Estilos de Respuesta y Correlaciones Residuales: Efectos Correcciones y Consecuencias

Identificador:  TDX:4303

Handle: https://hdl.handle.net/20.500.11797/TDX4303

Autors:  Hernández Dorado, Ana

Resum:
As of today, the interest in controlling variance related to content in personality measures remains prominent. However, studies typically focus on examining a single determinant, and the assessment of the combined impact of multiple sources of error seems to have taken a back seat. That is why it is necessary to determine the potential impact of two non-content-related determinants when they occur simultaneously. The most common sources of error are acquiescence and correlated residuals. Both individually have the potential to distort parameter estimation and assessment of fit, and their presence is plausible. To study the joint impact, two procedures were developed and implemented in R to detect and control acquiescence (SIREN) and correlated residuals (MORGANA). Starting from these two methods, a combined procedure was proposed, and the three types of detection were compared in databases that simultaneously presence acquiescence and residuals.