You are on page 1of 8

Homework M3

1. We are going to examine the association of high systolic blood pressure (SBP) in two health
assessment clinics on a given day. In Clinic A there were 30 patients, 5 with high SBP. In Clinic B,
there were 24 patients, 6 with high SBP.
a) Fill in the following 2x2 table with the frequency for each cell

Clinic A Clinic B Total

High 5 6 11
SBP

Low 25 18 43
SBP

Total 30 24 54

b) What is the odds of having high blood pressure in Clinic A?


Solution
High SBP 5
odds of having high blood pressure∈Clinic A= = =0.2
Low SBP 2 5
c) What is the odds of having high blood pressure in Clinic B?
Solution
High SBP 6
odds of having high blood pressure∈Clinic B= = =0.333
Low SBP 18

d) What is the odds ratio of having high blood pressure if you are from Clinic A compared to
clinic B?

Solution

5
25
odds ratio of having high blood pressure for Clinic A compared ¿ clinic B= =0.6
6
18

e) What is the odds ratio of having high blood pressure if you are from Clinic B compared to
clinic A?
Solution
6
18
5
odds ratio of havinghigh blood pressure for Clinic B compared ¿ clinic A= =1.667
25
f) What is the odds ratio of having LOW blood pressure if you are from Clinic A compared to
clinic B?

Solution

25/5
odds ratio of having low blood pressure for Clinic A compared ¿ clinic B= =1.667
18
6

2. The World Health Organization (WHO) recommends HPV vaccines as part of routine vaccinations in
countries that can afford them. HPV vaccine can prevent most genital warts and most cases of
cervical cancer. In 2006, the FDA approved the first HPV to protect females between the ages of 9
and 26. In effort to gauge the acceptance of the HPV vaccine in the US, data was collected in 2007-
2008 as part of a large national survey called National Health and Nutrition Examination Survey
(NHANES) to describe the types of variables that may be related to receiving the vaccine. The table
below summarizes the results of 381 females aged 19-26 years who responded to questions about
receiving at least 1 dose of the HPV vaccine. Demographic and sexual characteristics were
evaluated in bivariate analyses.
a) What do you think is considered the dependent variable in this study? Is this variable
quantitative or qualitative? If quantitative specify whether it is continuous or discrete and if
qualitative specify whether it is ordinal or nominal.
Solution
Dependent- acceptance of the HPV vaccine
The variable is qualitative-nominal
b) What is the population of interest for this study?
Solutions
All females between the ages of 9 and 26.
c) From the table above, in 2007-2008 what percent of US women aged 19-26 said they had
received a dose of the HPV vaccine? Is this a parameter or a statistic?
Solution
10.5%, this is a statistic since calculated from the sample.
d) Looking at the table, what category for what variable appears to have the lowest level of the
HPV vaccine initiation?
Solution
No category for Ever had sex variable
e) Calculate an odds ratio using the prevalence in the table to compare vaccine initiation of
HPV between white non-Hispanic women compared to black non-Hispanic women?
Interpret the odds ratio.
Solution
HPV for white non-Hispanic women= 12.5 %
No HPV for white non-Hispanic women= 100%-12.5%=87.5%
HPV for black non-Hispanic women=6.5%
No HPV for black non-Hispanic women=100%-6.5%=93.5%
12.5
87.5
Odds white non−Hispanic women compared ¿ black non−Hispanic women= =2.055
6.5
93.5

There are 2.055 times higher to accept HPV by white non-hispanic women than black non-hispanic
women

f) Calculate an odds ratio using the prevalence in the table to compare vaccine initiation of
HPV between women with private health insurance compared to women with no health
insurance? Interpret the odds ratio.
Solution
NPV for women with private health insurance=16.5%
No NPV for women with private health insurance=100%-16.5%=83.5%
NPV for women with no health insurance=4.5%
No NPV for women with no health insurance=100%-4.5%=95.5%
16.5
83.5
Odds for private healthinsurance compared ¿ no health insurance= =4.11
4.5
95.5

There are 4.11 times higher to accept NPV by women with private health insurance than women with no
health insurance.

3. Our goal is to estimate the association between 2 quantitative variables. Six scatterplots of two
quantitative variables are shown below.

Choose the letter that most accurately can measure the association shown in each scatterplot below
and write it underneath each scatterplot, based on the following choices:

a) Pearson Correlation
b) Spearman Correlation
c) Neither Pearson nor Spearman correlation
Solution

Left to right

1st figure- a) Pearson Correlation

2nd figure - c) Neither Pearson nor Spearman correlation

3rd figure- a) Pearson Correlation

4th figure - c) Neither Pearson nor Spearman correlation

5th figure- a) Pearson Correlation

6th figure- b) Spearman Correlation


4. Consider the following graphs A, B and C and answer the questions below:

a) Which graph(s) has a negative Pearson correlation? C


b) Compare graphs A and B. Which graph will have the smaller Pearson correlation? B
c) Do any of the graphs have a close to zero correlation (if yes, which one)? No

5. Use following data on 5 subjects

Person Y X

1 20 10

2 30 30

3 40 30

4 50 30

5 60 10

a) Draw a scatterplot of Y verses X.

Solution
Relationship between x and y
70
60
50
40
y

30
20
10
0
5 10 15 20 25 30 35
x

b) Find the Pearson Correlation between Y and X

Solution

n=5

pearson Y X y^2 x^2 xy


1 20 10 400 100 200
2 30 30 900 900 900
3 40 30 1600 900 1200
4 50 30 2500 900 1500
5 60 10 3600 100 600
total 200 110 9000 2900 4400
n ∑ xy −∑ x ∑ y
r p=
√¿ ¿ ¿
5∗4400−110∗200 22000−22000 0
r p= = = =0. 0
√ [ 5∗2900−110 ]∗[ 5∗9000−200 ]
2 2
√ 2400∗5000 3464.102
c) What important point about the Pearson correlation does this exercise illustrate?

Solution

Correlation measures the strength of the relationship. There is no relationship between x and y

5. A college newspaper interviews a psychologist about student ratings of the teaching faculty
members. The psychologist says, “The evidence indicates that the correlation between research
productivity and teaching rating of faculty members is close to zero.” The paper reported the
following the next day: “Dr. X said that good researchers tend to be poor teachers and vice versa.”
Explain why the paper’s synopsis of the evidence is wrong and what Dr. X meant. Use plain
language and avoid the use of the word correlation.
Solution
The paper’s synopsis of the evidence is wrong because between research productivity and teaching
rating are unrelated because correlation is close to zero. The means that faculty members with high
research productivity does not imply that they have lower or higher teaching ratings. Research
productivity and teaching rating of faculty members are not related or do not influence the other.

7. What is wrong with the following statements?

a) There is a high Pearson correlation between the age of American workers and their
occupation.-occupation is a categorical variable
b) We found a high correlation (r=1.19) between students’ ratings of faculty teaching and
ratings made by other faculty members.-correlation coefficient is greater than 1.0
c) The correlation between gender identity of a group of students and the color of their cell
phone was r=0.23.-the two variables are not quantitative (they are categorical/qualitative)

You might also like