Social Science and Medicine 52 (2001) 53–69 Diffusion of ideas about personal hygiene and contamination in poor countries: evidence from Guatemala Noreen Goldman a, *, Anne R. Pebley b , Megan Beckett c a Office of Population Research, Princeton University, 21 Prospect Avenue, Princeton, NJ 08544-2091, USA b School of Public Health, UCLA, 650 Charles E Young Drive South, Los Angeles, CA 90095-1772, USA c Population Center, RAND, PO Box 2138, 1700 Main Street, Santa Monica, CA 90407-2138, USA Abstract In this paper, we explore the diffusion of beliefs pertaining to the causes of childhood diarrhea in rural Guatemala. The analysis focuses on the importance of interpersonal and impersonal contacts as conduits for information and norms related to hygiene and contamination. Estimates from multivariate models reveal that there is evidence of a diffusion process through social contacts, primarily through interpersonal ones. The analysis also identifies striking differences between (1) the diffusion process related to hygiene (e.g. dirtiness) and that related to contamination (e.g. pathogens); and (2) beliefs about the causes of diarrheal illness among children in general and those among respondents’ own children. # 2000 Elsevier Science Ltd. All rights reserved. Keywords: Diffusion; Hygiene; Contamination; Diarrhea; Guatemala Introduction The large and sustained reduction in mortality that has taken place in virtually all countries in the past 200 years must be counted as one of the most dramatic social changes in human history. Much of the decline is due to decreased mortality from infectious causes, i.e. from diseases that are caused by the growth of pathogenic microorganisms within the body and that are usually transmittable to others. Although there was considerable controversy in the past about the reasons for this mortality decline, demographers are now in general agreement that the declines can be linked to two broad causes: (1) increases in income and associated improvements in nutrition and standards of living; and (2) factors generally thought to be exogenous to income levels, such as public works (e.g. sanitation systems and water quality improvements), public health interventions (quarantines, mosquito eradication, vaccination), and improvements in thera- peutic medical techniques (e.g. development of effective drugs, adoption of aseptic practice). For example, Preston (1976, p. 83) concludes that approx. 10–25% of the growth in life expectancy worldwide between the 1930s and the 1960s was attributable to increases in real income, while the remaining 75–90% was due to ‘‘the activities of ‘exogenous’ medical and public health factors’’. Palloni (1990) reports that results of a similar analysis for 23 countries in Latin America show that growth in real income accounts for 45% of mortality decline in the period 1945–65 and 75% in the period 1965–85. These explanations generally cast individuals and families as relatively passive actors in a process controlled primarily by macro-level forces, including macro-economic changes and programs developed and implemented by social and political elites. 1 However, *Corresponding author. Fax: +1-609-258-1039. E-mail address: ngoldman@opr.princeton.edu (N. Gold- man). 1 Exceptions include recent work by Nathanson (1996) which considers the role of ‘‘grass roots’’ social movements in the United States in pushing for and implementing public health programs. 0277-9536/00/$ - see front matter # 2000 Elsevier Science Ltd. All rights reserved. PII:S0277-9536(00)00122-2