Research to the Classroom: Daily Review - Part One - The Research

Subscribe to the Podcast

00:00:00
We Are Here for The Long Haul.
Hi there, and welcome to the Structured Literacy Podcast. It's Jocelyn here, coming to you from Pataway Burnie in beautiful Tasmania. I'd like to begin today's episode by saying thank you to you; thank you for being part of our world. Thank you for letting us be a part of yours. Thank you for the work you do with your students on the road to every child being able to read and write. The other day, my colleague Cass was asked how long our work would continue. Cass's response was, "Well, Jocelyn is going to keep going until every child can read and write." and that about sums it up. We are here for the long haul and won't rest until you have all the support you need to make great instruction happen. I know that it's the work on the ground that is the most important factor in shifting practice. It's the modelling, the planning, the coaching and the support to reflect on data that really gets things going in student outcomes.

00:00:56
What is the Research to Classroom Mini Series?
This podcast is intended to be helpful for you as you do the important work in your school, be it teaching in the classroom, talking about instruction with the teacher next door, or leading the team. My hope is that the work we do makes the work you do that bit easier. To that end, I have launched a new type of mini series inside the podcast called Research to the Classroom. These mini series are designed to help you move away from practice as a series of fads or tasks to really understanding why you are doing what you are doing. 

As a teacher, you have likely not had the advantage of examining research in your teaching degree and now that you're in the classroom or leadership position, you probably find it hard to find the time and head space to spend time reading papers and articles. I know that when I was in school full-time as a teacher and a leader, I really didn't have a chance to do that. So, I hope that these mini series of episodes help you bridge the gap between research and practice. They're made up of three episodes. Number one is an outline of the research. Number two is a discussion of the practical applications of the findings of the research, and number three is a chat with a real-life teacher who shares the positive aspects and challenges of being right there with kids in front of them in the classroom.

This current Research to the Classroom mini-series focuses on the practice of daily review and retrieval, why we might do it, what it might look like, and what it's like to implement it in a school. Let's begin by examining the reason that we do daily review or retrieval and it's not because someone told you to. There are some errors being made in how this is delivered out there in school land, and I think it's undermining the effectiveness of the practice. I'm not suggesting that teachers aren't clever, just that when we implement something new, it's tempting to race ahead, run a brief PL session in a staff meeting, drop some materials in teacher's laps and say, "There you go, job done, box ticked." That's not the best way to get lasting results. So, I'd like to go back to the research and theory that should be driving this practice. 

00:03:13
Today's Topic - Why Daily Review is Such a Powerful Practice?
The foundation of understanding why daily review is a powerful practice lies in the nature of the way that human memory works. It is inevitable that we forget things. Think about how long it takes for children to learn times tables or to bed down automatic phoneme-grapheme correspondences (or letter sound correspondences). Wouldn't it be wonderful if we could just have kids work with something once, and they remember it forever? Oh, if only. Human memory is fragile and things are so easily forgotten. We all know this, but what we do about it has not reflected research for far too long. Think back to your own study habits at uni or in high school. How often did you re-read notes, copy down text word for word, or cram? How well did this work to help you learn material in the long term? I'm going to bet that it didn't. Even knowing this, we still have so many practices based on these principles that just don't hold water. 

00:04:15
What Does it Mean to Have Learned Something?
In their book from 2014, Making It Stick, Brown, Henry, Roediger and McDaniel begin with a discussion of what learning is. When these authors, two of whom are cognitive scientists, write about learning, they say learning is acquiring knowledge and skills and having them readily available from memory so that you can make sense of future problems and opportunities. Kirschner, Sweller, and Clark (2006) summed this up as "learning is a change in long-term memory."

I've said many times on this podcast that there is a lot of doing in classrooms, but how much learning takes place? Well, that's sometimes questionable. You might dispute this, but what if I said that we can't talk about learning in our classrooms unless all of our students are learning? What does it mean to have learned something? Does it mean that we covered something? Does it mean that we gave an explanation? That we walked students through something (even if we did that step by step) or does learning mean that the children can talk about what they have learned and apply it in context? If we make that distinction, how does that change your feelings? Now, it's true that different children learn with different levels of ease and speed, (and that comes from Dehaene in his book from 2020) and I don't think that we can pat ourselves on the back until all students have experienced the permanent change in long term memory that Kirshner, Sweller and Clark are talking about. To explore what research has to say about bringing this change about, we'll explore three research papers. Two are single studies, and the other is a meta-analysis. 

