Testing, the Chinese Way
nytimes.com
nytimes.com
Personal experience: I am from Turkey and had to go through two testing periods that determined my life: One, at 11, that determined which secondary school I could go to. Every private school had its own exam, so as a kid you had to take 6-8 exams in about two weeks (the system has changed since then, now there's a single exam).
The worse one was the university entrance exam (actually two exams, a couple of months apart): close to a million people took it with me. The top department in the top university admitted only about 50 students each year! These 2-3 hours really determined your future, since if you couldn't get admitted to the 1-2 popular universities your job prospects were pretty much doomed. I still remember the hordes of anxious parents waiting outside the building where the exam was held. The level of math and physics you had to know/memorize was mind boggling; compared to that, the math part of SAT is a joke.
I used to both envy the US students who didn't have to go through such a grueling, trial by fire selection process to go to good universities and also think that our selection process was much superior, since it it made you learn much more. After finishing my PhD in the US, teaching classes here, and working professionally, I now know that this is not true. This forced feeding approach to eduction mostly produces socially awkward, nerdy people who are experts but are not creative. I know, this is such a cliche, but believe me, I can back it up from personal observation in the workspace.
I think, neither the relaxed, "it's the job of the professor to make me learn" US college attitude nor the extremely competitive test based approach elsewhere is ideal. However, I think the correct approach is much closer to the former than the latter.
EDIT: Just remembered a personal experience that illustrates the point: I recently interviewed at a software company. One of my interviewers of Asian origin asked me an algorithmic question. I came up with an approach (admittedly not very good) and tried to give my reasons; he wasn't satisfied. At the end, he got up and declared "The answer is ..." He wasn't interested in a method or approach, but the answer. I think the worst thing that a test-based education can enforce is this "answer worship", since that is such a warped way to view problems.
I really liked the Prof's Cizek's quote in the article: "What’s best for kids is frequent testing, where even if they do badly, they can get help and improve and have the satisfaction of doing better,” he said. “Kids don’t get self-esteem by people just telling them they are wonderful.”
Frequent summative (non-formative, i.e. stuff that determines which uni you go to, or whether you can progress to the next level) tests are bad and dangerous, as they force the kids to conform every day of their lives.
If kids want to blow off a formative test, then it's no big deal. If a kid makes a few silly mistakes, they don't have to beat themselves up. They shouldn't have to (or even want to) stay up to 2am cramming. They should be having fun with friends, learning to cook, playing sports, learning an instrument, or hacking on their graphics calculators. But if they flunk, it means they aren't learning, and something needs to be done.
This is already gaining significant traction in the US, as is said in the article.
The article is saying that summative tests are their not to help the student but to help the university. It really should be the other way around.
They said frequent testing, and the more tests you have, the less important each individual test becomes. If you have a test every week in every subject, then at the end you won't need a gargantuan test that supposedly summarizes a couple of years of knowledge.
You have a ton more data to look at a student from.
I don't see why you would ever want NO testing. What you want is CONSTANT testing, like we have Google Analytics on all the time, not just for 2 hours to 'test' some situation.
While this is not what the article is about, university entrance examinations seem to be commonplace around the world. It may not be perfect but it's hardly "stupid". In fact I find the American system rather bizarre.
No, it is not always better. There are many situations where more data leads people to make non-optimal decisions. This is well known.
(By "tribe", I mean certain personality types and cultural groupings, not ethnic tribes. Though of course, the other data points are used to reward certain ethnic groups as well.)
Spain ends up with people who regurgitate facts, but can't think.
However, I do not endorse current "ESO+Bachillerato", as in my opinion it is easier than previous "BUP+COU". From my point of view, Goverment of Spain should copy the Finnish or German secondary education system, as it's been proved to be better.
Here's another point: You have to learn to be more civil in arguments, even those you don't quite agree.
The article doesn't talk about huge tests that determine your future, but you are. You say:
> part of my comment was about testing for young children
> One, at 11, that determined which secondary school
> I could go to. Every private school had its own exam,
> so as a kid you had to take 6-8 exams in about two
> weeks
You claim:
> constant testing, as the article proposes, leads
> to a toxic educational atmosphere where students
> learn greedy optimization for each test.
Finally, you say:
> You have to learn to be more civil in arguments,
> even those you don't quite agree.
I agree with your comments. I think most people here will. The huge pressure, incredibly important, major testing periods to determine your future are toxic.
But that's not what the article is about.
Let me make my argument more clear: You (and the article) seem to believe that it's an either/or situation with the 1-2 large tests and having a lot of small tests (e.g. every week). I don't think this is the case. In fact, the constant testing arises as a byproduct of the large, life determining tests. I didn't make this very clear perhaps.
