Discovering Statistics; Chapter 5

Question 1

Under what contexts does bias need to be considered?

Answer 1

1) Things that bias parameter estimates (including effect sizes)
2) Things that bias standard error and confidence intervals
3) Things that bias test statistics and p-values.

Question 2

Most potential sources of bias occur when what assumptions are violated?

Answer 2

1) Additivity and linearity
2) Normality of something or other
3) Homogeneity of variance
4) Independence.

Question 3

Are parameter estimates affected by having a non-normal distribution?

Answer 3

Yes.

Question 4

Is the accuracy of confidence interval influenced by having a non-normal distribution?

Answer 4

Yes, the accuracy decreases is there is not a normal distribution.

Question 5

What is another term for homogeneity of variance?

Answer 5

Homoscedasticity.

Question 6

What is another term for heterogeneity of variance?

Answer 6

Heteroscedasticity.

Question 7

How does Levene’s test indicate homogeneity of variance?

Answer 7

If Levene’s is significant, assumption of homogeneity of variance has been violated.

Question 8

What is transforming the data?

Answer 8

A shift in the data to display a normal distribution.