The role of social support and social influence in maternal infant feeding behavior: Examination of social support as a multifaceted concept

This dissertation was concerned about the role of social environment, that is actual and perceived social support and social influence, in structuring behavior, in this instance, breastfeeding behavior in Istanbul, Turkey. Primarily, it analyzed the effects of social environment on intention to breastfeed and on exclusive breastfeeding behavior in the first two month of infants\u27 life. In addition, it assessed mothers\u27 knowledge with regard to exclusive breastfeeding, examining how knowledge translated into behavior under the influence of social environment. Data were collected in three phases as part of a prospective design. The first phase of data collection was conducted in the hospital with mothers after the delivery. Two follow-up questionnaires were administered at the end of the first and second months. Logistic regression models and multiple regression analyses indicated that measures taken at birth of knowledge and social environment variables did not predict end of first month behavior. Measures taken at the end of the first month of knowledge and positive actual influence were found to be positively associated with behavior at the end of the second month. When data were cross-sectionally analyzed knowledge was found to be positively associated with behavior at both times. Yet, none of the social environment variables contributed to the variance in behavior at the end of the first month. Lack of significant findings was attributed to the prospective nature of design at the methodological level. At the theoretical level, the definition and construction of outcome measures were questioned. The outcome variable, almost exclusive breastfeeding, turned out to be a measure which was not culturally meaningful to the sample of this study. The mothers were not aware of the distinctions between several types of breastfeeding. Intensity of breastfeeding was not considered to be an issue as long as the infant was breastfed. Consequently, the social environment variables constructed with regard to “almost exclusive breastfeeding” behavior failed to show any associations