Through the review of creativity research I have found that studies lack certain crucial parts: a) a theoretical framework for the study of creativity in science, b) studies considering the unique components related to scientific creativity, and c) studies of the interactions among key components through simultaneous analyses. The primary purpose of this study is to explore the dynamic interactions among four components (scientific proficiency, intrinsic motivation, creative competence, context supporting creativity) related to scientific creativity under the framework of scientific creativity. A total of 295 Korean middle school students participated. Well-known and commonly used measurements were selected and developed. Two scientific achievement scores and one score measured by performance-based assessment were used to measure student scientific knowledge/inquiry skills. Six items selected from the study of Lederman, Abd-El-Khalick, Bell, and Schwartz (2002) were used to assess how well students understand the nature of science. Five items were selected from the subscale of the scientific attitude inventory version II (Moore & Foy, 1997) to assess student attitude toward science. The Test of Creative Thinking-Drawing Production (Urban & Jellen, 1996) was used to measure creative competence. Eight items chosen from the 15 items of the Work Preference Inventory (1994) were applied to measure students' intrinsic motivation. To assess the level of context supporting creativity, eight items were adapted from measurement of the work environment (Amabile, Conti, Coon, Lazenby, and Herron, 1996). To assess scientific creativity, one open-ended science problem was used and three raters rated the level of scientific creativity through the Consensual Assessment Technique (Amabile, 1996). The results show that scientific proficiency and creative competence correlates with scientific creativity. Intrinsic motivation and context components do not predict scientific creativity. The strength of relationships between scientific proficiency and scientific creativity (estimate parameter=0.43) and creative competence and scientific creativity (estimate parameter=0.17) are similar [Δx²(.05)(1)=0.670, P > .05]. In specific analysis of structural model, I found that creative competence and scientific proficiency play a role of partial mediators among three components (general creativity, scientific proficiency, and scientific creativity). The moderate effects of intrinsic motivation and context component were investigated, but the moderation effects were not found.

In their exploratory study Muniz & O’Guinn (2001) found three markers of a brand community: a sense of belonging, rituals and tradition and a sense of duty toward the community. Two of the three markers of community have been included in conceptual models on brand communities. However, the third marker (sense of duty) has not been implemented up to now. Hence, the objective of this thesis is to extend Bagozzi & Dholakia’s (2006) brand community model by incorporating the construct “sense of duty”. In this research, a conceptual model of brand communities is developed. Overall, the findings support the conceptual model. The results show that sense of duty is a decisive mediator of brand community behaviours and that sense of duty is divided into three distinct components: new member integration, product usage and member retention. Further, this research indicates that community-related behavioural intentions are not significantly related to purchase intentions.

Structural equation models (SEMs) are commonly used in behavioral, educational, medical, and social sciences. Lots of software, such as EQS, LISREL, MPlus, and WinBUGS, can be used for the analysis of SEMs. Also many methods have been developed to analyze SEMs. One popular method is the Bayesian approach. An important issue in the Bayesian analysis of SEMs is model selection. In the literature, Bayes factor and deviance information criterion (DIC) are commonly used statistics for Bayesian model selection. However, as commented in Chen et al. (2004), Bayes factor relies on posterior model probabilities, in which proper prior distributions are needed. And specifying prior distributions for all models under consideration is usually a challenging task, in particular when the model space is large. In addition, it is well known that Bayes factor and posterior model probability are generally sensitive to the choice of the prior distributions of the parameters. Furthermore the computational burden of Bayes factor is heavy. Alternatively, criterion-based methods are attractive in the sense that they do not require proper prior distributions in general, and the computation is quite simple. One of commonly used criterion-based methods is DIC, which however assumes the posterior mean to be a good estimator. For some models like the mixture SEMs, WinBUGS does not provide the DIC values. Moreover, if the difference in DIC values is small, only reporting the model with the smallest DIC value may be misleading. In this thesis, motivated by the above limitations of the Bayes factor and DIC, a Bayesian model selection criterion called the Lv measure is considered. It is a combination of the posterior predictive variance and bias, and can be viewed as a Bayesian goodness-of-fit statistic. The calibration distribution of the Lv measure, defined as the prior predictive distribution of the difference between the Lv measures of the candidate model and the criterion minimizing model, is discussed to help understanding the Lv measure in detail. The computation of the Lv measure is quite simple, and the performance is satisfactory. Thus, it is an attractive model selection statistic. In this thesis, the application of the Lv measure to various kinds of SEMs will be studied, and some illustrative examples will be conducted to evaluate the performance of the Lv measure for model selection of SEMs. To compare different model selection methods, Bayes factor and DIC will also be computed. Moreover, different prior inputs and sample sizes are considered to check the impact of the prior information and sample size on the performance of the Lv measure. In this thesis, when the performances of two models are similar, the simpler one is selected. / Li, Yunxian. / Adviser: Song Xinyuan. / Source: Dissertation Abstracts International, Volume: 72-04, Section: B, page: . / Thesis (Ph.D.)--Chinese University of Hong Kong, 2010. / Includes bibliographical references (leaves 116-122). / Electronic reproduction. Hong Kong : Chinese University of Hong Kong, [2012] System requirements: Adobe Acrobat Reader. Available via World Wide Web. / Electronic reproduction. Ann Arbor, MI : ProQuest Information and Learning Company, [200-] System requirements: Adobe Acrobat Reader. Available via World Wide Web. / Abstract also in Chinese.

