At the moment I am taking a great course on reading and the research behind how we read. I just finished the part on reading assessments and it made me want to come here and write a post on it.
This is a very frustrating topic for me because I think this is one of the most misunderstood concepts in teaching, yet pivotal to the process of learning. As a teacher, this is one of the most important concepts I hammer home every year with my students and honestly, given the ingrained perspective most kids have of tests by 4th grade, I think it has been a pretty successful strategy.
In the course, it talks about different types of test you can give.
It talks about how tests need to be:
Reliable - giving consistent results each time it is given.
Valid - tests what it is supposed to test
These are all really important ideas we need to think about. Especially considering how subjective many assessments are. The more subjective a test is the more it allows internal biases and hidden desires to lead to inaccurate results that are highly biased by the individual giving them.
Furthermore, at least in my experience, this leads to hours upon hours of time each year where teachers are discussing these inaccurate results, trying to find common connections and assessment understandings, but in every case never really finding common ground.
For example, if teachers are trying to decide where some work falls on a 5 point scale, some will say 2, some will say 3, some will say 4. The number system is ok, but the subjectivity is because the vague wording for the question or the outcome... "Students will begin to show understanding of ......" .
Tests, especially when given across classrooms need to be more reliable and valid than what we currently use many times. This requires more objective standards that provide specificity and context.
But that this is not even the most important part of the conversation. The real issue is the downright wrong understandings and academically harmful anti-testing culture that is prolific in education.
While I understand that this is a direct result of poorly created and administered tests, and lack of understanding by teachers, parents, and students of how to use and how testing plays a vital part in learning, it still has had a very negative impact on education culture. There needs to be a change in direction.
Test, quiz, assessment. These are all just terms for what is basically retrieval. Simply put retrieval is when we access memories or from our long term memory. Ex. What sound does the letter T make? If the student can say it, they have retrieved it, or they know it. If they can't... well then the question is why not? Not taught, not remembered, not understood, context is too vague. This is where true "teaching" comes in.
So what is the problem: Like everything in life it's complicated. From a teacher's point of view here is where I see some problems.
Time: This process takes time. If kids test and don't learn, teachers need to assess, and then decide where to go. The chart is overly simplified.
Most tests teachers do are One and done: Some assessments like "standardized assessments" may be occasional, and more for large scale assessment of students. But it is my opinion that an assessment should never be one and done. Especially in an environment where the idea is encouragement to learn, assess as many times as you like or feel is necessary. Again, this is where understanding the learning process (retrieval, spacing, schema), development, growth mindset, etc. are all balanced.
Teachers don't know about the testing effect. The testing itself is a learning opportunity (possibly one of the most important ones). The act of recalling the information in different ways strengthens and expands the connections to that information in the brain. What if they get it wrong: That also can be a very important learning moment, the "Oh Yeah!" moment, that helps the information stick.
Kids don't develop Growth Mindset : (This is a deep rabbit hole all to itself so I will keep it simple) Students can and need to fail. Life and learning full of failures and some of the most successful people are the ones that have learned how to deal with and learn from failure. Our current culture in education is almost the polar opposite of this in many cases.
We celebrate the outcome rather than the process, the work.
We (many times unintentionally) promote that hard work is bad or unnecessary to achieve. This comes through in things like spelling, handwriting, homework, reading. We make all kinds of excuses for the kids not learning when the reality is they just didn’t put the “time on task” needed to be successful. Yet we still move them on.
We misunderstand “growth mindset”. Growth mindset takes time and requires many more steps than we actually do in school. Many teachers think it is, “Instead of “I don’t know how to do this.” say, “How do I figure this out”. Now, this is the first part to growth mindset, but the reality is you have to then guide them through the process of figuring it out or they haven’t done “growth mindset”.
Teachers wrongly emphasize activities over-testing. You constantly see this when it comes to discussion groups, or hands-on or discovery learning. In these activities, there are so many things going on, it is hard to tell just what kid has learned what. Too many confounding factors.
It's ingrained into the culture from the beginning "tests are bad" promoted by schools, teachers, and kids. Kids are primed from the beginning to avoid hard work, and taking tests is part of that. A test can be hard. A test's purpose is to say, you have it or you don't. It requires kids to fail sometimes. To pass a test you have to practice (a.k.a. work) at whatever you are doing. Even one of my favorite books "Powerful teaching" that I promote here on this blog seems to misrepresent testing. In the book, they are saying the word "retrieval" but they say that retrieval practice is different than testing because it is "no stakes"... but honestly, it's not, it is just different sides of the retrieval spectrum from higher stakes to lower stakes. referring. In one sentence they say "retrieval practice" should have "no stakes", but again I find this counterproductive to learning. There are always stakes, we just want to make those stakes as low as possible for kids.
