LA needs to stay on education reform course

The fault in Louisiana’s continuing drop in ACT Test scores lies in us, even in the face of a similar national trend.

Louisiana’s decline has gone from 19.5 (out of 36 points) in 2017 to 19.2 in 2018 to 18.8 in 2019 and 18.7 in 2020. However, that trend also appears nationally. Among the other 11 states where all students took the ACT in this period, only two had an increase from 2017-18, one had an increase from 2018-19, one had an increase from 2019-20, and only Nevada’s rose in this period – and it’s the lowest performer. Nationally, which includes all students including those in states they may take it voluntarily (which disproportionately excludes lower performers), scores also dropped from 21 to 20.6.

This can’t be written off as an artifact of fewer students taking it – which has seen a drop every year from 2017 of 2.03 million to 1.67 million in 2020 – because less able students typically eschew it. As well, the pattern downward replicated across the must-take states.

So, the Louisiana decline does seem part of larger trends. Some will argue the acceptance of Common Core starting a decade ago (with scores starting to fall around 2013) has something to do with it. But the problem in conflating the events is that in most states these curricula didn’t gain steam until just a few years ago, meaning these would have limited impact on test-takers of today and certainly in years past, as well as critics often caricature Common Core as teaching to a test, which would imply that standardized test scores should go up even as actual education breadth and intelligence developed from that would decline.

And, there’s the overblown argument that such a standardized test reveals little more than test-taking skills, despite research proving otherwise, or reflects too much differential access to educational resources, which is irrelevant to an argument about whether better or worse learning is occurring. Thus, as the test hasn’t changed a decline must show deteriorating performance across the board.

Still, the proportional fall of 4.2 percent made it the third-biggest loser among the all-student states, which in most cases lost 2 percent or less proportionally. Therefore, even if a national trend pushed scores downwards, it appears Louisiana experienced something else.

That could be from the adjusted cohort graduation rate (ACGR), or what proportion of public high school freshmen graduate with a regular diploma within four years of starting ninth grade. The fewer that graduate, the fewer that take the ACT. If Louisiana’s increase from 2013 – when it first began requiring the exam for all intending to graduate and the secular decline in scores started in earnest nationally – of seven percent through 2018 (the last year data are available) outpaced proportionally other states, that would mean disproportionately marginally-performing students who might not have graduated in years past did through 2020.

Reviewing data from the other states that requires taking the ACT, Alabama, Mississippi, and Nevada all had higher rates of ACGR increase (assuming it carries forward into 2020), and their change in scores all showed less deterioration (as noted, Nevada’s increased) than Louisiana’s. Most of the states with lower rates of increase (Oklahoma actually fell) declined proportionally less in scores relative to their increased rate of completion.

Test-taking did increase according to the ACT organization in Louisiana, about 2.8 percent in the 2017-20 period. But it rose the same among the other states. So, this also doesn’t shed light on the disproportionate Louisiana score decline.

Turning from exogenous factors, the explanation therefore must extend from internally. From 2017-20, completion of college core coursework recommended by the ACT organization fell nine percentage points in Louisiana. But in the other states, it fell 12 points – and they take about 11 percentage points fewer such courses anyway. (Making these data a bit murkier, nonreporting of the kind of class increased substantially in all 100 percent test-taking states in this period.) So, it’s not a matter of proportionately fewer students taking such courses in Louisiana.

Nor do reduced higher education aspirations seem to explain. These fell as well in Louisiana – but not nearly as much proportionally as in the other states. (In both, the proportion of respondents giving no response sharply rose, more in other states, which could mean either no intent going beyond high school or they didn’t care to answer.) Nor is it a dramatically differential change in the mix of parental educational achievement, where a drop in overall greater such educational attainment could trigger less achievement (because greater value placed on achievement by parents strongly influences child achievement), which actually remains higher in Louisiana and didn’t decline as much (again, since 2017 nonresponse increased in all substantially).

But these dynamics change when considered STEM-(science, technology, engineering, mathematics) interested students and those not. The latter drove the changes seen among all (including the soaring nonresponse rates for coursework and parental achievement and no aspiration given), as the former cohort, in all states, showed roughly the same patterns throughout 2017-20. Particularly interesting is that a significantly higher proportion of non-STEM Louisiana students (and to a lesser degree with STEM students), because of the state-mandated curriculum, take college preparatory courses, which the ACT organization says is key to higher test scores.

And, much as we who teach outside of STEM are loath to admit it, STEM coursework inherently demands more in critical thinking skills (although you can structure a non-STEM course to demand a lot of critical thinking; take one of my courses to find out that), which may explain the fall in ACT scores. Among the mandatory ACT test states excluding Louisiana, STEM-interested student totals dropped 36 percent over 2017-20 while they increased 34 percent among the non-interested.

So, if Louisiana numbers showed a greater STEM drop and greater non-STEM rise, that would explain the higher relative fall in scores, right? It might, except that Louisiana STEM-interested students fell by 14 percent while the non-STEM cohort increased by 6 percent. That would indicate that Louisiana should have bucked the trend, instead of the opposite.

As we eliminate some explanations, what’s left must explain, even if difficult to measure. And, what’s left seems to point at instructional quality, which can be addressed through two paths. One is in enhancing delivery, a more micro-level concern where it will take time for reforms enacted within the past decade to bear fruit. The other is in policy that encourages quality, a more macro-level issue that also more often than not takes time to have an impact but also can provide quicker stimuli to evoke change.

For example, after a flurry of educational reform activities in Louisiana in the first half of last decade, momentum slowed considerably in the second half, particularly with the ascension of a governor hostile to objective school and teacher accountability measuring. This could translate fairly quickly into worse outcomes.

If this downward trend has a message, it’s that Louisiana needs to stay the course yet increase its demand for higher quality and accountability from its delivery of elementary and secondary education.