Contexto: devido à crescente necessidade de melhor conhecer os elementos organizacionais, as medições subjetivas estão ganhando um espaço cada vez maior. Problema: podemos avaliar o quão confiável em relação à realidade do atributo de um mensurando é uma amostra de dados obtidos por meio de medição subjetiva bivalorada? Hipóteses: Hipótese 1 - já existe um método científico que consiga realizar eficazmente a avaliação citada no problema; Hipótese 2 - caso a primeira hipótese não seja satisfeita, é possível criar um método que possa realizar eficazmente aquela avaliação. Objetivo: após confirmada a inexistência de um método consagrado cientificamente que consiga realizar a avaliação citada no problema, então, desenvolver um método que realize eficazmente aquela avaliação. Métodos: elaboração do embasamento teórico sobre os fundamentos da medição subjetiva; revisão bibliográfica sobre os métodos existentes que trabalham com medição subjetiva bivalorada; elaboração e validação de um método que solucione o problema proposto. Escopo e limitações: as publicações utilizadas para elaborar a fundamentação teórica sobre medição subjetiva e o novo método são multidisciplinares, porém o enfoque dado aos estudos foi para a gestão organizacional, com ênfase nos assuntos pertinentes à Engenharia de Produção e áreas afins. Resultados obtidos: (1) foram elaborados os postulados sobre medição subjetiva; (2) foram delineadas as principais fontes de erros de medição subjetiva baseada naqueles postulados; (3) não foi encontrado na literatura pertinente qualquer método científico ou expressão matemática que solucione o problema proposto; (4) foi elaborado um método para trabalhar com medição subjetiva bivalorada. Análise dos resultados: (1) os postulados elaborados serviram de base para o levantamento das principais fontes de erros de medições subjetivas; (2) o embasamento teórico serviu como fundamentação para a elaboração do método preliminar que utilizava a medição subjetiva bivalorada; (3) novos estudos permitiram um aperfeiçoamento do método preliminar, elaborando o novo método; (4) um estudo de caso mostrou a aplicabilidade do novo método. Conclusões: os estudos realizados permitiram comprovar a eficácia do método para medições subjetivas bivaloradas. / Background: Due to the growing need to better understand the organizational elements, subjective measurements are gaining more and more space. Problem: can we assess how reliable a data sample collected via 2-value subjective measurement is in relation to the reality of the attribute of a measurand? Hypotheses: Hypothesis 1 - there is already a scientific method that can effectively perform the assessment mentioned in the problem; Hypothesis 2 - in the even the first hypothesis is not met, it is possible to create a method that can effectively perform that assessment. Purpose: After confirming the absence of a scientifically established method that can perform the evaluation mentioned in the problem, the purpose is to develop a method to effectively perform that evaluation. Methods: Development of theoretical basis on the grounds of subjective measurements; literature review on existing methods that work with 2-value subjective measurement; development and validation of a new method that solves the problem proposed. Scope and limitations: the publications used to develop the theoretical framework of subjective measurements and the new method are multidisciplinary, yet the focus given to the studies was more related to organizational management, with emphasis on matters pertaining to Production Engineering and related fields. Results: (1) postulates on subjective measurements have been elaborated; (2) the main sources of subjective measurement errors based on those postulates have been outlined; (3) no scientific method or mathematical expression that may address the problem proposed has been found on the literature; (4) a method was developed to work with two-value subjective measurements. Analysis of results: (1) the postulates developed served as a basis for raising the main sources of subjective measurement errors; (2) the theoretical framework served as the basis for establishing the primary method that used two-value subjective measurements; (3) new studies allowed for the improvement of the primary method by developing the new method; (4) a case study showed the applicability of the new method. Conclusions: The studies have allowed us to effectiveness of the method for 2-value subjective measurements.

