Evaluation of the Hosmer -Lemeshow global goodness -of-fit statistic for mixed variable logistic regression models

The performance of the Hosmer-Lemeshow global goodness-of-fit statistic for logistic regression models was explored in a wide variety of conditions not previously fully investigated. Computer simulations, each consisting of 500 regression models, were run to assess the statistic in 23 different situations. The items which varied among the situations included the number of observations used in each regression, the number of covariates, the degree of dependence among the covariates, the combinations of continuous and discrete variables, and the generation of the values of the dependent variable for model fit or lack of fit.^ The study found that the $\rm\ C$g* statistic was adequate in tests of significance for most situations. However, when testing data which deviate from a logistic model, the statistic has low power to detect such deviation. Although grouping of the estimated probabilities into quantiles from 8 to 30 was studied, the deciles of risk approach was generally sufficient. Subdividing the estimated probabilities into more than 10 quantiles when there are many covariates in the model is not necessary, despite theoretical reasons which suggest otherwise. Because it does not follow a X$\sp2$ distribution, the statistic is not recommended for use in models containing only categorical variables with a limited number of covariate patterns.^ The statistic performed adequately when there were at least 10 observations per quantile. Large numbers of observations per quantile did not lead to incorrect conclusions that the model did not fit the data when it actually did. However, the statistic failed to detect lack of fit when it existed and should be supplemented with further tests for the influence of individual observations. Careful examination of the parameter estimates is also essential since the statistic did not perform as desired when there was moderate to severe collinearity among covariates.^ Two methods studied for handling tied values of the estimated probabilities made only a slight difference in conclusions about model fit. Neither method split observations with identical probabilities into different quantiles. Approaches which create equal size groups by separating ties should be avoided. ^

Comprehensive Validation Of A Measure Of Student School Engagement: A Pilot Study Of Middle School Students

Although it may sound reasonable that American education continues to be more effective at sending high school students to college, in a study conducted in 2009, The Council of the Great City Schools states that "slightly more than half of entering ninth grade students arrive performing below grade level in reading and math, while one in five entering ninth grade students is more than two years behind grade level...[and] 25% received support in the form of remedial literacy instruction or interventions" (Council of the Great City Schools, 2009). Students are distracted with technology (Lei & Zhao, 2005), family (Xu & Corno, 2003), medical illnesses (Nielson, 2009), learning disabilities and perhaps the most detrimental to academic success, the very lack of interest in school (Ruch, 1963). In a Johns Hopkins research study, Building a Graduation Nation - Colorado (Balfanz, 2008), warning signs were apparent years before the student dropped out of high school. The ninth grade was often referenced as a critical point that indicated success or failure to graduate high school. The research conducted by Johns Hopkins illustrates the problem: students who become disengaged from school have a much greater chance of dropping out of high school and not graduating. The first purpose of this study was to compare different measurement models of the Student School Engagement (SSE) using Factor Analysis to verify model fit with student engagement. The second purpose was to determine the extent to which the SSE instrument measures student school engagement by investigating convergent validity (via the SSE and Appleton, Christenson, Kim and Reschly's instrument and Fredricks, Blumenfeld, Friedel and Paris's instrument), discriminant validity (via Huebner's Student Life Satisfaction Survey) and criterion-related validity (via the sub-latent variables of Aspirations, Belonging and Productivity and student outcome measures such as achievement, attendance and discipline). Discriminant validity was established between the SSE and the Appleton, Christenson, Kim and Reschly's model and Fredricks, Blumenfeld, Friedel and Paris's (2005) Student Engagement Instruments (SEI). When confirming discriminant validity, the SSE's correlations were weak and statistically not significant, thus establishing discriminant validity with the SLSS. Criterion-related validity was established through structural equation modeling when the SSE was found to be a significant predictor of student outcome measures when both risk score and CSAP scores were used. The third purpose of this study was to assess the factorial invariance of the SSE instrument across gender to ensure the instrument is measuring the intended construct across different groups. Conclusively, configural, weak and metric invariances were established for the SSE as a non-significant change in chi-square indicating that all parameters including the error variances were invariant across groups of gender. Engagement is not a clearly defined psychological construct; it requires more research in order to fully comprehend its complexity. Hopefully, with parental and teacher involvement and a sense of community, student engagement can be nurtured to result in a meaningful attachment to school and academic success.

