Abstract
<title>Abstract</title> <p>Urban India has undergone a rapid fertility transition, with fertility declining below replacement level and the two-child family becoming the dominant reproductive norm. Whether son preference persists under these conditions remains an important demographic question. This study examines how the relationship between child sex composition and progression to a third birth has evolved during three decades of urban fertility transition and investigates whether stated fertility preferences support observed reproductive behaviour. The analysis combines data from all five rounds of the National Family Health Survey (1992–93 to 2019–21) with a primary survey of 400 currently married women in Delhi. Parity progression ratios, binary logistic regression, and descriptive analyses were employed. The findings show that progression to a third birth has declined substantially, reflecting the consolidation of the two-child norm. However, women without sons remain significantly more likely to continue childbearing than those with at least one son, even after controlling for socioeconomic characteristics. Evidence from the primary survey further demonstrates that stated fertility preferences are consistent with observed reproductive behaviour, indicating the persistence of son preference despite sustained low fertility. Educational attainment reduces higher-order childbearing but does not completely eliminate son preferences. The study makes three contributions. It provides a comprehensive assessment of parity progression across all five NFHS rounds, integrates revealed reproductive behaviour with stated fertility preferences, and introduces the concept of compressed son preference, arguing that fertility decline concentrates rather than eliminates the expression of son preference within increasingly selective higher-order fertility decisions.</p>