Module 4

Question 1

normal distribution

Answer 1

happens in nature, most obs close to mean. symmetric around mean
-continous
-prop. is the area under curve
-symmetrical
-single-mode all mean, median and mode are same
-2/3 area under the cuvre is 1 std away from mean
-95% of the area of a normal distribution lies with in 1.96

Question 2

standard normal distribution

Answer 2

measures how far an observation y is from the population mean

Question 3

t-distribution

Answer 3

use z score to estimate the probability of obtaining a particular sample mean given the population of means from which we are sampling

Question 4

single sample t-test

Answer 4

compares the mean of a random sample(from normal population) to population mean proposed in a null hypothesis

Question 5

paired sample t-test

Answer 5

difference between two paired observations. increases the power. ignoring the pairing= reduced power

Question 6

when can we ignore the violations of assumptions?

Answer 6

1-small deviations
2-large sample size
3- non normality if deviations are small and sample size is large

Question 7

what can we do if we can’t ignore the violations?

Answer 7

change the scale of measurement( transformation)
must be convertible and cannot keep trying transformations until p<0.05.

Question 8

what is log transformation good for?

Answer 8

its good for rations, products, counts, right skewed distributions, can use base 10 or 2 or even e(ln). doesn’t work with zero