This study examines the effects of various factors on diarrhea morbidity and mortality among infants/children in Bobo-Dioulasso, Burkina Faso. These factors have been grouped into two sets of variables: Those associated with the child's background and the environment in which it is living, and identified as background variables and those associated with the child per se, and identified as proximate variables. / Breaking from traditional approaches that focus on either death or disease processes, and examines the effects of either background or proximate variables, this dissertation, based on a conceptual framework developed by Mosley and Chen (1984), attempts to integrate the study of both morbidity and mortality by assessing the effects of both types of variables and focusing on the "survival process." / The data used for this study were recorded longitudinally by following a cohort of 8,491 births for the first twenty-four months of life. Therefore, event history techniques are used for analysis. Relationships are estimated between independent and dependent variables, through four logistic regression models. As a complement, qualitative data based on interviews of a sub-sample of 76 mothers are also analyzed. / The findings suggest, first, that both types of variables exhibit significant links to morbidity and mortality. Interestingly, these variables are differently related to these two phenomena. This implies that morbidity and mortality are related to different processes captured by the same variables. Second, reporting of morbidity events and selectivity in going to health centers among mothers seemed to be a problem in the study area. This suggests that the understanding of the implications of childhood diseases is not the same for all mothers. Third, it seems that the framework developed for this study is an improvement over that of Mosley and Chen because it is simpler and uses variables that are straightforward and easily measurable in LDCs. Furthermore, the results support the findings of past research on childhood survival and suggest new orientations for research and policy. Finally, the analysis of the qualitative data implies that there are traditional beliefs regarding childhood illnesses and treatment in the study area. / Source: Dissertation Abstracts International, Volume: 54-08, Section: A, page: 3222. / Major Professor: Isaac W. Eberstein. / Thesis (Ph.D.)--The Florida State University, 1993.

This dissertation examines the association between race and infant mortality in the United States in the most comprehensive manner to date. Beginning with a sociological conceptualization of race, a framework is developed which delineates six sets of individual-level factors thought to be instrumental in the continuing association between race and infant mortality. The specific objectives of the project are to test the ability of the framework to explain differential mortality across racial groups and to understand the effects of each set of factors on infant mortality by group. / The 1988 National Maternal and Infant Health Survey (NMIHS) is used for the analysis. The NMIHS is the most comprehensive data set ever collected to study poor pregnancy outcome in the United States. Multinomial logit regression procedures are used to most appropriately model the effects of race and other independent variables on cause-specific infant mortality. / Descriptive findings illustrate that the racial gap in infant mortality is nearly identical for endogenous and exogenous causes of death, with the overall rate of infant mortality among African-Americans about 2.2 times higher than non-Hispanic Whites. The regression analysis confirms that the racial distribution of socioeconomic, demographic, health care, maternal health, and infant health factors are all instrumental in understanding the excess mortality experienced by African-Americans. The behavioral factors examined here do not account for any of the excess mortality among African-Americans. Separate analysis by group suggests that several of the variables work in unique ways to affect infant mortality by group. Theoretical and empirical contributions of the study are highlighted. / Source: Dissertation Abstracts International, Volume: 54-02, Section: A, page: 0691. / Major Professor: Isaac W. Eberstein. / Thesis (Ph.D.)--The Florida State University, 1993.

From its inception in 1967, the Kenyan national family planning program has suffered from a lack of popular support and confidence within the general population, absence of active local participation at all levels and, above all, the absence of male participation or neglect of men in the provision of family planning services in a country whose societies are predominantly patriarchal. / The present study uses a conceptual and theoretical framework based on data from the 1989 Kenya Demographic and Health Survey and measures the effects of the male role in family planning activities in Kenya. The study also provides analyses that can be used to identify the conditions under which opportunities for men to participate in family planning programs can be expanded. / The findings indicate that lack of communication between husbands and wives may be a greater obstacle to family planning than male opposition. Couples who discuss the subject of family planning are not only more likely to use contraception, but are also more likely to want to space, postpone or cease childbearing altogether and to use contraception to achieve these fertility goals. Hence, the largest gain in contraceptive use can occur if married couples openly, freely and regularly discuss the subject of family planning. Contraceptive use in Kenya is also a function of (1) residence in regions whose constituent communities adhere less to traditional patterns of reproductive behavior (Nairobi and the Central and Eastern regions), (2) higher levels of education for the husbands, (3) employment in higher status occupations, (4) higher levels of education for the wives, (5) women being in the middle of their reproductive ages, and (6) rural residence. Other factors seem to have little or no significant effect on contraception. / The study concludes that inclusion of male factors in the family planning program is not only crucial for more effective program performance, but is also an area of policy that should be among those consciously considered in the deliberative process of lowering fertility and reducing population growth. / Source: Dissertation Abstracts International, Volume: 53-07, Section: A, page: 2572. / Major Professor: Bruce Stiftel. / Thesis (Ph.D.)--The Florida State University, 1992.

