The present study tried to examine the important determinants of the nuptialityrnvariables 'age at first marriage' and 'the marital stability' ofreproductivernwomen in urban Addis Ababa for the time period of 1990 and 1995. Further, alirnattempt is also made to see whether there is any change in the importance ofrnvariables in affecting the above two dependent variables ofnuptiality over time.rnIt is hypothesised that educational level, age at first marriage and migrationrnstatuses of women are the three important factors that affect nuptiality.rnThe analysis is based on data from the 1990 National Family and FertilityrnSurvey and the 1995 Fertility Survey of Urban Addis Ababa conducted by thernCentral Statistics Authority. A bi-variate analysis of background variables onrnthe nuptiality variables clearly showed a great variation in nuptialtiy by theirrnbackground variables such as age, place of birth, migration status, educationalrnlevels, religion, ethnicity and work of the female respondents.rnSelected important background variables that can be considered for studying therndeterminants of 'age at first marriage' and 'marital stability' is decided byrnapplying the ANOYA and Chi-square tests on the independent variables. WhilernOLS regression model is applied to study the determinants of 'age at firstrnmarriage ', Logistic regression model is applied to study the determinants ofrn'marital stability' in the present thesis. Application of the multivariaterntechniques is decided based on the type of the dependent and independentrnvariables considered in the model. As stated in the hypothesis, the study isrncarried out on the nuptiality data collected in period 1990 and 1995.The independent variables considered in the multiple regression analysis inrnboth periods were place of birth, migration status, educational levels, religion,rnand husband literacy level. Similarly, the independent variables considered inrnthe logistic regression analysis were age at first marriage, place of birth,rnmigration status, educational levels, etlmicity and duration of marriage.rnAs such a comparative analysis is made in the entire study. Results of thernmultiple regression analysis carried out in 1990 and 1995 data clearly show thatrneducation of females (Grade 9+) is the most important variable that determinesrnthe age at first marriage offemales in the study area in both 1990 and 1995.rnBesides, place of birth is the second important variable in detelmining age atrnfirst marriage in 1995.rnResults of the logistic regression on ' marital stability ' seem to show that afterrnethnicity, the most important variable that affect the 'marital stability' is age atfirst marriage and then followed by education of women (Grade 9+) in bothrn1990 and 1995. The results also show that migration is also one of thernimportant variables that is affecting age at first marriage and marital stability inrnurban Addis Ababa during the study period. However, its affect is seen to be onrnthe opposite direction, unlike the affect of education of women, which is in thernexpected direction in affecting age at first marriage and marital stability.