The evolution of health status and chronic conditions in Catalonia, 1994-2006: the paradox of health revisited using the Blinder - Oaxaca decomposition

[EN] Background: The paradox of health refers to the improvement in objective measures of health and the increase in the reported prevalence of chronic conditions. The objective of this paper is to test the paradox of health in Catalonia from 1994 to 2006. Methods: Longitudinal cross-sectional study using the Catalonia Health Interview Survey of 1994 and 2006. The approach used was the three-fold Blinder - Oaxaca decomposition, separating the part of the differential in mean visual analogue scale value (VAS) due to group differences in the predictors (prevalence effect), due to differences in the coefficients (severity effect), and an interaction term. Variables included were the VAS value, education level, labour status, marital status, all common chronic conditions over the two cross-sections, and a variable for non-common chronic conditions and other conditions. Sample weights have been applied. Results: Results show that there is an increase in mean VAS for men aged 15-44, and a decrease in mean VAS for women aged 65-74 and 75 and more. The increase in mean VAS for men aged 15-44 could be explained by a decrease in the severity effect, which offsets the increase in the prevalence effect. The decrease in mean VAS for women aged 65-74 and 75 and more could be explained by an increase in the prevalence effect, which does not offset the decrease in the severity effect. Conclusions: The results of the present analysis corroborate the paradox of health hypothesis for the population of Catalonia, and highlight the need to be careful when measuring population health over time, as well as their usefulness to detect population's perceptions.

The Blinder-Oaxaca decomposition for linear regression models

The counterfactual decomposition technique popularized by Blinder (1973, Journal of Human Resources, 436–455) and Oaxaca (1973, International Economic Review, 693–709) is widely used to study mean outcome differences between groups. For example, the technique is often used to analyze wage gaps by sex or race. This article summarizes the technique and addresses several complications, such as the identification of effects of categorical predictors in the detailed decomposition or the estimation of standard errors. A new command called oaxaca is introduced, and examples illustrating its usage are given.

Standard Errors for the Blinder-Oaxaca Decomposition

The decomposition technique introduced by Blinder (1973) and Oaxaca (1973) is widely used to study outcome differences between groups. For example, the technique is commonly applied to the analysis of the gender wage gap. However, despite the procedure's frequent use, very little attention has been paid to the issue of estimating the sampling variances of the decomposition components. We therefore suggest an approach that introduces consistent variance estimators for several variants of the decomposition. The accuracy of the new estimators under ideal conditions is illustrated with the results of a Monte Carlo simulation. As a second check, the estimators are compared to bootstrap results obtained using real data. In contrast to previously proposed statistics, the new method takes into account the extra variation imposed by stochastic regressors.

OAXACA: Stata module to compute the Blinder-Oaxaca decomposition

-oaxaca- computes the so-called Blinder-Oaxaca decomposition, which is often used to analyze wage gaps by sex or race. Older versions of this routine are available as -oaxaca9- and -oaxaca8-.

A Stata implementation of the Blinder-Oaxaca decomposition

The counterfactual decomposition technique popularized by Blinder (1973) and Oaxaca (1973) is widely used to study mean outcome differences between groups. For example, the technique is often used to analyze wage gaps by sex or race. The present paper summarizes the technique and addresses a number of complications such as the identification of effects of categorical predictors in the detailed decomposition or the estimation of standard errors. A new Stata command called -oaxaca- is introduced and examples illustrating its usage are given.