Psychosocial Distress Screening in a Community Cancer Center: An Exploratory Study of Distress Screening Across the Cancer Treatment Continuum

Purpose: The primary goal of this exploratory study is to demonstrate that distress screening across the course of cancer treatment is possible and provides valuable information about patient needs over time. Distress screening is aligned with guidelines from national accrediting organizations and may lead to improved health-related quality of life, satisfaction with medical care, and possibly survival.Methods: Medical, surgical, and radiation oncology patients completed a screening instrument before their appointments during a six-month period. Patients indicated their level of distress on four domains (practical, emotional, health and social concerns). De-identified data was collected, aggregated and descriptive statistics were analyzed.Results: Approximately 3000 screens were collected and 1500 cancer patients were screened. Of patients who indicated distress, 54% demonstrated a distress level of five or greater. Distress level eight was the most frequent level of distress indicated. The Cancer Dietitian was the most commonly requested healthcare team provider. The Health Concern domain was most frequently endorsed.Conclusion: NCCN, IOM and COC guidelines recommend distress screening in all cancer treatment centers, however implementation has proven difficult. This study adds to the literature about distress in cancer patients, demonstrates the feasibility of repeated distress screening and provides a model program demonstrating the implementation of repeated distress screening at a community cancer center. Findings highlight the importance of supportive oncology services due to the prevalence of high levels of distress. Findings demonstrate the importance of the Cancer Dietitian in supportive cancer care. Additionally, the research reveals a potential perceived stigma in seeking psychosocial oncology services.

Kinetic Model for Micro Algae Cell Concentration and Size Distribution: Application to Nannochloropsis gaditana

The cell concentration and size distribution of the microalgae Nannochloropsis gaditana were studied over the whole growth process. Various samples were taken during the light and dark periods the algae were exposed to. The distributions obtained exhibited positive skew, and no change in the type of distribution was observed during the growth process. The size distribution shifted to lower diameters in dark periods while in light periods the opposite occurred. The overall trend during the growth process was one where the size distribution shifted to larger cell diameters, with differences between initial and final distributions of individual cycles becoming smaller. A model based on the Logistic model for cell concentration as a function of time in the dark period that also takes into account cell respiration and growth processes during dark and light periods, respectively, was proposed and successfully applied. This model provides a picture that is closer to the real growth and evolution of cultures, and reveals a clear effect of light and dark periods on the different ways in which cell concentration and diameter evolve with time.

Life Is Getting Worse in ESS Data: Is This Due to Micro or Macro Factors? Bruges European Economic Research (BEER) Papers 28/November 2013

In order to evaluate the success of a society, measuring well-being might be a fruitful avenue. For a long time, governments have trusted economic measures, Gross Domestic Product (GDP) in particular, to assess their success. However GDP is only a limited measure of economic success, which is not enough to show whether policies implemented by governments have a positive perceived impact on the people they represent. This paper belongs to the studies of the relationship between measures of well-being and economic factors. More precisely, it tries to evaluate the decrease in happiness and life satisfaction that can be observed in European countries in the 2000-2010 decade. It asks whether this deterioration is mainly due to microeconomic factors, such as income and individual characteristics, or rather to environmental (macroeconomics) factors such as unemployment, inflation or income inequality. Such aggregate factors could impact individual happiness per se because they are related to the perception of an aggregate risk of unemployment or income fall. In order to strengthen this interpretation, this paper checks whether the type of social protection regime existing in different countries mediates the impact of macroeconomic volatility on individual well-being. To go further, adopting the classification of welfare regimes proposed by Esping-Andersen (1990), it verifies whether the decreasing pattern of subjective well-being varies across these regimes. This is partly due to the aggregate social protection expenditure. Hence, this paper brings some additional evidence to the idea that macroeconomic uncertainty has a cost in terms of well-being. More protective social regimes are able to reduce this cost. It also proposes an evaluation of the welfare cost of unemployment and inflation (in terms of happiness and life satisfaction), in each of the different social protection regimes. Finally different measures of well-being, i.e. cognitive, hedonic and eudaimonic, are used to confirm the above mentioned result.

