Christian-Muslim Theological Dialogue: The Case of Catholic Universities of East Africa

Christians and Muslims interact on a daily basis but as far as their beliefs and practices are concerned, there is a general mutual apprehension, suspicion, stereotyping, mistrust, insulting and even physical confrontations. The aim of this paper, therefore, is to examine how Catholic universities in East Africa can help Christian students and others to rethink their attitudes towards Muslims in view of the official teaching of the Church. The findings from the reviewed formal Christian and Muslim statements show that theological dialogue is valid and necessary for fostering peaceful relations. It therefore urges Catholic universities and other learning institutions to assist students to know more about their own religious traditions and those of others through formal theological training, seminars and provision of reading materials in order to participate effectively in this kind of dialogue.

Estimation of Population Mean of Sensitive Variable Using Auxiliary Information

Warner (1965) introduced Randomized Response Technique (RRT) for qualitative variable, which helps interviewers extract reliable data corresponding to sensitive questions while maintaining respondent anonymity. RRT allow respondents to mark their actual response by giving a scrambled response which the researcher is later able to unscramble at an aggregate level but not at an individual level. In the thesis, we focus on generalized quantitative optional randomized response models and some generalized ratio-type estimators have been proposed for a finite population mean of a sensitive variable based on RRT by using non-sensitive auxiliary variable. These estimators have been proposed under different sampling designs, such as single-phase sampling, two-phase sampling, stratified sampling and stratified two-phase sampling. In Chapter 1, the discussion has been made about randomized response technique. Different methods, related to randomized response models and RRT have been discussed. Furthermore, uses of two-phase sampling, auxiliary information, stratified sampling, ratio and regression estimators based on randomized response technique have been illustrated. In Chapter 2, literature on the randomized response models and mean estimators based on randomized response techniques is discussed. Chapter 3 contains the quantitative randomized response models, already developed in full and optional randomized response devices. The procedure of ratio and regression estimators for sensitive variables by using non-sensitive auxiliary variable based on RRT have also been given in Chapter 3. The major contribution of this dissertation starts from Chapter 4. In this chapter, three new quantitative randomized response models to estimate the mean of a sensitive variable have been proposed. We also improved mean estimation utilizing information from a non-sensitive auxiliary variable by way of ratio and regression estimators in three models. These three models have been constructed under single sample approach, split sample approach and generalized linear combination scrambling approach. Some special cases and privacy protection for generalized quantitative models have also been discussed. Estimators for the mean, ratio and regression of the variable of interest are developed for each model and simulation studies are also done to analyze them. These proposed models are generalization of additive, subtractive and multiplicative models under different parameter assumptions. In Chapter 5, a class of generalized estimators of population mean have been suggested for randomized response technique under simple random sampling. The estimators have been proposed for the two different situations illustrated as: Situation-I: Single Phase Sampling. Situation-II: Two-Phase Sampling. The mean square error and bias expressions have been derived for generalized estimators under the single and two-phase sampling. An extensive simulation study and an application to a real dataset are done for all the study estimators to evaluate their performance. In Chapter 6, a class of generalized estimators of population mean have been suggested for randomized response technique under stratified sampling. The estimators have been proposed for the two different situations illustrated as: Situation-I: Single Phase Sampling. Situation-II: Two-Phase Sampling. The procedure of stratified sampling in randomized response technique in two situations for single and two-phase sampling have been discussed. The mean square error and bias expressions have been derived for generalized estimators under the single and two-phase stratified sampling. We use the simulation studies for efficiency comparison by empirically and theoretically.