Retrieval Practice: I said it above. Tests are just a form of retrieval and retrieval practice. Some are higher stakes, and some are lower stakes. If we give good tests: reliable and valid (see above) and near transfer (see below)
Far vs Near transfer of information: Transfer is how far we have to carry the information. An example of near is 4 + 4 = 8. Now tell me what 4+4 equals. An example of far transfer is ____ + 4 = 8 or "I have 4 muffins I make 4 more muffins how many muffins do if have. Many of our tests are asking kids to transfer the information too far, to questions or situations they don't understand. The concept is the same but it is not how they understand it. It is my opinion as a teacher that tests, in general, should not be a place of trickery or asking for new ways to use the information, but very direct. Here is the idea we want you to know, show you know A, B, C, etc...
Testing is absolutely and widely misunderstood and misused in education today. Testing is really the most powerful tool a teacher has, and it is the most powerful tool to teach students to harness for themselves as independent learners. It is just Retrieval of information. Retrieval and Retrieval practice are going to be entries into this blog all on their own but really that is what makes tests so important because that is all they are. As teachers, we need to make good tests that are clear and specific. We need to test in many different ways, and just explain the different forms of tests we are giving. In the end, in a good education system, no test is ever just one and done. The point is always to say, "Ok what does this tell me about what I know and what I need to keep working on."
Below are some good articles I have read to help me understand this idea.
Butler, A.C. (2010). "Repeated testing produces superior transfer of learning relative to repeated studying". Journal of Experimental Psychology: Learning, Memory, and Cognition. 36 (5): 1118–1133.
Roediger, H. I & Pyc, M. A. (2012). Inexpensive techniques to improve education:
We identify three general principles that are inexpensive to implement and have been shown in both laboratory and field experiments to improve learning: (1) distribution (spacing and interleaving) of practice in learning facts and skills; (2) retrieval practice (via self testing) for durable learning; and (3) explanatory questioning (elaborative interrogation and self-explanation) as a study strategy. We describe each technique, provide supporting evidence, and discuss classroom applications. Each principle can be applied to most subject matters from kindergarten to higher education. Applying findings from cognitive psychology to classroom instruction is no panacea for educational problems, but it represents one helpful and inexpensive strategy.
Retrieval practice produces more learning than elaborative study with concept mapping.
Karpicke, J. D., & Blunt, J. R. (2011).
Taking a memory test not only assesses whatone knows, but also enhances later retention, a phenome-non known as the testing effect. We studied this effect witheducationally relevant materials and investigated whethertesting facilitates learning only because tests offer an op-portunity to restudy material. In two experiments, studentsstudied prose passages and took one or three immediatefree-recall tests, without feedback, or restudied the mate-rial the same number of times as the students who receivedtests. Students then took a final retention test 5 min, 2 days,or 1 week later. When the final test was given after 5 min,repeated studying improved recall relative to repeatedtesting. However, on the delayed tests, prior testing pro-duced substantially greater retention than studying, eventhough repeated studying increased students’ confidence intheir ability to remember the material. Testing is a powerful means of improving learning, not just assessing it.
Promising principles: Translating the science of learning to educational practice
David B. Daniel∗
Department of Psychology, James Madison University, 91 E. Grace St, MSC 7704, Harrisonburg, VA 22807, United States
It is also important to insure that the laboratory practices align with the goals of the educators. Recommendations for practice need to take into account artifacts of experimental design that may not align with the demands and/or goals of other contexts. If the goals or methods are not aligned with those deployed by educators, a promising strategy could be resisted by the educator for very good reasons, few of which are on the researcher’s radar. Most studies of the testing effect, for example, tend to use quiz questions identical to the questions on the exam. One would be hard-pressed to find an educator who would give such a practice high value. Several studies have demonstrated the potential for the testing-effect to increase learning for related questions and far transfer (see Carpenter, 2012 for a review), but these paradigms have not been fully developed for broad educational application. While studies in this area are beginning to move into representative classrooms, many have not done so in an authentic fashion.