Coming back to my example: The private school entrance exams (taken at 11) were few and concentrated in a couple of weeks. Yet, since everybody practically started preparing for those exams from the start of primary school (at 7), you had to, too. Understanding wasn't important, creativity wasn't important, it all boiled down to how good a test taker you are. Ditto, for the university entrance exams. The only difference is that by then the test culture was very well ingrained. So, to reiterate, my point was that critical testing periods lead to constant testing.
My second point was that this is bad. This is debatable and, of course, I don't have scientific data to back me up, only anecdotal data gained from observing first my classmates and now my colleagues from different countries (esp. China) who were raised in a similar test-based culture.
I think judging students predominantly by tests (which is what constant testing is about, i.e. the "Google Analytics" approach) are bad for a couple of reasons:
(i) Some people just don't perform well under stress. This is something I've seen some people have a very hard time learning, even after years of test taking.
(ii) It creates a culture where you are reduced to a performance number. We (in Turkey) learned this when we were very young, you are what you do on the test, and don't get me wrong, I thought this was the best way (since I was good). In my undergrad, professors use to hang exam scores on their office doors and we made fun of people with low scores. We didn't find this appalling, we were the best of the best (50 out of a million) and anyone who cannot do good deserved the ridicule. All we thought were the scores, not what we wanted to do in life, if we had an entrepreneurial streak, etc. Just how well did you do in the last test. We were raised like prize race horses, who had to run the fastest.
(iii) So, my final point. Actually, until the end of my PhD in the US the state I was in didn't bother me, after all being an alpha test taker had served me well. I crushed my American classmates in exams, did well, etc. Only after I started my career that I sensed something was wrong. First, years of testing had trained me on a stimulus-response basis: I learned that if I studied well, I would do well on the exam (most of the time). I found that success in real life wasn't as simple, there are office politics, stupid managers, back stabbing galore. The fact that I couldn't do best when I worked the most frustrated me greatly. But what was worse is that many problems I encountered did not fit into simple question/answer approach I knew how to tackle so well after decades of training. I found that complex tradeoffs between many factors was in fact the essence of engineering. I can say that years of test taking didn't really prepare me for my job.
After I wizened to this fact I looked around and saw similar patterns in my foreign colleagues, except most were nor retrospective enough to see what they lacked. I've been trying to correct for my decades of miseducation ever since.
P.S. As for the "uncivil" comment, I think asserting that a commentor "clearly did not read the article" is uncivil, if not on Reddit on HN.
It seems to me that you're seeing all these small tests as preparation for the big tests. You seem to see the small tests as requiring cramming, and that the results from them hang over you. You seem to see them as boxing you in, converting you to a number.
That's not how it's supposed to go.
The point is not to have the tests arising from the culture of the large, life determining tests. The point is to have small, fast, fun tests where it's possible to get an answer wrong and not care. It should be possible to try something off-the-wall in a test, and should it work you get a reward, and if it doesn't work it doesn't really matter. Do it differently next time.
All the points you make are valid, but I think they are still to one side of the intention of the article, and other discussions. I agree entirely that pure "teaching to the test" and "the one true answer - or else" are deeply destructive to the attitudes that are more creative, inventive, and possible more useful in the real world. But I don't think that's the point.
There's a shed-load of stuff for kids to learn, and while exploratory, enabled learning can be effective, it can also be patchy and slow. It can be deep, but it often isn't wide, and it doesn't serve all children equally well.
And to some extent, that's the point. No single system fits every teacher-student pair, and yet that's what people tend to end up with. Teachers are forced to follow the latest fad, even if it doesn't suit their style. Children are exposed to current "best practice" with little of no feedback.
It's the feedback that counts.
If tests are normal, fun, expected and de rigueur then the kids get less stressed, and the teachers get more feedback about their understanding and abilities. Then the best teachers can adapt their presentation to the kids.
It's a tool, a technique, an option. It's not a silver bullet. But it's also not what you are arguing against. You are making valid, but different, points.
My point was that although lots of tests->feedback->better performance is a valid and strong argument, in practice it degenerates into what I describe, at least from what I've seen. If the barrage of tests are well balanced with interactive activities that build on creativity and teamwork, that'll be a perfect combination. Hard to achieve, though, I think.
So what does this mean? Clearly there's more to it than just "more testing is necessary!" or the reverse. Is the Anglosphere in some "unhappy middle", where moving towards either more OR less testing would help? Is the amount of testing in fact orthogonal to school performance? I know both Finland and China are said to have high levels of societal respect for education, perhaps that's the crucial factor—but I'm not aware of any good measurements of this, maybe it's anecdotal.
