics working model. In a final third phase we integrated both strands, agreed on the knowledge base by stepwise discursive consensus and created the CLD.Results:The CLD model includes 73 unique factors directly or indirectly influencing cognitive functioning. Those factors show the basic causal structure of the interplay of variables present in the development and trajectory of dementia. These factors could be grouped into six thematic/disciplinary clusters, such as personal factors, socioeconomic factors, lifestyle factors, neuro -bio-medical factors, environmental factors and social health factors. The model indicates a significant role of social health indicators for cognitive health in dementia.Conclusion:The generic CLD model reflects the knowledge of a multidisciplinary group of researchers, merged with results from a systematic literature overview and supplemented by discussion and iterative feedback processes. The mixed methods approach as well as the integration of system thinking methods turned out to be a reasonable approach to develop and graphically represent the complex structure of factors influencing cognitive functioning in dementia. This model facilitates the development of novel hypotheses about causal relationships between social health and dementia.

Deciphering the link between procurement systems and sustainable building success in Hong Kong

Sustainable buildings are regarded as one of the most effective approaches to saving resources, reducing greenhouse gas emissions and supporting sustainable development. In the absence of prior research, Delphi as a group model-building approach for building a system dynamics model is appropriate for capturing the knowledge of construction professionals to identify the relationship between procurement systems and sustainable building success by facilitating efficient group communication. The concepts of the Delphi method were reviewed. Two rounds of Delphi questionnaires were carried out to identify the causal relationships between procurement system variables, critical success factors and success criteria. A causal-loop diagram was built based on the results of the two rounds of Delphi study; this will serve as a basis for building a system dynamics model of sustainable building success in a future study. The results show that some critical success factors related to sustainable building in Hong Kong include ‘market planning/business case for sustainability as sustainable objectives’; ‘top management commitment and support’; ‘sustainable strategic planning’; ‘green timing: early evaluation and adoption of sustainability considerations’; and ‘training and education’.

Structural interpretation of cryo-EM image reconstructions

The productivity of single-particle cryo-EM as a structure determination method has rapidly increased as many novel biological structures are being elucidated. The ultimate result of the cryo-EM experiment is an atomic model that should faithfully represent the computed image reconstruction. Although the principal approach of atomic model building and refinement from maps resembles that of the X-ray crystallographic methods, there are important differences due to the unique properties resulting from the 3D image reconstructions. In this review, we discuss the practiced work-flow from the cryo-EM image reconstruction to the atomic model. We give an overview of (i) resolution determination methods in cryo-EM including local and directional resolution variation, (ii) cryo-EM map contrast optimization including complementary map types that can help in identifying ambiguous density features, (iii) atomic model building and (iv) refinement in various resolution regimes including (v) their validation and (vi) discuss differences between X-ray and cryo-EM maps. Based on the methods originally developed for X-ray crystallography, the path from 3D image reconstruction to atomic coordinates has become an integral and important part of the cryo-EM structure determination work-flow that routinely delivers atomic models.

An Examination of Ethnic-Racial Identity and U.S. American Identity among Black, Latino, and White Adolescents

ABSTRACT In the United States (U.S.), adolescent identity development occurs within a socio-historical context characterized by an ethnic-racial hierarchy, as well as an unequal distribution of power and privilege. The current study examined the associations among two ethnic-racial identity components (i.e., exploration, resolution), perceived ethnic-racial discrimination, and U.S. American identity among White, Black, and Latino adolescents in the U.S. A cross-sectional sample of self-identified White, Black, and Latino adolescents (N = 1,378) completed self-administered surveys (M age = 16.16 years; SD = 1.12, 52.8% female). A sequential model-building approach using multiple-group path analysis revealed that both exploration and resolution were positively associated with U.S. American for White and Black adolescents, but no association emerged for Latino adolescents. Furthermore, among Black adolescents, the association between exploration and U.S. American identity was moderated by perceived ethnic-racial discrimination, such that these two identity dimensions were positively associated only among those who reported higher levels of discrimination. Overall, ethnic-racial identity and U.S. American identity were more strongly associated among White and Black adolescents compared to Latino adolescents. Future research is needed to better understand the intersections between ethnic-racial identity and U.S. American identity.

