This paper is housed on contraceptive switching in indonesia. Since a woman might use more than one method of contraception during her reproductive period she can contribute more than one contraceptive use interval to the data. Some studies of contraceptive switching have used statistical techniques which assume independence between switching but this assumption can be violated because the likelihood of switching might be correlated. Hence contraceptive switching varies not only across groups of contraceptive users but also across contraceptive users with the same socioeconomic characteristics. Hence the consequences of this contraceptive users' eject for the relationship between socioeconomic. demographic and contraceptive-related characteristics of users and contraceptive switching in indonesia is analysed through the use of random-ejects logistic models. The data used for the analysis is from the 1991 Indonesia Demographic and Health Survey. The findings confirm the results of previous studies that a number of socioeconomic and demographic factors influence the likelihood of switching contraception. Among these factors. contraceptive-related /actors such as the method being used before switching, the duration of use at switching and the reason for switching, affect contraceptive switching most. The results also indicate that there is correlation between intervals for the same woman. This implies that there are other factors which have not been observed and contribute to the differentials of switching contraception across women with the saute socioeconomic characteristics.
