Results of ordered logistic regression of completed fertility for women and men and for various adjustment variables: coefficient of sibship size by birth cohort. Additional adjustment variables are respondent’s level of education, country of birth, occupational status, mother’s and father’s country of birth, and their occupational status (see online appendix C). The sample is women aged 45–89 and men aged 45–79 at survey. In the 1922–1926 birth cohort and with all controls, for a one-unit increase in the number of siblings, the number of children is expected to change by 0.117 + 0.007 = 0.124 (see Model 4 for women in Table 3) in the log-odds scale, while holding the other variables in the model constant. Full results are available in Table A2 in the online appendix. Source: Enquête Famille et Logements 2011 (EFL, INED-INSEE).