Primary health care was identified as the ideal model of health care delivery during the Alma Atta conference in USSR in 1978. The World Health Organization urged member countries to adopt it as a model of health care service delivery to their people. Since then many countries have adopted a primary health care model including Tanzania. In countries such as Cuba and Costa Rica where primary health care has successfully been implemented, access to affordable and quality health care service to all has resulted in lowered morbidity, mortality, improved quality and longevity of life. The success story behind primary health care implementation in these countries has been attributed to family physicians taking a leading role. In 2008 the World Health Organization (WHO) challenged all countries to train more family physicians and emphasized the need to fully implement the primary health care model. Despite this call, there has been a decline in the number of doctors applying for family medicine residency programmes especially in developed countries and in Africa many countries are yet to start the training of family physicians. In Tanzania the Aga Khan University has been offering family medicine residency for more than five years. The number of local applicants to this program has remained low. To explore the reasons as to why this is the case, a qualitative study was designed. The study question was “what are the undergraduate medical students’ perceptions towards family medicine?” Methodology: This was a qualitative study carried out in two universities; Muhimbili University of Health and Allied Sciences (public) and the Hubert Kairuki Memorial University (private) among final year undergraduate medical students. A self-administered open-ended questionnaire was constructed, validated and used as the study tool to collect the data. The questionnaire was administered to all 212 final year undergraduate medical students in both institutions and the return rate was 84.5%. The questionnaires were administered by a research assistant at each institution. The final year class in each institution had four clinical rotation groups to which the questionnaire was administered and collected at different times. The answers of the open-ended questionnaire were treated as qualitative data (texts) and were analyzed through a data-led four-step method of analysis as described by Giorgi’s methods of analysis (1985). Themes were identified through the analysis and further validated through researcher triangulation. The study was approved by the Aga Khan University Ethics and Research Committee, Muhimbili University Research and