This paper studies 18s the economic efficiency of small holder maize farmers in Zambia. It uses a two stage analysis approach. In the first stage, efficiency indices are generated using a non parametric technique, the Data Envelopment Analysis (DEA). In the second stage the indices generated are regressed onto farm and farmer characteristics to identify factors affecting small holder technical and allocative efficiency in maize production. The DEA results indicate very low levels of technical and allocative efficiency among smallholder maize farmers. On average, farmers are 30 percent technically and 12 percent allocatively efficient. Regression results have shown that use of certified hybrid seed, livestock ownership, and education attainment of household head can improve both technical and allocative efficiency. Other factors that could enhance technical efficiency include involvement in community agricultural activities and the use of mechanized tillage methods and organic fertilizer. These low efficiency levels signaled by the results confirm the long-held view that most Zambian maize farmers operate far below the potential of existing maize production technologies. The results also suggest that productivity (and efficiency) is as much a public issue as it is private. Human and physical capital attributes of the farm household, for example, are hugely significant. This suggests the need for more attention to broad based, productivity- and output market-enhancing investments.