Immigrant wage differentials and the role of self-employment in Australia

Immigration has played an important role in the historical development of Australia. Thus, it is no surprise that a large body of empirical work has developed, which focuses upon how migrants fare in the land of opportunity. Much of the literature is comparatively recent, i.e. the last ten years or so, encouraged by the advent of public availability of Australian crosssection micro data. Several different aspects of migrant welfare have been addressed, with major emphasis being placed upon earnings and unemployment experience. For recent examples see Haig (1980), Stromback (1984), Chiswick and Miller (1985), Tran-Nam and Nevile (1988) and Beggs and Chapman (1988). The present paper contributes to the literature by providing additional empirical evidence on the native/migrant earnings differential. The data utilised are from the rather neglected Australian Bureau of Statistics, ABS Special Supplementary Survey No.4. 1982, otherwise known as the Family Survey. The paper also examines the importance of distinguishing between the wage and salary sector and the self-employment sector when discussing native/migrant differentials. Separate earnings equations for the two labour market groups are estimated and the native/migrant earnings differential is broken down by employment status. This is a novel application in the Australian context and provides some insight into the earnings of the selfemployed, a group that despite its size (around 20 per cent of the labour force) is frequently ignored by economic research. Most previous empirical research fails to examine the effect of employment status on earnings. Stromback (1984) includes a dummy variable representing self-employment status in an earnings equation estimated over a pooled sample of paid and self-employed workers. The variable is found to be highly significant, which leads Stromback to question the efficacy of including the self-employed in the estimation sample. The suggestion is that part of self-employed earnings represent a return to non-human capital investment, i.e. investments in machinery, buildings etc, the structural determinants of earnings differ significantly from those for paid employees. Tran-Nam and Nevile (1988) deal with differences between paid employees and the selfemployed by deleting the latter from their sample. However, deleting the self-employed from the estimation sample may lead to bias in the OLS estimation method (see Heckman 1979). The desirable properties of OLS are dependent upon estimation on a random sample. Thus, the 'Ran-Nam and Nevile results are likely to suffer from bias unless individuals are randomly allocated between self-employment and paid employment. The current analysis extends Tran-Nam and Nevile (1988) by explicitly treating the choice of paid employment versus self-employment as being endogenously determined. This allows an explicit test for the appropriateness of deleting self-employed workers from the sample. Earnings equations that are corrected for sample selection are estimated for both natives and migrants in the paid employee sector. The Heckman (1979) two-step estimator is employed. The paper is divided into five major sections. The next section presents the econometric model incorporating the specification of the earnings generating process together with an explicit model determining an individual's employment status. In Section 111 the data are described. Section IV draws together the main econometric results of the paper. First, the probit estimates of the labour market status equation are documented. This is followed by presentation and discussion of the Heckman two-stage estimates of the earnings specification for both native and migrant Australians. Separate earnings equations are estimated for paid employees and the self-employed. Section V documents estimates of the nativelmigrant earnings differential for both categories of employees. To aid comparison with earlier work, the Oaxaca decomposition of the earnings differential for paid-employees is carried out for both the simple OLS regression results as well as the parameter estimates corrected for sample selection effects. These differentials are interpreted and compared with previous Australian findings. A short section concludes the paper.

Wage differentials between Permanent and Temporal Contracts: Further Evidence

The main objective of this paper is to estimate wage differentials between permanent and temporal workers for different qualification levels and decompose such differentials to see which factors contribute more to explain them. The data we use is the "Encuesta de Estructura Salarial", a survey carried out in 1995 in all countries of the European Union, which contains very detailed information on wages and other characteristics for about 180.000 workers. The empirical results indicate that (a) the wage gap between permanent and temporal workers increases with qualification and it is smaller for females than for males, (b) when decomposing average wage differentials for each qualification level, we observe that the vast majority of such differential is explainedby differences in characteristics, and in particular by differences in tenure and occupation. With respect to differences in returns, our results indicate that the returns to tenure are higher for temporal workres relative to permanent workers for both males and females and for every qualification level. Returns to occupation are higher for permanent than for temporal workers, and the differenceis particularly high for females.

La performance économique des immigrants du Canada : une analyse régionale

Mémoire numérisé par la Division de la gestion de documents et des archives de l'Université de Montréal

L’évolution de la qualité d’emploi des immigrants du Canada par rapport aux natifs : une comparaison interprovinciale

