IS THERE A PLACE FOR ROTE MEMORIZATION, WITHIN STRUCTURED LITERACY INSTRUCTION?
Rote memorization was once seen as the most important pedagogical tool for student instruction; however, over the last several decades, the concept has not only fallen out of favor but been largely demonized. Modern education scholars have looked at rote memorization as an authoritarian relic of a bygone age, as it has been largely rejected within the post-modern age for being too oppressive. Part of this rejection stemmed from the introduction of a host of new child friendlier pedagogies that were aimed to replace it and part of this rejection likely stemmed from modern scholars personally having unpleasant memories of rote learning from their own childhood. Undoubtedly, rote memorization can be one of the most boring ways to teach a child. However, I would argue that it does not have to be. Rote memorization can be easily integrated with game-based learning. Moreover, I find students sometimes enjoy short bursts of rote memorization, when it is used to help them to learn challenging concepts.
Within the realm of literacy instruction, rote memorization has been especially demonized, as both of the two largest opposing literacy camps have largely rejected the concept. Many Structured Literacy advocates claim students do not need to memorize words if they possess the phonetic ability to decode. Whereas, Balanced Literacy scholars are a part of a post-modern education movement, which seeks to minimize direct instruction methods and replace them with inquiry-based learning and implicit instruction methods. This all being said, I would ask what does the research say?
John Hattie did a secondary meta-analysis of Rote Memorization in 2020. His analysis looked at 132 studies of the topic and 3 meta-analyses. His research found an effect size of .73. However, this effect size was for rote-memorization in general and not specifically for reading instruction. I attempted to find a meta-analysis specifically looking at rote memorization in language instruction. Despite looking through multiple academic databases, I was unable to find a meta-analysis of the topic. However, I was able to find several meta-analyses of individual rote memorization strategies, as they pertain to reading.
Realistically, the most commonly used form of rote learning in reading instruction is Repeated Reading. Repeated Reading has students read the same text over and over again, with the goal of increasing their reading accuracy and reducing their time to read the text. This intervention is one of the most studied fluency interventions. A meta-analysis of the topic by Therrion et al, in 2004 found that Repeated Reading was a high yield strategy for students with Reading Difficulties. Their study found that students reading the same text just three times, had an effect size of .95 using a Cohen’s d effect size calculation. Comparatively, phonics only has an effect size of .70, according to Dr. John Hattie’s 2018 meta-analysis and .86 if we look at the 2006 NRP meta-analysis led by Dr. Timothy Shanahan. Moreover, The NRP 2006 study finds an effect size of .86 for Repeated Reading as well. Whereas Hattie’s 2018 Meta-analysis found Repeated Reading to have an effect size of .75. Ultimately though we do not need to compare Repeated Reading to phonics, though as they are complementary ideas, not competing ideas.
To get a real understanding of the value of rote learning and repetition within fluency interventions, we need to compare Repeated Reading to non-repetitive fluency interventions. Fortunately, there have been several meta-analyses of this topic and they consistently find the same general result. Repeated Reading is more effective than non-repetitive fluency interventions. Zimmermann et al, did a meta-analysis of non-repetitive reading fluency interventions in 2019, for students with reading difficulties, and their study found a cohen’s d effect size of .105. The Repeated Reading effect size in the 2004 Therrion et al, meta-analysis is more than 9 times higher.
In 2006 Webb and et al, did a meta-analysis of several other rote memorization teaching strategies for language instruction. However, I did not include this data with the above data, because they did not calculate an effect size, nor did they publish enough of their own raw data for me to try and calculate those effect sizes independently. However, their results were significant and still worth sharing. Looking at 2 studies on flashcards, they found a mean difference of .77. Looking at 5 studies, they found a .73 mean difference for word lists. Looking at 11 studies they found a mean difference of .54 for rote memorization through writing. And lastly, looking at 8 studies they found a mean difference, of .43 for fill in the blanks. It is important to note, however, that these mean differences would likely be slightly lower if properly calculated into an effect size. However, rote memorization through flashcards and word lists would likely still be high to moderately large effect sizes. That being said, it does not appear that rote memorization by writing or fill in the blanks is very effective, according to this meta-analysis.
Ultimately, I think if we objectively examine the data of the topic of rote memorization within language instruction, it is almost impossible to claim that the strategy is ineffective as a learning strategy. Indeed, phonetic instruction, which is considered one of the most important and effective language instruction strategies, actually has a marginally, albeit insignificantly, lower effect size than rote memorization strategies such as repeated reading. However, as I previously stated these concepts are complementary concepts, not competitive ones. While Structured Literacy advocates might not advocate for the memorization of vocabulary, especially for emerging readers, they might very likely advocate for the memorization of phonetic sounds, suggesting they do believe there is someplace for rote-memorization.
Personally, I believe this comes down to timing. We want to focus on different language instructional strategies at different phases for our students. For early readers, we need students to learn their basic phonetic sounds and as students become more advanced at decoding, we want to focus on more advanced components of phonics such as blends and word families. However, students will eventually reach a point in their language learning process, in which we want them to start transitioning out of the decoding stage of reading into the automatic stage of reading. As Dr. Shanahan, previously pointed out on my podcast, the 2006 NRP study showed phonics was a high yield strategy for students below grade 3; however, these results were not replicated in later grades. Moreover, the more advanced students became, the less important phonics instruction became. This is why I have previously suggested the below model of literacy instruction. Personally, I do not believe there is any reason why rote memorization should not have a place within the Structured Literacy movement.
