Estimates
Very early teen marriage and you may losing away from highschool one another features large outcomes with the probability a female becomes for the poverty. Brand new estimates indicate that marrying young is actually of a 30.6 fee area rise in the possibilities of located in poverty. Dropping off high-school is regarding the a keen eleven.2 fee area rise in impoverishment.
We now establish a series of choice estimation remedies for assess the robustness of your baseline effect. Desk step one indicated that aggregation made an enormous huge difference to possess OLS estimates: anyone-top efficiency suggested a tiny impact off very early adolescent relationships for the poverty, because category-top OLS rates suggested a massive effect. The first line into the Desk 5 repeats the newest baseline IV analysis, but now having classified research. The info try aggregated at the condition out of beginning ? seasons off birth ? census year level. The fresh new classified-studies IV quotes are interestingly just like the personal-peak IV estimates (0.314 instead https://datingmentor.org/gay-dating-chicago-illinois/ of 0.306 to have very early teenager matrimony and you can 0.112 in the place of 0.112 to possess dropout). The resemblance of your own coefficient quotes is not stunning due to the fact tool are ongoing for everyone people in a state-cohort category, effectively aggregating the personal-height and you may class-level prices. The standard problems as well as transform very little when using the classified study, broadening from the about ten% to possess very early teen matrimony rather than anyway to possess dropout. Since the aggregated investigation supplies comparable point estimates and you can somewhat alot more conventional basic mistakes, with what comes after, We establish outcomes for aggregated research until otherwise detailed. sixteen
Table 5
Notes: Standard problems, adjusted having clustering by condition off delivery, get for the parentheses. Every regressions is dummy details getting census year, race, many years, state regarding beginning, and cohort regarding delivery, and you will area for beginning style. Investigation is actually aggregated to state away from birth ? 12 months out-of birth ? census year cellphone setting. Understand the notes so you can Dining table 1 . Brand new migration-modified approach is revealed throughout the text while the control form method is actually revealed throughout the text message as well as the appendix.
The following activity is always to assess the impact migration has on the assignment regarding county guidelines to own relationships, education, and functions and the further IV quotes. As an indication, laws and regulations are tasked considering a great female’s condition out of beginning, whether or not preferably we should use the state a female lived-in at the years fifteen. Just like the some females has actually migrated out of their delivery condition and you may toward your state having a separate number of rules of the ages fifteen, the various tools try measured which have error. I evaluate just how that it has an effect on the IV estimates into the line 3 away from Dining table 5 .
To see how I examine the issue, notice that the expected value of the ideal (but unobserved) state laws can be calculated if migration probabilities are known. Let zij * be a dummy variable indicating the state law woman i faces at age 15, given she was born in state j. The asterisk indicates that this variable is not observed, given that she may have moved from her birth state by age 15. However, if migration probabilities are known, the expected value of this variable can be calculated as
where pjk represents the probability that a woman will live in state k at age 15 given that she was born in state j, and wk is the law in force in state k for the relevant year. The same logic applies when there are several variables for the state laws.
It is straightforward to show that substituting in E[zij * ] for zij * yields consistent estimates in an IV framework. The remaining issue is how to consistently estimate the conditional migration probabilities, pjk. Although this information is not available for all women, the migration patterns for women who were age 15 at the time of the census enumeration can be estimated because the census records both state of birth and state of current residence. I use 15-year-old women in the 1960 census to estimate these migration probabilities. I then calculate the expected value of the laws based on the state a woman lived in at age 15 as outlined above and use these expected laws as instruments. 18 The migration-adjusted estimates in column 3 of Table 5 are very similar to the baseline estimates, indicating that the assignment of state marriage laws based on state of birth is a reasonable approach.