Il est bien connu que les immigrants rencontrent plusieurs difficultés d’intégration dans le marché du travail canadien. Notamment, ils gagnent des salaires inférieurs aux natifs et ils sont plus susceptibles que ces derniers d’occuper des emplois précaires ou pour lesquels ils sont surqualifiés. Dans cette recherche, nous avons traité de ces trois problèmes sous l’angle de la qualité d’emploi. À partir des données des recensements de la population de 1991 à 2006, nous avons comparé l’évolution de la qualité d’emploi des immigrants et des natifs au Canada, mais aussi au Québec, en Ontario et en Colombie-Britannique. Ces comparaisons ont mis en évidence la hausse du retard de qualité d’emploi des immigrants par rapport aux natifs dans tous les lieux analysés, mais plus particulièrement au Québec. Le désavantage des immigrants persiste même lorsqu’on tient compte du capital humain, des caractéristiques démographiques et du taux de chômage à l’entrée dans le marché du travail. La scolarité, l’expérience professionnelle globale et les connaissances linguistiques améliorent la qualité d’emploi des immigrants et des natifs. Toutefois, lorsqu’on fait la distinction entre l’expérience de travail canadienne et l’expérience de travail étrangère, on s’aperçoit que ce dernier type d’expérience réduit la qualité d’emploi des immigrants. Dans ces circonstances, nous trouvons incohérent que le Canada et le Québec continuent à insister sur ce critère dans leur grille de sélection des travailleurs qualifiés. Pour valoriser les candidats les plus jeunes ayant peu d’expérience de travail dans leur pays d’origine, nous suggérons d’accroître l’importance accordée à l’âge dans ces grilles au détriment de l’expérience. Les jeunes, les étudiants étrangers et les travailleurs temporaires qui possèdent déjà une expérience de travail au Canada nous apparaissent comme des candidats à l’immigration par excellence. Par contre, les résultats obtenus à l’aide de la méthode de décomposition de Blinder-Oaxaca ont montré que l’écart de qualité d’emploi entre les immigrants et les natifs découle d’un traitement défavorable envers les immigrants dans le marché du travail. Cela signifie que les immigrants sont pénalisés au chapitre de la qualité d’emploi à la base, et ce, peu importe leurs caractéristiques. Dans ce contexte, la portée de tout ajustement aux grilles de sélection risque d’être limitée. Nous proposons donc d’agir également en aval du problème à l’aide des politiques d’aide à l’intégration des immigrants. Pour ce faire, une meilleure concertation entre les acteurs du marché du travail est nécessaire. Les ordres professionnels, le gouvernement, les employeurs et les immigrants eux-mêmes doivent s’engager afin d’établir des parcours accélérés pour la reconnaissance des compétences des nouveaux arrivants. Nos résultats indiquent aussi que le traitement défavorable à l’égard des immigrants dans le marché du travail est plus prononcé au Québec qu’en Ontario et en Colombie-Britannique. Il se peut que la société québécoise soit plus réfractaire à l’immigration vu son caractère francophone et minoritaire dans le reste de l’Amérique du Nord. Pourtant, le désir de protéger la langue française motive le Québec à s’impliquer activement en matière d’immigration depuis longtemps et la grille de sélection québécoise insiste déjà sur ce critère. D’ailleurs, près des deux tiers des nouveaux arrivants au Québec connaissent le français en 2011.

Do Proctored Exams Matter In Online Classes?

Does the format of assessment (proctored or un-proctored exams) affect test scores in online principles of economics classes? This study uses data from two courses of principles of economics taught by the same instructor to gain some insight into this issue. When final exam scores are regressed against human capital factors, the R-squared statistic is 61.6% for the proctored format exams while it is only 12.2% for the un-proctored format. Three other exams in the class that had the proctored final were also un-proctored and also produced lower R-squared values, averaging 30.5%. These two findings suggest that some cheating may have taken place in the un-proctored exams. Although it appears some cheating took place, the results suggest that cheating did not pay for these students since the proctored exam grades were 4.9 points higher than the un-proctored exam grades although this difference was significantly different at only the 10% level. One possible explanation for this is that there was slightly higher human capital in the class that had the proctored exam although this must have occurred by chance since the students did not know if the exams were going to be proctored in advance so there is no issue of selection bias. A Oaxaca decomposition of this difference in grades was conducted to see how much was due to human capital and how much was due to the differences in the rates of return to human capital. This analysis reveals that 17% of the difference was due to the higher human capital with the remaining 83% due to differences in the returns to human capital. It is possible that the un-proctored exam format does not encourage as much studying as the proctored format reducing both the returns to human capital and the exam scores.

