sapiens

I don't think that intelligence is a universal psychological construct. What is intelligent in one culture may very well be considered clueless in another. So, there could never be a "true intelligence test" for everyone in the world.

I would consider using American IQ tests to evaluate the intelligence of members of non-western cultures to be culturally biased. Within the United States, scores on an IQ test predict a variety of outcomes regardless of your race/ethnicity. Because the IQ test predicts social outcomes regardless of race or ethnicity within the united states, I don't consider it to be culturally biased when used within the united states.

If there was such a thing as a "True intelligence test" (again, most evidence suggests that there is not), I'm not sure what benefit would come from testing everyone.

Incidently, there is debate within the field of intelligence about whether intelligence is a domain general ability (like a general problem solving ability), or a number of domain specific abilities. (Few deny the predictive power of IQ scores, but they do debate what an IQ score represents). Thurstone forwarded Primary Mental Abilities theory which argued that we don't have one ability (as represented by an IQ score), rather we have many specific intellectual abilities. He included verbal comprehension, word fluency, number, spatial ability, associative memory, perceptual speed, and reasoning or induction. Others, like Gardener, have argued for even more specific abilities. These conceptualizations suggest that individuals in diverse cultures would be considered "intelligent" to the extent that the individual possesses the specific abilities necessary to succeed in that in those specific environments.

There are reputable studies done by reputable researchers at reputable schools and published in reputable peer-reviewed journals. I don't have time to give you examples right now. Off the top of my head, the National Longitudinal Study of Youth is one reputable data set. The raw data is available, but what would you do with the raw data? And how would double-blinding help? Most IQ tests are administered via a paper/pencil test. The outcome data is objective and demographic. Plus, this is all correlational research, not experimental.

I don't particularly care whether self-identified members of one ethnic group score differently than self-identified members of another ethnic group, but my experience with people of many races is irrelevant - personal experience is not aparticularly good source of data.