Social capital plays a more significant role today compared to other forms of capital, such as physical and human capital, in societies. Among the scholars emphasizing the importance of social capital, James Coleman assigns a critical role to family and educational systems as sources that generate social capital alongside other sources. This study, based on James Coleman’s theoretical framework, investigates the relationship between the social capital within families, the formation of restricted and elaborated language codes, and their connection to academic performance among students enrolled at Payame Noor University. The research employed a survey method, with data collected through questionnaires accompanied by interviews. The statistical population consisted of all male and female students studying at Payame Noor universities in East Azerbaijan Province. A sample size of 452 was determined using simple random sampling and Fadi’s proposed formula. The reliability of the research scales was assessed using the internal consistency method and Cronbach’s alpha coefficient, while their validity was measured using the construct validity method. To analyze the research data, two-way analysis of variance (ANOVA) tests were employed. The findings revealed that students with moderate or higher levels of family social capital and elaborated language codes demonstrated better academic performance compared to students with restricted language codes.