Multiple Membership Multilevel Models Research Articles

Can less be more? Effects of study load on science achievement among Chinese eighth-grade students

With students’ study load becoming increasingly controversial around the world, we investigated whether “less can be more” regarding the relationship between study load and science achievement (i.e., lighter study load, higher science achievement). Using data from the 2016 (Chinese) Program for Regional Assessment of Basic Education Quality (N = 40,536 students under 946 science teachers within 118 schools), we constructed seven measures of study load. We developed multiple membership multilevel models to accommodate students being taught by multiple science teachers. After adjustment for student, teacher, and school characteristics, six measure did not indicate any statistically significant relationship between study load and science achievement, with one indicating a statistically significant but negative relationship, leading us to conclude that heavier study load was not associated with higher science achievement. Meanwhile, some school climate variables could function as moderators of the relationship with academic pressure and parental involvement as the highlights. Under the reduction in these variables, lighter study load began to associate with higher science achievement.

A multiple membership multilevel negative binomial model for intersection crash analysis

Many intersections belong to more than one zone, but most research has not considered the effects of multiple zones in intersection crash analysis. This issue is known as a boundary problem. Unobserved heterogeneity between zones can lead to model misspecification which can result in biased parameter estimates and poor model fitting performance. This study investigated the issue using five years of intersection crash data from the City of Regina, Saskatchewan, Canada. The study developed three multiple membership multilevel negative binomial models to reduce unobserved zonal-level heterogeneity. Each multiple membership multilevel model used a different weight strategy. When the fitting performance of the three multiple membership multilevel models was compared with two additional models, a traditional single level model and a conventional multilevel model, all three multiple membership multilevel models had a better fitting performance. Five individual-level and seven group-level variables were statistically significant (90% confidence level) in all the models with five of the individual-level and four of the group-level variables statistically significant at the 99% confidence level. The multiple membership multilevel models also helped to prevent the underestimation of group-level variance and type I statistical errors that tend to occur with single level models and conventional multilevel models. In particular, the three multiple membership multilevel models produced more accurate results for intersections with a large AADT. As intersections with a large AADT are known to have more crashes, multiple membership multilevel models are likely to be more useful than single level models and conventional multilevel models when selecting intersections for safety improvement. The study recommends the adoption of a multiple membership multilevel model to improve fitting performance and reduce the boundary problem for intersections affected by more than one zone.

Students’ and teachers’ critical thinking in science education: are they related to each other and with physics achievement?

ABSTRACT Background Although there is an increasing recognition of the importance of critical thinking in science education across the world, the relevant research literature is extremely thin. There is an urgent need to generate more empirical evidence. Purpose We examined two essential issues concerning critical thinking (the relationship between students’ and teachers’ critical thinking as well as the relationship between critical thinking and science achievement) in the context of physics education. Sample We utilized data from the 2017 (Chinese) Program for Regional Assessment of Basic Education Quality with 46,820 students under 547 teachers from 128 schools. Design and methods We developed multiple membership multilevel models to account for students from different schools and under multiple teachers (in each school). Results After control over student, teacher, and school characteristics, although Chinese eighth graders demonstrated statistically significant critical thinking, there was not any statistically significant relationship between students’ and teachers’ critical thinking. With the same control, there was a statistically significant relationship between students’ critical thinking and students’ physics achievement among Chinese eighth graders. Meanwhile, there was not any statistically significant relationship between teachers’ critical thinking and students’ physics achievement (over and above the relationship between students’ critical thinking and students’ physics achievement). Conclusion We suggest that critical thinking may need to become an essential and explicit outcome of science education to help establish a relationship between students’ and teachers’ critical thinking and promote the correlation between critical thinking of students and teachers and students’ science achievement.

Associations Between Community Social Capital and Preservation of Functional Capacity in the Aftermath of a Major Disaster.

