Abstract:
This qualitative study used interview transcripts to find clues to attrition from a
longitudinal study among nonpartnered, rurally-located, poor mothers, a
generally vulnerable population with characteristics commonly associated with
attrition (low-income, low levels of educational attainment, and rural location).
Drawing on Rural Families Speak data, the author used the number of family
members and friends living in close proximity to mothers, and mothers'
descriptions of their family members and friends (including boyfriends) and the
quality of their relationships with these individuals to predict whether mothers
would drop out of the study after Wave 1 (attriters), stay in the study through
Wave 3 despite moving (continuer-movers), or stay in the study through Wave 3
while remaining in place (continuer-nonmovers). Analyzing data through the lens
of the affect theory of social exchange, the author was unable to consistently
predict the groups to which mothers belonged. Trends supported the use of
affect theory of social exchange as preferable to classic social exchange theory
in predicting outcomes with this sample. In ad hoc investigations, the author
found that attriters and continuer-movers differed significantly with respect to age
and education, and that these two demographic variables were useful in
predicting outcomes for these two groups. The findings also revealed the
importance of establishing consistent protocols in longitudinal research,
particularly when data are collected from widely dispersed geographic locations.
The findings led the researcher to recommend using a more contextual
framework and a mixed methods approach to studying attrition and informal
social support. The author recommends that future, similar research include
interviews of continuers to learn what factors contribute to their decisions to stay
in longitudinal studies, and that adequate funding for the implementation and
oversight of consistent protocols be provided.
