Class Notes (1,100,000)

US (480,000)

UCLA (10,000)

STATS (200)

STATS 10 (100)

Zanontian, L.A. (8)

Lecture 5

Department

StatisticsCourse Code

STATS 10Professor

Zanontian, L.A.Lecture

5This

**preview**shows half of the first page. to view the full**3 pages of the document.**Stephanie Sanchez

Stats 10 Lec 4 Sec A

904652912

Assignment: Lab 5 - Compared to What?

TA Name: Min, S.

Question 1: Using visual and numerical summaries, describe the differences between the

distributions of weights for babies born to smokers vs. non-smokers in the actual data. If a

physician told you that a mother’s smoking affected the birth of her baby, (a) how much,

typically, would you say the mother’s smoking changes a baby’s weight at birth?

a) We could say that smoking mothers’ babies have a lower birth weight by just looking at

the medians of the table and graph of the actual data. As a result, we can say that

babies of smoking mothers weigh 6 ounces less than babies of nonsmoking mothers.

Question 2: We will work with a chance model that assumes that smoking has NO effect on the

babies’ weights. Under this model, simulate a set of data and calculate the difference in median

birth weights between the smoking and non-smoking groups. Repeat this 100times. According

to the output of your chance model (your 100 simulations), (a) what would you estimate is the

typical difference in medians between the two groups? (b) What would you consider to be an

unusually large difference, according to the chance model?

a) We can estimate the typical difference in medians between the two groups to be 0,

according to the output of the chance model

b) It would be unusual to see a difference over 2 standard deviations, such as 6, according

to the chance model.

###### You're Reading a Preview

Unlock to view full version