calculate confidence intervals

we’ve learned about “margins of error” and “confidence intervals”, which allow us to estimate not just quantities we care about, but also our level of uncertainty about those quantities.

*Open your data in Excel or Statgraphics and answer the following in complete sentences.

1) Explain why this data is a population, rather than a sample. Remember that we can generally describe a population using a phrase like, “this is a list of all of ___________.”

2) Like last week’s exercises, we’ll calculate confidence intervals using random samples of this data. Choose 30 rows at random (it’s fine to use the same random 30 rows you picked last week). Find the standard deviation of depdelay of your 30 selected flights. Is this standard deviation a sample standard deviation (s) or population standard deviation (σ)?

Hint 1: You may find Hint 2 in Week 2 helpful for randomly selecting 30 rows.

Hint 2: In Excel, use =stdev.s() to find the standard deviation.

3) Based on your findings in (2), calculate a confidence interval for the average flight departure delay at 95% confidence level, being sure to show your calculations clearly.

Hint 3: If you’re using Word, Insert > Equation will make your life easier as you show your work.

4) Repeat (3), using the same 30 rows, find a confidence interval for the proportion of flights, which are on Southwest airlines. Again, be sure to show your work clearly.

Hint 4: remember a confidence interval for the proportion uses a different equation than a confidence interval for an average!

Hint 5: To decipher carrier codes (including Southwest airlines and many others), click on this PDF document

Hint 6: If your randomly selected 30 rows do not contain any Southwest airlines flight, that’s okay. What’s your sample proportion (p¯) then?

5) Finally, pick 20 Southwest flights at random. Calculate the average departure delay for Southwest airlines using your sample of 20 Southwest flights.