The strength of social connections in the community ("social capital") is hypothesized to be a crucial ingredient in disaster resilience. We examined whether community-level social capital is correlated with the ability to maintain functional capacity among older residents who experienced the 2011 Great East Japan Earthquake and Tsunami. The baseline of our cohort (mean age, 74 years) was established in 2010, 7 months before the disaster in Iwanuma, a Japanese city located 80 km from the epicenter. Disaster-related personal experiences (e.g., housing damage or relocation) were assessed through a follow-up survey (n=3,594; follow-up rate, 82.1%) conducted in 2013, 2.5 years after the earthquake. Multiple membership multilevel models were used to evaluate the associations between functional capacity, measured by the Instrumental Activities of Daily Living scale, and 3 subscales of community-level social capital: social cohesion, social participation, and reciprocity. Community-level social participation was associated with a lower risk of functional decline after disaster exposure. The average level of social participation in the community also mitigated the adverse impact of housing damage on functional status, suggesting a buffering mechanism.

Discrimination as a frame-of-reference effect in overlapping friendship communities of ethnically diverse youth.

To what extent is the frame of reference of overlapping friendship communities important for young people's feelings of discrimination and subjective well-being? That is, do youth feel better or worse to the extent that they feel less or more discrimination than their friends? Participants (N = 898; Mage = 14.13; SDage = 3.37; 46% females; 46% Whites; 20% Indigenous; 34% other minorities) were high school students of three ethnically diverse, low socioeconomic status public schools in New South Wales, Australia. Cross-sectional data were collected to measure felt discrimination, mental health, subjective well-being, social support, and nominations of close friends. A state-of-the-art method of clustering links was used to identify overlapping friendship communities, and multiple membership multilevel models were run to examine whether community-level discrimination moderated the link between individual-level discrimination and well-being. When the community level discrimination was low, there was no well-being related cost or benefit of individual-level discrimination. But when the community-level discrimination was high, individuals in those communities who themselves felt low discrimination had better well-being than individuals who themselves felt high discrimination. We provide evidence for a frame-of-reference effect involving discrimination. Individuals' relative standing in their friendship communities with high group-level discrimination reliably predicted the individuals' well-being levels, regardless of ethnicity. The results highlight the importance of identifying overlapping friendship communities for understanding the dynamics of discrimination and well-being of ethnically diverse youth. (PsycINFO Database Record (c) 2020 APA, all rights reserved).

The Impact of Sample Size in Cross-Classified Multiple Membership Multilevel Models

The Impact of Sample Size in Cross-Classified Multiple Membership Multilevel Models

Assessment of Multiple Membership Multilevel Models: An Application to Interviewer Effects on Nonresponse

ABSTRACTMultilevel multiple membership models account for situations where lower level units are nested within multiple higher level units from the same classification. Not accounting correctly for such multiple membership structures leads to biased results. The use of a multiple membership model requires selection of weights reflecting the hypothesized contribution of each level two unit and their relationship to the level one outcome. The Deviance Information Criterion (DIC) has been proposed to identify such weights. For the case of logistic regression, this study assesses, through simulation, the model identification rates of the DIC to detect the correct multiple membership weights, and the properties of model variance estimators for different weight specifications across a range of scenarios. The study is motivated by analyzing interviewer effects across waves in a longitudinal study. Interviewers can substantially influence the behavior of sample survey respondents, including their decision to participate in the survey. In the case of a longitudinal survey several interviewers may contact sample members to participate across different waves. Multilevel multiple membership models are suitable to account for the inclusion of higher-level random effects for interviewers at various waves, and to assess, for example, the relative importance of previous and current wave interviewers on current wave nonresponse. To illustrate the application, multiple membership models are applied to the UK Family and Children Survey to identify interviewer effects in a longitudinal study. The paper takes a critical view on the substantive interpretation of the model weights and provides practical guidance to statistical modelers. The main recommendation is that it is best to specify the weights in a multiple membership model by exploring different weight specifications based on the DIC, rather than prespecifying the weights.

Using Weighted Hospital Service Area Networks to Explore Variation in Preventable Hospitalization.

