Are our Disadvantaged Students Really Learning? – an Analysis of the 2017 US News High School Rankings

download

The latest US News & World Report High School Rankings (2017) indicate that disadvantaged populations of students are underperforming in Douglas County, and that high graduation rates have been achieved in large part by expedient means of moving such students through the system. This post will explain briefly the ranking’s methodology and then analyze some of the data to support this conclusion.

Only 1 of 5 high schools – Douglas County High – qualified for a ranking. To qualify for the lowest ranking – “Bronze” – schools must meet two criteria. First, their students as a whole must perform better on reading and math exams, administered by the state, than what is statistically expected based on the school’s poverty level (typically measured by the % of the student body on free and reduced lunch).  Second, the “disadvantaged” student groups (includes African-American, Hispanic, and poor students) must perform better than the state average for those populations.

To qualify for “Silver” or “Gold” ratings, there is an additional criteria: schools must exceed the median College Readiness Index (CRI). The CRI is comprised of both AP participation (how many students take an AP exam) and AP performance (how many students pass an AP exam). AP exams are used since these are college level exams through which students can earn college credit by passing.

To understand why 4 of 5 high schools did not qualify for a ranking, let’s look at Alexander High School as a case study (AHS consistently has the top CCRPI score in the district). In reading and math, Alexander students underperformed statistical expectations, having a gap score of negative 18 (a negative score means the actual performance was lower than expectations). Also, its disadvantaged populations scored 8.9 percentage points below the averages for disadvantage students throughout the state (only 12.7% of disadvantaged students were proficient in reading and math).

To understand why Douglas County High School earned a Bronze but not a Silver or Gold, let’s look at its statistics. Its gap score was positive 6.8, and its disadvantaged populations scored 6.5 percentage points above the state average. Thus, the first two criteria were met. Given the fact that its students with disabilities reading and math scores were on par or worse than the district as a whole, it is reasonable to assume that these test scores were boosted by the school’s population of International Baccalaureate students (at least 25% of test takers), which sees significant participation from African-American students.

DCHS did not qualify for Silver, though, because of low performance on AP exams. Even though the participation rate was fairly high (39%), the passing rate on all exams was only 17%, with 28% of seniors passing at least one exam (this is higher than 17% because it only requires one passed exam and excludes multiple failed exams). It is worth noting that in years past, DCHS ranked higher (recall the top 10% ranking?) because the rankings used IB exams instead of AP.

How does this relate to the high graduation rate spike in 2015? These latest rankings use 2015 academic data.  The fact that disadvantaged students are performing less than their peers while graduating at higher rates indicates clearly that the standards for graduation were lower. The high participation of disadvantaged students in the county’s suspect online credit recovery program, e2020, further supports this conclusion.

High Graduation Rates: An Illusion of Success?

illusions

The graduation rates for 2016 were released this week. While Douglas County’s rates dipped slightly from 2015 (88.5 to 87.1%), it still surpasses the state average significantly (79.2%) and remains higher than all surrounding districts. Since graduation standards are now completely established internally by the local district, one should critically examine whether or not higher rates are simply due to a district lowering its standards required for graduation.

Douglas County high schools have relied heavily on online credit recovery (OCR) courses to attain and maintain their high graduation rates. The sharp increase in 2015 was attributed by the district to “personalized learning programs,” which is a euphemism for OCR courses (this was one of the reasons given in the GA State Senate commendation the district received for the increases). Hundreds of DCSS graduates each year receive graduation credits from these courses.

senate-commendation

Yet, as an Atlanta Journal Constitution investigative report has shown, students learn very little in these courses (as indicated by proficiency rates on standardized tests). Proficiency rates statewide are around 10% while the failure rate is above 50%. In a sample of over 200 online students in Douglas County, the failure rate was over 70% in 2015.

online_classes_graphic_web

Earning a high school diploma is supposed to mean that a student is college ready. Perhaps a better metric of student achievement then is how graduates perform in college. According to data on the state DoE website, 37% of DCSS graduates enrolled in a public college or university in Georgia had to take at least one remedial class, and this number increased from the previous year. Compare this to a 25% rate across the state (which is still a problem!).

Our graduation rate is higher than the state average, yet more of our students have to take remedial classes in college.  How can this be?

 

 

A Response to a Defense of Online Credit Recovery

A few weeks ago, at the August 15th Board of Education meeting, I submitted a nearly 20 page report documenting problems with the online credit recovery program, e2020. The main reason I felt compelled to do this is that this program has been the major driver behind dramatic graduation rate increases, yet allows students to earn credits without really having to learn. So with the report, I made a public statement, calling for transparency and accountability. At the next BoE meeting, the school system issued a response to my claims. The response was incoherent in places, and filled with misleading hyperbole (the program is “the best in the state, if not the country”), downright deceptive claims (it was “setup to maximize learning”), and even personal attacks (that I was criticizing the program to “cause hurt to others”).  The local paper reported both on my initial questions and on the system’s response.

Exercising my blood-bought liberties of free speech and of a free press, I wrote a letter to the editor of the local paper.  I am grateful that they published it:

Dear editor,
I would like to recommend a few questions to the community to ask our school system in response to the claims that were made in defense of their online credit recovery program (e2020). One claim was that e2020 is setup to “maximize learning.” What is the evidence that most students are learning at high levels, if at all, in these courses? Another claim was that the program has contributed to graduation rate increases, but is not the only factor. Exactly what percentage of graduates have been earning these credits in recent years? And if the program really is “the best in the state, if not the nation,” why downplay the significance of its contribution? Instead, why not broadcast loudly and invite the nation to look at how this program helped create a quasi educational miracle in 2015? (Interestingly, I was told by an administrator in private that they were “thinking about scrapping e2020 altogether” – why even consider this? And why are they exploring new vendors?) Finally, the issues I raised were valid enough to be “investigated and acted upon when deemed necessary.” If so then, what were these issues that demanded action, and why would a top notch program warrant such significant changes?

Facts and evidence were conspicuously missing from this public statement. I had submitted to the board a lengthy document of evidence about problems and abuses in e2020, including, but not limited to, very low test scores (6% proficiency on state exams; and a 7-8th grade reading level, on average) in contrast with very high course grades (over 90% of the same sample making As or Bs); specific ways students were regularly using the Internet to cheat on tests; how students were allowed to change incorrect answers before submitting tests; and how some took fast track courses whereby full-year courses were completed in a few weeks or even days so that they could graduate on time.

Instead of trying to discredit critics by attacking motives (how is that relevant to whether students are learning in e2020?), I would hope the leadership would instead refute claims by focusing on the facts and constructing their own fact-based arguments. As one senior administrator told me, “If there is a problem, there is a problem.” What is needed is transparency about the problem so that there can be public accountability for solving it, for our children’s sake.

Jeremy Noonan