kitchen table math, the sequel: Speaking of millennials

Wednesday, February 18, 2015

Speaking of millennials

Which I was ....

Education Week reports that a new set of dismal international comparisons--of millenials, this time--is coming out next week. I can't tell whether the test is good, but, then again, I'd just as soon U.S. high school graduates not score lower than high school graduates in every country except France no matter how possibly lousy the test.

[pause]

Yikes.

Here are 3 sample math questions.

Down the rabbit hole:
It's far from the first study to suggest American students are falling behind their international peers. But the analysis of U.S. millennials—those born after 1980, ages 16 to 34 during the study—specifically highlights that the skills gap goes beyond young people who are typically seen as more "at-risk," like immigrants and high school dropouts.

[snip]

The ETS study, to be released this week, compares millennials in 22 industrialized countries, including the United States, who took part in the Program for the International Assessment of Adult Competencies, or PIAAC, in 2012, the last time it was given. The Organization for Economic Cooperation and Development, which runs the school-based Program for International Student Assessment, also runs the PIAAC, which is given directly to adults in their homes. Unlike the PISA, which measures academic content skills, the PIAAC measures practical, career-oriented literacy and numeracy skills, and, as of 2012, "problem-solving in technology-rich environments."

[snip]

Across the board, young Americans fared poorly compared to those in the other countries studied. They tied for last, with Italy and Spain, in math skills. In problem-solving, they again performed at the bottom of the pack, with Ireland, Poland, and the Slovak Republic. U.S. millennials also had lower literacy scores than peers in 15 out of 22 countries, tied with a few, and outperformed only peers in Italy and Spain.

[snip]

. . . the skills gaps persisted among students who are least likely to be considered academically at risk. Those who performed in the top 10 percent of all Americans in their age group still performed worse than the top performers in 15 other countries, including Germany and the Republic of Korea.

While a higher proportion of U.S. millennials versus those in other countries had earned a college degree, those with a four-year degree in the United States still showed lower math skills than those with college degrees in any country studied but Poland and Spain.* Moreover, the percentages of Americans who demonstrated the lowest-level math skills increased from 2003 to 2012, regardless of what level of education they had achieved.

[snip]

Even those with a master's or doctoral degree demonstrated lower numeracy skills than their counterparts in all but a few countries. The average U.S. math score for millennials with a postbaccalaureate degree, 308, was not only below the average for countries studied who are in the Organization for Economic Cooperation and Development, but was below the average score for young adults with just a bachelor's degree in several countries, and near the score for top-performing students with less than a bachelor's degree in a few countries.

[snip]

American millennials with a high school diploma or less performed lower than those with a secondary credential in every country but France.

[snip]

. . . in the United States, native-born millennials showed a greater decline in skills from the 2003 to 2012 cohorts than did their immigrant peers.

[snip]

Racial and ethnic performance gaps continued, with 12 percent of white and Asian young adults in America showing advanced levels of math skill, versus only 3 percent of Hispanic and 1 percent of black millennials. Yet on average, OECD countries had about 15 percent of all millennials performing at an advanced level, and white and Asian students in the United States performed below their counterparts in most other countries studied.

From the department of unintentional irony:
While a higher proportion of U.S. millennials versus those in other countries had earned a college degree, those with a four-year degree in the United States still showed lower math skills than those with college degrees in any country studied but Poland and Spain.
If a higher proportion of U.S. millennials finish college, then you would expect college-educated U.S. millennials as a group to have a lower average math score than college-educated millennials elsewhere. [Copy edit courtesy of Glen]

I wonder how the numeracy skills of U.S. education reporters stack up compared to those of education reporters in France and Spain.
.

6 comments:

Glen said...

If a higher proportion of U.S. millennials finish college, then you would expect U.S. millennials as a group to have a lower average math score.

(Presumably you meant "you would expect *college-educated* U.S. millennials" to do worse than college-educated millennials elsewhere.)

Not if our colleges maintained high and unwavering standards of math proficiency for graduation regardless of who they admitted. If the truth were really that college took "raw recruits" and hammered them into educated scholars, then having a higher proportion of college grads would just mean more people who were really good at math.

Of course, if that is not the case, and most of your math proficiency as an adult is the result of factors other than your college experience, then you would expect the math ability of college grads to fall as colleges lowered admissions standards to admit a larger portion of the population.

Catherine Johnson said...

True!

Unknown said...

The edworld, overall, has little to no interest in challenging the top kids. The dreaded GAP rules all. Let no child get ahead.

Unknown said...

The edworld, overall, has little to no interest in challenging the top kids. The dreaded GAP rules all. Let no child get ahead.

Anonymous said...

"If the truth were really that college took "raw recruits" and hammered them into educated scholars, then having a higher proportion of college grads would just mean more people who were really good at math.

Of course, if that is not the case, and most of your math proficiency as an adult is the result of factors other than your college experience, then you would expect the math ability of college grads to fall as colleges lowered admissions standards to admit a larger portion of the population."

Since none of the three example question involve college level math(*), I'm pretty confident that this is measuring a drop in admissions standards.

-Mark Roulo

(*) The first question asked the kids to properly read an analog thermometer with both F and C scales and then subtract 30 from 25.

The second question required converting Gigawatt-hours to Megawatt-hours and then dividing one number into a second.

The third question required reading a chart with six data points on it and knowing which points were lower than the point to the left.

SteveH said...

I want to see the full test and the raw percent correct scores, not relative comparisons of scores where I can't see how they were derived from the raw score. It could be that the whole world stinks on an absolute basis. However, if you only do relative comparisons (which is the case in the entire ed world) then you will probably fix the wrong things.

When I was on a parent/teacher committee long ago that analyzed NCLB data to figure out why our students' "problem solving" skills were down that year, the solution was to work harder on problem solving.

Testing higher level skills is the wrong place to start. Why can't an adult read and understand a graph? How can you tell? Do they know about lines and slopes? Is it a lack of experience? Do they test with a simple graph and then try one where a graph of births is placed on top of a photo of baby holders/carts - as in one of the test problems?

Everyone knows there is a problem, so what is the purpose of this test - to allow everyone a chance to claim that their solution will fix the problem? What problem, exactly?

Is the goal of education to fix this relative difference even if there is still a huge absolute problem? How about the problem that the ed world has officially admitted with CCSS that K-6 is a complete NO-STEM zone. It's right there in black and white. What happened to all of the calls for STEM education? They're gone. Many have moved on to STEAM, thereby weakening and confusing the issue.

Meanwhile, many urban parents can easily tell that school 'B' will be better for their child than school 'A', but educators fight to keep them from having that choice.