/sci/ - Science & Math

[Anonymous No.10402949]

Would an SAT score be relatively reliable in deducing IQ if it was taken by a regular high school student in a 3rd world country (that being they have no specific preparation for the exam itself)?

[Anonymous No.10402966]

Yes, but you wouldn't be able to use a SAT -> IQ score model developing on a sample of Western students as the SAT is very easy to prepare for and scores are inflated. You'd have to build a model using 3rd worlders

[Anonymous No.10402999]

>>10402949
Your performance on the SAT could change over time (which is why SAT scores become invalid after a few years)

[Anonymous No.10403002]

>>10402966
In what way would the modelling change?

>>10402999
Wouldn't the results be particular to the current focuses and style of the education system at the time? In that sense, wouldn't be at least somewhat accurate in telling cognitive ability? Or am I wrong?

[Anonymous No.10403035]

>>10403002
>Wouldn't the results be particular to the current focuses and style of the education system at the time? In that sense, wouldn't be at least somewhat accurate in telling cognitive ability? Or am I wrong?
Your second sentence contradicts your first. Most if not all of the abilities in the SAT are learned and drillable, not inherent. You'd have to find a way to isolate all confounding variables and have an entirely equal environment spanning literal decades to ensure that there is nothing except inherent cognitive ability affecting the results, which cannot be feasibly done. So you'll always be in a state of not knowing how far off you are.

[Anonymous No.10403095]

>>10403035
I see. That makes sense. Thanks for the clarification.

[Anonymous No.10403140]

>>10403002
>In what way would the modelling change?
Western students will have a higher SAT score than their IQ should predict, because of the prevalence of test prep.

[Anonymous No.10403404]

>another IQ thread