The Promises and pitfalls of making national educational assesment adaptive: America's assessment as an example

• Autores: Howard Wainer
• Localización: Metodología de las ciencias del comportamiento, ISSN 1575-9105, Vol. 5, Nº. 2, 2004 (Ejemplar dedicado a: Validez de los test / coord. por Gerardo Prieto Adánez), págs. 211-224
• Idioma: inglés
• Texto completo no disponible (Saber más ...)
• Resumen

  • The most contemporary view of the validity of a measurement instrument (Mislevy, Steinberg & Almond, 2003) is that it is the extent to which that instrument provides relevant and credible evidence for the inferences that will be drawn from that instrumento In this essay 1 examine the advantages and disadvantages of making the administration of America's National Assessment of Educational Progress (NAEP) adaptive both with respect to its validity (in the sense just mentioned) and its efficiency. 1 conclude that adaptivity, taken in the usual sense in which individual items are chosen to maximize the potential information gained about a specific examinee's proficiency, is likely to be inappropriate in this context, but that a different form of sequential item choice mar provide some help in improving the efficiency of the survey. However, if making NAEP adaptive means administering it by computer, a substantial number of problems emerge that mar be very expensive to overcome. One scheme for making NAEP adaptive without using computers is suggested.