The phrase "average differences in" is correct and usable in written English.
It can be used in contexts where you are discussing statistical comparisons or variations between groups or datasets.
Example: "The study aimed to analyze the average differences in test scores between students from different educational backgrounds."
Alternatives: "mean variations in" or "typical disparities in".
What is scientifically untenable - and indeed scientifically meaningless - is to claim that average differences in intelligence, as measured by IQ scores between different population groups, are caused by genetic differences between those groups - the core thesis of pseudoscientific racism.
Researchers have long debated what IQ tests actually measure, and whether average differences in IQ scores--such as those between different ethnic groups--reflect differences in intelligence, social and economic factors, or both.
The average differences in height are about 0.5 mm (about 3%).
Take a photo of women and men in any moment in time, and, apart from those matters related to reproductive physiology, you will find average differences in, say, fondness for romance novels, vampire movies, power tools, and fistfights.
The average differences in relative humidity and dry bulb temperature between ambient and inside the cooler were 54.8% and 15.1 °C, respectively.
We estimated a multilevel mixed-effects linear regression model by comparing the level of total monthly diabetes drug expenditure for drugs subject to PVAP and existing drugs after adjusting the average differences in drug expenditure before and after the PVAP.
Separate subspecies are sometimes recognised in the Atlantic and Indian Oceans, based on average differences in the breast-band and overall depth of the upperparts coloration, but the species is arguably better characterised by its having two colour morphs, of which the dark morph is much less numerous.
Also, we calculated average differences in seismic intensity between two stations for scalar correction.
A recent study of students in the Netherlands explores one possible aspect of that complexity: the influence of average differences in competitiveness on students' choices of curricula.
Indeed, we show that accounting for academic background can fully account for average differences in switching behavior between blacks and whites.
We found not clear evidence for significant average differences in elasticity sizes, and the critiques of the Hausman model might be overstated.
