Rating changes for last rounds are temporarily rolled back. They will be returned soon. ×

asrinivasan007's blog

The relationship between IQ and CF Rating (continued)
By asrinivasan007, history, 3 years ago, In English

Hello Codeforces,

Sorry for the delay on the update blog.

This is a continuation of my blog on the relationship between IQ and CF Rating. Previous Blog Post

First off, I would like to begin by expressing my gratitude to everyone who submitted the form. In the end, I received over 140 submissions, which is a crazy number. Thank you to everyone.

After putting all the data into excel, I saw there was almost no correlation in any of the categories.

Here are the scatter plots for all the categories:

Overall:

The R^2 value here is very low, and although you can see a general increase in rating as IQ increases, it is slight and full of outliers.

Memory IQ:

The Memory IQ section also had a low R^2 value, and although rating once again increases slightly with IQ, there are many outliers.

Verbal IQ:

This section will have three separate graphs.

This graph is the combined results of both those fluent in English and not:

An even lower R^2 value is present here, implying no correlation.

For non-English speakers:

There is another weak correlation here. No trend can really be observed.

English speakers:

Surprisingly, there is a lower correlation for English speakers than non-English speakers.

Spatial IQ:

A general upward trend can be observed here, but once again with many outliers.

General Observations:

The verbal IQ section lowered many people’s scores, both English speakers and non-English speakers. Also, given it had the least correlation, it is easy to see that it affects rating less than all the other categories, which was expected.

Spatial and Memory IQ had more impact on rating than verbal.

What does this mean?

As some people were commenting on the previous blog, IQ can speed up improvement in CP, but is not a direct indicator of rating in any way.

Although these results were not what I expected, this does still provide hope that despite your IQ, hard work and dedication will drive you further.

  • Vote: I like it
  • +257
  • Vote: I do not like it

Comments Comments (10)
| Write comment?
»
3 years ago, # |
  Vote: I like it +20 Vote: I do not like it

Great Work Indeed!

→ Reply
»
3 years ago, # |
  Vote: I like it +8 Vote: I do not like it

Nice beginning. Now lets wait for the part with Test validation process )

→ Reply
»
3 years ago, # |
← Rev. 2   Vote: I like it +20 Vote: I do not like it
Surprisingly, there is a lower correlation for English speakers than non-English speakers.

I wouldn't say that's so surprising; knowing lots of words in a non-native language requires more effort than knowing lots of words in a native language.

Also, wasn't too surprised by the low correlation; CP, aside from memorizing algorithms, is nothing close to rote memorization.

→ Reply
»
3 years ago, # |
← Rev. 2   Vote: I like it +84 Vote: I do not like it

yet again proving that IQ is astrology

edit: also, reporting bias

→ Reply
»
3 years ago, # |
  Vote: I like it +18 Vote: I do not like it

cf is full of high iq 5Head people

→ Reply
»
3 years ago, # |
  Vote: I like it +73 Vote: I do not like it

It's particularly interesting that you have barely any people with sub-100 composite IQ. Is it reporting bias, does CF only have above-average people or does the IQ test inflate everyone's IQ like so many online tests?

→ Reply
»
3 years ago, # |
  Vote: I like it +8 Vote: I do not like it

Rip. Above the line on all graphs.

→ Reply
»
3 years ago, # |
← Rev. 3   Vote: I like it +101 Vote: I do not like it

Note that CF is a self-selecting community (plenty of people who aren't as good at programming just stop doing it), so you can easily run into things like Berkson's paradox with analysis like this. This self-selection probably also contributes to the fact that the IQ averages are much higher than 100.

→ Reply
»
3 years ago, # |
  Vote: I like it 0 Vote: I do not like it

How about taking a larger test case? I would suggest you to wait until you have at least a thousand entries, and then analyse the results.

→ Reply
  • »
    »
    3 years ago, # ^ |
      Vote: I like it +15 Vote: I do not like it

    I could reopen the form submission, but I doubt there would be any major changes.

    → Reply