There is a government-sponsored study of student achievement in large urban school districts. They give a reading test scored from 0 to 500. A score of 232 is a "basic" reading level and a score of 270 is "proficient." Scores for a random sample of 1400 eighth-graders in a city had x = 237 with standard error 1.0.

We don't have the 1400 individual scores, but use of the t procedures is surely safe. Why?

A. Test scores are always normally distributed.
B. A random sample should be normally distributed.
C. The sample size is very large, so the only potential hazard is extreme skewness.
D. A small sample size of a simple random sample should be normally distributed.