From a 2015 ETS publication:
This report was written by:
Madeline J. Goodman
Anita M. Sands
Richard J. Coley
Educational Testing Service
This report, the first in a series to be produced by Educational Testing Service using data from the Programme for the International Assessment of Adult Competencies (PIAAC), is an attempt to focus attention on a topic of interest to a broad range of constituencies. The subject of this report is our nation’s millennials, those young adults born after 1980 who were 16–34 years of age at the time of the assessment. …
How do the average scores of U.S. millennials compare with those in other participating countries?
• In literacy, U.S. millennials scored lower than 15 of the 22 participating countries. Only millennials in Spain and Italy had lower scores.
• In numeracy, U.S. millennials ranked last, along with Italy and Spain.
• In PS-TRE, U.S. millennials also ranked last, along with the Slovak Republic, Ireland, and Poland.
• The youngest segment of the U.S. millennial cohort (16- to 24-year-olds), who could be in the labor force for the next 50 years, ranked last in numeracy along with Italy and among the bottom countries in PS-TRE. In literacy, they scored higher than their peers in Italy and Spain.
How do U.S. top-performing and lower-performing millennials compare to their
international peers? What is the degree of inequality in the score distribution?
• Top-scoring U.S. millennials (those at the 90th percentile) scored lower than top-scoring millennials in 15 of the 22 participating countries, and only scored higher than their peers in Spain. …
We will focus solely on the numeracy scale in our examination. We do so for the following reasons.
First, a number of reports on U.S. performance have already covered the ground on literacy.
Second, the PS-TRE assessment—while an innovative and important measure of problem-solving skills—has a more limited number of participating countries, a more restricted sample of participants overall, and (because the measure is new) no trend data.
Third, researchers have found that numeracy is a robust predictor of labor market success.
Finally, the relatively poor performance of the U.S. on the numeracy scale—compared to the previous assessment year and in relation to other countries in 2012—calls for greater scrutiny of how different demographic subgroups in the U.S. performed on this measure and what these patterns might suggest in terms of policy recommendations. …
For the most part, the racial/ethnic gaps identified in educational achievement and attainments at the K-12 level are borne out in the PIAAC skills data on millennials. Gaps in average numeracy scores are evident (table 6). White and Asian millennials outperform their Black and Hispanic peers, though the scores for Asian and White millennials do not differ significantly from one another as they do in many of the K-12 national assessment results. The overall demographic trends evident in the K-12 population—with an increasingly more diverse school age population—are present when we look at age segments of the adult population, with millennials decidedly more diverse than older adults. While Whites comprise 70 percent of the population of adults over the age of 35, they are only 58 percent of millennials.
In what ways do race/ethnicity influence our understanding of the overall performance of U.S. millennials? As a means of comparison, 64 percent of millennials in the U.S. performed below the minimum standard (below level 3) in numeracy, compared to 47 percent of millennials in the OECD average. Fifty-four percent of White millennials and 52 percent of Asian millennials performed below this level, as compared to 83 percent of Hispanic and 88 percent of Black millennials.
The performance of White and Asian millennials, however, still does not reach the level of the top performers internationally and remains below the OECD average. In fact, average scores and percentages of U.S. White millennials that performed below level 3 are similar to those of millennials in France (average score of 267 and 54% scored below level 3), which ranks near the bottom internationally. While a greater percentage of White and Asian/Pacific Islander millennials (12%) performed at the highest proficiency level (level 4/5) compared to Hispanic (3%) or Black (1%) millennials, these percentages are still lower than the OECD average (15%) and the percentages of millennials at this level in top-performing countries (Finland at 26% and the Netherlands at 21% ). The issues of race and ethnicity clearly impact our understanding of how skills are distributed among our young adult population and deserve further attention and research.
Race/ethnicity and educational attainment. As with adults overall, differences in the performance within racial/ethnic groups in the U.S. are associated with different levels of educational attainment. (Note: Due to sample sizes in the PIAAC data for race/ethnicity in the U.S., performance for racial/ethnic groups by levels of educational attainment could not be estimated for millennials and is reported here for adults 25–65). Across all racial/ethnic groups, those with greater levels of educational attainment scored higher than adults with less education (table 7).
However, the distribution of the population among levels of educational attainment differed by race/ethnicity.
For example, 57 percent of the Hispanic adult population reported having an upper secondary education or less, compared to 38 percent of White adults, 46 percent of Black adults, and 23 percent of Asian/Pacific Islander adults. In terms of post-secondary education, a greater percentage of White (34%) and Asian/Pacific Islander (34%) adults age 25–65 reported that their highest level of educational attainment was either a post-secondary non-bachelor’s degree or a four year bachelor’s degree as compared to Black (27%) or Hispanic (16%) adults in this age group. Given the previously noted association between educational attainment and skill level, it is therefore not surprising that Hispanic and Black adults age 25–65 (as well as millennials) performed worse than their White and Asian peers.
Variation in performance among racial/ethnic groups persists, however, even for those with similar levels of education. For example, 95 percent of Black adults age 25–65 reporting their highest level of education as upper secondary and 86 percent of Hispanic adults at this educational level performed below the minimum standard (below level 3) in numeracy compared to 71 percent of White adults. This differential in performance, particularly the gap in percentages of White and Black adults that scored below level 3, was noted at each level of educational attainment (with reportable results). In fact, Black adults age 25–65 consistently scored about 50 points lower than their White peers across most educational attainment categories, where sample sizes were adequate to allow for a reliable estimate. Equally alarming, there is no difference between the percentages below level 3 for Black adults that report having only a high school credential (95%) and those that report having some post-secondary (nonbaccalaureate) education (91%).