Understanding the Impact of Machine Learning on Labor and Education by Joseph Ganem

Author:Joseph Ganem
Language: eng
Format: epub
ISBN: 9783031310041
Publisher: Springer Nature Switzerland

For example, if Q = SAT score for high school seniors, all four conditions above are met. (1) SAT scores are distributed nearly normally about an average μ = 1500 so that Eq. (3.6) is a good model for f(x; z) with z = 300. (2) The function described by Eq. (3.6) is normalizable. (3) There are millions of high school students that take the SAT, which means that the addition or subtraction of any one student with any given score has no effect on the overall distribution. (4) Even though each year a different population of students take the SAT, the test is intended to provide a standard such that the distribution of scores each year remains about the same—normally distributed about μ = 1500 with z = 300.

Over long periods of time this final condition does break down for many traits. Average SAT scores have decreased in recent years. Over many decades the average height for adult females in the United States has increased. The distribution of other possible Q’s, such as age also evolves slowly in time—life expectancy has slowly increased which has increased average age. But these are slow drifts and not relevant to the arguments in this chapter.

Download

Copyright Disclaimer:
This site does not store any files on its server. We only index and link to content provided by other sites. Please contact the content providers to delete copyright contents if any and email us, we'll remove relevant links or contents immediately.