More than a decade ago I devised a short quiz designed to test the chess thinking skills of children rated up to about 1500 ELO/100 ECF. There were eight questions in which my pupils were invited to choose a move for White and give reasons for their choice either using a short sentence or a variation.
The results were written up in an article which was published in various places.
Two of the original questions were slightly unsatisfactory so were omitted from the article. I’d rather forgotten about the whole project, partly because I hadn’t had the opportunity recently to teach in an environment where the test would be appropriate. But following a recent discussion about the article with my online friend from across the Atlantic, Paul Swaney, I decided to revive it for the Intermediate Group at Richmond Junior Club, making some minor changes to two of the other positions in the process. This group is for children of primary school age who have mastered the basics and understand notation, but who are not yet ready for serious competition. Their ratings would be up to about 800/1000 ELO and their ECF grades up to about 40/50. My previous experience is that players of about 1500 ELO/100 ECF will get most of the questions right, but anyone much below that will struggle to get more than a few correct.
My interest in these questions is not so much the answers that the children give but the reasons for their answers. I have to bear in mind, of course, that young children are not always very good at putting their thoughts into words and their words onto paper.
This is Q1: a basic test of endgame knowledge. I repeat this over and over again with my pupils, so some of them will get it right. Others will choose a random move, saying that if Black replies with c2 they’ll be able to capture the pawn. This is an error in differentiation. I explain to them it’s like me asking what the difference is between Jack and Joe, and getting the reply “Jack’s a boy”.
It occurred to me that there’s a slight problem with interpreting the reasons they give for their answers to this question. Many children think ‘stalemate’ is just another word for draw and announce ‘stalemate’ when they reach a position with king against king. If they say that they’ll play Kc1 because they’re more likely to get stalemate they might have the right reason, or they might just be saying it because they’re anticipating c2 in reply.
In Q2 we have a pawn on the seventh rank about to promote. But there’s a problem: Black is threatening mate in 1, which, because it relies on a pin, is not easy to see at this level. The other problem is that, because promotions are very common in games played by young children and a queen is almost always chosen, they find it very difficult even to consider the idea of promoting to anything other than a queen.
As expected, most children at this level promote to a queen here, overlooking the mate. They will often point out that next move they intend to play Q(either)g8#. Some children will play something else instead even though they haven’t seen Black’s mate threat, thinking that promotion can wait. So Rd1, for instance, is sometimes chosen. Some children move the rook, explaining that they’ve seen the mate and want to provide an escape square for their king on g1, overlooking that they’re just allowing mate in 2. Some children will stop the mate by playing a move like c4 or Kh2. Only a few will even consider promoting to a knight rather than a queen because the idea of promoting to a queen is so ingrained. When I ask children how many possible moves they have with the pawn on f7 they’ll usually say ‘one’: it takes me a very long time to persuade them that the answer is actually ‘four’.
When I first devised Q3 I expected it to be a straight choice between captures on d4. I was wondering how many would choose the rook capture because they wanted to avoid doubled pawns. At this level at least half the children, typically, will give an incorrect answer. A few will mention doubled pawns but most will not: children will usually see the rook capture first because pawns move and capture in different ways and play it with no further thought.
Many children at this level are familiar with the back rank mate and some of them will notice the problem with Rxd4. You’d expect them all to play cxd4 instead, but not all of them do. Some of them will instead play a move such as h3 to stop the back rank mate, intending to capture the knight next move. This didn’t actually happen with the small RJCC group, but when I tested some of my private pupils later, one of them did play h3. Again, this demonstrates that children at this level tend, if they’re thinking ahead at all, to think “I go there, then I go there, then I go there” rather than “I go there, then you go there, then I go there”. The idea that they have an opponent who is going to try to find the best move is a very difficult concept for most young children whose theory of mind skills are, as yet, insufficiently developed for them to consider their opponent’s perspective in any meaningful way.
Q4 might be considered the hardest question in this set. The idea, which mostly works, is that weaker players will play the right answer for the wrong reason, and will take the queen without any further thought. The intermediate players will spot the potential back rank mate but the idea of meeting Re1+ with Rf1 rather than Rxa1 won’t occur to them. So they’ll play a move such as Qd1 or h3, planning to capture the queen next move.
At this level, only a few players will get the question right for the right reason, pointing out that after 1. Rxf6 Re1+? they’re going to play Rf1. There will also be a few who don’t notice that they can capture the black queen.
I’ll consider the other four questions next week. If any of my readers teach at this level and would like to try this quiz out or devise suitable questions of their own I’d love to hear from them.