Equity weights in the allocation of health care: the rank-dependent QALY model

This paper introduces the rank-dependent quality-adjusted life-years (QALY) model, a new method to aggregate QALYs in economic evaluations of health care. The rank-dependent QALY model permits the formalization of influential concepts of equity in the allocation of health care, such as the fair innings approach, and it includes as special cases many of the social welfare functions that have been proposed in the literature. An important advantage of the rank-dependent QALY model is that it offers a straightforward procedure to estimate equity weights for QALYs. We characterize the rank-dependent QALY model and argue that its central condition has normative appeal. (C) 2003 Elsevier B.V. All rights reserved.

Relations between coping and positive and negative outcomes in carers of persons with multiple sclerosis (MS)

This study examined relations between stress and coping predictors and negative and positive outcomes in MS caregiving. A total of 222 carers and their care-recipients completed questionnaires at Time 1 and three months later, Time 2 ( n = 155). Predictors included care-recipient characteristics ( age, time since diagnosis, course and life satisfaction), and Times 1 and 2 carer problems, stress appraisal and coping. Dependent variables were Time 2 negative ( anxiety, depression) and positive outcomes ( life satisfaction, positive affect, benefits). Regressions indicated that, overall, the hypothesised direct effects of stress appraisal and coping strategies on positive and negative outcomes were supported. The hypothesised stress-buffering effects of positive reframing coping were also supported. All but one of the coping strategies were related to both positive and negative outcomes; specifically, practical assistance coping emerged as a unique predictor of distress. Of the model predictors, care-recipient life satisfaction emerged as the strongest and most consistent predictor of both positive and negative outcomes except benefit finding. Findings support the role of care-recipient characteristics and the carer's appraisal and coping processes in shaping both positive and negative outcomes. The guiding framework and findings have the potential to inform interventions designed to promote well-being in carers.

A stochastic metapopulation model accounting for habitat dynamics

A stochastic metapopulation model accounting for habitat dynamics is presented. This is the stochastic SIS logistic model with the novel aspect that it incorporates varying carrying capacity. We present results of Kurtz and Barbour, that provide deterministic and diffusion approximations for a wide class of stochastic models, in a form that most easily allows their direct application to population models. These results are used to show that a suitably scaled version of the metapopulation model converges, uniformly in probability over finite time intervals, to a deterministic model previously studied in the ecological literature. Additionally, they allow us to establish a bivariate normal approximation to the quasi-stationary distribution of the process. This allows us to consider the effects of habitat dynamics on metapopulation modelling through a comparison with the stochastic SIS logistic model and provides an effective means for modelling metapopulations inhabiting dynamic landscapes.

Relations between social support, appraisal and coping and both positive and negative outcomes for children of a parent with multiple sclerosis and comparisons with children of healthy parents

Objective: To examine adjustment in children of a parent with multiple sclerosis within a stress and coping framework and compare them with those who have 'healthy' parents. Subjects: A total of 193 participants between 10 and 25 years completed questionnaires; 48 youngsters who had a parent with multiple sclerosis and 145 youngsters who reported that they did not have a parent with an illness or disability. Method: A questionnaire survey methodology was used. Variable sets included caregiving context (e.g. additional parental illness, family responsibilities, parental functional impairment, choice in helping), social support (network size, satisfaction), stress appraisal, coping (problem solving, seeking support, acceptance, wishful thinking, denial), and positive (life satisfaction, positive affect, benefits) and negative (distress, health) adjustment outcomes. Results: Caregiving context variables significantly correlated with poorer adjustment in children of a parent with multiple sclerosis included additional parental illness, higher family responsibilities, parental functional impairment and unpredictability of the parent's multiple sclerosis, and less choice in helping. As predicted, better adjustment in children of a parent with multiple sclerosis was related to higher levels of social support, lower stress appraisals, greater reliance on approach coping strategies (problem solving, seeking support and acceptance) and less reliance on avoidant coping (wishful thinking and denial). Compared with children of 'healthy' parents, children of a parent with multiple sclerosis reported greater family responsibilities, less reliance on problem solving and seeking social support coping, higher somatization and lower life satisfaction and positive affect. Conclusions: Findings delineate the key impacts of young caregiving and support a stress and coping model of adjustment in children of a parent with multiple sclerosis.

