Richard Innes

1/ I become more and more dismayed by the huge number of supposed education experts who regularly use data from the NAEP improperly. In this thread I plan to outline some of the issues so the rest of us will have a better understanding of how people mislead with NAEP.

But the problem with such studies is they don't check the quality of remedial programs for the retained students. Holding a student back only to receive more of the same approach that already failed him or her doesn't really tell us much. Can you cite studies that include such analysis?

The majority of published studies and decades of research indicate that there is little to be gained, and much harm being done when retaining students. tultican.com/2026/03/22/gra… #EducationMatters

@rpondiscio No state is really excelling in Grade 8 Reading, but Mississippi is making a good start at it. You have to break NAEP down by race to see this.

This point can't be made enough, particularly in response to "gotcha" claims calling into question gains made in Mississippi, et al. No one has cracked the code on 8th grade reading. And the answer likely has less to do with policy than pedagogy: aei.org/op-eds/crackin…

4/Also note that Thomas presents the first table below of Black student scores in 2024 in Grades 4 and 8 in his substack to claim that Mississippi isn’t performing in Grade 8 for Black students. This table is fundamentally misleading, however. To start, there are statistical sampling errors present in all NAEP scores, and a proper analysis must consider this. Once those errors are considered, the picture is quite different from the Thomas table. Some of the states Thomas color codes as scoring below Mississippi in Grade 4 NAEP Reading are actually statistically tied with Mississippi. The second table, generated with the NAEP Data Explorer from NCES, shows a far more accurate analysis. FL, GA and MA are tied with MS for Grade 4 NAEP Reading in 2024 as close as the assessment can confidently tell us. The color coding for them is wrong. NAEP only confirms that just LA and CA scored statistically significantly lower (Department of Defense Education Agency schools are a special case, nothing like a state, and are unfairly compared in Thomas’ table). Regarding Black student results in Grade 8 Reading, the third table below shows NAEP can really only be used to claim that Black students in the states of Colorado and New Jersey truly outperform those in Mississippi. Again, the color coding is misleading. The entire construction of Thomas’ table is technically flawed, rendering it misleading and inaccurate.

The False Promise of Education "Miracles" and Misunderstanding Standardized Test Scores open.substack.com/pub/paulthomas…

3/ Thomas’ agenda becomes clearer after reading this passage in the substack: “Some have begun adjusting test scores by poverty and demographics, establishing Mississippi as the top state in the US. The ugly aspect of this is that a statistical technique doesn’t remove those students from poverty and racism, but this approach feeds not addressing poverty and racism.” Thomas essentially tries to use poverty and racism to excuse for poor public school performance. That lets schools off the hook for making any improvement as these factors are beyond school control.. The facts are that Mississippi directly challenges Thomas’ assertions. Schools can make differences despite poverty, but he won’t admit that.

2/ The substack claims, “Mississippi (like Florida) has exceptional grade 4 reading scores, but by grade 8, those scores are back at the bottom.” That’s no longer true. Check the Grade 8 tables below for a better picture. Note in particular the columns labeled “significantly higher,” which show the number of states scoring statistically significantly higher that the state listed for each row of data. White students in a total of 43 states outscored Mississippi’s in 2013 by a statistically significant amount. By 2024, only whites in 7 states could make a similar claim. Black students in Mississippi did even better. Back in 2013, Black students in 27 of the 42 states with scores outscored Mississippi’s Blacks by a statistically significant amount. By 2024, Black students in only 2 states could make a similar claim.

1/ A comment on the troubled “On education miracles in general (and those in Mississippi in particular)” paper mentioned in the substack. This paper has already seen one correction (In February 2026) (academic.oup.com/jrssig/article…). This is for an obvious gross error in the original (statmodeling.stat.columbia.edu/wp-content/upl…). More corrections are needed. For example, the original paper claims that in MS, “Prior to 2013, a higher percentage of third-graders moved on to the fourth grade and took the NAEP fourth-grade reading test. After 2013, only those students who did well enough in reading moved on to the fourth grade and took the test.” Evidence from the NAEP about the ages of test takers shows that is also incorrect. In fact, the percentage of students over modal age in the NAEP Grade 4 Reading Assessments has stayed rather constant for decades. The paper’s authors were working from assumptions by others that also were wrong.

