1. Calculate the probability of death (i.e. COVID-19 infection fatality rate) and conduct t-tests of mean or proportion using a hypothesized value. Supposed the national population COVID-19 infection fatality rate during the same period was 0.042. Was the infection fatality rate in Georgia statistically different from the hypothesized national population rate during this period? Use prop.test() or binom.test() for testing the proportion in R.
Repeat the test using t.test( ), which conducts the test of means. This can be done because proportion of death is equal to the mean of a binary variable death. Do you obtain larger or smaller p-values? Which of the three t-test commands [prop.test( ), binom.test( ), and t.test( )] appear to be the most conservative?
Provide the test results in a table and several paragraphs interpreting the results.
2. Conduct multiple one-sample tests of proportion (prop.test or binom.test) for multiple groups. Note that you may need to create two vectors to isolate a group. For example, conducting a test for a male sample and another test for a female sample requires filter( ) into two vectors, and then run a test for each sample. Are they different from the hypothesized national population rate (0.042)? Make decision based on p-value. Do this in R. Provide the null and the alternative hypothesis for each test. Provide the test results in a table and several paragraphs interpreting the results.
