A cross-sectional , comparative study was conducted in rural Sulultarnworeda between September I and January 30, 1994 to assess gender bias in thernallocation of household resources for children 5-14 years of age. Data werernobtained through a questionnaire survey and focus group discussions. Caretakersrnfor a total of 1785 children, including 903 boys (50.6%) and 882 girls (49.4%),rnwere interviewed during a survey of every fifth household in 17 randomlyrnselected farmers associations.rnNutritional practices were found to favour boys, with male children morernlikely to receive highly valued foods including meat (risk ratio [RR] = 1.24, 95 %rnconfidence intervals [CI] = 1.19-1.28), milk (RR=2.74, CI=2.43-3.09), eggrn(RR=2.75 , CI=1.39-5.44), butter (RR=1.45, CI=1.38-1.53), and lentilsrn(RR= 1.02, CI = 1.00-1.03). No statistical differepce hy gender was ohserved forrnduration of breast feeding or age at introduction of complement;n)' foods.rnBoys were given greater access to educa!io:!al ~pportunities, w;th enrolmentrnin school significantly higher among m?J!" r.bilciren (RR=1.27. CI=1.05-1.53).rnBoys were more likely to be immunized (RR=1.28, CI=1.03-1.58), and morernlikely to receive the care of a health professional when ill (RR= 1.12, CI= 1.05-rn1.19). Girls were more likely to work at household chores for 12 or more hoursrnper day (RR=1.31 , CI=1.16-1.49) and were more often expected to marry atrnless than 20 years of age (RR=25 .65 , CI=18. 15-36.25).rnIVrnOther parental expectations for the future of their children, including earningrnpower and inheritance patterns, also favoured male children. A special programrnto improve the status of the girl child is recommended.