Managerial job satisfaction:a longitudinal study

The study examines the job satisfaction of supervisors and managers in four organisations over time. It also considers the importance which they attached to different facets of their job. The major objectives were: To examine the constituent dimensions of job satisfaction at intervals over one year. To examine reasons for change inthe level of job satisfaction at intervals over one year. To provide information on job satisfaction for those concerned with job satisfaction policies. The sample consisted of one hundred and eight people. Each was interviewed on at least three occasions over the course of a year. Interviews took place at predetermined time intervals. The study shows that job satisfaction is dynamic over a relatively short period of time. The ratings which supervisors and managers gave to aspects of their job did not, however, all change by equal amounts or in the same direction. Changes in job satisfaction were associated with events experienced but it was the meaning of those events to correspondents which appeared to be particularly important. People tended to adopt a localised frame of reference when considering their work situation. Certain job variables, such as variety, were consistently and positively correlated with job satisfaction. With some other variables, the relationship varied across time. Frequently, age and job level moderated the association between independent variables and job satisfaction. Links were found between the quality of life and job satisfaction. There was a consistent positive association between job satisfaction and life satisfaction. However, the job was rarely considered to be the main factor contributing to a person's quality of life. The research highlights the difficulties and desirability of introducing standardised job satisfaction policies in the light of individual differences. In addition, it demonstrates that merely correlating variables with job satisfaction at one point in time may conceal complex relationships and meanings. A new measure of job satisfaction - whereby facets are assessed and rated relative to each other was also developed as part of this study.

Help in finding the right balance:leadership, work-family balance and employee outcomes

Although much research has examined employees’ experience of the work-family interface, its conceptualization has been rather problematic, ranging from work and family as mutually constraining through to mutually enriching and, more recently, to work-family balance (WFB). Building on Greenhaus and Allen’s (2011) conceptualization of WFB as comprising satisfaction and effectiveness components, I proposed and tested a model of he antecedents and outcomes of WFB. Based on work-family border theory, I hypothesised that family-supportive supervisor behaviours (FSSB) facilitate WFB and hat the relationship is stronger when the organisation also offers formal support (availability of family-friendly practices (FFPs); enhancement effect). Furthermore, I integrated the leadership and work-family interface literatures by proposing authentic eadership as an antecedent of FSSB. Based on role accumulation theories, I proposed life satisfaction and health as outcomes of WFB satisfaction and WFB effectiveness and job performance as an outcome of only WFB effectiveness. I tested my hypotheses with individual-level data in Study 1 (two waves of data; employees from Germany and the UK) and nested data (individuals nested in teams; two waves of data; employee and supervisor ratings; Germany and the UK) in Study 2. The obtained findings largely supported the hypothesized model and showed that both authentic leadership (Study 1) and team authentic leadership (Study 2) predicted FSSB which, in turn, increased WFB satisfaction and WFB effectiveness. Contrary to my prediction, both studies revealed that FSSB and (team) availability of FFPs compensated for each other, only impacting WFB satisfaction/effectiveness if the other form of family support was not available. Furthermore, both components were positively related to life satisfaction and health, while WFB effectiveness was only related to self-rated performance (Study 1) and not supervisor-rated performance (Study 2). Lastly, the serial moderated mediation model hat tested the conditional indirect effect of (team) authentic leadership on the outcomes received mixed support.

