Statistics
Permanent URI for this collection
Browse
Browsing Statistics by Author "Abegaz, Fentaw (PhD)"
Now showing 1 - 8 of 8
Results Per Page
Sort Options
Item Correlates of Harmful Traditional Practices: The Case of Women in Ethiopia Path(Addis Ababa University, 2009-06) Yohannes, Zemene; Abegaz, Fentaw (PhD)This study attempts to identify the correlates of harm/iii traditional practices on Elhiol'ian women and to assess the predictive validity of some selected variables such as age, region. place of residence, educational level, ethnicity, religion, Fequency of reading news pOIJer or magazine, Fequency of watching TV, .Fequency of listening radio. weallh index. knowledge and allitude an harm/iii traditional practices. In this study, the dala colieeled by Central Statistical Agency (CSA) for the 2005 Demographic and [{eailh Survey (DIIS) were used. The study consists of 14,070 randomly selected women in Elhiopio. Preliminary analysis is carried out through bivariate analysis. Then combined elkel ollhe variables is analyzed using logislic and probit regression models. The resull shows Ihol age, region, educational level, religion, place of residence, wealth index. ethnicil)!, and attitude on harmjiil traditional practices are significanl correlates of harln/it! Imdil ional practices. In order to decrease or discontinue harm/iii traditional praC/ices like Ielnale circumcision and marriage by abduction, strategies should be designed in a such way IhCil would focus on and address the most important correlates identified inlhis stud),Item Determinants of Nutritional Status of Children in Ethiopia Using Multilevel Analysis(Addis Ababa University, 2008-06) Tesfaw, Dejen; Abegaz, Fentaw (PhD)TIllS study attempts to identify the determinants of nutritional status of children in Ethiopia. The study focuses on children aged less than 59 months based on data from 2005 Ethiopia Demograplllc and Heald, Survey. A total of 3095 children were included in dlls study. As the outcome measure of nutritional status, stunting (an anthropometric index of height for age) defined as a binary outcome is taken as the dependent variable. A preliminary univariate analysis was used to see the percentage share of background variables wid, respect to the nutritional status of children. In order to exarnine the clifferent factors of nutritional status, multiple logistic regressions were employed. Moreover, a multilevel logistic regression analysis is considered to identify the determinants of stunting in clllidren within regions as well as between regions. The multivariate results using three logistic regressions for urban, rural and national level samples revealed that the most important socioecononllc, demographic, health and environmental factors related to nutritional status of children are individual, parental, household and regional factors . The results inclicated that nutritional status varies across regions and urban and rural divisions. Furthermore, the results indicated that household economic status, mother's employment status, age of child, prececling birth interval and source of drinking water are some of the factors that affect nutritional status of children direcdy or indirecdy.Item Determinants of Stigmatization of Families of Psychiatric Patients: The Case of Butajira District(Addis Ababa University, 2008-07) Kebede, Dejene; Abegaz, Fentaw (PhD); Deyessa, Negussie (PhD)Considerable research has documented the stigmatization of people with mental illnesses and its negative consequences. Recently it has been shown that stigma may also seriously affect families and caregivers of psychiatric patients, but a few empirical researches have addressed this problem. In this study we examine the determinants of stigmatization of families of psychiatric patients 111 Butajira district. A total of 514 caregivers' of individuals who were diagnosed as suffering from schizophrenia and major affective disorders in a community based survey were interviewed. The result reveals that 66.1 % of the fami lies of the psychiatric patients have reported that they are stigmatized due to the presence of psychiatric patient in the family. Statistical methods such as logistic regression, factor analysis and alternating logistic regression were applied and results fi'om logistic regression reveals that sex of patient, age of the informant, quality of subjects' cun'ent dwelling, economic position of patients' house-hold relative to catchments ' area, and type of disease are significant factors that determine the stigmatization of families, and results from factor analysis provides us the two major components of stigma (threat from extemal environment and self feeling) explaining 64.72% of total variance in all variables and finally alternating logistic regression was used that provide odds ratio as a measure of association between these two factors. Perceiving stigmatization with respect to self feeling are 4.338 times that of with respect to threat from external environment.Item Factors Affecting Use of Contraceptive Methods Among Married Women in Ethiopia(Addis Ababa University, 2008-07) Adugna, Getahun; Abegaz, Fentaw (PhD)The increasing use of family planning methods seems to be the intermediate determinant which mostly influences the fertility decline in developing countries. Ethiopia is characterized by very different geographical realities and by strong regional heterogenei ties. Large scale surveys often follow a nested structure of data because infonnation is collected from different levels of hierarchy. It is thus necessary to elicit the inherent variations, which occur for a particular phenomenon among these different levels. The present study seeks to detect variations resulting from such nested structure in the case of contraceptive practice in Ethiopia, as well as to identify sociodemographic factors affecting the individual's behavior. Applying multilevel analysis, models are developed for current use of (;ontraception. The study used data collected in the Ethiopian Demographic and Health Survey, 2005. The final multilevel model reflects the existence of significant variations in the mean effects of the predictors of current contraceptive practice at regional level. The study shows that running intensive family planning programs along with other awareness programs in those regions where the particular determinant of contraceptive use has been found to be not very effective.Item Predictors of Burden on Caregivers of Patients of Schizophrenia and Bipolar Disorder: The Case of Butajira, Ethiopia Merga BeJilla(Addis Ababa University, 2008-07) Belina, Merga; Abegaz, Fentaw (PhD)The aim of the study was to investigate the prevalence and nature of the objective burden experienced by caregivers of schizophrenia and bipolar disorder