This study is intended to test two major hypotheses linking socioeconomic variables, intervening variables, and fertility. The first hypothesis states that under a higher level of socioeconomic development, the socioeconomic-fertility association is expected to be negative. The second hypothesis asserts that this association is mediated by several intervening variables (child loss, marital stability, age at marriage, and contraception). The 1987 Indonesian Demographic Health Survey data are analyzed, using regression analysis to test these hypotheses. / The study found that the socioeconomic variables included in this study, except for female labor force participation, did not behave as expected by the first hypothesis. Education has a curvilinear association with fertility, where fertility is low at zero years of education, then moves up at the incomplete primary level, declines a bit at the complete primary level, and finally declines again beyond the primary school level. This result is different from that found from the 1976 WFS data, which showed a positive education-fertility association. With respect to urban-rural residence, a positive relationship of urban residence to fertility was observed; the relationship is similar to that found in 1976 from the WFS data. The last socioeconomic variable, female labor force participation, was found to be negatively related to fertility; however, the differential effect of this variable has decreased. / Turning to the second hypothesis, the study found that three intervening variables incorporated into the final regression equation behaved as expected by the hypothesis. They showed a significant direct effect on fertility. In terms of the order of significance of these intervening variables, child loss experience ranks first, followed by marital stability and age at marriage in that order. The findings with regard to contraception are indeterminate. / The results suggest that social changes in Java are transforming the pattern of Javanese differential fertility to some extent. However, neither a modernization perspective or a family planning perspective can fully account for the transformations. It is necessary to explore other theoretical approaches in order to achieve a greater level of explanation of trends in differential fertility in Java. / Source: Dissertation Abstracts International, Volume: 54-09, Section: A, page: 3602. / Major Professor: Charles B. Nam. / Thesis (Ph.D.)--The Florida State University, 1993.

In the research regarding family and fertility, little attention has been paid to the couple unit. The majority of fertility studies have focused upon the wife only. Where data are available for both husband and wife, it is often clustered under the headings 'males' and 'females;' again ignoring the specific husband and wife dyad. In addition, the attitudes of husbands and wives are often assumed to be essentially the same This study investigates couple agreement on desired family size. The following hypotheses are examined. (1) Agreement on desired family size is related to similarity of background; the more homogeneous the couple in their background characteristics, the more they are likely to agree on their family size desires. (2) Agreement on desired family size will be greater among those who have discussed their family size desires and goals than among those who have not. (3) The longer the duration of the family of procreation, the greater will be the agreement concerning family size preference between husbands and wives The study is based on the secondary analysis of data from the New Orleans Family Survey of 1965. The data are from a subsample of 357 couples. Separate interviews were held with husband and wife; their responses to the survey questions were subsequently compared The data concerning demographic homogeneity are inconsistent. Sometimes those similar in characteristics agree more highly on family size preferences than those dissimilar in characteristics; sometimes they do not. The zero order relationships are weak, rarely manifesting chi squares large enough to reject chance. Multiple regression of the twenty-six explanatory variables indicates homogeneity on these variables explains only fourteen percent of the variance in agreement on desired family size Discussion of family size also yields inconclusive data. Couples who have discussed family size preferences indicated a slightly higher percent agreement than did those who had not discussed preferences. The highest percent agreement was shown, however, among couples who disagreed concerning whether they had ever discussed family size preferences Length of marriage was found to have no relationship to agreement on desired family size In all, the data show a high percent of disagreement concerning family size preferences among the couples; slightly over sixty percent of the couples disagree. This indicates that researchers should be cautious in interpreting information from one spouse as indicative of the opinions of the other. In addition, spouses who are homogeneous in characteristics should not be assumed to be homogeneous in outlook / acase@tulane.edu