Contexto: devido à crescente necessidade de melhor conhecer os elementos organizacionais, as medições subjetivas estão ganhando um espaço cada vez maior. Problema: podemos avaliar o quão confiável em relação à realidade do atributo de um mensurando é uma amostra de dados obtidos por meio de medição subjetiva bivalorada? Hipóteses: Hipótese 1 - já existe um método científico que consiga realizar eficazmente a avaliação citada no problema; Hipótese 2 - caso a primeira hipótese não seja satisfeita, é possível criar um método que possa realizar eficazmente aquela avaliação. Objetivo: após confirmada a inexistência de um método consagrado cientificamente que consiga realizar a avaliação citada no problema, então, desenvolver um método que realize eficazmente aquela avaliação. Métodos: elaboração do embasamento teórico sobre os fundamentos da medição subjetiva; revisão bibliográfica sobre os métodos existentes que trabalham com medição subjetiva bivalorada; elaboração e validação de um método que solucione o problema proposto. Escopo e limitações: as publicações utilizadas para elaborar a fundamentação teórica sobre medição subjetiva e o novo método são multidisciplinares, porém o enfoque dado aos estudos foi para a gestão organizacional, com ênfase nos assuntos pertinentes à Engenharia de Produção e áreas afins. Resultados obtidos: (1) foram elaborados os postulados sobre medição subjetiva; (2) foram delineadas as principais fontes de erros de medição subjetiva baseada naqueles postulados; (3) não foi encontrado na literatura pertinente qualquer método científico ou expressão matemática que solucione o problema proposto; (4) foi elaborado um método para trabalhar com medição subjetiva bivalorada. Análise dos resultados: (1) os postulados elaborados serviram de base para o levantamento das principais fontes de erros de medições subjetivas; (2) o embasamento teórico serviu como fundamentação para a elaboração do método preliminar que utilizava a medição subjetiva bivalorada; (3) novos estudos permitiram um aperfeiçoamento do método preliminar, elaborando o novo método; (4) um estudo de caso mostrou a aplicabilidade do novo método. Conclusões: os estudos realizados permitiram comprovar a eficácia do método para medições subjetivas bivaloradas. / Background: Due to the growing need to better understand the organizational elements, subjective measurements are gaining more and more space. Problem: can we assess how reliable a data sample collected via 2-value subjective measurement is in relation to the reality of the attribute of a measurand? Hypotheses: Hypothesis 1 - there is already a scientific method that can effectively perform the assessment mentioned in the problem; Hypothesis 2 - in the even the first hypothesis is not met, it is possible to create a method that can effectively perform that assessment. Purpose: After confirming the absence of a scientifically established method that can perform the evaluation mentioned in the problem, the purpose is to develop a method to effectively perform that evaluation. Methods: Development of theoretical basis on the grounds of subjective measurements; literature review on existing methods that work with 2-value subjective measurement; development and validation of a new method that solves the problem proposed. Scope and limitations: the publications used to develop the theoretical framework of subjective measurements and the new method are multidisciplinary, yet the focus given to the studies was more related to organizational management, with emphasis on matters pertaining to Production Engineering and related fields. Results: (1) postulates on subjective measurements have been elaborated; (2) the main sources of subjective measurement errors based on those postulates have been outlined; (3) no scientific method or mathematical expression that may address the problem proposed has been found on the literature; (4) a method was developed to work with two-value subjective measurements. Analysis of results: (1) the postulates developed served as a basis for raising the main sources of subjective measurement errors; (2) the theoretical framework served as the basis for establishing the primary method that used two-value subjective measurements; (3) new studies allowed for the improvement of the primary method by developing the new method; (4) a case study showed the applicability of the new method. Conclusions: The studies have allowed us to effectiveness of the method for 2-value subjective measurements.

As the learning outcomes movement gains strength, the need to effectively measure learning outcomes becomes more important. This study looked at the effectiveness of self-reported mastery in measuring learning outcomes by examining the correlations between (a) self-reported mastery, (b) self-reported gains, and (c) objective measures of learning outcomes. The objective measures of learning outcomes were final exams for two classes, Calculus (consisting of two forms) and Statistics. The self-reported mastery and self-reported gains items were taken from the pilot student ratings form and the old student ratings form. A total of 848 undergraduate students completed the final exam and the two student ratings forms. The summed total of the self-reported mastery items correlated at a medium strength with objective measures of learning outcomes (Calculus Form A: r = .436; Calculus Form B: r = .361; Statistics: r = .416). The relationship between self-reported gains and objective measures of learning outcomes was weaker than that of self-reported mastery and objective measures of learning outcomes (a difference of .276 for Calculus Form A, .138 for Calculus Form B, .110 for Statistics). The relationship between self-reported gains and self-reported mastery was stronger than the other two relationships (Calculus Form A: r = .473, Calculus Form B: r = .500, Statistics: r = .628). A confirmatory factor analysis produced even stronger relationships between the three latent variables, including differences between the two forms of the Calculus exam. Self-reported mastery may be more effective at measuring objective measures of learning outcomes than self-reported gains, but self-reported mastery cannot completely serve as a proxy for objective measures of learning outcomes. Administrators or researchers measuring learning outcomes on a large scale may benefit by administering self-reported mastery items instead of self-reported gains items.

