Rosa Montaño

How Should We Assess the Fit of Rasch-Type Models? Approximating the Power of Goodness-of-Fit Statistics in Categorical Data Analysis.

We investigate the performance of three statistics, R1, R2 (Glas in Psychometrika 53:525–546, 1988), and M2 (Maydeu-Olivares & Joe in J. Am. Stat. Assoc. 100:1009–1020, 2005, Psychometrika 71:713–732, 2006) to assess the overall fit of a one-parameter logistic model (1PL) estimated by (marginal) maximum likelihood (ML). R1 and R2 were specifically designed to target specific assumptions of Rasch models, whereas M2 is a general purpose test statistic. We report asymptotic power rates under some interesting violations of model assumptions (different item discrimination, presence of guessing, and multidimensionality) as well as empirical rejection rates for correctly specified models and some misspecified models. All three statistics were found to be more powerful than Pearson’s X2 against two- and three-parameter logistic alternatives (2PL and 3PL), and against multidimensional 1PL models. The results suggest that there is no clear advantage in using goodness-of-fit statistics specifically designed for Rasch-type models to test these models when marginal ML estimation is used.

Sintomas depresivos y distres laboral en trabajadores chilenos: condiciones diferenciales para hombres y mujeres

This article assessed depressive symptoms associated with work-related psychosocial risk factors according to gender in Chile, using the demand-control model (Karasek) and effort-reward imbalance (Siegrist). A cross-sectional study was conducted in a random sample of 3,010 workers (35% female and 65% male) from the country. Data analysis determined prevalence and associations through various statistical techniques (χ2, logistic regression). Exposure to psychosocial risk factors at work and prevalence of depressive symptoms were higher in women than men (15% vs. 5%). The adjusted analysis highlighted that female workers exposed to Isostrain (OR = 2.34; 95%CI: 1.42-3.85) and low rewards (OR = 2.13; 95%CI: 1.41-3.21) and male workers exposed to psychological demands (OR = 3.04; 95%CI: 1.94-4.76) and effort-reward imbalance (OR = 2.19; 95%CI: 1.39-3.46) had increased risk of depressive symptoms. Exposure to work-related psychosocial risk factors was associated with depressive symptoms in Chilean workers. Effective prevention in key aspects of work organization is thus needed.O objetivo foi avaliar os sintomas depressivos associados com fatores de risco ocupacionais psicossociais, por sexo, no Chile, por meio do modelo demanda-controle (Karasek) e desequilibrio esforco-recompensa (Siegrist). O estudo foi transversal, com uma amostra aleatoria de 3.010 empregados (35% mulheres e 65% homens) em todo o pais. Prevalencia e associacoes foram determinados por diversas tecnicas estatisticas (eg. χ 2 ; regressao logistica). A exposicao ao fatores de risco ocupacionais psicossociais e prevalencia de sintomas depressivos e maior em mulheres do que em homens (15% vs. 5%). Analise ajustada salienta que as mulheres expostas a Isostrain (OR = 2,34; IC95%: 1,42-3,85) e baixa recompensa (OR = 2,13; IC95%: 1,41-3,21) e homens expostos a demandas psicologicas (OR = 3,04; IC95%: 1,94-4,76) e desequilibrio esforco-recompensa (OR = 2,19; IC95%: 1,39-3,46) tem chance maior de sintomas depressivos quando comparado com o nao exposto. A exposicao a fatores de risco ocupacionais psicossociais foi associada com sintomas depressivos em trabalhadores chilenos. A geracao de cuidados preventivos em aspectos de organizacao do trabalho e fundamental.

