Background: Nutritional status especially that of preschool children is a sensitive indicator of health and nutritional status of a community. Aim: To estimate the prevalence of undernutrition among tribal preschool children and to assess the factors associated with variation in nutritional status. Methods: A cross-sectional survey was conducted among 438 tribal preschool children in 10 clusters of Wayanad district of Kerala. Height, weight, mid-arm circumference and hemoglobin level were measured. Children more than two standard deviations (SDs) below the standard median of World Health Organization Multi Centric Growth Reference Study were considered underweight (weight-for-age), stunted (height-for-age) and wasted (weight-for-height) respectively. Qualitative variables such as prevalence of underweight, stunting and wasting were summarized using percentages. Mean (SD) was used, to summarize, quantitative variables such as height and weight. Generalized estimating equation models were constructed to assess associations. Adjusted models included social factors and child morbidities. Results: More than half of the children say 58.7% (257/438) had deficits in at least one of the three anthropometric indicators. The prevalence of underweight, stunting and wasting was 39% (171/438), 38% (167/438) and 20.5% (90/438) respectively. The prevalence of anemia was 95.7% (419/438). Bivariate analysis showed significant associations between undernutrition and educational status of parents, tribe to which the child belonged, diarrheal episode and low birth weight. In adjusted analysis, lower educational status of mother (adjusted odds ratio 1.9, 95% confidence intervals 1.2-3) and an episode of diarrhea (1.8, 1.03-3.2) independently predicted undernutrition in a tribal preschool child. Conclusion: Undernutrition in the form of stunting, wasting and underweight is very high among the tribal preschool children. There is an urgent need to improve health care services to the tribal population and tribal children.