Abstract Background Traditional birth attendants (TBAs) are likely to deliver lower quality maternity care compared to professional health workers. It is important to characterize women who are assisted by TBAs in order to design interventions specific to such groups. We thus conducted a study to assess if socio-economic status and demographic factors are associated with having childbirth supervised by traditional birth attendants in Iraq. Methods Iraqi Multiple Indicator Cluster Survey (MICS) data for 2000 were used. We estimated frequencies and proportions of having been delivered by a traditional birth attendant and other social characteristics. Logistic regression analysis was used to assess the association between having been delivered by a TBA and wealth, area of residence (urban versus rural), parity, maternal education and age. Results Altogether 22,980 women participated in the survey, and of these women, 2873 had delivery information and whether they were assisted by traditional birth attendants (TBAs) or not during delivery. About 1 in 5 women (26.9%) had been assisted by TBAs. Compared to women of age 35 years or more, women of age 25–34 years were 22% (AOR = 1.22, 95%CI [1.08, 1.39]) more likely to be assisted by TBAs during delivery. Women who had no formal education were 42% (AOR = 1.42, 95%CI [1.22, 1.65]) more likely to be delivered by TBAs compared to those who had attained secondary or higher level of education. Women in the poorest wealth quintile were 2.52 (AOR = 2.52, 95%CI [2.14, 2.98]) more likely to be delivered by TBAs compared to those in the richest quintile. Compared to women who had 7 or more children, those who had 1 or 2 were 28% (AOR = 0.72, 95%CI [0.59, 0.87]) less likely to be delivered by TBAs. Conclusion Findings from this study indicate that having delivery supervised by traditional birth attendants was associated with young maternal age, low education, and being poor. Meanwhile women having 1 or 2 children were less likely to be delivered by TBAs. These factors should be considered in the design of interventions to reduce the rate of deliveries assisted by TBAs in favour of professional midwives, and consequently reduce maternal and neonatal mortality rates and other adverse events.