00:05:55
It's better to present concepts in multiple contexts
Let's begin with the single study by Hayley Vlach and Catherine Sandifer. Vlach and Sandifer published their findings in 2012 in the journal, Child Development, which is a peer reviewed journal from Wiley. In this paper, titled 'Distributing Learning Over Time, the Spacing Effect in Children's Acquisition and Generalisation of Science Concepts.' the authors sought to examine whether space practice resulted not just in better learning but in a greater capacity to generalise concepts across tasks. That is, could children apply their new learning in unfamiliar situations? To answer this question, they conducted a study with a group of 36 primary school children. That's about the average size for educational research. The children were aged between five and a half and seven and a half years old, (so year one and two), and were randomly assigned to one of three groups. Group one contained 12 children who were given massed practice. That is, they were taught four lessons, one after the other, on the same day. Group two contained 12 children who were assigned to the clumped condition. That is, two lessons on day one and two lessons on day two. The third group also contained 12 children, but these children were given spaced learning. That is, one lesson per day for four days. All four lessons were the same across the groups. The only difference was the timing of the lessons. The researchers used a pre and a post-test to determine learning. Each of the four lessons delivered was approximately five minutes long and was about a particular biome in the environment: arctic, desert, grassland, ocean, or swamp. The biome in the pre-test was not included in the instruction and was the same biome that was used in the post-test. All post-tests were given one week after the children's final lesson, regardless of which group they were in. So, all children had the same amount of time between the end of the learning phase and the testing. Instruction in the biomes focused on food chains. Children were taught about what each animal in the food chain ate and then were asked more complex questions about the impact of things like interruptions to the food chain, such as using poisons.

The result of this study showed that children's learning was significantly stronger in the spaced learning group than the other two groups. Not only did they recall more information, but they performed better on the complex generalisation task in the post-test than the other children. The authors suggested that an examination of generalisation is an area for future research and thought about a couple of ideas about why perhaps the results they found showed up. But I'm actually not sure that this finding needs to be mysterious, and some further reading reinforced this for me. 

Bradsford and Schwartz's paper from 2001 discussed that presenting concepts in multiple contexts can increase transfer to unfamiliar tasks, and that's what these researchers did. They taught a concept over several days with a similar lesson structure, varying the context of the learning slightly (that is a different biome) but the core content about food chains and how they operated was the same. Further, the idea that developing semantic knowledge or facts and information enables us to generalise knowledge from one situation to another is really prominent in the literature. So, I'm not particularly surprised by the findings of this research. What this paper adds to the picture is that the combination of spaced learning and presenting learning in multiple concepts in a similar way across multiple days leads to stronger learning. Now, that was not the intent of these authors, but it's just something that occurs to me as I think about the sorts of learning we see in classrooms. I find this really quite affirming because that's how we approach instructional design in our text-based units inside the Resource Room and in our Spelling Success in Action program for Year 3 and beyond.

00:10:05
Should we be Rethinking the Use of Tests?
The other research paper that I'd like to share is a meta-analysis titled Rethinking the Use of Tests, a Meta Analysis of Practise Testing. This paper was published in a peer-reviewed journal, The Review of Educational Research, in 2017. In this meta analysis, the authors examined 118 articles with a total of 15427, participants included in the various trials. The authors began the paper with a discussion of the perception of tests in modern education. There's a feeling out there, wherever out there is, that we're testing children too much. This can lead to a reluctance on the teacher's part to use quizzes as part of standard practice, but the testing that leads to student stress that is really unhelpful is the high-stakes summative testing that we can often see. The authors suggest an alternative, and I'm reading directly from the paper here...