A real empirical study was conducted to illustrate the application of the unconstrained approach. The genuine study focused on the interaction effect of music self-concept and music-domain importance on the global self-concept. The result showed that structural equation analysis was advantageous over the traditional regression analysis, while such superiority of the structural equation modeling approach was more prominent in higher-order structural equation models. As expected, the estimated main and interaction effects in the quasi-standardized solution obtained using the centered data coincided with those using the standardized data in different order structural models. / Concepts and issues related to standardized solutions for the latent interaction model were discussed. Quasi-standardized solution was proposed and formulated by using the estimates of the original solution and the ordinary standardized solution. Some properties of the quasi-standardized solution were mathematically derived and proved, which included the demonstration that the main and interaction effects were scale free, so were the loading and the Chi-square of model fit, while t statistics of main and interaction effects were approximate scale free. / Six simulation studies, four for the latent interaction models, and two for the latent quadratic models were conducted to compare the performances of the four approaches. Results generally showed that the QML approach and the constrained approach behaved similarly, while the performance of the unconstrained approach was close to that of the GAPI approach. Under the normal distribution condition, the QML approach performed the best among the four approaches in terms of lack of bias, precision, and power. However, with moderate and large sample sizes (N=200 or above), the differences among the four approaches were systematically smaller, with similar bias and precision. Under nonnormal conditions, the unconstrained approach was more robust, with a smaller bias and predictable type I error rate (near the significant level), and its precision and power increased as the sample size increased. These results supported the use of the unconstrained approach for the analyses of latent interaction and quadratic models. / The unconstrained approach was extended to estimate interaction effects in latent growth models. With the indicators of the interaction term formed by the products of differences (rather than using the usual indicator product strategy), a simplified full interaction model for the latent growth model was proposed. The model was further simplified when only the interaction between change rates was considered. Importantly, the unconstrained approach was an appropriate method for analyses of such simplified full interaction model for latent growth model, which also constituted a unique contribution of this dissertation. / Through a series of related studies, the research attempted to identify better estimation approaches and modeling techniques for latent interaction and quadratic effects. The literature review provided a conceptual framework for the unconstrained approach which was recommended for its simplicity and robustness. Three other approaches, namely, the constrained, the partially constrained (i.e., the generalized appended product indicator, GAPI), and the quasi-maximum likelihood (QML) approaches were selected and compared with the unconstrained approach. / 溫忠粦. / 論文(哲學博士)--香港中文大學, 2005. / 參考文獻(p. 191-203). / Adviser: Kit-Tai Hau. / Source: Dissertation Abstracts International, Volume: 67-01, Section: A, page: 0160. / Electronic reproduction. Hong Kong : Chinese University of Hong Kong, [2012] System requirements: Adobe Acrobat Reader. Available via World Wide Web. / Electronic reproduction. [Ann Arbor, MI] : ProQuest Information and Learning, [200-] System requirements: Adobe Acrobat Reader. Available via World Wide Web. / Abstracts also in English. / School code: 1307. / Lun wen (Zhe xue bo shi)--Xianggang Zhong wen da xue, 2005. / Can kao wen xian (p. 191-203). / Wen Zhonglin.

