Few studies have explored whether or how nutrition knowledge interacts with education--do they act as substitutes or complements? Exceptions include research in Nicaragua (Lamontagne Engle and Zeitlin 1998) which found that maternal education and certain types of nutrition knowledge are significantly but independently associated with child outcomes. Another study in Brazil (Thomas Strauss and Henriques 1990) found that most of the correlation between maternal knowledge and child height could be explained by mothers’ access to media messages (on TV and radio) and that formal schooling and messages gained through community health services acted as substitutes. Similarly Glewwe (1999) found that maternal knowledge (rather than schooling) in Morocco strongly influences child height-for-age (a measure of longer-term child nutritional well-being) and that such knowledge is obtained mainly outside the classroom through the media from relatives and via public service messages. Variyam et al (1999 p. 381) also report that maternal schooling in the United States has a strong impact on children’s diets “wholly through its positive effect on maternal . . . nutrition knowledge.” Thus some of the contribution of schooling to child nutrition appears to come through its interaction with nutrition knowledge--possibly by enhancing women’s ability to acquire knowledge or by enabling them to put such knowledge into practice. However since nutrition knowledge has been shown to generate nutritional improvements even among illiterate populations and since formal education remains severely limited in most poor developing countries the potential for targeted transfers of nutrition information to assist in nutritional improvements may be large. This article offers preliminary support for such an argument. The article is organized as follows. Section II describes the survey data used in the analysis as well as the construction of our proxy for nutrition knowledge. Section III presents nonparametric evidence of the effects of nutrition knowledge maternal education and per capita expenditures on nutritional outcomes. Section IV presents supporting parametric results and Section V concludes with programmatic and research implications. (excerpt)