After a brief review of the literature on inbreeding in human populations, the historical and political circumstances that led to the formation of a unique, isolated demographic unit indigenous to southwestern Louisiana are described. Details of the frontier boundary disputes in western Louisiana and east Texas, first between the French and Spanish (1715-1762) and later between the United States and the Spanish (1803-1821), are reviewed with special attention given to the events associated with the establishment of the Neutral Ground and the make-up of the first settlers to that region. Specifically, the history of the settlers of the former Cherry Winche country, known by the nickname Ten Milers, is updated and consolidated from widely scattered sources. As unbiased assessment of the possible origins of this alledged tri-racial isolate leads to the conclusion that there is no objective evidence that this population is of tri-racial ancestry Demographic and genetic data were compiled from censuses of local cemeteries and marriage records for this area. Inbreeding in this population was quantified using Crow and Mange's (1965) isonymy technique. The average inbreeding coefficient was estimated to be 7.938 x 10('-3), which is much higher than the average inbreeding estimates for the United States (9.0 x 10('-5); Freire-Maia, 1968) and is comparable with other human isolates. As part of a descriptive demographic profile, lifetable analyses were performed on the cumulative survival distribution in this population and various sub-populations (males vs. females, Ten Milers vs. non Ten Milers, etc.) are compared. Overall childhood mortality was high with 16% of the population failing to reach age one and 20% failing to reach age five. Relative measures of endogamy and migration confirm the endogamous nature of this population Investigation of the possible adverse effects of inbreeding, through the analysis of mortality distributions across sub-populations in different time intervals and in different seasons, suggests that differences in mortality distributions among sub-populations are the result of cultural rather than biological circumstances. Data on the relative age of marriage partners is also suggestive of a confounding influence on childhood mortality. Certain types of cancer show anomalous incidence levels in the region. No unique genetic diseases were found in association with this population / acase@tulane.edu

Sociologists and policy analysts depict many poor and low-income communities as matriarchal female domains. Nevertheless, many men reside and participate in these communities. Unfortunately, researchers and policy makers often misconceive of public housing as female-headed, male-absent households. So this dissertation ask whether men are really missing or are social scientists and policy makers just unable to find them. This dissertation sheds light on this question by examining residential and affiliation patterns of black men in a low-income public housing neighborhood in New Orleans. Using qualitative methods that include participant observation, in-depth interviews, focus groups and genealogies, this research explores the presence of men in a public housing neighborhood and their financial and non-financial contributions to families, households and the neighborhood at large / acase@tulane.edu

A major factor in the social and economic development of Vietnam is population growth and its determinants. In order to provide information to assist in the organization and planning of future policy decisions several important research questions were formulated. We intended to estimate the level of unintended pregnancy in Vietnam, to identify the determinants of unintended pregnancy, non-use and ineffective use of contraceptive methods in relation to the supply environment of contraceptive methods and family planning service. Of particular interest was the objective of identifying the determinants of abortion among unintended pregnancies. The conceptual framework included socio-demographic characteristics of both women and their husbands, selected predicting factors on the supply side of contraceptive methods and services at the community, commune and district levels. An innovative methodological element of the study which utilized the data of the 1997 Vietnam Demographic and Health Survey. Data was the calculation of 13,540 'segments' of pregnancy outcomes and contraceptive practice, based on the retrospective reproduction calendar of the women, and analyzed using bivariate and multivariate techniques This study found evidence of high levels of unintended pregnancy among married women in Vietnam (40 percent). An increase of one contraceptive method in the supply environment at the district level significantly reduced the risk of unintended pregnancy by 30 percent (p < 0.01). One increment in supply environment at community level significantly reduced the risk of non-use of contraception by 8 percent (p < 0.05). IUDs were found to be an important contributor to successful contraceptive practice, the availability of IUDs at commune health facilities significantly reduced the risk of unwanted pregnancy by 50 percent. Withdrawal accounted for 35 percent of ineffective contraceptive use, and, combined with condom use, accounted for 50 percent of ineffective use. Besides supply side factors, unintended pregnancy was significantly associated with: age, early marriage, spousal age difference, number of sons alive, past unintended pregnancy, region, and contraceptive use. Factors found positively related to non-use and ineffective use of contraception included: region, age, ethnic minority, low socio-economic status, and spousal age difference. Factors found related to abortion were: region, high socio-economic status, woman over 40 years of age, spousal age difference, husband's education, past abortion, and number of sons alive. The study draws some conclusions for policy modifications focused on improving contraceptive method supply and method mixed at the community level, targeting high risk groups, and promoting communication programs / acase@tulane.edu

