Skip to main content

Table 5 CLM marginal effects by subsample

From: Common law marriage and couple formation

  In couple probit Married probit Cohabiting, among unmarried probit
Women Men Women Men Women Men
No college -0.031 -0.038 -0.016 -0.018 -0.027 -0.024
  [0.012]*** [0.010]*** [0.010] [0.011]* [0.009]*** [0.008]***
College educated -0.006 -0.108 -0.034 -0.103 0.036 -0.03
  [0.025] [0.025]*** [0.025] [0.022]*** [0.028] [0.018]*
Low sex ratio -0.042 -0.084 -0.024 -0.046 -0.01 -0.031
  [0.017]** [0.020]*** [0.012]** [0.012]*** [0.013] [0.012]***
High sex ratio -0.024 -0.029 -0.024 -0.025 -0.032 -0.021
  [0.017] [0.018] [0.015] [0.020] [0.013]** [0.007]***
No children present -0.035 -0.041 -0.027 -0.017 -0.015 -0.023
  [0.012]*** [0.009]*** [0.009]*** [0.007]** [0.010] [0.007]***
Children present -0.015 -0.012 -0.007 -0.013 -0.018 -0.044
  [0.016] [0.010] [0.019] [0.012] [0.019] [0.051]
Age 18-25 -0.011 -0.046 0.001 -0.015 -0.012 -0.031
  [0.013] [0.009]*** [0.011] [0.007]** [0.010] [0.007]***
Age 26-35 -0.022 -0.038 -0.021 -0.039 -0.02 -0.01
  [0.015] [0.016]** [0.014] [0.016]** [0.017] [0.020]
  1. Notes: The table reports CLM marginal effects in regressions for probability of being in a couple. High sex ratio areas are those with sex ratio >1 and low are sex ratio < =1.
  2. The table shows full estimates of the first model from Table 3, our preferred specification with demographics, state characteristics, year & state fixed effects, and six dummy variables for years before and after repeal of the law.
  3. All pairs of coefficients estimated using two subsamples are statistically different from each other at less than 1% level.
  4. Sample sizes in the first four columns vary between N = 55,908 for men with college degrees to N = 249,659 for women with less than college education.
  5. Sample sizes among unmarried adults in the last two columns vary between N = 10,450 for men with children to N = 159,883 for men with less than college education.
  6. *significant at 10%; **significant at 5%; ***significant at 1%.