CKLA:

CKLA is an English Language Arts curriculum developed by the Core Knowledge Foundation and published by Dr. E.D. Hirsh. “The Core Knowledge Sequence is a detailed outline of recommended content knowledge and skills to be taught in language arts, history and geography, visual arts, music, mathematics, and science from preschool through grade eight. The Sequence complements the general skills and objectives typically found in state and local curriculum guides.” The curriculum is very detailed at 357 pages long. It includes explicit instruction on phonemic awareness, phonics, morphology, spelling, vocabulary, fluency, and comprehension. Fundamental to this curriculum is its systematic development of background knowledge instruction. Individuals such as Daniel Willingham, and Hirsh Jr. have theorized that systematic knowledge building can improve reading comprehension. For more information about this theory click here: https://www.pedagogynongrata.com/_files/ugd/237d54_ba203a114c7a4081932e7d84810d807e.pdf

I have been publicly critical of the strength of the claims often made surrounding this theory. However, I think it should also be noted that there is no evidence of harm from a curriculum that systematically builds background knowledge. Moreover, there is certainly a theoretical benefit. 

Amplify CKLA:

Is based on the CKLA program but has more diverse topics, as well as several computer-based improvements that include: readymade slide decks, Dynamic live review tools, learning management systems, animated videos, recorded read-alouds, a professional development website, and real time support. I interviewed an Amplify CKLA teacher about the program and you can listen to it here: https://podcasts.apple.com/ca/podcast/interview-jennifer-cyr-on-amplify-ckla/id1448225801?i=1000550395984 I was in particular impressed with how powerful their formative assessment tools were.

​

Research Efficacy:

To analyze the efficacy of CKLA, I looked at studies on the topic with control groups and analyzed their results. I searched for studies on both company websites, on Education Source, and on Google. I also reached out to Dr. Sonia Cabell, who is currently conducting research on CKLA (and is a phenomenal researcher) to ask her if she had any studies that she could recommend to me. In total I found 20 studies. However, only 4 of these studies had a control group and enough reporting detail for me to find an effect size.

​

Studies:

Grismer 2023:

Grissmer, et al. published a four-year-long experimental working paper with eight other scholars, on the impact of the Core Knowledge Curriculum on standardized test scores. The study employed an RCT study design, had a sample size of 6652 (after attrition), and used standardized, norm referenced testing. In this study, students entered into lotteries to try and attend highly competitive charter schools, which also used the Core Knowledge curriculum. The researchers then compared the students who entered the lottery and won, vs the students who entered, but did not win and had to attend public school education. The study did have a high attrition rate of 34.1%; however, high attrition rates are often quite normal for longer studies. The students in the treatment group showed a mean effect size for English proficiency, of .47. This would suggest that the study results showed a strong positive result on English proficiency. The study also showed particularly positive benefits for low-income students with an effect size of 1.29 on English proficiency.

That said, it does need to be pointed out that, students in the treatment group attended highly competitive charter schools. Indeed, these schools were so competitive that parents entered lottery pools for their students to attend. Conversely, students in the comparison group stayed in the public school system and received business as usual instruction. This means that there is no real way to know what instruction looked like in the comparison group. Ostensibly, this is not really a study on just the CKLA program, but also a study on the impacts of specific charter schools. In order to assume that the effect size here represents was caused by the Core Knowledge curriculum, we also have to assume that there were no other significant differences between these charter schools and the public schools. This assumption does seem to be quite a stretch. None of this is to say that this research is not important or that it does not build on the scientific understanding of the topic. However, it cannot be used in isolation as definitive proof that the CKLA program works.

Cabell & Hwang, 2020: Sonia Cabell and Dr. HyeJin Hwang published preliminary results on 2 large scale RCTs on the efficacy of CKLA: Knowledge, in 2020. While the full results are not yet publicly available, Dr. Cabell was kind enough to send me her submitted manuscript for the full study details. The two studies were essentially identical as the second study was a replication of the first. Both studies compared CKLA to a business-as-usual control group, across 23 schools. The studies focused on kindergarten students and each lasted 1 semester. Both proximal (researcher designed) and distal (standardized) assessments were used. Overall, a mean effect size of .39 was found, .59 for proximal assessments and .10 for distal assessments.

​

Amplify 2018: Amplify published their own cohort study in 2018, without peer-reviewing it. The study included 2801 grade 5 students and compared students' ELA state test scores for the year before Amplify CKLA was implemented to the year after. The original authors found a small effect size of .26.

​

Synthesis Methodology: I synthesized the results of these four studies to calculate a mean overall Cohen’s d effect size, for the Core Knowledge curriculum. 95% confidence intervals were also used to calculate the probable range of study results. The study effect sizes were weighted by their inverse variance. However, the 2018 Amplify study only listed the full study sample size and not the sample size for the treatment and control groups respectively. To include the 2018 study, we had to make the theoretical assumption that both groups had the same sample size.

​

Results:

All 4 of these studies showed overall statistically significant results, when comparing the treatment group performance to the control group. However, only (Grissmer, 2023) showed moderate or higher results on a standardized test. The overall weighted mean effect size was .30 and the raw-unweighted mean was .33 [.23, .43]. The below graph displays the moderator variables, to model how these outcomes changed across assessed variables.

​

Figure 1: CKLA Moderator Analysis

​

​