Another hypothesis is that there's a fairly hard limit to how well children can be educated at this point in time, and that the best schools in Finland, China and the U.S. all near this limit. In China we only look at the best schools and students, and with so many people in China the best will look pretty good. Finland has very high equality and homogeneity, so the worst schools aren't so far off from the best, and the whole country comes out looking good. The U.S., however, has many badly funded schools and marginalized populations, so that could make the difference. Then the question becomes whether we can approach Finland's success, or if the Anglo countries are too inherently different. Again, this entire paragraph is truly rampant speculation, and my mind and gut are both skeptical.
Thoughts, anyone?
PS: I'm aware that I can't give justice to the stuff Kahl writes, what I have written is quite incomplete. But one has to read it on its own. I'm also aware that some of this stuff is only relevant for german schools, yet I think the broader questions is relevant for all countries.
PPS: Refrain from calling me a hippy, a dreamer, a communist or any other stereotype _you_ think is negative. I went through the normal (german) school and university system and it gradually became clear to me, that things as they were and are working are not as good as we want to think, actually, things are rather bad. I'm looking for new ways and possible solutions, trying to find the right questions.
I'm curious how he would explain the success of schools in places such as Eastern Europe and China. More testing, regimented learning, little focus on individual needs...these seems like the opposite of what he's recommending, but it still works somehow.
PS: I think everyone on HN is a bit of a dreamer, no? :)
The test taking tradition is centuries old in China - one of the emperors (forget which one) made many of the civil servant type positions require exams in order to cut down on the cronyism that was rampant at the time. It worked and the tradition stuck for quite a long time.
Reminds me of an article I read about South Korea, where the police escort students to the big high school exam.
http://www.npr.org/templates/story/story.php?storyId=1203397...
Of course, the test degraded (over centuries) to a several day exam, where students were expected to write a massive essay. So they just competed to smuggle in lots of stock essays that they could just copy an appropriate one out.
Why is it that different for kids that need to know, for example, that they suck at multiplication tables and so need to improve for their next exam?
Sometimes we overthink things way too much.
It reduces children (who are sensitive to messages like that) from a full person with a unique character to their performance as tested. If you haven't yet developed the necessary cynicism to deal with this onedimensional assessment of personhood, you will quickly. If that is actually desirable is another matter.
To do well, a student must be motivated to learn. And one of the biggest motivators at this level is that "I always am at the top of my class in this subject, so I enjoy it." Not everyone can be at the top, so naturally the others won't care as much. They will stop trying to get an A and settle with "not failing." And the students who start failing will stop caring completely.
If you hide a student's ranking, then that initial motivation, "I do better than the others," disappears and we are left with students who must be entirely self motivated. And that happens so rarely that it isn't a practical motivator.
So, the solution is to find some kind of measurement that accurately reflect somewhat student progress. This way, teachers can reliably test education experiments. Of course, the larger problem is probably the teachers' possible resistance to the scientific method and poor training in devising experiments.
This is way too much discussion about way too little. Leave it up to the teacher to choose how and when to drill, measure, and stretch their pupils.
How convenient that the "research" confirms his relevance.
"These are not the big once-a-year or once-in-a-lifetime exams, like the SATs, but a stream of smaller, less monumental tests, designed in theory, at least, primarily to help students and their teachers know how they’re doing"
How profound. How exactly does that differ from current exams!? K-12 students don't have once-in-a-lifetime exams - they have less monumental tests that already take the focus away from what school is for - learning - not assessing. More assessments will lead to more focus on passing assessments and less focus on students and then less learning. Who can argue this? The fact that Chinese schools may possibly (which I'm suspicious of in the first place) have more tests, should in now way be seen as a cause for better learning. Here are a number of things the Chinese do better than most in the states that would lead to superior education:
1. Parents are more involved in their kids education. 2. Students are more involved in their own education. 3. Teachers are more involved in quality teaching/mentoring. 4. Teaching/learning methodologies are better.
How ridiculous.
I find this kind of thinking dangerous in that this question is essentially asking another person to think for you. Sometimes the answer should be a string of questions, not always a test.
What do you feel you have learnt? Do you remember anything? Do you feel it has changed you? Was anything memorable? Did you enjoy the class? Do you enjoy doing problems? Do they interest you? Do you feel you are learning well?...Please think about it, child. What do you feel?
Many schools today have eliminated all tests for younger children. Think about this. No tests at all, IMO, cannot be a good thing.
Children do need to be assessed and challenged. Sitting a child down with a pencil and paper is not the only method of assessing a child, and most definitely not the most effective.