There are several limitations to this study. First,
pain measured at four time points may have been a
burden to some women during the latter part of labor,
when increased unreliability of the VAS scales was
found. Approximately 10 participants marked their
VAS scales quickly, seemingly without much thought.
Second, the restriction to only five types of music also
may have had an effect on the choices of a few
women; the music that they selected may not have
helped as much as a more preferred selection. Third,
Music for Labor Pain 59
the ceiling effect on the VAS in the latter part of the
study may have occurred because the primiparas reported
their pain was fairly severe during early labor,
but had not experienced the greater pain of advanced
labor; they may have prematurely marked close to “the
most pain imaginable” anchor. Finally, significant box
tests indicated that the assumption of equality of variance
was violated for both sensation and distress
across time, however the repeated measures ANCOVA
was considered robust because the groups were randomized
and equal in size (Green, Salkind, & Akey,
2000; Stevens, 1996).