To test or not to test, what’s the benefit?

Had I been asked the question above, I believe my answer would be sort of evasive. I’m not particularly in favour of using tests as the only way to assess learning, but I also have a hard time believing tests are all bad. What I believe in, though, is that there are well-written tests and poorly made tests – and I also believe that fewer and fewer people have actually learned how to write a test. Much on the contrary, it seems we, educators, are all onboard with the view that tests do very little to help learners go beyond the remembering stage of learning. What I get from most talks about tests is that we equate tests with standardised testing, and that as fewer teachers are brave enough to say that tests may be useful, we are witnessing a time in which tests are getting simpler and simpler to score, and, consequently, demanding more memorisation skills than learning from students.

Reality check, teacher. Abandoning tests alone won't change that kind of attitude. // Photo by Richard Phillip Rücker

However, as William Shakespeare said, “There is a tide in the affairs of men…”, and we have to take advantage of things as they happen, so I decided to take advantage of a recommended article by one of the members of my PLN I grew to admire due to her insights and always amazingly packed with important information yet straight to the point comments and blog posts, @ssuzip. And even though Susi hasn’t made clear what her opinion is regarding this matter, I’m going to put myself out in the open and grant any kind of criticism that may come so I can maybe change my mind on the topic, or at least have better arguments the next time I write or talk about it.

The article was published on The New York Times on January 20th, and it’s title is To Really Learn, Quit Studying and Take a Test. As you’ve got it by now, it is in favour of testing as a learning tool. What comes to mind is that most of our parents and grandparents were educated in a system that praised tests. Furthermore, and I don’t think this is only nostalgia, standards were higher in terms of actually learning what you were supposed to learn. There weren’t things such as ADHD, as Sir Ken Robinson pointed out well here, and the whole concept of an affective side to education wasn’t really that big a deal. Nowadays, with the decline of another very important institution – family – schools have to bend over backwards to try to cater for both the cognitive and affective growth. On the one hand, this has helped us see there are numerous benefits in taking the learner as a whole human being. On the other hand, well, it may at times drive us away from our main goal of teaching and making sure our students achieve what they can fully achieve. Now, if all teachers were just as the ones who are likely to actually read this blog, I’d be completely off my rocker to say such a thing. Nevertheless, no matter how hard our PLN is, we’re still far from being the majority.

Unfortunately, there are many teachers who claim to be against tests without having ever read anything from, say, Alfie Kohn to back them up. I myself can’t say I’m familiar with his work, but most people whose blogs I visit and who are firm in their opinion of abandoning tests tend to consistently citing Kohn. The problem is not saying tests are good or bad – the problem is not being able to reason on your own whether they are good or bad. Coming back to the article and my beliefs towards testing, I must say I saw myself arguing with the text many times while reading it. For instance,

“I think that learning is all about retrieving, all about reconstructing our knowledge,” said the lead author, Jeffrey Karpicke, an assistant professor of psychology at Purdue University. “I think that we’re tapping into something fundamental about how the mind works when we talk about retrieval.”

The very first thing to think about there regards the concept of learning. Had Mr. Karpicke stopped at retrieving, I’d probably stop reading the article right there. However, I agree with his view of learning as the ability to using what we’ve already seen and being able to adapt and make effective use of such memory. Shall we compare it to Bloom’s taxonomy of learning domains and focus on the cognitive domain? We need to remember what we have learned, but this has got to come into being with some sort of purpose, which will lead us to the following step – making use of what we’ve just remembered. In this regard, the very first thing (and, unfortunately, the only thing in some cases) tests do is make us retrieve information. Moving on with the article, we find out that students who are going to be tested have the feeling they don’t know as much as they should. In comparison, those students who are allowed to create diagrams and concept maps for the sake of learning, but who know they are not going to be tested, have the feeling they know pretty much everything they need to know. It turned out in research that the result was exactly the opposite when students were asked to remember the subject one week later.

Why retrieval testing helps is still unknown. Perhaps it is because by remembering information we are organizing it and creating cues and connections that our brains later recognize.

However, in order for such things to take place, we’ve got to be able to write good tests. It can’t all be about rote learning and memorising passages. Tests can be used in education; let’s not throw out the baby with the bathwater. I guess the very first thing is keeping an open mind. As Howard Gardner, and education professor at Harvard said,

the results “throw down the gauntlet to those progressive educators, myself included. Educators who embrace seemingly more active approaches, like concept mapping,” he continued, “are challenged to devise outcome measures that can demonstrate the superiority of such constructivist approaches.”