The effects of parental nurturance and involvement on peer relationships and psychosocial functioning of young adults

This study examined peer relationships and psychosocial functioning as a function of maternal and paternal involvement and nurturance along with the moderating effects of gender, family form, and ethnicity. Prior research has shown the influence of mother’s involvement on peer relationship quality but not of fathers. Further, previous studies did not examine moderation by family form, gender, or ethnicity. The sample consisted of 1359 students who identified their biological mother and father as the most influential parental figures in their lives. Their ages ranged from 18 to 26; Sixty-one percent of the sample was Hispanic, 13% non-Hispanic Black, 25% non-Hispanic White; 76% female and 70% from intact families. The analytical strategy included using bivariate correlations and structural equation modeling to examine these relationships. ^ All dimensions of maternal and paternal nurturing and involvement were positively related to positive characteristics of peer relationships, self-esteem and life satisfaction consistent with the multicultural findings of PARTheory (Rohner, Khalique, & Cournoyer, 2005). A structural model was developed that was able to adequately account for the relationship between parental influence, peer relationships, and psychosocial functioning. These effects of both maternal and paternal influence were strongly moderated by culture, family form, and gender. Finally, a differential effect was found among parental influence with fathers having a greater influence on friendship quality and importance than mothers, despite greater maternal involvement. ^ These findings have theoretical, clinical, and social implications as they call for a socially based theoretical perspective within which to study these relationships. Such a perspective would better inform clinicians when using impaired social functioning as indicative of axial diagnosis, and for the implementation of social policy to encourage paternal involvement. ^

Making Life More Satisfying for Hospitality Managers

Because of the considerable amount of time that hospitality managers spend at work, the relationship between life satisfaction and job satisfaction is of particular interest. Dissonance may result when the role at work conflicts with the role at home of with the family. Food service managers indicate that the top inter-role conflicts area is that work time takes up time that they would like to spend with family and others; that they are too tired to do some of the things they would like to do after work; and that the job makes it difficult to be the kind of friend, spouse, or parent that they would like to be.

Identity, purpose, and well-being among emerging adult Hispanic women

Few studies apply the Eriksonian model of identity formation to cross-cultural samples (3), even though issues of ethnicity and culture may inform a Hispanic woman's self-concept (Phinney, 1996). Hispanic women may also be influenced by traditional gender role behaviors such as passivity or dependence that are outlined by marianismo (Stevens, 1973). A recent study of a multiethnic sample of emerging adult women and men found that purpose commitment mediated the effects of identity commitment on hope and life satisfaction (Burrow & Hill, 2011). The current research consists of two studies that replicate and expand upon the work of Burrow and Hill (2011). Study I replicated the work of Burrow and Hill (2011) among a sample of emerging adult Hispanic women, in order to assess the extent to which the original findings would replicate in a culturally distinct sample. Study II examined the role of marianismo, ethnic identity, and acculturation on identity commitment among emerging adult Hispanic women. Both studies utilized a sample of 532 female undergraduate psychology students, age 18 to 25, who self-identified as Hispanic and submitted data via online surveys. Both studies used self-report, quantitative data, which was analyzed using structural equation modeling. Results from Study I indicated good model fit and replicated the findings from Burrow and Hill (2011). Specifically, the direct effect of identity commitment on hope was fully contingent upon an individual's level of purpose commitment, while the effect of identity commitment on life satisfaction was not contingent upon an individual's level of purpose commitment. Results from Study II indicated that marianismo, Spanish proficiency, familiarity with Latino culture, and familiarity with American culture demonstrated statistically significant direct effects on identity commitment among emerging adult Hispanic women. Results indicated cultural convergence regarding the association of an individual's identity with well-being through a sense of purpose. Findings also revealed the role of cultural factors in the extent to which Hispanic women commit to a personal identity. Future studies should employ mixed method research designs as a means to better ascertain implications of findings. ^

A recessionary debate: who suffered more? A study of sectoral well-being and economic insecurity in Ireland

