Yes, structurally if not deliberately (it is unlikely that mainstream IQ tests are anything but diligently designed to avoid overt bias, unlike those of the middle 20th century). If you measure something that has a large SES causative component, and SES is for whatever reason itself highly correlated with race, then your measurements will encode a racial bias.
The big debate in academic psychology is the extent to which factors like SES (and factors downstream of SES, like general health) confound the measurement of some kind of general intelligence. The evidence we have right now pretty clearly points to an effect, so it's down to what the size of that effect is.
Your second statement is incorrect.
Having a racial bias would mean that IQ tests are incorrect (having significantly different errors) for certain samples/groups. But no one is saying that.
Imagine a test to measure the resistance of various materials. Is the test biased because it shows that concrete is stronger than rotten wood? No, it just measures the resistance of different materials. If the measurement error (not the measurement itself) was significantly different for different materials then yes you could say the test is biased.
You are just saying that SES affects intelligence, which is most likely true and immediately prompts the question of what affects SES.
A large percentage of the population seems to believe that IQ and similar cognitive ability tests are something like a credit score, having no purpose but to direct society to the high scorers so that they can be showered with opportunity, while denying the same to low scorers. And it's true that tests like the SAT can have a big impact on the opportunities we get in life. But they can also be used in a more win-win way, to match people to opportunities they're ready for. And for that, you need an honest assessment of where the person is actually at, separate from how they got there.
A poor kid may score worse than they would have scored if they weren't poor, but if we want to know whether they're ready for MIT or a course above their grade level, it might do them a disservice to confound our perceptions of their actual current level with our personal indignation over socioeconomic inequalities and our hopes in the potential they'd realize if those inequalities were eliminated.
Bias has many different meanings. Now, correct me if I'm mistaken, but in this context, I believe we are talking about bias as a difference between a measured value and a true value.
What is the true value in this case? Is it the counterfactual performance a person would have if their SES were not low? Would that be the "right" true value in all applications? Might it not be the case that, in some contexts, the "right" true value is the person's actual current ability level and not some counterfactual of how they would do in other circumstances that were perhaps more just and fair? For example, what if you are trying to give them the learning opportunity best suited to their actual current ability level, not trying to rank people and only give the top-ranked some special status or credential?
Does this mean that today's IQ tests have racial bias?
> There’s also a historical association with IQ assessment and eugenics.
Does this mean that today's IQ tests have racial bias?