In measurement, validity is traditionally defined as an argument in support of a construct, made using data. Although this definition might seem simple, its meaning changes as investigators adopt different scientific philosophies, descriptions of the entities to be measured, and norms for acceptable data. Uncertainty regarding what is to be measured has taught researchers about the fallibility of all forms of measurement. Rather than expect perfection, most researchers accept that new knowledge can be generated only if important ideas and events are assessed well enough to support meaningful conclusions. Assessment guidelines from three general approaches to measurement—labeled here as psychometric, developmental, and interpretive—reveal how investigators generate valid evidence and respond to the inevitable threats that might undermine even the strongest research design if left unchecked. ...
Looks like you do not have access to this content.