Quiz 3 over chapter 12.

The same study among rural Bangladeshis also looked at the association between arsenic in the well and obesity. The SPSS results are below. Which p-value should the investigators report?

Because two cells have expected cell counts less than 5 (see message at the bottom on the chi-square tests table), we should use the Fisher's exact test instead of the chi-square test p-value (option A). Options C and D are not p-values but instead at test statistics.

A study was conducted to look at the association between exercise and obesity. The 2 × 2 table is presented below. What is the probability of being obese among those who exercise frequently (ie, the probability of obesity conditional on exercising frequently)?

Conditional probability is calculated by taking the number of people with the outcome of interest (obesity) and dividing that by the total number with the condition of interest (frequent exercisers). Therefore, the probability of being obese conditional on exercising frequently is 8/50 = 0.16.

A cross-sectional study among rural residents of one town in Bangladesh looked at the association between the level of arsenic in the well water of each household and the presence of high blood pressure in the head of the household. The SPSS results are below. What is the best interpretation of these results?

CorrectC. Those with arsenic in their well are significantly more likely to have high blood pressure compared to those without arsenic in their well.

Under which of the following conditions would you need to use the Fisher's exact test instead of the chi-square test?

The Fisher's exact test is used when one or more expected cell counts in the cross-tabulation are less than 5. When the groups are not independent (option C), McNemar's test is used.

What types of variables are needed to perform a chi-square test?

The chi-square test is used to look at the association between two categorical variables. The variables can be dichotomous or polytomous. For option A, one could use Pearson correlation; for option C, one could use one-way ANOVA; and for option D, one could use ANCOVA.

A cross-sectional study among older adults (>65 years old) looked at the association of exposure to second hand smoke in the home with chronic obstructive pulmonary disease (COPD) and found an odds ratio of 1.9. How would you interpret these results?

The odds of having COPD among those exposed to second hand smoke is 1.9 times higher than that of those NOT exposed to second hand smoke.

A study was conducted among 50 employees of a grocery store looking at the probability of bladder cancer among those who consume artificial sweeteners compared to those who do not. The study found that the odds ratio was 5.9, and the p-value for the chi-square test was 0.98. What would you conclude?Those who drink artificial sweeteners have higher odds of getting bladder cancer compared to those who do not, but the difference is not statistically significant.

The odds ratio (5.9) is greater than 1, and therefore those who drink artificial sweeteners had higher odds of getting bladder cancer. However, the p-value (0.98) is greater than 0.05, and therefore this difference in probability is not statistically significant.

9 Which of the following are assumptions required for the chi-square test to be valid?

There are three assumptions that must be met in order to use the chi-square test—(1) the data are frequency data, (2) there is an adequate sample size, and (3) independence.

A case-control study was conducted looking at the association between pet ownership and adult-onset asthma. Because environment during childhood is thought to have an impact on the risk for allergy and asthma, the study used siblings without asthma as matched controls and compared the probability of current pet ownership in the newly diagnosed asthmatic sibling to that of the nonasthmatic sibling. Which of the following tests should be used to calculate the p-value for this association?

Siblings share a number of characteristics including genes and childhood environmental factors and, therefore, are not independent. The correct statistic to use for nonindependent samples such as a matched case-control study is McNemar's test.

The following 2 × 2 table presents data from a longitudinal study looking at the association between high level of stress and the condition known as dry eye. From this data, what is the odds ratio

OR = 25 × 95/75 × 5 = 6.33