Читать книгу A Companion to Medical Anthropology онлайн

215 страница из 242

Systematic observation differs from participant observation because it imposes more structure on sampling and measurement (Johnson and Sackett 1998). For that reason, methods of systematic observation – including continuous monitoring, spot sampling, and time allocation – are best suited to confirmatory research questions. Systematic observation deserves wider use in medical anthropology, because many research questions concern what people do, not just what they say. And there is ample evidence that what people say is seldom a good proxy for what they do (Bernard et al. 1984).

Vitzthum (1994) studied concordance between maternal recall and systematic observation of breastfeeding in the Peruvian Andes. She interviewed 30 Nuñoa women with children under three years of age and asked each woman to estimate how often and how long her child breastfed each day. Vitzthum then monitored a subset of 10 women over a total of 86 hours and recorded the frequency and duration of breastfeeding to the nearest second. She found little association between observational and recall data: Women generally underreported frequency and overreported duration of breastfeeding, but not in a consistent pattern. If we rely only on maternal recall – as epidemiologic studies of breastfeeding often do – we are likely to make mistakes (see also Li et al. 2005; Miller et al. 2013).