Keywords: Bayesian analysis, Finite mixture, Gibbs sampler, Langevin-Hasting sampler, MH sampler, Model comparison, Nonrecursive nonlinear structural equation model, Path sampling. / Structural equation models (SEMs) have been applied extensively to management, marketing, behavioral, and social sciences, etc for studying relationships among manifest and latent variables. Motivated by more complex data structures appeared in various fields, more complicated models have been recently developed. For the developments of SEMs, there is a usual assumption about the regression coefficient of the underlying latent variables. On themselves, more specifically, it is generally assumed that the structural equation modeling is recursive. However, in practice, nonrecursive SEMs are not uncommon. Thus, this fundamental assumption is not always appropriate. / The main objective of this thesis is to relax this assumption by developing some efficient procedures for some complex nonrecursive nonlinear SEMs (NNSEMs). The work in the thesis is based on Bayesian statistical analysis for NNSEMs. The first chapter introduces some background knowledge about NNSEMs. In chapter 2, Bayesian estimates of NNSEMs are given, then some statistical analysis topics such as standard error, model comparison, etc are discussed. In chapter 3, we develop an efficient hybrid MCMC algorithm to obtain Bayesian estimates for NNSEMs with mixed continuous and ordered categorical data. Also, some statistical analysis topics are discussed. In chapter 4, finite mixture NNSEMs are analyzed with the Bayesian approach. The newly developed methodologies are all illustrated with simulation studies and real examples. At last, some conclusion and discussions are included in Chapter 5. / Li, Yong. / "July 2007." / Adviser: Sik-yum Lee. / Source: Dissertation Abstracts International, Volume: 69-01, Section: B, page: 0398. / Thesis (Ph.D.)--Chinese University of Hong Kong, 2007. / Includes bibliographical references (p. 99-111). / Electronic reproduction. Hong Kong : Chinese University of Hong Kong, [2012] System requirements: Adobe Acrobat Reader. Available via World Wide Web. / Electronic reproduction. [Ann Arbor, MI] : ProQuest Information and Learning, [200-] System requirements: Adobe Acrobat Reader. Available via World Wide Web. / Abstracts in English and Chinese. / School code: 1307.

Submitted by Inácio de Oliveira Lima Neto (inacio.neto@inpa.gov.br) on 2018-10-01T14:38:01Z No. of bitstreams: 2 João_Gabriel_Rafaelli.pdf: 2157835 bytes, checksum: 8d523aee7395e3c81de2736e95f7010b (MD5) license_rdf: 0 bytes, checksum: d41d8cd98f00b204e9800998ecf8427e (MD5) / Made available in DSpace on 2018-10-01T14:38:02Z (GMT). No. of bitstreams: 2 João_Gabriel_Rafaelli.pdf: 2157835 bytes, checksum: 8d523aee7395e3c81de2736e95f7010b (MD5) license_rdf: 0 bytes, checksum: d41d8cd98f00b204e9800998ecf8427e (MD5) Previous issue date: 2018-03-02 / Fundação de Amparo à Pesquisa do Estado do Amazonas - FAPEAM / Agroforestry system is a land use technique practiced a long time around the world. Currently, an attention has been paid to this practice, with several projects and organizations working with the optimization of this technique. Among many ways of establishing an agroforestry system, taking into account the species to be used, the environmental conditions and the type of system to be implemented; the understanding of the specific factors that can influence the plantations are of utmost importance for its success. The present work aimed to evaluate the initial development of seedlings from 16 species planted in different lands, influenced by environmental factors and management of local farmers. The study was carried out at the Andirá-Marau Indigenous Reserve (Amazonas - Brazil) in eight different plantations with different agricultural practices. The parameters of plant performance analyzed were: Carbon Stock (CARBON), Absolute Diameter Increment (ADI), Relative Growth Rate (RGR) and Specific Leaf Area (SLA). As factors influencing the performance of seedlings were analyzed: Soil Quality (chemical and physical), percentage of Vegetation Cover (VC), Forest Distance (FD), Above Ground Biomass (AGB), Species Richness (SP_RICH), Litter Biomass, Charcoal Biomass, Nearest Next Neighbor and Competition Index. As a descriptive analysis, the soils were analyzed using ANOVA (two-factor) testing soil depths and areas. Also, the plantations were analyzed in terms of: area size, seedling survival, species composition, spacing, height and biomass. As an exploratory data analysis we used linear regressions between each performance trait and each