Bullying y calidad de vida relacionada con la salud en adolescentes escolares chilenos

BACKGROUND Bullying has a negative impact on the perception of adolescents of their quality of life. AIM To analyze the association between being bullied and health related quality of life in Chilean adolescents. MATERIAL AND METHODS The bullying module of the KIDSCREEN 52 survey and the KIDSCREEN 10 survey to assess health related quality of life were applied to 7737 students aged 10 to 18 years. Social and demographic data, information about disability and type of school were also gathered. RESULTS Fifteen percent of surveyed students were bullied. A lower quality of life perception increased by a factor of 2.6 among bullied students. It also increased by a factor of 4.4 among students with a low self-esteem, by a factor of 3.1 among those who perceived their health status as regular or bad and by a factor of 1.4 among women. CONCLUSIONS Bullying is associated with a lower quality of life perception among Chilean students.BACKGROUND: Bullying has a negative impact on the perception of adolescents of their quality of life. AIM: To analyze the association between being bullied and health related quality of life in Chilean adolescents. MATERIAL AND METHODS: The bullying module of the KIDSCREEN 52 survey and the KIDSCREEN 10 survey to assess health related quality of life were applied to 7737 students aged 10 to 18 years. Social and demographic data, information about disability and type of school were also gathered. RESULTS: Fifteen percent of surveyed students were bullied. A lower quality of life perception increased by a factor of 2.6 among bullied students. It also increased by a factor of 4.4 among students with a low self-esteem, by a factor of 3.1 among those who perceived their health status as regular or bad and by a factor of 1.4 among women. CONCLUSIONS: Bullying is associated with a lower quality of life perception among Chilean students. Language: es

Structure and content of messages in an online environment

This study examines the participation of 24 university professors in the training process that uses on-line environment, which has a duration of 36?h in classroom and 30?h online. The main aim of the course is to provide theoretical and methodological elements that allow teachers to investigate and innovate in their educational practice. With this purpose, the messages were analysed regarding their nature, information level, depth level, and the relation between each of the indicators. The data corpus corresponds to ten discussion forums which were part of a training program in research and innovation in university teaching. The methodology used for the analysis considered the combination of quantitative indicators of participation with an analysis of message content. The results show that the messages are organized mainly around the content or learning task, which in turn tend to concentrate on levels with less complexity and depth. The data also indicate that there is a relation between message length, information level, and depth level achieved by the contributions of the participants when they collaborate in an online environment. The nature of the messages, which is associated mainly with the tasks content.The results show the relation between the information level and the depth of the message.The information level, the depth of it and the length of the messages are related.There is a relationship between the estructural indicators and the content indicators.

Psychosocial Risk at Work and Hazardous Alcohol Consumption Among Chile’s Working Adults

ObjectivesKarasek’s demand-control model and Siegrist’s effort-reward imbalance model have accumulated solid evidence regarding the associations between exposure to psychosocial risk at work (PSRW) and mental health of workers. However, there is scarce such evidence with regard to its associations with alcohol consumption. This study proposed to estimate the associations between exposure to PSRW and hazardous alcohol consumption (HAC) in Chile’s working adult population.MethodThe study was cross-sectional and a nationally representative survey was applied to 3,010 workers (65% male and 35% female, ages 20 to 65). The analysis included prevalences and logistic regression controlling for covariables.ResultsThe adjusted analyses show that male workers exposed to low social support (OR=1.47; 95% CI:1.14–1.89), low reward (OR=1.38; 95% CI:1.07–1.78) and effort-reward imbalance (OR=1.34; 95% CI:1.04–1.73) have a higher chance of presenting with HAC compared to those who are not exposed. Female workers exposed to effort-reward imbalance (OR=2.34; 95% CI:1.10–5.58) have twice the risk of HAC compared with their reference group.ConclusionThis study shows evidence of the associations between HAC and exposure to a set of psychosocial risk factors from the Karasek and Siegrist models. For future research, it is recommended that HAC and PSRW factors be examined in a longitudinal study in order to control for possible confounding effects on these associations.RésuméObjectifsLe modèle ªDemande-Latitude« de Karasek et le modèle ªDéséquilibre Efforts/Reconnaissance« de Siegrist accumulent des preuves solides sur les associations entre l’exposition aux risques psychosociaux au travail (RPT) et la santé mentale des travailleurs. Cependant, il existe peu de preuves de la sorte en ce qui a trait aux associations entre les RPT et la consommation d’alcool. Nous avons voulu estimer les associations entre l’exposition aux RPT et la consommation dangereuse d’alcool (CDA) dans la population active d’âge adulte du Chili.MéthodeUne enquête représentative nationale transversale a été menée auprès de 3 010 travailleurs (65% d’hommes et 35% de femmes, 20 à 65 ans). L’analyse a inclus le calcul des prévalences et de la régression logistique après ajustement pour tenir compte des effets des covariables.RésultatsLes analyses ajustées montrent que les travailleurs masculins exposés à un faible soutien social (RC=1,47; IC de 95%: 1,14–1,89), à une faible reconnaissance (RC=1,38; IC de 95%: 1,07-1,78) et à un déséquilibre efforts/reconnaissance (RC=1,34; IC de 95%: 1,04–1,73) ont une probabilité supérieure de présenter une CDA comparativement aux travailleurs non exposés. Les travailleuses exposées à un déséquilibre efforts/reconnaissance (RC=2,34; IC de 95%: 1,10–5,58) ont un risque deux fois supérieur de présenter une CDA comparativement à leur groupe témoin.ConclusionCette étude prouve qu’il existe des associations entre la CDA et l’exposition à un ensemble de facteurs de risque psychosociaux tirés des modèles de Karasek et de Siegrist. Pour pousser la recherche, il est recommandé d’examiner les facteurs de la CDA et des RPT dans une étude longitudinale afin de tenir compte des effets de facteurs de confusion possibles sur ces associations.

