Intelligence quotient (IQ) tests produce highly variable results. Do you think it is reasonable to base the placement of children in special classes, such as for the gifted, on the single score? Support your opinion.

Almost without exception, I do not support differentiating people based on any single measure. Human beings are simply too complex and measures of human attributes too unreliable to make such single factor decision making prudent. 

Having said that, there are definitely areas where IQ scores could reasonably be used as a significant factor in decision-making. To the extent that tiering students by ability in school is a goal (ignoring the question of whether such tiering is itself a good idea), IQ is most likely a useful tool. In the first place, it was designed to predict academic success and studies indicate that it does this reasonably well. As Bee & Boyd (2007) say, “Children with high IQ scores are more likely than their peers with average and low scores to be among the high achievers in school” (p. 191). Therefore, it should be clear that for this purpose, IQ scores are “valid.” Further, if IQ scores were to be used in this fashion, they would need to be “reliable,” which is to say that individuals’ results must be stable over time. According to Bee & Boyd, “IQ scores are, in fact, very stable” (p. 189). Finally, I would have ethical qualms if IQ were somehow misrepresenting predictive outcomes across social or ethnic groups. Not so, say Bee & Boyd: “These predictive relationships hold within each social class and ethnic group in the United States” (p. 192). Therefore, to the extent that faculty is attempting to differentiate students by likely educational performance, IQ can morally and practically be reasonably used as one consideration.

However, it is important to understand what IQ scores are not. They are not the sole determinant of likely success. Many factors influence educational outcomes, from culture to heredity to birth order to early childhood environment. As Bee & Boyd (2007) say, “Some children with high IQ scores don’t shine in school while some children with lower scores do” (p. 191). A much strong tool to use in placing students in gifted class would be actual prior educational outcomes, particularly those in the subject in question. IQ tests are also predictors of very specific measures of performance, academic performance. So depending on the type of class, IQ could be very misleading. For example, it would be unwise to use them to populate varsity sports team or the honors drama classes.

Finally, there are troubling correlations between IQ scores, ethnicity, and social class, among others. It is true that these factors correlate across the entire gamut of student achievement. However, given this, I would much prefer direct measures of capability (i.e. grades in prior similar classes) to be the main determinant in tiering classes.

Reference 

Bee, H., & Boyd, B. (2007). The developing child (11th ed.). Boston, MA: Pearson Education, Inc.