The final word on testing in education, in my humble opinion, is yet to be said. I am, however, not in favour of abandoning tests altogether. I don’t think tests should be the only source of assessment in the classroom, either. My final quote from the article comes from Dr. Marcia Linn, “Some tests are just not learning opportunities. We need a different kind of testing than we currently have.” I guess she’s just touched the crux of the matter. As a teacher, I have to say that I had to learn about writing tests on my own, and I did take the time to read about it and look for data on the matter. This is not what most people who started teaching with me have done, though. It seems to be a big gamble and, in this hit-or-miss game of ours, more misses are made than hits, which only proves to strengthen the view that tests aren’t beneficial for learning. Add to that the fact that tests have become a million dollar business, and you’ll see that most people who advocate for tests aren’t exactly worried about learning. This, obviously, make them an easy prey for educators who, once again, see others not only making money out of something they consider serious – children’s learning – but also see those people doing things that make us, educators, look bad. We want to help children learn, testing companies want to make children take as many tests as possible so they can profit from them.

As my final words, I add that I agree with Penny Ur when she says that tests are one of the many sources of extrinsic motivation that teachers have control over in class. Just as anything, it can’t be overused. If tests are well-written, they can certainly help in the learning process. To what extent? This seems to be uncertain at the moment, but if we don’t keep an open mind, we might end up doing more harm than good by abandoning tests just because we want to follow the trend.

16 thoughts on “To test or not to test, what’s the benefit?

  1. As a Cambridge ESOL Examiner, I am in favour for tests as long as they are meaningful and useful for language learners. I have recently written a journal review which looked at fluency of L2 speakers and the speakers had a few minutes to prepare a monologic story. To me, fluency is assessed by natural and meaningful language and involves a speaker and a listener both taking turns to complement to the conversation. Obviously, monologic testing is an easy way to assess fluency but it is not natural. So, I agree with your post that well written (as well as well prepared) tests are suitable as long as students find them useful.

    1. Hi Martin,

      Thanks for your contribution. I agree with you when you say that monologic testing is unnatural – no way you can assess whether someone is able to hold a conversation based on a soliloquy. 🙂
      The biggest issue is coming up with well prepared tests that do allow for washback.


  2. I think your question sets up a false dichotomy. Tests are part of the feedback that learners need in order to continue to develop. As you write, testing should be a learning opportunity. Standardized testing seems to contribute little to learning. I remember getting the “results” of my children’s standardized tests and seeing numbers that put the child into a group that was achieving or was not achieving a certain result. Those tests helped neither me nor my child. I agree that writing good and fair tests is difficult especially if the goal is a learning opportunity. Difficult is okay with me.

    1. Hi John,

      Let me see if I got the false dichotomy bit while I elaborate on my reply to your comment. Tests are indeed one of the sources of feedback, but they will not fulfill their needs if all they do is grant students a grade or place them in a curve. Tests will be effective when their results are taken into account for a lot more than a grade – finding out areas that need further work, or areas in which students excel is one of the good things tests should be able to do. However, tests cease to be a learning opportunity when the final grade is more important than what was tested.
      I like the bit about how hard it is to write good and fair tests. This kind of difficult should be OK with all teachers as this is what we do. If we think about it, no matter how hard it might be to write good tests, it still is a lot easier than facing the outcome of poorly written tests. I’m very much in favour of being fair, and part of being fair is showing students that they all have weaknesses as well as strengths, right?
      Thanks for the comments! Looking forward to more contributions. 🙂



  3. Hi Henrick,

    Thanks for sharing your views on tests. I couldn’t agree more. I also think tests and exams should be well made and should have a rationale behind.
    Hugs from Argentina!

    1. Hi Marisa,

      Thanks for your words! I only hope more teachers not only think likewise, but they are also allowed to write their tests based on sound pedagogy and not just because the educational system they’re inserted in asks for a grade.

      Hugs from Brazil!! 🙂

  4. Hi Henrick,

    Loved this post as I do think we sometimes pay lip-service (oh dear, think some people might want to disagree violently with me here…) to a philosophy which goes like this: I don´t teach to test; my tests do measure learner progress and it reflects the learning which takes place in class.

    But here´s the question: how many of us are actually prepared to create tests which are valid and reliable? Good test construction is really difficult – it´s not just a question of having a data-base with items for test questions which you can pick and choose from (so many publishing houses are now selling LMS which come with a freebie test-builder…).

    So I’m glad you wrote about this tricky topic to make us all reflect a bit more deeply about it.

