White minus black achievement gaps continue problematic in new NAEP results – Grade 4 Reading

The 2017 National Assessment of Educational Progress (NAEP) results for Grade 4 and Grade 8 mathematics and reading have finally been released, and it is time to examine a key problem in Kentucky’s schools – the chronic achievement gaps between the state’s white and black students.

Here is how the NAEP Grade 4 Reading assessment proficiency rates have trended for Kentucky’s white and black students since NAEP first reported state results in 1992.This isn’t a happy graph. Click the “Read more” link to learn why.

G4 Reading White Minus Black Gaps on NAEP by Year

G4 Reading White Minus Black Gaps on NAEP by Year

First, a few comments about how the graph is designed.

All the proficiency rate numbers come from the NAEP Data Explorer web tool. The Data Explorer put a superscript number 1 after the number for each year (1992 and 1994) where the NAEP didn’t offer any testing accommodations. More recent NAEP results include scores for some learning-disabled students who did receive NAEP-accepted accommodations.

White proficiency rates are found along the blue line and black scores are found along the red line.

The proficiency rates for white and black students for each year are in the yellow shaded boxes and the gaps between those proficiency rates for each year are shown by the numbers in the middle of the arrows for each test year. For example, in 1992 whites scored 24 percent proficient on NAEP Grade 4 reading while only eight percent of the black students met this level of performance. That difference was a proficiency rate achievement gap of 16 points.

Now, let’s explore what this graph shows. There are some pretty disturbing things in NAEP’s messages about reading in Kentucky.

To begin, the 2017 proficiency rate achievement gap of 25 percentage points is the second largest ever reported. However, due to the relative insensitivity of NAEP to gap changes due to the small numbers of black students tested by the assessment and the correspondingly large measurement errors, the NAEP Data Explorer’s gap significance test feature indicates that the latest gap isn’t larger than the measurement errors and cannot be declared statistically significantly different from the gaps in any of the previous years. That is a real shame because a 9-point proficiency rate change since 1992 – especially going in the wrong direction – is actually a fairly large performance difference.

Here’s another point: the NAEP Data Explorer indicates that the black Grade 4 NAEP Reading proficiency rate in 2017 is essentially no different from rates all the way back to 1992 once the sampling errors in the scores are considered. But, forget the statistics for a moment. It is clear that after more than a quarter of a century of KERA, if Kentucky’s average reading proficiency for black students is only 16 percent, then our public school system has failed to deliver on a promise to better serve these students’ need to learn to read.

Furthermore, as we have discussed before, there are concerns that the NAEP’s transition to digital testing in 2017 might have created some problems for comparison to earlier testing, and the impacts might be more severe for less advantaged students. Could that be borne out for the NAEP Grade 4 Reading Assessment by the 2015 to 2017 score changes in the graph? White reading proficiency in Kentucky dropped by only 3 percentage points, but black reading tumbled down by 7 percentage points. That is a pretty big drop.

I looked at the KPREP fourth grade reading scores for both races for 2015 and 2017. The results are in this table.

KPREP G4 Reading for 2015 and 2017

KPREP G4 Reading for 2015 and 2017

Like the NAEP, KPREP reported drops in both white and black Grade 4 reading scores in Kentucky between 2015 and 2017. Also like the NAEP, the black scores dropped more. The magnitude of the drops for both races were only about half the drops NAEP reported, but once you consider the sampling error in the NAEP scores, there isn't enough evidence to show a definite problem with NAEP in 2017.On the other hand, there appears little doubt that there were real performance drops in Grade 4 reading in Kentucky between 2015 and 2017, and the drop was larger for the state’s black students.

After more than a quarter of a century of KERA, when KPREP says only about one in four of the state’s black students read proficiently and the NAEP says that the proficiency for blacks is even lower than that – and that the trends are getting worse – Kentuckians need to ask some very serious questions about the performance of their schools. Clearly, it is time for some changes, and given the current economic climate in Kentucky, it doesn’t look like just throwing still more money at this problem is either affordable or likely to produce the desired results. Something else – like expanding school choice – certainly seems in order.