Skills and Earnings in the Full-Time Labor Market
Neeta Fogg, Paul Harrington, and Ishwar Khatiwada
Educational Attainment and Literacy and Numeracy Proficiencies
Formal education is the most readily available measure of the human capital of individuals. Human capital represents the skills, abilities, knowledge, and work experience of individuals. Attaining formal education adds to their stock of human capital.
Although education is readily available, it is not a complete measure of human capital because it only serves, along with other mechanisms, as an important channel through which human capital traits are developed. A fundamental role of schooling is to increase some of the most important personal characteristics valued by employers, especially basic skills and occupational knowledge. The influence of formal schooling on literacy and numeracy proficiencies and labor market outcomes works through two distinct channels. First, since student literacy and numeracy skills at lower grades influence the likelihood that students will attain higher levels of educational attainment, the likelihood of students dropping out of high school is closely connected.20 Second, the likelihood of postsecondary enrollment, persistence, and completion are also closely related to high school students' scores on basic skills measures.21 It thus comes as no surprise that basic skills scores and levels of educational attainment are closely connected.
The likelihood of further education is closely connected to the present level of basic skills of students. College enrollment of graduating high school students is heavily influenced by students' basic skill development as measured by standardized test scores as well as student grade point average in high school. Indeed, both of these measures are powerful predictors of the likelihood of college enrollment, retention, and completion.22 Young people who fail to develop solid basic skill proficiencies by the secondary school level have only a remote chance of achieving further education beyond high school. Indeed, earning a high school diploma itself is associated with the development of reading, writing, and math skills. The likelihood of an entering ninth grader completing high school is related to his or her level of basic skills as measured by standardized test scores and grade point average.23 Measures of math and English language proficiency at the middle school level are also thought to be important predictors of the risk of dropping out of high school.24
Mean scores on the PIAAC literacy and numeracy proficiencies of prime-age, full-time employed U.S. workers included in our analysis reveal a close positive association between both literacy and numeracy scores and the highest level of educational attainment of these workers (Figure 5). The mean literacy score of 25- to 54-year-old full-time employed high school dropouts was just 208, a score that placed them at level 1. The mean score of 256 among high school graduates with full-time jobs fell within level 2 score range. Mean literacy scores of 288, 307, and 316 among workers with associate's, bachelor's, and master's or higher degrees, respectively, placed the average literacy proficiency of these three groups of full-time employed college graduates in level 3. The difference between the mean literacy proficiency score of workers without a high school diploma and those with a master's degree or higher was sizable: 108 points.
Among workers with graduate level college degrees, the mean literacy proficiency scores were 313, 320, and 327 for workers with master's, professional, and doctorate degrees, respectively. Only the doctorate degree group of prime-age, full-time employed workers had a mean score that placed them within the literacy score range that defines level 4.
The mean numeracy proficiency scores among educational groups of these workers also exhibited a strong positive connection with their level of education. Mean numeracy scores among prime-age, full-time workers increased from 199 among high school dropouts and 241 among high school graduates to 274 among workers with an associate's degree, and 296 and 308 among those with bachelor's and master's or higher degrees, respectively, yielding a yawning gap between the mean numeracy score of workers with the lowest and highest level of education (high school dropouts and master's degree or higher) of 109 points. Workers with graduate degrees (master's, professional, and doctorate degrees) had higher scores on the numeracy proficiency scale: 305, 308, and 326, respectively.
While there is a clear positive connection between mean basic skill scores and the level of education, we also found considerable variation in literacy and numeracy scores among workers with the same level of educational attainment. Earning a high school or college credential is not necessarily a signal of basic skills achievement. Indeed, we find that about 9 percent of prime-age, full-time employed high school dropouts scored in higher literacy and numeracy levels than 18 percent of their counterparts who earned a bachelor's degree and, even more surprising, higher than 13 percent of those who earned an advanced degree.
A closer look at the distribution of 25- to 54-year-old full-time employed workers in each educational group by the level of their literacy and numeracy proficiencies can be found in figures 6 and 7. The concentration of workers in literacy proficiency level 1 or below and levels 4/5 varied sharply by educational attainment. Workers with literacy proficiency scores at or below level 1 accounted for 58 percent of high school dropouts and 23 percent of high school graduates, but fewer than 6 percent among college graduates with an associate's degree and 2 percent or less among those with a bachelor's or higher degree. Workers with literacy proficiency scores in levels 4 or 5 comprised much lower shares of those without a college degree than those with a college degree. Fewer than 5 percent of high school graduates and less than 1 percent of high school dropouts had literacy proficiency scores in level 4/5, compared to 13 percent among those who had completed some college without a credential, 9 percent among workers with a college/trade school certificate, and 16, 30, and 41 percent among those with associate's, bachelor's, and master's or higher degrees, respectively.
The results for numeracy proficiency were similar to the results for literacy proficiency. The share of workers with high levels of numeracy proficiency (levels 4/5) rose with education from just less than 1 percent among high school dropouts and 3 percent among high school graduates to 10, 24, and 34 percent among workers with associate's, bachelor's, and master's or higher degrees, respectively. Workers with numeracy proficiency scores at or below level 1 comprised 64 percent of high school dropouts and 37 percent of high school graduates and progressively smaller shares of workers with higher levels of education (Figure 7).
