This report provides detailed information on levels and trends of fertility in the 78 districts of Malaysia for 1970 and 1980. Although Malaysia's vital statistic system produces accurate data on fertility at the local level, these statistics have, until recently, been classified by the place of the occurrence of the birth, rather than the usual residence of the mother. This tends to exaggerate the fertility of districts with centrally located maternity hospitals. The fertility rates presented in this report do not have this drawback. The report will be of interest for its use of complex statistical techniques of estimation, which have rarely been used before in the developing countries of the region. The main objectives of the study were: to use the census data for 1970 and 1980 to estimate fertility levels for small geographical areas of the particiapting countries, and to utilize such estimates and other such data to study the spatial aspects of fertility changes at the sub-national level. Chapter I provides background information on the socioeconomic characteristics of Malaysia, on its statistical system and on the principal data sources used in the study. Chapter II describes the 4 techniques used in making the district fertility estimates presented in the report. These are the Bogue-palmore, Rele, synthetic and synthetic-regression methods. Chapter III presents the results of the study as wll as the residence-based vital registration data for 1982. On the whole, the final estimates were mainly those derived from the sythetic and synthetic-regression methods. The results obtained show that the estimates for 1982 based onthe district-level regressions are more in line with the residence-based vital registration data. Further work is needed to provide a more complete understanding of the marked differences in the Bogue-Palmore regression coefficients developed from district-level data and those developed from cross-national level data. A main issue raised is the development of suitable criteria for the selection of a preferred set of estimates.