L’utilisation des mesures subjectives en épidémiologie s’est intensifiée récemment, notamment avec la volonté de plus en plus affirmée d’intégrer la perception qu’ont les sujets de leur santé dans l’étude des maladies et l’évaluation des interventions. La psychométrie regroupe les méthodes statistiques utilisées pour la construction des questionnaires et l’analyse des données qui en sont issues. Ce travail de thèse avait pour but d’explorer différents problèmes méthodologiques soulevés par l’utilisation des techniques psychométriques en épidémiologie. Trois études empiriques sont présentées et concernent 1/ la phase de validation de l’instrument : l’objectif était de développer, à l’aide de données simulées, un outil de calcul de la taille d’échantillon pour la validation d’échelle en psychiatrie ; 2/ les propriétés mathématiques de la mesure obtenue : l’objectif était de comparer les performances de la différence minimale cliniquement pertinente d’un questionnaire calculée sur des données de cohorte, soit dans le cadre de la théorie classique des tests (CTT), soit dans celui de la théorie de réponse à l’item (IRT) ; 3/ son utilisation dans un schéma longitudinal : l’objectif était de comparer, à l’aide de données simulées, les performances d’une méthode statistique d’analyse de l’évolution longitudinale d’un phénomène subjectif mesuré à l’aide de la CTT ou de l’IRT, en particulier lorsque certains items disponibles pour la mesure différaient à chaque temps. Enfin, l’utilisation de graphes orientés acycliques a permis de discuter, à l’aide des résultats de ces trois études, la notion de biais d’information lors de l’utilisation des mesures subjectives en épidémiologie. / Recently, subjective measurements have increasingly been used in epidemiology, alongside the growing will to integrate individuals’ point of view on their health in studies on diseases or health interventions. Psychometrics includes statistical methods used to develop questionnaires and to analyze questionnaire data. This doctoral dissertation aimed to explore methodological issues raised by the use of psychometric techniques in epidemiology. Three empirical studies are presented and cover 1 / the validation stage of a questionnaire: the objective was to develop, using simulated data, a tool to determine sample size for internal validity studies on psychiatric scale; 2 / the mathematical properties of the subjective measurement: the objective was to compare the performances of the minimal clinically important difference of a questionnaire, assessed on data from a cohort study, computed using the classical test theory (CTT) framework or the item response theory framework (IRT); 3 / its use in a longitudinal design: the objective was to compare, using simulated data, the performances of a statistical method aimed to analyze the longitudinal course of a subjective phenomenon measured using the CTT or IRT framework, especially when some of the available items used for its measurement differ at each time of data collection. Finally, directed acyclic graphs were used to discuss the results from these three studies and the concept of information bias when subjective measurements are used in epidemiology.

L’utilisation des mesures subjectives en épidémiologie s’est intensifiée récemment, notamment avec la volonté de plus en plus affirmée d’intégrer la perception qu’ont les sujets de leur santé dans l’étude des maladies et l’évaluation des interventions. La psychométrie regroupe les méthodes statistiques utilisées pour la construction des questionnaires et l’analyse des données qui en sont issues. Ce travail de thèse avait pour but d’explorer différents problèmes méthodologiques soulevés par l’utilisation des techniques psychométriques en épidémiologie. Trois études empiriques sont présentées et concernent 1/ la phase de validation de l’instrument : l’objectif était de développer, à l’aide de données simulées, un outil de calcul de la taille d’échantillon pour la validation d’échelle en psychiatrie ; 2/ les propriétés mathématiques de la mesure obtenue : l’objectif était de comparer les performances de la différence minimale cliniquement pertinente d’un questionnaire calculée sur des données de cohorte, soit dans le cadre de la théorie classique des tests (CTT), soit dans celui de la théorie de réponse à l’item (IRT) ; 3/ son utilisation dans un schéma longitudinal : l’objectif était de comparer, à l’aide de données simulées, les performances d’une méthode statistique d’analyse de l’évolution longitudinale d’un phénomène subjectif mesuré à l’aide de la CTT ou de l’IRT, en particulier lorsque certains items disponibles pour la mesure différaient à chaque temps. Enfin, l’utilisation de graphes orientés acycliques a permis de discuter, à l’aide des résultats de ces trois études, la notion de biais d’information lors de l’utilisation des mesures subjectives en épidémiologie. / Recently, subjective measurements have increasingly been used in epidemiology, alongside the growing will to integrate individuals’ point of view on their health in studies on diseases or health interventions. Psychometrics includes statistical methods used to develop questionnaires and to analyze questionnaire data. This doctoral dissertation aimed to explore methodological issues raised by the use of psychometric techniques in epidemiology. Three empirical studies are presented and cover 1 / the validation stage of a questionnaire: the objective was to develop, using simulated data, a tool to determine sample size for internal validity studies on psychiatric scale; 2 / the mathematical properties of the subjective measurement: the objective was to compare the performances of the minimal clinically important difference of a questionnaire, assessed on data from a cohort study, computed using the classical test theory (CTT) framework or the item response theory framework (IRT); 3 / its use in a longitudinal design: the objective was to compare, using simulated data, the performances of a statistical method aimed to analyze the longitudinal course of a subjective phenomenon measured using the CTT or IRT framework, especially when some of the available items used for its measurement differ at each time of data collection. Finally, directed acyclic graphs were used to discuss the results from these three studies and the concept of information bias when subjective measurements are used in epidemiology.