patients. Also to compare the results from the two statistical methods, polytomous logi stic regression and discriminant analysis in correctly classifying and selecting important explanatory variables of burden on caregivers in psychiatric patients. A total of 514 patients were included in this study. Burden on the caregivers was assessed in telms of how much the caregiver's life was affected by the presence of the illness in the family. Results indicate that 36.5% of the caregivers experienced no burden, whereas 36.9% of them experienced moderate burden. The polytomous ordinal logistic regression classified 67.2% of the caregivers in their respective burden group, whereas the discriminant analysis classified 64.7% of the caregivers correctly. The polytomous ordinal logistic regression shows a slight improvement in correctly classifying caregivers over multiple discriminant analysis. Based on the findings of the present study we conclude that high burden on the caregivers is signi ficantly characterized by dependency status of the patient, emotional helps given to the care giver from the society, negative and positive symptom of the patient. The results call for more attention to be given to reduce patients ' positive and negative symptoms, to create awareness in the society to give emotional helps to the caregivers and to economically support the patients so as to minimize their dependency on their caregivers.Item A Production Function Analysis for Private Peasant Holdings Crop Farms in Ethiopia: An Application of Robust Regression(Addis Ababa University, 2009-06) Kassahun, Taddesse; Abegaz, Fentaw (PhD)This study applied production function analysis for private peasant holdings crop farms in Ethiopi a. Four major crop producing regions viz., Tigray, Amhara, Oromia and SNNP were included in the study. Three regression models for production fun ction namely, linear, exponential and Cobb Douglas were considered and thoroughly assessed {'or stati stical model diagnostics. The statistical model diagnostics and checking suggested that crop production function for each of the regions was found to be appropriately represented by the Cobb-Douglas production function based on data from the 2007/08 (2000 EC) agricultural sample survey. The Cobb-Douglas production function was first fitted for each region using ordinary least squares (OLS) regression. As expected, the parameter estimates using OLS were mi sleading due to the occurrence of several outlying cases and hence robust regression was taken as a viable alternative. Based on the results of robust regression, many of the parameter est imates took on the expected signs, the R2 values were substantially increased and the standard errors of parameter estimates were decreased at large. In general, robust regression results indicated that farm size, fe rtili zer, seed, oxen power and human labor were playing a pivotal role for the maximization of crop yield in each of the studied regions. From among these variables, the great contribution was found to be due to farm size in each of the regions with SNNP an exception in whi ch the great share was due to human labor. However, the contribution of education variable for crop yie ld was found to be stati sti cally insignificant and received negative sign in Tigray and Amhara regions. Counter to expectations, the coefficient estimate for irrigation variable in Tigray, Amhara and SNNP regions had come to obtain negative sign though it was not found to be stati stically significant. The product ion elasticities for each of the inputs at each region except farm size in Tigray, Amhara and Oromia suggested that the relation between inputs and output was inelastic, i.e., ho lding other {actors constant, the marginal return to each factor wi ll decrease as more of the factors are used. Additionally, crop production functions revealed that returns to scale were estim ated to be greater than unity in each of the regions indi cating increasing returns to scale.Item Survival Analysis of Twin and Singlcton infants in South-Wcst Ethiopia(Addis Ababa University, 2009-08) Asefa, Henok; Abegaz, Fentaw (PhD)In this study the effect of socioeconomi c, demographic and biological va riables on infant survival is in vestigated using data li'om "'fhe .limma in f~\I1 t survival differential longitudinal stud y"' conducted in 46 urban and 64 ru ral ' kebe les' in South-West Ethi op ia during September 1992 and October 1994. The study uses 7,851 singleton and 89 twin infants. Matched pair analysis is used to assess the survival experience between twins, and parametric regression model is used to ,examine the association between infant mortality and socioeconomic, demographic and bio logica l va riables. Data analyses wcre made using STATA 10 and SI'SS 16. The result of the matehcd pair analysis showed that, thc hazard ratcs for thc lirst and the second born twins are different at 5% leve l of significance. The Kaplan-Meier survi val estimate far the first born twi ns is 0.7143 (95% CI: 0.6069 to 0.7971) and ror the second born I twins is 0.5634 (95% CI: 0.4523 to 0.6604). In regard to the survival of singleton in fants , the result of Kaplan-Meier estimator shows that the survival probability is 0.9 12 with standard error or 0.0033. That is, infant mortality rate or about 88 deaths per 1000 live birth s. Dirferent parametric regression models were fitted to select the s ignificant factors afreet ing infant mortality in South-West Eth iopia. Almost all the models considered provided good fit l'or the data but log-no rma l regression model was round to be the best fit l'or the data. Ilased on the result br the final log-normal regression model, in fant mortality was significantly (p < 0.05) associated with birth weight, sex of the infant, mother's education, lamily size and marita l status and the interaction between ethnie ity and place of res idence.Item Woreda Level Estimates of Agricultural Area and Production: Small Area Estimation Using Auxiliary Data(Addis Ababa University, 2009-08) Jemal, Seid; Abegaz, Fentaw (PhD)This study applies small area estimation technique to analyze available agricultural data to estimate the total of basic agricultural variables such as total cultivated area in hectares and total production in quintals ofTeJJfor woredas in the Tigray Region using cultivated area and production data from 2001102 National Agriculture Sample Census Enumeration and from Ministry of Agriculture and Rural Development as an auxiliary variable. An area level model for small area analysis is used to produce estimate on cultivated area and production for 33 woreda in the Tigray Region. The estimation process shows significant woreda-specific random eJJect which leads to acquire a beller estimate for cultivated area and production.