This study assessed the relationship between teachers' perception of population change and their implementation of population education conditioned by such background factors as sociodemographics, school environment, and population education program facilities. The general hypothesis was that teachers with a positive perception of population change were more inclined to implement population education in their courses. Theoretically, the scope of population education can be explained through a perception-behavior model which was presented as a perception-implementation relationship derived from the results of this study. / A self-administered questionnaire was used to collect data from the general public secondary schools in North Sulawesi, Indonesia. With a multi-stage purposive sampling design, a sample of 455 teachers were selected to respond to the questionnaire. Data were analyzed utilizing a two-stage least squares method with additional descriptive procedures. / The findings indicated a strong relationship between perception and implementation with r =.316. Using first stage least squares on the seven background variables; exposure, place of residence, and teaching assignment shared a great amount of the variation in perception. These variables, together with educational background and school level, affected implementation. / Teachers who were more exposed to population education program facilities, who lived in rural areas, and who taught geography, civics, economics, and biology had higher perceptions of population change. Together with those having low academic backgrounds and working in the middle schools, they effectively implemented population education. / Although the background variables influenced implementation more than perception in the ordinary (first stage) least squares analysis, the second stage analysis showed that estimated perception was a more powerful predictor than estimated implementation. This suggests a background-perception-implementation model to be recommended for population education programs, research, and evaluation. / Source: Dissertation Abstracts International, Volume: 53-03, Section: A, page: 0769. / Major Professor: Byron G. Massialas. / Thesis (Ph.D.)--The Florida State University, 1992.

Ethnic differentials in birth outcome are significant. Currently, a black infant is at twice the risk of dying before age one than either a non-Hispanic white or an Hispanic infant. The role of culture has received little attention as a determinant of infant mortality. / Culture is defined as an "evaluative conversation constructed by actors out of the raw materials afforded by tradition and ongoing experience" (Hammel, 1990). This definition is consistent with that of race and ethnicity as socially constructed concepts (Bean and Tienda, 1987; Hummer, 1993a). Thus, it is argued that some observed differences related to infant mortality among ethnic groups result from different cultural backgrounds. / Family structure, arguably a cultural characteristic, is expected to have an impact on birth outcome. The linkages between family structure and birth outcome are availability of financial resources and role models for observational learning (Bandura, 1977). Ethnic differences in the likelihood of births to unwed mothers is also viewed as culturally based. Leslie (1995) suggests that African and African American culture views out-of-wedlock births as "mistakes", not as "sins". Thus, unwed motherhood represents a "normative" state when viewed from an Afrocentric perspective. The final cultural variable, financial aid from relatives, is included to partially compensate for the fact that our primary family composition measure is household-based. / Cultural, ethnicity, and socioeconomic status are seen as prior determinants of birth outcome. Intermediate factors include socio-demographic characteristics, maternal behaviors and attitudes, and maternal health. Proximate determinants are health care and infant birth weight. The outcome variable is infant mortality which is grouped into three categories: (1) lived, (2) died from endogenous causes, and (3) died from exogenous causes. / The model is tested using the 1988 National Maternal and Infant Health Survey. Logistic regression is used to fit the data to the model. / This dissertation contributes to the study of infant mortality by conceptually linking cultural practices and behaviors to birth outcomes. It develops a multidimensional set of household-based family structure variables and conceptually links them to birth outcome. It tests the existence of these linkages on samples of non-Hispanic whites, blacks, and Mexican Americans, the three largest ethnic groups in the U.S. / Source: Dissertation Abstracts International, Volume: 56-08, Section: A, page: 3326. / Major Professor: Isaac W. Eberstein. / Thesis (Ph.D.)--The Florida State University, 1995.