Validity of Tests

RESUMO Objetivo: Identificar, na literatura, o uso da Psicometria na área da Fonoaudiologia e os processos utilizados para a busca de evidências de validade dos instrumentos da área. Estratégia de pesquisa: As bases pesquisadas foram MEDLINE (acessado via PubMed), LILACS, Scopus e SciELO. Os descritores foram “Validation studies”, “Validity of tests”, “Speech, Language and Hearing Sciences” e “Valid” (seguido de elemento de truncagem), em português, inglês e espanhol. Critérios de seleção: Foram incluídos os estudos que realizavam algum tipo de validação de testes referentes a áreas da Fonoaudiologia. Resultados: Foram encontrados 296 artigos e destes, apenas 48 foram incluídos. A maioria dos estudos foi publicada por periódicos internacionais, da área da Fonoaudiologia e com amostra de ampla faixa etária. A principal área avaliada pelos instrumentos foi linguagem (20 estudos), seguida por audiologia (13 estudos). O ano de maior publicação foi 2014 e o principal tipo de busca de evidências de validade foi com base na estrutura interna. Conclusão: O uso dos princípios de busca de evidências de validade de instrumentos da área fonoaudiológica ainda é escasso. Porém, observa-se que a maior parte dos estudos foi desenvolvida nos últimos anos, demonstrando tendência atual para atenção à necessidade de aprimoramento dos instrumentos.

Download Full-text

A Test of a Proposed Method for Estimating Validity of a Multivariate Composite Predictor: Extending the Job Component Validity Model

Psychological Reports ◽

10.2466/pr0.105.3.900-916 ◽

2009 ◽

Vol 105 (3) ◽

pp. 900-916 ◽

Cited By ~ 2

Author(s):

Calvin C. Hoffman ◽

David Morris ◽

Gypsi Luck

Keyword(s):

Job Analysis ◽

Test Battery ◽

Validity Study ◽

Multiple Correlation ◽

Position Analysis ◽

Supervisory Ratings ◽

Validity Of Tests ◽

Correlation Procedure ◽

Job Component Validity

In this study, a proposed extension to the job component validity model from the Position Analysis Questionnaire was tested. Job component validity, a form of synthetic validation, allows researchers to select useful predictors and to estimate the criterion-related validity of tests based on conducting a job analysis which includes the Position Analysis Questionnaire. Morris and colleagues described a method for estimating the multiple correlation of a test battery assembled via job component validity estimates. In the current study, job component validity estimates, derived from the multiple correlation procedure proposed by Morris, et al., were compared to unit-weighted validity estimates obtained in a criterion-related validity study of six job progressions. The multivariate job component validity estimates were comparable to unit-weighted validity coefficients obtained using supervisory ratings as criteria. Multivariate job component validity estimates were conservative compared to corrected unit-weighted validity coefficients.

Download Full-text

GEL METHODS FOR NONSMOOTH MOMENT INDICATORS

Econometric Theory ◽

10.1017/s0266466610000137 ◽

2010 ◽

Vol 27 (1) ◽

pp. 74-113 ◽

Cited By ~ 13

Author(s):

Paulo M.D.C. Parente ◽

Richard J. Smith

Keyword(s):

Empirical Likelihood ◽

Method Of Moments ◽

Jacobian Matrix ◽

Generalized Method Of Moments ◽

Gmm Estimator ◽

First Order ◽

Generalized Empirical Likelihood ◽

Finite Samples ◽

Class Of Estimators ◽

Validity Of Tests

This paper considers the first-order large sample properties of the generalized empirical likelihood (GEL) class of estimators for models specified by nonsmooth indicators. The GEL class includes a number of estimators recently introduced as alternatives to the efficient generalized method of moments (GMM) estimator that may suffer from substantial biases in finite samples. These include empirical likelihood (EL), exponential tilting (ET), and the continuous updating estimator (CUE). This paper also establishes the validity of tests suggested in the smooth moment indicators case for overidentifying restrictions and specification. In particular, a number of these tests avoid the necessity of providing an estimator for the Jacobian matrix that may be problematic for the sample sizes typically encountered in practice.

Download Full-text