    1. Hi Valéria,

      I share your view. Many times I’ve seen teachers going with the flow – I myself have already been like that (I like to think I can say “have been” here) – and blaming students low performance on tests on the common belief that tests don’t actually do what they claim to do: test. However, these very same people have never read about assessment in their lives and seem to be looking for an easy way out of writing tests.

      How many of us are actually prepared to create tests? Gee… based on what I see, I’d say very little people. And the fact that there’s this new trend in education in which only formative assessment matters is, IMHO, doing more harm than good in most cases.

      I’ll ask another question, then: Don’t you think that publishers have been selling test-builder CDs and etc because they noticed that most teachers are: a) not prepared to write tests; b) not willing to add one more thing to their already extensive to-do list; and c) a potential market, obviously.

      If teachers simply decided not to use those test-builders and stopped buying them, my feeling is that publishers would either raise the bar, or stop selling such material. And we’re back to the business of teacher training, which is a lot more complicated, right? Teachers will hardly see the need for professional development as long as schools are OK with what they have.

      Many thanks for your comments! They made me think a lot more about the matter. 🙂

  5. Hi Henrick,

    Good points. I was thinking that if teachers could devise tests in such a way that couldn’t just be passed by cramming the night before but rewarded students who had developed understanding over the duration of a course, standards would be raised. Regurgitating for a test doesn’t usually uncover if any deep, long lasting learning has taken place.


  6. The main point of institutionalised testing is to position the student, in comparison with other students. Ignoring large institutions and examination boards and thinking on a micro level, I believe that the main purpose of evaluation in language teaching is to help the student compare his or herself to a previous self, so they can see their progress and consider whether they could achieve more by making any changes to their learning habits.

    In global terms there is a lot more to it, obviously. Some students enjoy being better than classmates, parents often want comparison with the class average, employers often demand results based on testing, over here evaluation often needs to match the council of Europe framework. However, I consider evaluation that helps the student observe their own progress is the most important objective.

    There are a few of my thoughts at the moment!

  7. Good post Henrick. I eliminated tests at my last school because they were simply being used as a filter. You got a 70% so you can pass and you didn’t, so you can’t. What did 70% even mean to the teacher or learner? On top of that, tests were often of mediocre quality.

    Tests are ok if they are used for assessment. Assessment helps the learners reflect on what they don’t know and it shows the teacher both what the students need and what they aren’t teaching effectively. Tests are basically a diagnostic tool and can be used to great effect in that way.

    Really though, how many schools use tests as diagnostics rather than filters? How often do schools make any kind of change to the class based on test results? Not many in my experience.

    The backwash effect can also be positive as it encourages study and retrieval, but it can also create that dreaded “teach to the test” effect that is so harmful to learning.

  8. Very interesting post, Rick. I flippantly state that tests aren’t good, but I need to have a better arsenal to back it up. A number of times in the comments, people say making a test is difficult. Why? What are the considerations? If you’ve got time, please take a look at my website. If you go to “Have a go”, go to Branches and then go to Puzzle, would you say that Puzzle a test? If so, is it a good one? Why (not)?

    1. I had an email from a distressed parent who follows my blog yesterday. Her son has been diagnosed as having a very poor working memory although he is bright when it comes to understanding concepts. Every week he’s given a spelling test. Mother and son work every night on these five words he has to learn for a test on Friday. He gets very upset and stressed about them. If they work hard he gets 5/5 on the test. Brilliant! Result! BUT she says he’s forgotten them two days later. What, pray, is the point? Wouldn’t he be better off just reading for pleasure or playing with words rather than stressing about them? Wouldn’t he be better understanding why, rather than what (which is what so many tests test)? Tes

    2. I had an email from a distressed parent who follows my blog yesterday. Her son has been diagnosed as having a very poor working memory although he is bright when it comes to understanding concepts. Every week he’s given a spelling test. Mother and son work every night on these five words he has to learn for a test on Friday. He gets very upset and stressed about them. If they work hard he gets 5/5 on the test. Brilliant! Result! BUT she says he’s forgotten them two days later. What, pray, is the point? Wouldn’t he be better off just reading for pleasure or playing with words rather than stressing about them? Wouldn’t he be better understanding why, rather than what (which is what so many tests test)? Tests may motivate some to study for teh test itself, but do they really help deep meaningful learning?

Leave a Reply

Fill in your details below or click an icon to log in: Logo

You are commenting using your account. Log Out /  Change )

Google photo

You are commenting using your Google account. Log Out /  Change )

Twitter picture

You are commenting using your Twitter account. Log Out /  Change )

Facebook photo

You are commenting using your Facebook account. Log Out /  Change )

Connecting to %s