Investigating gender differences in chess pattern recognition

by ChessBase
2/21/2020 – Here's an opportunity to support chess research. Tom Koolen bills himself as an "interdisciplinary data scientist" and a chess player, who's attempting to get a research project off the ground using the crowdfunding platform GoFundMe. Commentator WIM Fiona Steil-Antoni has promoted the project on social media. | Photo: Michal Vrba on Unsplash
YOUR PERSONAL CHESS COACH - Whether you’re taking your first steps into the world of club chess, or already playing at a tournament level: with FRITZ, you can train more efficiently, intelligently and with a more personalised approach than ever before.
FRITZ is more than just a chess engine – it’s a training revolution! Whether you’re taking your first steps into the world of club chess, or already playing at a tournament level: with FRITZ, you can train more efficiently, intelligently and with a more personalised approach than ever before.

A foray into crowdfunded research

We have covered the gender gap in chess on a number of occasions and it always provokes interesting (if sometimes heated) conversation in the comments. While there are heated opinions on all sides, there remains a dearth of serious research on the subject.

This new project is endorsed by chess influencer and commentator Fiona Steil-Antoni:

Tom Koolen is organizing the fundraiser, and describes it as follows:

Because of the strength difference between the top male and female chess players, only male experts have been analysed regarding pattern recognition. More relevant should be the exact study of gender difference regarding the perception of the game in order to be able to raise female participation at the high level. Thus the aim of this experiment is to research the key component of chess strength (pattern recognition) for female players as well in comparison to male players.

In (Bilalic, 2009) the higher strength of top male chess players is attributed mostly to participation rate and male predominance at the top. (Bilalic, 2010) is a very interesting and visual research on the difference in pattern recognition (which is seen as the key strength factor of chess masters) between chess experts and novices. Female chess players were not included in the research group.

Other variables attributed are memorization and psychological factors that lead different strategic decision-making between the two genders. This projects wants to create a homogeneous test for each of these variables to present a clear visualization of the specific strengths of both genders.

The choice to only look at male chess players leaves a wide research gap open to explore. In order to increase participation rate amongst female chess players, chess education needs to be tailored to their specific needs, like in many sports and disciplines. Measuring the differences in the above-mentioned variables between the two genders will lead to significant insights into the gender difference in perception of the game of chess, which can be used to improve training approach and raise the participation rate amongst females.

In a psychological game like chess, it is of importance to research these differences. The insights could furthermore lead to evidence for further fMRI studies into generalizing gender brain differences in decision-making.

The goal of this research is to conduct the above-described homogeneous test with a target group of chess players representing equally as many male as female players and with a spread in playing strength. The results will be visualised and published openly, while also being presented on well-known chess media. A more practical goal is to provide research data that can improve training focuses for female chess players.

What will be the end product?

A research paper will be published open-access and the research will be provided to major chess media. A video presentation will also be shared online. The product can be used freely by all chess media and federations and aims to clarify unclear misconceptions about the gender difference in chess.

Tom Koolen may be reached via email.

Links

Topics: Gender, Gender Gap
Reports about chess: tournaments, championships, portraits, interviews, World Championships, product launches and more.

Discuss

Rules for reader comments

 
 

Not registered yet? Register

tomohawk52 tomohawk52 2/25/2020 02:54
"... since they are often so unrightfully prejudiced. ..."

If you mean that females receive preferential treatment in the chess world then we are in agreement.
gmtum gmtum 2/25/2020 11:26
(continued) That is why I prefer a quantitative method, as you are correct about social science, bias should be reduced and coding of qualitative interviews would probably have a lot of bias in there. A questionnaire could be conducted on the side into the psychological part for sure, but neurological data following from eye-tracking should be key as has been in the earlier comparison of novices and experts.

I thank you Shakey for reaching out to me with great advice and I want you to know that I take it very serious as your professional opinion and will think about all these things when conducting my research :)

Kind regards,
Tom
gmtum gmtum 2/25/2020 11:24
Thank you so much for your helpful reply, I totally agree with the points that you have stated and have elaborated these in emails with people who asked for them. For full transparancy, I will be more than glad to share them with you:

Funding:
The first $500 (which the fundraiser will probably not even cross) are for travelling costs (each travel in Netherlands, Belgium, Germany would cost at least $50, and I will go to a number of locations in March) to the locations where the players will be tested, the costs for research equipment (eye-tracking hardware and software) and indeed private statistical software license. If more money would have been raised, the plan was to travel by plane abroad to a large tournament and try to get some of the very top female players involved in the research. As this is not reachable anymore, I will cut off the fundraiser were it to pass the first $500, as no more money will be required. The research will go on anyway as already $250 have been raised and the rest of the costs I would cover myself if necessary!

Sampling:
I am recruiting participants over the whole ELO spectrum; the test population will control for the variable of age as well. So there will be pair-wise comparisons between male and female players of ELOs 1200, 1300, 1400, 1500, 1600 and so on whereas there will also be comparisons between age groups, to check if differences in decision-making from pattern recognition occurs at birth or develops later on. Evidence there could give a future recommendation for a future longitudinal study in pattern recognition with specific training plans.