influence factor. Later, we used the Structural Equation Modeling analysis to test how the factors together influence the performance of the seedlings. Finally, to test how the general models for all species influence in a specie-specific way, we tested models for Carapa guianensis Aubl. As results, it can be observed that during the project 45% of the seedlings died (9% between Dec 2014 and Dec 2015, 32% between Dec 2015 and Aug 2016, 11% between Aug 2016 and Feb 2017), mainly due to large dry season on the 2nd semester. Ingá, Urucum, Andiroba and Graviola species showed great variability in biomass accumulation, while Acerola, Cumaru, Guaraná and Mahogany varied much less. The exploratory analysis showed that RGR was positively influenced by FD and soil nutrients Mn, Ca, Mg, K and CPB and negatively by Fe; the Carbon Stock was positively influenced by FD, Coal Biomass, Nearby Neighbor, by soil nutrients Al, Mg, K, C, N, CEC and negatively by soil Density and Fe; ADI was positively influenced by FD, Neighbor Next, by nutrients Al, Mg, K, C, N, CEC and negatively by soil Density and Fe; SLA was positively influenced by VC and AGB. In the general models for all species, in SEM, we could observe that Carbon Stock had a 14% variation explained by the model, ADI had 14%, RGR had 11% and SLA had 10%. For the specific model (Carapa guianensis) the percentage of variation explained by the models was: Carbon Stock with 31%, ADI with 27%, RGR with 10% and SLA with 71%. We can conclude that soil factors (C, N, P, Mg, CEC, pH, Texture and Density), Biomass of Serrapilheira, VC and Next Neighbor, had greater influence on the initial performance of the seedlings planted in different agroforestry systems, which have to be taken into account for management practices. / Sistemas agroflorestais são formas de uso da terra utilizadas por muito tempo ao redor do mundo. Atualmente tem-se prestado maior atenção a essa prática, com diversos projetos e organizações trabalhando com a otimização das técnicas, para um melhor desenvolvimento desses sistemas. Dentre as diversas maneiras de se estabelecer um sistema agroflorestal, levando em consideração as espécies a serem utilizadas, o ambiente em que se encontra e o tipo de sistema a ser implementado, o entendimento dos fatores que possam influenciar os plantios são de suma importância para um maior sucesso desses. O presente trabalho teve como objetivo avaliar o desenvolvimento inicial de mudas de 16 espécies plantadas em diferentes ambientes, pela influência de fatores ambientais e do manejo de agricultores locais. O estudo foi realizado na Terra Indígena Andirá-Marau (Amazonas - Brasil) e contou com oito diferentes plantios em ambientes com diferentes práticas agrícolas. Foram avaliados os parâmetros de desempenho vegetal o Estoque de Carbono (CARBON), o Incremento Absoluto do Diâmetro (ADI), a Taxa de Crescimento Relativo (RGR) e a Área Foliar Específica (SLA). Como fatores de influência no desempenho das mudas foram avaliados: a Qualidade dos Solos (química e física), a porcentagem de Cobertura Vegetal (VC), a Distância dos plantios para a Floresta (FD), Biomassa Acima do Solo (AGB), Riqueza de Espécies (SP_RICH), Biomassa da Serrapilheira, Biomassa de Carvão, Vizinho Próximo das mudas plantadas e Índice de Competição. Como análise descritiva, os solos foram analisados utilizando o teste ANOVA (two-factor) entre as profundidades e a áreas. Também de forma descritiva foram analisados os plantios quanto: sobrevivência das mudas, composição de espécies, espaçamento, altura e biomassa. Para análise dos dados foram utilizadas regressões lineares entre cada medida de desempenho e cada fator de influência, como forma exploratória dos dados. Posteriormente utilizo-se a análise Structural Equation Modeling (SEM) para testar como os fatores influenciam o desempenho das mudas de forma conjunta. Por fim, para testar como os modelos gerais para todas as espécies influenciam de maneira específica, foi testado modelos para a espécie Carapa guianensis Aubl. Como resultado, pode-se observar que ao longo do projeto 45% das mudas morreram (9% entre Dez 2014 e Dez 2015, 32% entre Dez 2015 e Ago 2016, 11% entre Ago 2016 e Fev 2017), devido principalmente a grande seca no 2o. semestre de 2015. As espécies Ingá, Urucum, Andiroba e Graviola mostraram grande variabilidade no acumulo de biomassa, enquanto Acerola, Cumaru, Guaraná e Mogno variaram muito menos. Já a análise exploratória nos mostrou que o RGR foi influenciado positivamente por FD e pelos nutrientes do solo Mn, Ca, Mg, K e CEC e negativamente por Fe; o Estoque de Carbono foi influenciado positivamente