...regrettably, the heavy emphasis on tests as summative assessment tools used to make high stakes decisions often obscures some very important functions of tests. These include the opportunity to use test results for low stake formative assessment and provide students with feedback on their strengths and weaknesses, as well as helping teachers improve their instruction.

and I agree with the authors here. Testing, otherwise known as retrieval, can be an incredibly useful tool if we know why we're using it and how to use it. The findings of this meta analysis were consistent with other findings that using retrieval or testing to learn was more effective than simply reading. Interestingly, this meta-analysis revealed that multiple-choice practice tests were more effective than short-answer practice tests but that a combination of the two was even more effective than just the multiple choice. So, potentially, an implication could be for us to use a combination of question types in our reviews. When it came to the age that retrieval or testing was effective for in terms of our students, the study showed that while 83 per cent of the studies they analysed included participants who were post-secondary students, the effectiveness of retrieval or testing was robust across a wide range of educational settings and ages. It was effective for the classroom and effective in the laboratory; it was also effective for students of different ages. The findings of this meta-analysis are frankly unsurprising, considering how robust the research is in this area.

Test or retrieval as a tool to enhance learning is not new, we may think of it as being the latest fad, but research into this technique has been occurring now for over 100 years.

00:13:05
So, what are the practical implications?
It's difficult to understand why a practice that is so well supported and so well researched is not used more widely. In this episode, I've shared a variety of research papers that are all saying the same thing. There is an abundance of research supporting retrieval and testing to improve learning and that spacing learning out over time gets better results than engaging in mass practice or learning where things are done all at once. Spaced learning results in stronger learning that is able to be transferred to new situations much better than learning that occurs in a one off lesson. It's not about the total time spent on the learning. It's the timing. So, what are the practical implications from the papers explored in this episode?

Implication One: It's better to do small amounts of learning across several days than do more learning on one day. 

Implication Two: When it comes to practice, retrieval is much better than simply reading. So if you're teaching phonics, you're likely to get better outcomes by saying a phoneme and asking students to write a grapheme than you are to just show them the grapheme and ask them to say the phoneme. The same goes for older students all the way across the curriculum. Rereading doesn't give bang for your buck; questioning and retrieval does. 

Implication Three: Potentially, we could look at varying the question types used in retrieval practice. So some multiple choice and some short answer. 

Implication Four: Keep retrieval practice and testing low stakes and let students know why you're doing what you're doing. Pitch it at a point where you expect them to be successful and let them know what's going to happen from there; that it will help to inform you as the teacher about how best to support them. You are not using it to make a judgment about their intelligence or their ability. 

00:15:01
What's Next?
In the next episode of the podcast, I'm going to unpack more practical suggestions about retrieval, review, and space learning so that you can take decisive, simple action in your classrooms and in the third episode of the mini series, we'll talk with a real teacher and instructional leader about her experiences of using retrieval and daily review in her school and her classroom. 

So until then, take care. Happy teaching, everyone. Bye. 

References:

Bransford, J. D., & Schwartz, D. L. (1999). Rethinking Transfer: A Simple Proposal with Multiple Implications. Review of Research in Education, 24, 61–100. https://doi.org/10.2307/1167267

Brown, P. C., Roediger, H. L. III, & McDaniel, M. A. (2014). Make it stick: The science of successful learning. The Belknap Press of Harvard University Press.

Adesope, O. O., Trevisan, D. A., & Sundararajan, N. (2017). Rethinking the Use of Tests: A Meta-Analysis of Practice Testing. Review of Educational Research, 87(3), 659–701.
https://www.researchgate.net/publication/315706448_Rethinking_the_Use_of_Tests_A_Meta-Analysis_of_Practice_Testing

Dehaene, S. (2020). How we learn: why brains learn better than any machine ... for now (First American edition.). Viking.

Paul A. Kirschner , John Sweller & Richard E. Clark (2006) Why Minimal Guidance During Instruction Does Not Work: An Analysis of the Failure of Constructivist, Discovery, Problem-Based, Experiential, and Inquiry-Based Teaching, Educational Psychologist, 41:2, 75-86, DOI: 10.1207/s15326985ep4102_1

https://www.researchgate.net/publication/27699659_Why_Minimal_Guidance_During_Instruction_Does_Not_Work_An_Analysis_of_the_Failure_of_Constructivist_Discovery_Problem-Based_Experiential_and_Inquiry-Based_Teaching 

Vlach, H. A., & Sandhofer, C. M. (2012). Distributing Learning Over Time: The Spacing Effect in Children’s Acquisition and Generalization of Science Concepts. Child Development, 83(4), 1137–1144. http://www.jstor.org/stable/23255684