- Here is a data set that measures population growth rates in the US from 1910-1920
|2.1 ||1.56 ||1.56 ||1.96 ||1.92 ||1.44 ||1.4 ||1.27 ||-0.06 ||1.26 ||1.85
|1910 ||1911 ||1912 ||1913 ||1914 ||1915 ||1916 ||1917 ||1918 ||1919 ||1920
- 1918 would drag the average down from the median
- 1918 would drag the average up from the median
- 1918 would leave the average alone
What happened in 1918?
What can you say about the data from the median to q3 of the reaction times?
- The cell phone users did better because the data is more tightly clustered together
- The control group did better because the data is lower on the graph
- Assume little to no bias and truly a random sample. If a polling company conducted
100 such polls with a 95% confidence interval, then about how many of them are likely to include the true population
- Is there any way to know which intervals contain the true percentage and which ones don't?
- Is there any way to know for sure if it is a representative sample?
- How should we interpret the margin of error if the sample is very biased?
- It is still valid as is
- Garbage in garbage out, so the margin of error would not represent
the entire population, although it would still be useful to interpret whatever biased sample it did
- In which of the following examples will the margin of error be the smallest? Assume each refers
to a random sample that is not biased for a 95% confidence interval.
- Sample A: a sample of n = 1000 from a population of 10 million
- Sample B: a sample of n = 2500 from a population of 200 million
- Sample C: a sample of n = 400 from a population of 50,000
- Write down in your own words what 95% confidence interval means.
- Write down in your own words what margin of error means.