por FD, Biomassa de Carvão, Vizinho Próximo, pelos nutrientes do solo Al, Mg, K, C, N, CEC e negativamente por Densidade do Solo e Fe; o ADI foi influenciado positivamente por FD, Vizinho Próximo, pelos nutrientes Al, Mg, K, C, N, CEC e negativamente por Densidade do Solo e Fe; SLA foi influenciada positivamente por VC e AGB. Nos modelos gerais para todas as espécies, em SEM, vimos que o Estoque de Carbono teve 14% de variação explicada pelo modelo, ADI teve 14%, RGR teve 11% e SLA teve 10%. Para o modelo específico (Carapa guianensis) a porcentagem de variação explicada pelos modelos foram, Estoque de Carbono com 31%, ADI com 27 %, RGR com 10% e SLA com 71%. Podemos concluir que os fatores do solo (C, N, P, Mg, CEC, pH, Textura e Densidade), Biomassa da Serrapilheira, VC e Vizinho Próximo, tiveram maior influência no desempenho inicial das mudas nos sistemas agroflorestais implantados, devendo ser levadas em consideração nas práticas de manejo.

This thesis bridges a gap in extant research by examining key factors that play a role in behavioral grocery shopping routines following minor and major disruptions. The present research involves two interrelated investigations incorporating mixed methodologies (Cresswell, 2003). Study 1 involves semi-structured in-depth interviews seeking to establish how goal-directed grocery shopping routines are developed over time. Utilizing a laddering approach of questioning (Gutman, 1997), respondents are probed on their routines (Brotherton, 2001) and goals, including end goals as described in the List of Values (Kahle & Kennedy, 1988). Three participants were interviewed on three occasions over an eight week period, until theoretical saturation was achieved. A significant contribution of Study 1 lies in the development of a conceptual framework for understanding factors associated with grocery shopping routines. This model reflects a working definition characterizing routines as goal-driven and value-guided heuristic strategies. It is proposed that routines are repetitive patterns of personal and private behavioral activities dependent upon situational and temporal contexts, and utilized for instrumental reasons. Risk-taking attitudes and personal values also shape goal-directed behaviors. Using structural equation modeling (SEM) procedures (Jöreskog & Sörbom, 1993), Study 2, an online experiment, aims to test and build upon the conceptual model emanating from Study 1. This study also investigates the impact of minor and major disruptions on routinized grocery shopping behavior. 612 participants were allocated across three experimental groups: situational contexts, anticipated temporal conditions, and repetitive value. Cohorts were assessed at baseline levels and received unique minor and major disruptions appropriate to their circumstance. Study 2 contributes through the large-scale SEM testing of a model of grocery shopping routinization. Overall, sound structural model fit demonstrates that the present model of grocery shopping routinization is explained by six distinct components including routinized behavior, goal-centeredness, situational contexts, anticipated temporal conditions, repetitive value, and risk-taking attitudes; and three dimensions of personal values: maturity, self-direction/achievement, and enjoyment. In terms of disruptions, findings indicate that routine strength is dependent on degree of situational, temporal, and instrumental interruptions. Disruptions can both facilitate and impede routines. Results demonstrate that regardless of goal stability, routines change when model components are disrupted. Findings suggest theoretical, research, and practical implications. This thesis expands decision making theory (Betsch, Fiedler, & Brinkmann, 1998) by demonstrating that, despite unwavering goals, new contexts arising from disruptions influence new behavioral deliberations. In relation to research implications, this thesis develops then subsequently tests a model of grocery shopping routinization. Despite routines becoming subconscious over time (Aarts & Dijksterhuis, 2000a), this study asserts that routines are intentional and involve goal-directed strategies for dealing with the environment. From an applied perspective, practitioners should be aware that routine-disrupted consumers remain goal-driven. Consumers are unlikely to forego focal goals (e.g., shop for weekly household meals) if these goals are non-negotiable. Present results suggest that consumers esteem maturity-related personal values, such as fostering and maintaining warm relationships with others and sense of belonging, when grocery sho pping.