Sintomas depressivos e estresse em trabalhadores chilenos: condições diferenciais para homens e mulheres

This article assessed depressive symptoms associated with work-related psychosocial risk factors according to gender in Chile, using the demand-control model (Karasek) and effort-reward imbalance (Siegrist). A cross-sectional study was conducted in a random sample of 3,010 workers (35% female and 65% male) from the country. Data analysis determined prevalence and associations through various statistical techniques (χ2, logistic regression). Exposure to psychosocial risk factors at work and prevalence of depressive symptoms were higher in women than men (15% vs. 5%). The adjusted analysis highlighted that female workers exposed to Isostrain (OR = 2.34; 95%CI: 1.42-3.85) and low rewards (OR = 2.13; 95%CI: 1.41-3.21) and male workers exposed to psychological demands (OR = 3.04; 95%CI: 1.94-4.76) and effort-reward imbalance (OR = 2.19; 95%CI: 1.39-3.46) had increased risk of depressive symptoms. Exposure to work-related psychosocial risk factors was associated with depressive symptoms in Chilean workers. Effective prevention in key aspects of work organization is thus needed.O objetivo foi avaliar os sintomas depressivos associados com fatores de risco ocupacionais psicossociais, por sexo, no Chile, por meio do modelo demanda-controle (Karasek) e desequilibrio esforco-recompensa (Siegrist). O estudo foi transversal, com uma amostra aleatoria de 3.010 empregados (35% mulheres e 65% homens) em todo o pais. Prevalencia e associacoes foram determinados por diversas tecnicas estatisticas (eg. χ 2 ; regressao logistica). A exposicao ao fatores de risco ocupacionais psicossociais e prevalencia de sintomas depressivos e maior em mulheres do que em homens (15% vs. 5%). Analise ajustada salienta que as mulheres expostas a Isostrain (OR = 2,34; IC95%: 1,42-3,85) e baixa recompensa (OR = 2,13; IC95%: 1,41-3,21) e homens expostos a demandas psicologicas (OR = 3,04; IC95%: 1,94-4,76) e desequilibrio esforco-recompensa (OR = 2,19; IC95%: 1,39-3,46) tem chance maior de sintomas depressivos quando comparado com o nao exposto. A exposicao a fatores de risco ocupacionais psicossociais foi associada com sintomas depressivos em trabalhadores chilenos. A geracao de cuidados preventivos em aspectos de organizacao do trabalho e fundamental.