The aim of this thesis is to identify the relationship between subjective well-being and economic insecurity for public and private sector workers in Ireland using the European Social Survey 2010-2012. Life satisfaction and job satisfaction are the indicators used to measure subjective well-being. Economic insecurity is approximated by regional unemployment rates and self-perceived job insecurity. Potential sample selection bias and endogeneity bias are accounted for. It is traditionally believed that public sector workers are relatively more protected against insecurity due to very institution of public sector employment. The institution of public sector employment is made up of stricter dismissal practices (Luechinger et al., 2010a) and less volatile employment (Freeman, 1987) where workers are subsequently less likely to be affected by business cycle downturns (Clark and Postal-Vinay, 2009). It is found in the literature that economic insecurity depresses the well-being of public sector workers to a lesser degree than private sector workers (Luechinger et al., 2010a; Artz and Kaya, 2014). These studies provide the rationale for this thesis in testing for similar relationships in an Irish context. Sample selection bias arises when a selection into a particular category is not random (Heckman, 1979). An example of this is non-random selection into public sector employment based on personal characteristics (Heckman, 1979; Luechinger et al., 2010b). If selection into public sector employment is not corrected for this can lead to biased and inconsistent estimators (Gujarati, 2009). Selection bias of public sector employment is corrected for by using a standard Two-Step Heckman Probit OLS estimation method. Following Luechinger et al. (2010b), the propensity for individuals to select into public sector employment is estimated by a binomial probit model with the inclusion of the additional regressor Irish citizenship. Job satisfaction is then estimated by Ordinary Least Squares (OLS) with the inclusion of a sample correction term similar as is done in Clark (1997). Endogeneity is where an independent variable included in the model is determined within in the context of the model (Chenhall and Moers, 2007). The econometric definition states that an endogenous independent variable is one that is correlated with the error term (Wooldridge, 2010). Endogeneity is expected to be present due to a simultaneous relationship between job insecurity and job satisfaction whereby both variables are jointly determined (Theodossiou and Vasileiou, 2007). Simultaneity, as an instigator of endogeneity, is corrected for using Instrumental Variables (IV) techniques. Limited Information Methods and Full Information Methods of estimation of simultaneous equations models are assed and compared. The general results show that job insecurity depresses the subjective well-being of all workers in both the public and private sectors in Ireland. The magnitude of this effect differs among sectoral workers. The subjective well-being of private sector workers is more adversely affected by job insecurity than the subjective well-being of public sector workers. This is observed in basic ordered probit estimations of both a life satisfaction equation and a job satisfaction equation. The marginal effects from the ordered probit estimation of a basic job satisfaction equation show that as job insecurity increases the probability of reporting a 9 on a 10-point job satisfaction scale significantly decreases by 3.4% for the whole sample of workers, 2.8% for public sector workers and 4.0% for private sector workers. Artz and Kaya (2014) explain that as a result of many austerity policies implemented to reduce government expenditure during the economic recession, workers in the public sector may for the first time face worsening perceptions of job security which can have significant implications for their well-being (Artz and Kaya, 2014). This can be observed in the marginal effects where job insecurity negatively impacts the well-being of public sector workers in Ireland. However, in accordance with Luechinger et al. (2010a) the results show that private sector workers are more adversely impacted by economic insecurity than public sector workers. This suggests that in a time of high economic volatility, the institution of public sector employment held and was able to protect workers against some of the well-being consequences of rising insecurity. In estimating the relationship between subjective well-being and economic insecurity advanced econometric issues arise. The results show that when selection bias is corrected for, any statistically significant relationship between job insecurity and job satisfaction disappears for public sector workers. Additionally, in order to correct for endogeneity bias the simultaneous equations model for job satisfaction and job insecurity is estimated by Limited Information and Full Information Methods. The results from two different estimators classified as Limited Information Methods support the general findings of this research. Moreover, the magnitude of the endogeneity-corrected estimates are twice as large as those not corrected for endogeneity bias which is similarly found in Geishecker (2010, 2012). As part of the analysis into the effect of economic insecurity on subjective well-being, the effects of other socioeconomic variables and work-related variables are examined for public and private sector workers in Ireland.