Bias:
I agree with you that I have stated that not-optimally. I meant to say that I want to give the female players an objective analysis since they are often so unrightfully prejudiced. But of course, if the objective statistical tests give significant advantages to males as results, these will be fully objectively reported as such. (continued in next comment)
Shakey Shakey 2/25/2020 05:39
It could be interesting research, if conducted effectively. Social science research is indeed notoriously difficult! People are tricky.

Queries:

1. FUNDS paying for...?
Curious. What are the actual costs of the research? If funding is requested, what for?
Is this for production of materials - copy costs? To pay grad reseachers for the data analysis? To buy SPSS?
It might seem polite and transparent to indicate what the money is actually for.
When writing a research grant, a breakdown is required. If asking for funds from chess players for something currently fairly nebulous, it appears appropriate to do similarly.

2. SAMPLING
How will you sample the population?

3. BIAS
Be wary of bias in your project. There appear to be certain assumptions underpinning this already. Try very hard to not bias this in advance.

Sorry if this appears tough. It is meant to be robust, but helpful and friendly. There are reasons that all academics go through a period of research training. It is so that the resultant published findings should be somewhat plausible at a minimum, and resistant to more obvious challenges.

Wishing you the best with your research,

Shakey
fede666 fede666 2/24/2020 03:34
males are just better than females in chess...don't care about the real reason..I just enjoy the game as it is...
gmtum gmtum 2/24/2020 12:10
The summary on the GoFundMe page was written with the intent to make the research overview accessible to readers of all education levels. In the research itself, the statistical methodology and population design will be explained fully transparant. Reliability and validity will be tested to academic standards. While I understand that academics immediately think about these things, the goal of this article was to inform a broader group of the research goals. Obviously I know that APA style citation is used in academic journals, but we are here on ChessBase making research accessible and not in a journal.

Furthermore, this research is not going to be in an academic journal, this is a practical study aimed to address unanswered questions that are on the mind on many of the people that reached out about this research. All of the technicalities will be fully transparant in the end product, which will be provided in full-text along with the presentation on chess media. I respect everyone's comments. and as I said, all methodological details will be there in the paper for those who want to validate it, but on the chess media, the intention is again to make it all accessible.
Shakey Shakey 2/24/2020 11:16
Alas. Yes, per the previous comments, this does not read as the efforts of someone who has undertaken research before. The methodology is not at all clear.
By the way, when writing citations it should be name(s) then year, assuming APA style. For example, 'As Bilalic (2009) earlier indicated...'. Name and year in brackets together is used *following* a reference to prior work.

This needs an undergraduate dissertation supervisor to fix lots of things. Go away, come back when this has all been tidied up.

My professional advice as an academic is that ChessBase should not be anywhere near this, and readers might be wary of funding the research of someone who appears not to know about writing research proposals, nor about conducting research.
karban karban 2/23/2020 06:14
I feel bit of the same. The point of the research seems to be to prove that women have similar playing strength as men and that only problem is their smaller participation.
But even very basic data, as well as common sense, suggest that there are fewer women because they, on average of course, don't want to play a war game which chess is all about. Look on teenagers in real world, where the statistical pool is much bigger- virtually everyone in the West has an access to computer/smartphone but only boys are glued to screens for hours to play strategical/fighting games.

There's even famous document done by some Norwegian performer where the so called 'paradox of equality' is presented - the more 'equal', in modern sense, the society is, the more people are free to choose according to their inborn preferences. Thus in Norway and other Scandinavian countries we have even higher ratio of women in kindergarten, school, hospital than in other countries and similar more men in egineering and other traditionally male professions. And all this despite ferocious and sometimes ridiculous efforts of these countries to make them as 'equal' as possible:)
So instead fighting with nature maybe we should accept it and try to make it work for us. I'm all for women to play chess but we don't make them stronger by basically forcing more of them to play the game or cutting of women tournament altogether. And you don't need to have phD to conceive these simple conclusions.
fgkdjlkag fgkdjlkag 2/23/2020 05:02
This reads like a high school project by someone who has never done research before. How is this worthy of publication on a leading international news portal?

It cannot be considered as scientific research as the author is biased (the aim of the "research" is to raise female participation to the highest level). If it was exploratory research, the objective could not be stated in advance. If it is not exploratory research, what is the hypothesis that is being tested?
What are the study methods? How do we know that any of the variables can be properly tested and that the study design is sound?
How is the study coordinator going to enroll subjects and complete the research in such a short period of time?
The explanation for higher male levels cited from a previous study (participation rate and male predominance at the top) are non-starters; it is well-known that people do what they are good at, so it would be expected that a group that is better at something will be higher represented. Male predominance at the top is a tautology, or maybe he is referring to increased variability.
If he finds that females are worse at pattern recognition than males, what does it explain? Maybe the males in this project studied more hours on pattern recognition?
Where is the link to the actual study proposal? It almost seems like the aim of this project is to make a quick buck.
1