Depressive symptoms and work-related stress in Chilean workers: differential conditions for males and females

This article assessed depressive symptoms associated with work-related psychosocial risk factors according to gender in Chile, using the demand-control model (Karasek) and effort-reward imbalance (Siegrist). A cross-sectional study was conducted in a random sample of 3,010 workers (35% female and 65% male) from the country. Data analysis determined prevalence and associations through various statistical techniques (χ2, logistic regression). Exposure to psychosocial risk factors at work and prevalence of depressive symptoms were higher in women than men (15% vs. 5%). The adjusted analysis highlighted that female workers exposed to Isostrain (OR = 2.34; 95%CI: 1.42-3.85) and low rewards (OR = 2.13; 95%CI: 1.41-3.21) and male workers exposed to psychological demands (OR = 3.04; 95%CI: 1.94-4.76) and effort-reward imbalance (OR = 2.19; 95%CI: 1.39-3.46) had increased risk of depressive symptoms. Exposure to work-related psychosocial risk factors was associated with depressive symptoms in Chilean workers. Effective prevention in key aspects of work organization is thus needed.O objetivo foi avaliar os sintomas depressivos associados com fatores de risco ocupacionais psicossociais, por sexo, no Chile, por meio do modelo demanda-controle (Karasek) e desequilibrio esforco-recompensa (Siegrist). O estudo foi transversal, com uma amostra aleatoria de 3.010 empregados (35% mulheres e 65% homens) em todo o pais. Prevalencia e associacoes foram determinados por diversas tecnicas estatisticas (eg. χ 2 ; regressao logistica). A exposicao ao fatores de risco ocupacionais psicossociais e prevalencia de sintomas depressivos e maior em mulheres do que em homens (15% vs. 5%). Analise ajustada salienta que as mulheres expostas a Isostrain (OR = 2,34; IC95%: 1,42-3,85) e baixa recompensa (OR = 2,13; IC95%: 1,41-3,21) e homens expostos a demandas psicologicas (OR = 3,04; IC95%: 1,94-4,76) e desequilibrio esforco-recompensa (OR = 2,19; IC95%: 1,39-3,46) tem chance maior de sintomas depressivos quando comparado com o nao exposto. A exposicao a fatores de risco ocupacionais psicossociais foi associada com sintomas depressivos em trabalhadores chilenos. A geracao de cuidados preventivos em aspectos de organizacao do trabalho e fundamental.

Análisis de la evolución y tendencia de las tasas de hepatitis A en Chile desde 197 5 a 2002

Las tendencias temporales de las tasas de hepatitis A en Chile, resultan ser erráticas, tanto si se analizan en forma global como separadas por rango etáreo. Esta tendencia no regular en el tiempo, hace muy complejo intentar predecir una futura tasa mediante métodos tradicionales de análisis de series temporales tales como modelos ARIMA o ARMA entre otros. La Figura 1 muestra la tendencia temporal de la tasa global y desagregadas por rango etáreo de hepatitis A en Chile. Un análisis muy simplista nos inclinaría a pensar que dichas evoluciones son constantes en el tiempo, resultado que puede obtenerse al descartar una tendencia creciente o decreciente en el tiempo, mediante un análisis de regresión lineal. Por ejemplo, si se opta por estudiar la tasa para el grupo etáreo de O a 4 años, el estudio de tendencia lineal arroja (Tabla 1) una tendencia estadísticamente significativa a la baja (p = 0,007), conclusión que no es satisfactoria, puesto que no considera los quiebres de la tendencia en determinados años. La Figura 2 induce a pensar que la tasa de hepatitis A, en este primer rango etáreo, tenderá a crecer el año 2004 y 2005. Sin embargo, el modelo estimado en la Tabla 1 predice una tasa de 84,2 y 79,4, respectivamente, predicciones muy discutibles si se analizan las Figuras 2 y 3. El análisis de la Figura 3, induce a modelar la evolución de esta tasa, mediante un modelo mínimo cuadrático de segundo grado, las estimaciones de dicho modelo se muestran en la Figura 4. Este modelo predice tasas para el rango etáreo de O a 4 años para los años 2004 y 2005 de 186,0 y 263,5, dada la tendencia al alza de dichas tasas estas predicciones resultan ser más creíbles que las entregadas por el modelo lineal, pero igualmente discutibles, pues no consideran las variaciones y quiebres que se observan a través del tiempo, o dicho de otro modo, no se