Researchers at the Perelman School of Medicine at the University of Pennsylvania and the University of Pennsylvania School of Nursing linked seventh-grade reading among 12,339 girls (average age 11.9 years) enrolled in Philadelphia Public Schools to subsequent live birth records between 1996-2002. . . . girls with a less-than-average reading skill were 2.5 times more likely to have a child in their teen years compared with those with average reading skill. Twenty-one percent and three percent of girls with below-average reading skill had either one or two (or more) live births respectively during the six-year assessment period. Meanwhile, 12 percent and 1 percent of girls with average reading skill and 5 percent and 0.4 percent with above average reading skill had such births.
The study also assessed racial disparities in literacy as a contributor to teen child bearing. Hispanic and African American girls were overrepresented in the below-average reading skill group. In addition, the effect of low literacy on risk of teenage parenting was stronger in Hispanic and African American girls than those who self identified as White. The researchers point out that poor reading skills in early grades are difficult to overcome and predictive of subsequent decisions to drop out of formal education.
From here, citing American Public Health Association (APHA) (2012, October 31). "Pre-teen literacy a strong predictor of pregnancy in U.S. teens."
It is worth noting, however, that U.S. teen pregnancies are currently near all time lows and have declined materially scine the 1996-2002 time period that was studied. It is also worth noting that while literacy is an important factor in the determining the likelihood that a Philadelphia Public Schools teenager will get pregnant, that one in twenty teenage girls with above average reading skills in the Philadelphia Public Schools got pregnant and had a child.
The article cited doesn't discuss the plausible possibility that lower frequencies of live births might be due to higher abortion rates among more literate teenage girls. But, there is evidence from sources other than those discussed in the study that suggest that this isn't actually happening, or at least that it is not happening to a great enough extent to alter the conclusion that literacy drives teenage birth rates mostly by having an effect via the number of unplanned pregnancies, rather than mostly due to the percentage of pregnant teenagers who choose to get abortions.