0376 Demographic and Socioeconomic Implications of Excessive Daytime Sleepiness in the Community

Abstract Introduction Daytime sleepiness impairs daily functioning and may be directly related to insufficient nighttime sleep. Previous studies have assessed disparities in sleep duration and quality, but community-level disparities in daytime sleepiness using validated measures are lacking. Methods Data were from the Sleep and Healthy Activity, Diet, Environment, and Socialization (SHADES) study of N=1007 adults age 22-60. Daytime sleepiness was assessed with the Epworth Sleepiness Scale (ESS). Predictors included age, sex, race/ethnicity, education, and social class (“Upper middle class or above,” “Middle class,” “Lower middle class,” “Poor,” or “Very Poor”). One-way ANOVAs evaluated group differences. Stepwise linear modeling evaluated ESS score relative to sociodemographic predictors. Final models included all variables entered together to evaluate independent effects. Finally, habitual sleep duration was entered as an additional covariate. Results ESS score was higher among racial/ethnic minorities (p=0.0006), men (p&amp;lt;0.0001), those with less education (p=0.008) and lower social class (p=0.0007), and those who are retired or unable to work (p=0.03); marginal differences were seen according to age (p=0.06). Using a model-building approach, age, sex, race/ethnicity, education, social class, and employment were evaluated. Only race/ethnicity (F=5.1, p=0.0004), education (F=4.8, p=0.003), and social class (F=2.14, p=0.046) incrementally added variance to model R2. No 2-way interactions were found. In the final model, significant predictors included Black/African-American race/ethnicity (B=0.94, p=0.01), some college (B=0.99, p=0.005), and being very poor (B=2.16, p=0.005). When controlling for nocturnal sleep duration, the increased sleepiness associated with being Black/African was attenuated (p=0.06), but the other relationships were still significant. Conclusion There is a “sleepiness disparity” in the population associated with race/ethnicity and socioeconomics. Daytime sleepiness in the community is associated with being Black/African-American, having some college, and being “very poor.” The race/ethnicity difference in daytime sleepiness may be partially explained by differences in total sleep time. Support This work was supported by a grant from Jazz Pharmaceuticals. The SHADES study was funded by R21ES022931. Dr. Grandner is supported by R01MD011600.

Trauma-related drinking to cope: A novel approach to the self-medication model.

Multiple etiological models have been proposed to explain posttraumatic stress disorder (PTSD) and alcohol use disorder (AUD) comorbidity, but the predominant model to date is the "drinking to cope" self-medication model. Despite its popularity, the self-medication model lacks rigorous empirical support related to inconsistencies and methodological limitations, particularly the failure to operationalize drinking to cope with trauma symptoms specifically. The present study sought to measure trauma-related drinking to cope (TRD) in order to provide a more specific test of the self-medication model among a representative sample of 1,896 undergraduates with a history of trauma exposure and alcohol use. Using a model-building approach in Mplus, a correlated multiple mediator model tested the association between PTSD symptoms and alcohol use problems (AUPs) through TRD and more generalized drinking to cope motives (as assessed by the coping subscale of the Drinking Motives Questionnaire; DMQ-Cope) as moderated by sex. Results indicated that, while accounting for the effects of generalized drinking to cope motives, TRD partially mediated the relation between PTSD symptoms and AUPs and that this relationship was stronger for males than for females. With the exception of moderation by sex, results were substantiated using longitudinal data. Findings were consistent with the self-medication model, suggesting that TRD motives may serve as a mechanism through which PTSD symptoms influence AUPs. TRD may serve as a more specific screening tool for AUP risk among individuals endorsing PTSD symptoms compared with the commonly used DMQ-Cope. (PsycInfo Database Record (c) 2020 APA, all rights reserved).

Assessing the efficacy of group model building workshops in an applied setting through purposive text analysis

AbstractGroup model building (GMB) approaches have been shown to improve participants' understanding of complexity by shifting and aligning individuals' mental models of the interconnections within complex systems. However, reviews of GMB literature have identified knowledge gaps for assessing the efficacy of GMB activities. To address these gaps, these studies recommend assessing multiple cases, shifting from controlled to applied settings, and reporting on objective measures. We address each of these items by comparing the outputs of multiple community‐based GMB workshops to participants' mental models elicited through pre‐workshop interviews. Using purposive text analysis, we developed causal loop diagrams for comparison to a group workshop model. Through a quantitative analysis, we find that individuals convened in GMB workshops have greater alignment on factors, causal links, and feedback. We believe these contributions can help other GMB practitioners better assess the efficacy of their activities with more rigor and detail.© 2020 System Dynamics Society