2/ A comment on the troubled “On education miracles in general (and those in Mississippi in particular)” paper you mentioned is also warranted. This paper has already seen one correction (In February 2026) (academic.oup.com/jrssig/article…). This is for an obvious gross error in the original (statmodeling.stat.columbia.edu/wp-content/upl…). More corrections are needed. For example, the original paper claims that in MS, “Prior to 2013, a higher percentage of third-graders moved on to the fourth grade and took the NAEP fourth-grade reading test. After 2013, only those students who did well enough in reading moved on to the fourth grade and took the test.” Evidence from the NAEP about the ages of test takers shows that is also incorrect. In fact, the percentage of students over modal age in the NAEP Grade 4 Reading Assessments has stayed very constant for decades. The paper’s authors were working from assumptions by others that also were wrong.

Grade 8 NAEP: A More Valid Reading Story (Although Still Incomplete) open.substack.com/pub/paulthomas…

1/ You claim “Black students in CA, GA, LA, (and) MA” outperform those in Mississippi for NAEP Grade 8 Reading. You are totally ignoring the fact that all NAEP scores have statistical sampling errors and what looks to you like a clear difference is in fact not so. Actually, among the states you mention, only MA definitely outscored MS for Black students in the 2024 Grade 8 NAEP Reading. See the bottom half of the tables below. Below are the scores and statistically significant difference information for all states in the 2024 NAEP Grade 8 Reading assessment, presented separately for white and Black students. We’ll concentrate on the bottom right table for Black student Grade 8 NAEP scores from 2024. Pay attention to the column marked "significantly higher." This shows the number of states that scored statistically significantly higher than the state listed at the beginning of each row. Mississippi is selected as the “focal state” in this table, so the column labeled “Cross-jurisdiction significant difference” shows states where Black students definitely performed better than MS in 2024. There are only two of them. CA, GA and LA are not in this listing. In 2024, only the Black students in CO and MA outscored those in MS by a statistically significant amount. Just those 2. If you refer to the bottom left table, which covers Black student results from 2013, you can see MS has made very notable progress for Grade 8 NAEP Reading compared to other states. Your claims about MS are wrong. Even in Grade 8, improvement compared to other states is becoming clear.

In Kentucky, a dominant hospital system can legally sue a competitor out of existence during the CON review process — before the new provider opens its doors. That's not patient protection. That's institutional self-protection. Rep. Proctor's legislation ends it. @MarianneProc60

EdNext reports standards for AP exams have been lowered by the College Board. Makes you wonder if standards for the College Board's SAT, the test the Kentucky Dept. of Ed. just purchased for all 11th graders, are also lowered. educationnext.org/dumbing-down-o…

Thomas' NAEP summary is on the left. It claims multiple states scored above MS on Grade 8 NAEP Reading. But he totally ignores the sampling errors in those scores. The bottom right table with blue highlighting shows that once the sampling errors in the NAEP are considered, only CO and MA scored statistically significantly higher than MS for Black students in 2024. It seems a lot of misunderstanding of test scores is out there.

The False Promise of Education "Miracles" and Misunderstanding Standardized Test Scores open.substack.com/pub/paulthomas…

2/ A comment on the troubled “On education miracles in general (and those in Mississippi in particular)” paper you mentioned is also warranted. This paper has already seen one correction (In February 2026) (academic.oup.com/jrssig/article…). This is for an obvious gross error in the original (statmodeling.stat.columbia.edu/wp-content/upl…). More corrections are needed. For example, the original paper claims that in MS, “Prior to 2013, a higher percentage of third-graders moved on to the fourth grade and took the NAEP fourth-grade reading test. After 2013, only those students who did well enough in reading moved on to the fourth grade and took the test.” Evidence from the NAEP about the ages of test takers shows that is also incorrect. In fact, the percentage of students over modal age in the NAEP Grade 4 Reading Assessments has stayed very constant for decades. The paper’s authors were working from assumptions by others that also were wrong.

