APA Dictionary of Psychology

an expression indicating that if the data used in analyses are not reliable or coherent, the results will not prove useful. The phrase was first used in computer science, cautioning users against trusting computer output derived from unreliable input. Although the term is not used much currently, it remains a relevant reminder to researchers to pay attention to how studies are designed and how data are collected, and not to depend on analyses to produce meaningful results when the input data are not themselves meaningful.