The rapid growth of secondary cities in Kenya is explored through an analysis of various factors related to development. Through a shift-share analysis of employment figures it is shown that, while employment expansion and diversification has occurred in some of the secondary cities, this does not fully explain the population growth which has taken place. Road improvements facilitated rural out-migration while perceived employment opportunities encouraged in-migration to smaller urban centers. While these factors do not fully correspond to the growth which has taken place, it is suggested that they do reveal certain growth patterns which are occurring. Finally, it is suggested that the growth of Kenyan secondary cities is a positive step towards a more equitable and efficient spatial distribution of the population and that, through their development, secondary cities can play an important role in diminishing rural-urban differences and unifying Kenya's space economy. (Abstract shortened with permission of author.)

Studies of African urban development have used the term “peri-urban" to refer to residential development at a city’s edge. No one, however, has developed a precise definition of peri-urban areas which guides this research. This study attempts to define peri-urban areas using an urban economic model and analyzes the African peri-urban literature in the context of this model. This review of the literature produces several hypotheses, which are then tested using data from a housing survey done in thirty-two Kenyan cities by the Kenyan government. The first set of hypotheses examined ways in which peri-urban areas differed from central city areas in their physical development. Houses in peri-urban areas were found to be newer, lower quality, built on land held informally, and with lower levels of urban services. The second set of hypotheses concerned the development of rental housing in peri-urban areas. Homeownership rates were higher in peri-urban areas than in central city areas, although a person renting a home in a peri-urban area was likely to have better accommodations than a peri-urban homeowner. Finally, the cities in the survey were grouped according to city size and growth rate, and the effects of these variables on the development of peri-urban areas were tested. The results of this section were somewhat inconclusive, as other factors, such as environment and politics, influence urban development. The conclusion of this research is that peri-urban areas are different from central city areas due primarily to the fact of their recent urban development on formerly agricultural land. This study was just a small step in understanding patterns of peri-urban development. The peri-urban zone can provide a valuable unit of analysis for studying the process of urban development, provided it is rigorously defined. / Master of Urban and Regional Planning