Grade 8 NAEP: A More Valid Reading Story (Although Still Incomplete) open.substack.com/pub/paulthomas…

1/ You claim “Black students in CA, GA, LA, (and) MA” outperform those in Mississippi for NAEP Grade 8 Reading. You are totally ignoring the fact that all NAEP scores have statistical sampling errors and what looks to you like a clear difference is in fact not so. Actually, among the states you mention, only MA definitely outscored MS for Black students in the 2024 Grade 8 NAEP Reading. See the bottom half of the tables below. Below are the scores and statistically significant difference information for all states in the 2024 NAEP Grade 8 Reading assessment, presented separately for white and Black students. We’ll concentrate on the bottom right table for Black student Grade 8 NAEP scores from 2024. Pay attention to the column marked "significantly higher." This shows the number of states that scored statistically significantly higher than the state listed at the beginning of each row. Mississippi is selected as the “focal state” in this table, so the column labeled “Cross-jurisdiction significant difference” shows states where Black students definitely performed better than MS in 2024. There are only two of them. CA, GA and LA are not in this listing. In 2024, only the Black students in CO and MA outscored those in MS by a statistically significant amount. Just those 2. If you refer to the bottom left table, which covers Black student results from 2013, you can see MS has made very notable progress for Grade 8 NAEP Reading compared to other states. Your claims about MS are wrong. Even in Grade 8, improvement compared to other states is becoming clear.

But Dan concludes that improvement is beginning to show in Grade 8. I agree that gaps remain a challenge, so I don't call MS a miracle. But some noteworthy things are happening, and I don't understand why there isn't an army of researchers camped out trying to figure out which elements in the reform are having the most impact so all can better benefit. Much of the reporting I have seen makes it obvious the writers have spent no time on the ground in MS or talking to those who are. Many reporters make mistakes about things like when the 3rd Grade Gate really started (no, it wasn't in 2013) and when retention increased (actually it was already high in the early 1990s and the age breakdown of NAEP takers didn't change much after the reform started). Also, the improvements are more than a "glimmer" for both white and Black students. In 2013, white students in 43 other states statistically significantly outscored MS' whites. In 2024 only whites in just 7 states could make that claim. For Black students, in 2013, those in 27 out of 42 states that got scores statistically significantly outscored those in MS. In 2024, only Black students in 2 states could make that claim. That's not just a "glimmer."

@Innes434 @Dale_Chu According to Dan McGrath, that glimmer of improvement is from the higher achieving students. Low achieving students, not so much. I recommend reading this deconstruction of MS 8th grade scores: danmcgrath2.substack.com/p/did-mississi…

“They used to say the progression in school was you learn to read and then you read to learn. Now people realize it needs to be both for much longer. ‘Reading to learn’ should start earlier, and ‘learning to read’ must continue well past third grade.”

@plthomasEdD Nice listing of Ed School type research. But, is it really valid? Lots of close reading needed. I wonder how many of these researchers know what is really happening in Grade 8 NAEP in Mississippi. Check this out:

Research Contradicts “Science of Reading” Media Story Being Sold open.substack.com/pub/paulthomas…

While the "science of" folk market their fake miracles and ed reform by statistically adjusting student test scores for poverty Those students STILL LIVE IN VERY REAL POVERTY [NOTE: Raw NAEP scores were "good" to declare MS a miracle, but "bad" when they prove it isn't]

@plthomasEdD No longer true for Mississippi. Gains now starting in Grade 8 NAEP Reading, too.

Close but no. Southern states boost early grade TEST SCORES, but scores drop in middle school because it was data manipulation, not learning paulthomas701128.substack.com/p/grade-8-naep…

The table on the left isn't really accurate. It fails to consider the sampling errors in all NAEP Scores. The table in the lower right part of those on the right show that once sampling errors are properly considered, only 2 states outscore MS in 2024 for Black student scores. The tables on the right also show MS is starting to show improvement in NAEP Grade 8 Reading.

@jillbarshay Grade 8 NAEP: A More Valid Reading Story (Although Still Incomplete) paulthomas701128.substack.com/p/grade-8-naep…

@palan57 You need to be updated. Mississippi's gains in Grade 4 NAEP Reading are starting to appear in Grade 8, as well. Pay particular attention to the changes in the number of states scoring statistically significantly higher than MS.

CURMUDGUCATION: The Mystery of Eighth Grade Test Results curmudgucation.blogspot.com/2026/03/the-my…

@WSJopinion @shadwhite Well, the Mississippi gains in Grade 4 NAEP Reading are now starting to appear in Grade 8, too. Especially note the changes in the number of states scoring statistically significantly higher than MS for both white and Black students.

The “Mississippi Miracle” will fizzle if the state fails to extend gains to later grades with school choice, writes @shadwhite on.wsj.com/3NbwmJG