Although workers with a college degree were very unlikely to score at or below level 1 of the literacy and numeracy proficiency scales, sizable shares of college-educated prime-age, full-time employed workers had literacy and numeracy skills below level 3, which is considered a minimum standard for literacy and numeracy proficiencies associated with more positive economic, social, and educational outcomes.25 Level 3 skills are more sophisticated and require the ability to integrate different sources of information and solve complex problems. Over 13 percent of 25- to 54-year-old full-time employed U.S. workers with a master's or higher degree scored at level 1 or 2 on the PIAAC literacy proficiency test. This ratio was 18 percent among workers with a bachelor's degree and 36 percent among those with an associate's degree. For workers with a college/trade school certificate, the ratio was 47 percent. It was 41 percent for those with some college without earning a credential, while two-thirds of high school graduates and over 90 percent of high school dropouts had literacy proficiency scores below level 3 (Figure 6).
The results on the numeracy scale reveal that one-fifth of prime-age, full-time workers with a master's or higher degree had a numeracy skill score below level 3. Nearly 30 percent among those with a bachelor's degree and nearly one-half among workers with an associate's degree had sub-level 3 numeracy scores (Figure 7).
So even though average proficiencies of workers increase with their level of education, these findings reveal that there is a sizable variation in the level of skills of workers with the same level of educational attainment. We find this variation in basic skill proficiencies particularly pronounced among better educated workers with college credentials. It follows that we expect earnings of workers to vary by not only the level of educational attainment, itself a measure of human capital, but also by the level of literacy and numeracy skills within educational groups. In fact, our analysis of the level of skills within educational subgroups of prime-age, full-time employed workers reveal the existence of surprisingly large shares of college graduates who have achieved the level of literacy and numeracy proficiencies that studies have found to be below skill score levels thought to be associated with positive economic, educational, and social outcomes.26
In the human capital measurement framework, education is defined with a standard measure—years of schooling completed and the level of credentials earned—yet it is quite clear that education is far from a standardized product. Not all education is created equal. The quality of education at each level of attainment is highly varied with respect to the basic skills development of students, their acquisition of occupational knowledge, social skills, and fundamental character traits, and, therefore, their future labor market outcomes.
To understand that not all education is the same, one needs only to look at standardized test scores across different public schools in the same school district. For example, an examination of standardized scores of public high school students in Philadelphia found that two public high schools in the city are ranked among the top five high schools in the state of Pennsylvania in basic skill scoring; a total of 5 high schools in the city are ranked in the top decile of all high schools in the state, accounting for about 6 percent of public high school enrollment in the city. In contrast, 22 public high schools in the city are ranked in the bottom 10 percent of Pennsylvania high schools; these schools account for 45 percent of all public high school students in the city.27
At the postsecondary level, long-term employment and earnings experiences of bachelor's degree holders with no additional schooling are closely connected to their undergraduate field of study and vary quite sharply by major field.28 College credentials are awarded by thousands of institutions of higher learning across numerous fields of study. The variation in the quality of education provided by institutions of higher learning partly underlies the variation in skills and knowledge within groups with the same level of education. While individual skills vary among those with the same level of education, it is also very likely that individual skills vary, albeit not as widely, among individuals with the same level of education from the same institution and even the same field of study. This suggests that the strength of the link between education and earnings (and other labor market outcomes) is contingent upon, among other factors, the quality of the education provided and the choice of field of study at the college and university level.
20 Russell Rumberger, Dropping Out: Why Students Drop Out of High School and What Can be Done About It (Cambridge, MA: Harvard University Press, 2011).
21 Neeta P. Fogg and Paul E. Harrington, From Diplomas to Degrees, Project U-Turn and the Philadelphia Youth Network, 2015.
23 See Russell Rumberger and Sun Ah Lim, Why Students Drop Out of School: A Review of 25 Years of Research, California Dropout Research Project (Santa Barbara, CA: University of California, August 2009).
24 Ruth Neild and Robert Balfanz, Unfulfilled Promise: The Dimensions and Characteristics of Philadelphia's Dropout Crisis, 2000-2005, Philadelphia Youth Network, Johns Hopkins University, and University of Pennsylvania, 2006.
25 Goodman et al., America's Skills Challenge.
27 Paul Harrington, Ishwar Khatiwada, and Laura Romano, The Hierarchy of Secondary Schools in Philadelphia, 1199c Education and Training Fund, April 2017.
28 See Neeta P. Fogg and Paul E. Harrington, Determinants of the Hourly Earnings of College-Educated Immigrants, prepared for U.S. Department for Education with NOVA Research Company, March 2012; Neeta P. Fogg and Paul E. Harrington, Findings from an Examination of the Labor Force Participation of College-Educated Immigrants in the United States, prepared for U.S. Department for Education with NOVA Research Company, May 2012; Neeta P. Fogg and Paul E. Harrington, Unemployment Problems among College-Educated Immigrants in the United States, prepared for U.S. Department for Education with NOVA Research Company, August 2012; Neeta P. Fogg and Paul E. Harrington, Involuntary Part-Time Employment Problems among College-Educated Immigrants in the United States, prepared for U.S. Department for Education with NOVA Research Company, August 2012; Neeta P. Fogg and Paul E. Harrington, Mal-Employment Problems among College-Educated Immigrants in the United States, prepared for U.S. Department for Education with NOVA Research Company, October 2012.