Abstract The communication and information technology, particularly the Internet, has dramatically changed the way we have done business before. Therefore. Some different work arrangements have emerged in the technology-based business, such as telecommuting, mobile office, hoteling, satellite office, etc. The number of telecommuters in the US today is up to 25 million. In Taiwan, China Productivity Center (CPC), Taiwan Xerox, Taiwan IBM, and Taiwan HP have implemented the mobile office system for several years. Arthur Andersen and Sun also built their flexible office last year. As the literatures indicated that telecommuting did increase organizational flexibility, job efficiency, employee satisfaction, productivity, customer satisfaction, while reduce commuting time and transportation costs, office spaces. However, the teleworkers felt more isolated as a result of working in a remote environment. Their interpersonal relationship and communication with supervisor or co-worker all got worse. Besides, managers also worried about telecommuting will reduce their authority and control power. The main purpose of this study is to understand how Internet affect the individual worker¡¦s work style, especially what are the key factors being considered in telecommuting. Finally, the proposed telecommuting model would be empirically examined in the selected information technology-based organizations.

The purpose of the study was to examine multilevel regression models in the context of multilevel structural equation modeling (SEM) in terms of accuracy of parameter estimates, standard errors, and fit indices in normal and nonnormal data under various sample sizes and differing estimators (maximum likelihood, generalized least squares, and weighted least squares). The finding revealed that the regression coefficients were estimated with little to no bias among the study design conditions investigated. However, the number of clusters (group level) appeared to have the greatest impact on bias among the parameter estimate standard errors at both level-1 and level-2. In small sample sizes (i.e., 300 and 500) the standard errors were negatively biased. When the number of clusters was 30 and cluster size was held at 10, the level-1 standard errors were biased downward by approximately 20% for the maximum likelihood and generalized least squares estimators, while the weighted least squares estimator produced level-1 standard errors that were negatively biased by 25%. Regarding the level-2 standard errors, the level-2 standard errors were biased downward by approximately 24% in nonnormal data, especially when the correlation among variables was fixed at .5 and kurtosis was held constant at 7. In this same setting (30 clusters with cluster size fixed at 10), when kurtosis was fixed at 4 and the correlation among variables was held at .7, both the maximum likelihood and generalized least squares estimators resulted in standard errors that were biased downward by approximately 11%. Regarding fit statistics, negative bias was noted among each of the fit indices investigated when the number of clusters ranged from 30 to 50 and cluster size was fixed at 10. The least amount of bias was associated with the maximum likelihood estimator in each of the data normality conditions examined. As sample size increased, bias decreased to near zero when the sample size was equal to or greater than 1,500 with similar results reported across estimation methods. Recommendations for the substantive researcher are presented and areas of future research are presented.

This dissertation focuses on results of multi-group SEM models estimated using data from the National Longitudinal Study of Adolescent Health (Add Health) in order to determine appropriate measurement and structural models for the relationship between socioeconomic status (SES) and health among six young adult U.S. social groups. Examining the links between SES and health during young adulthood is important because while there is a strong, documented link between lower SES and poorer health (Adler & Snibbe, 2003), young adults can exercise a considerable amount of agency with regard to their own SES and health. Young adults make critical decisions about pursuing post-secondary education, entering the workforce, and practicing healthy behaviors--activities which differ in their immediate and long-term economic and health payoff (Mirowsky & Ross, 2003; Elder, 1985; 1994). Yet, the nature of SES and its links with health for members of various gender and racial/ethnic groups is not entirely clear. Literature suggests that occupation, education, and income are neither defined nor linked among women in the same ways that they are for men (APA, 2007). Self-assessment of health is also thought to differ by gender and ethnicity (Krause & Jay, 1994). Moreover, limited research has addressed the unique mediating pathways by which aspects of SES affect health for specific social groups (Matthews, Gallo, & Taylor, 2010). In this work, I estimate measurement models for several aspects of SES among African American, Latina, and White men and women, then link aspects of SES with each other and with health using structural equation modeling. I also examine the unique mediating pathways by which aspects of SES are linked with health for these groups. / text