Random loudspeaker repositioning across listening seats eliminates room-position bias

Evidence

Rigorous tests used one listener at a time. After each sequence of music, during which listener(s) made notes and decided on “fidelity” ratings, the loudspeakers were randomly repositioned, and the listeners took different seats. This process was repeated until all loudspeakers had been heard in all locations. It was time-consuming, but it reduced the biasing influence of room position, which in some cases could be the deciding factor. — Toole, Floyd E, Sound reproduction- the acoustics and…, 1102

Categorization

• XREF: Connects to the classic experimental-design insight that randomization controls nuisance variables — parallels randomized block designs in statistics.