Online Format vs. Live Mode of Instruction: Do Human Capital Differences or Differences in Returns to Human Capital Explain the Differences in Outcomes?

Our paper asks the question: Does mode of instruction format (live or online format) effect test scores in the principles of macroeconomics classes? Our data are from several sections of principles of macroeconomics, some in live format, some in online format, and all taught by the same instructor. We find that test scores for the online format, when corrected for sample selection bias, are four points higher than for the live format, and the difference is statistically significant. One possible explanation for this is that there was slightly higher human capital in the classes that had the online format. A Oaxaca decomposition of this difference in grades was conducted to see how much was due to human capital and how much was due to the differences in the rates of return to human capital. This analysis reveals that 25% of the difference was due to the higher human capital with the remaining 75% due to differences in the returns to human capital. It is possible that for the relatively older student with the appropriate online learning skill set, and with schedule constrains created by family and job, the online format provides them with a more productive learning environment than does the alternative traditional live class format. Also, because our data are limited to the student s academic transcript, we recommend future research include data on learning style characteristics, and the constraints formed by family and job choices.

DEVCON: Stata module to apply the deviation contrast transform to estimation results

devcon transforms the coefficients of 0/1 dummy variables so that they reflect deviations from the "grand mean" rather than deviations from the reference category (the transformed coefficients are equivalent to those obtained by the so called "effects coding") and adds the coefficient for the reference category. The variance-covariance matrix of the estimates is transformed accordingly. The transformed estimated can be used with post estimation procedures. In particular, devcon can be used to solve the identification problem for dummy variable effects in the so-called Blinder-Oaxaca decomposition (see the oaxaca package).

Ethnic conflict and economic disparity:Serbians and Albanians in Kosovo

Using the Living Standards Measurement Study (LSMS) household survey from post-conflict Kosovo, we investigate the comparative economic well-being of Serbs and Albanians. An Oaxaca decomposition shows Serb households are both better endowed with income generating characteristics, such as education, and receive higher returns to these characteristics than Albanian households. Despite these advantages, Serb households have lower living standards, on average, than Albanian households. Most of the difference in living standards between Serb and Albanian households is due to unobserved non-economic factors. This result has serious implications for the political economy of policymaking in post-conflict Kosovo.

Oaxaca-Blinder type counterfactual decomposition methods for duration outcomes

Los métodos disponibles para realizar análisis de descomposición que se pueden aplicar cuando los datos son completamente observados, no son válidos cuando la variable de interés es censurada. Esto puede explicar la escasez de este tipo de ejercicios considerando variables de duración, las cuales se observan usualmente bajo censura. Este documento propone un método del tipo Oaxaca-Blinder para descomponer diferencias en la media en el contexto de datos censurados. La validez de dicho método radica en la identificación y estimación de la distribución conjunta de la variable de duración y un conjunto de covariables. Adicionalmente, se propone un método más general que permite descomponer otros funcionales de interés como la mediana o el coeficiente de Gini, el cual se basa en la especificación de la función de distribución condicional de la variable de duración dado un conjunto de covariables. Con el fin de evaluar el desempeño de dichos métodos, se realizan experimentos tipo Monte Carlo. Finalmente, los métodos propuestos son aplicados para analizar las brechas de género en diferentes características de la duración del desempleo en España, tales como la duración media, la probabilidad de ser desempleado de largo plazo y el coeficiente de Gini. Los resultados obtenidos permiten concluir que los factores diferentes a las características observables, tales como capital humano o estructura del hogar, juegan un papel primordial para explicar dichas brechas.

Sources of Regional Income Disparity in Rural Vietnam: Oaxaca-Blinder Decomposition

This paper investigates determinants of regional income disparity in rural Vietnam, with special emphasis placed on the roles of human capital and land. We apply a decomposition method, suggested by Oaxaca and Blinder. We found that returns to assets rather than endowments, especially those of human capital, are one of the leading factors to account for income differences across regions. We also found that substantial improvements of returns to human capital in the Red River delta region are a driving force to catch up with Mekong River delta region. Unexpectedly, differences in land endowment do not strongly correlate with regional income disparity because better access to land in a region was partially offset by lower returns.