ObjectiveTo demonstrate the use of multiple‐membership multilevel models, which analytically structure patients in a weighted network of hospitals, for exploring between‐hospital variation in preventable hospitalizations.Data SourcesCohort of 267,014 people aged over 45 in NSW, Australia.Study DesignPatterns of patient flow were used to create weighted hospital service area networks (weighted‐HSANs) to 79 large public hospitals of admission. Multiple‐membership multilevel models on rates of preventable hospitalization, modeling participants structured within weighted‐HSANs, were contrasted with models clustering on 72 hospital service areas (HSAs) that assigned participants to a discrete geographic region.Data Collection/Extraction MethodsLinked survey and hospital admission data.Principal FindingsBetween‐hospital variation in rates of preventable hospitalization was more than two times greater when modeled using weighted‐HSANs rather than HSAs. Use of weighted‐HSANs permitted identification of small hospitals with particularly high rates of admission and influenced performance ranking of hospitals, particularly those with a broadly distributed patient base. There was no significant association with hospital bed occupancy.ConclusionMultiple‐membership multilevel models can analytically capture information lost on patient attribution when creating discrete health care catchments. Weighted‐HSANs have broad potential application in health services research and can be used across methods for creating patient catchments.

Using weighted hospital service area networks to explore variation in preventable hospitalisation.

ABSTRACT
 ObjectivesMarkets of health care are created in health services research to attribute variation in performance to characteristics of the health system. Defining patient catchments to capture hospital-level variation poses particular difficulties, because many factors other than geography drive choice of hospital. Several methods using linked data have been developed to create patient catchments or ‘hospital service areas’ (HSAs), including patterns of patient flow and networks of patient and physician referrals, yet these discrete catchments often have poor patient loyalty and are unable to attribute variation to specific hospitals. This study sought to demonstrate the use of multiple membership multilevel models, which cluster people in one or more higher-level units (such as multiple hospitals), in exploring between-hospital variation of preventable hospitalisations.
 ApproachLinked hospital data from 267,014 participants in the 45 and Up Study, NSW Australia, with linkage by the NSW CHeReL, were used to create weighed hospital service area networks (HSANs) in which patterns of patient flow to large public hospitals within 593 postal areas were used to create a weighted probability of admission of participants to each hospital. Multiple membership multilevel Poisson models were used to explore variation in rates of preventable hospitalisation, clustering participants in hospitals using a weighted HSAN, and compared with models clustering participants in HSAs based on the most common hospital of admission.
 ResultsThe most common hospital of admission accounted for an average of 67% of all admissions in each postal area. There was significant variation in rates of preventable hospitalisation between all 79 large public hospitals when clustering participants in a weighted HSAN, which was more than twice as large as the variation between the 72 hospitals forming the basis of HSAs. The ranking of hospitals differed between modelling approaches, and the hospital with the highest rate of preventable hospitalisation wasn’t identified when using HSAs. There was no association between hospital bed occupancy rate and preventable hospitalisations when using either modelling approach.
 ConclusionQuantifying variation in health service use and outcomes is the cornerstone of creating accountable health care systems, yet much information is lost in creating discrete health catchments. Multiple membership multilevel models can help capture this uncertainty, and given they can be applied using extensions of current methodology, have potential to be used across a variety of methods for defining and analysing health care catchments.

OP03 Spatial Segregation and Socioeconomic Inequalities in Health in Brazilian Cities: Combining Spatial and Social Epidemiology

BackgroundSpatial and socioeconomic inequalities in health are well-documented across the world, with poorer people, areas and regions experiencing poorer health and higher mortality risks than richer people, areas and regions....

Modelling The Effect of Pupil Mobility on School Differences in Educational Achievement

SummaryThe recently introduced national pupil database in England allows the tracking of every child through the compulsory phases of the state education system. The data from key stage 2 for three local education authorities are studied, following cohorts of pupils through their schooling. The mobility of pupils among schools is studied in detail by using multiple-membership multilevel models that include prior achievement and other predictors and the results are compared with traditional ‘value-added’ approaches that ignore pupil mobility. The analysis also includes a cross-classification of junior and infant schools attended. The results suggest that some existing conclusions about schooling effects may need to be revised.