- In This Project You Will Be Using Your Results From Project 2 And A Second Sample To Test Whether The Distribution Gi 1 (69.79 KiB) Viewed 285 times
In this project, you will be using your results from Project 2, and a second sample, to test whether the distribution gi
-
- Site Admin
- Posts: 899603
- Joined: Mon Aug 02, 2021 8:13 am
In this project, you will be using your results from Project 2, and a second sample, to test whether the distribution gi
In this project, you will be using your results from Project 2, and a second sample, to test whether the distribution given by Benford's Law fits a population. Preparation: 1) Summarize your results from Project 2 regarding the First Digit variable. Example: "I took a sample of... and found that..." In particular, be sure to include the table of your results. Remember that one of the requirements of using a chi-squared test is that the expected value for each cell is at least 5. But that presents a problem for us, since our sample size was only 20: X (First Digit) 1 2 5 6 8 7 0.058 0.051 9 0.046 0.301 3 4 0.176 0.125 0.097 0.079 3.52 2.5 1.94 1.58 Probability Expected 0.067 6.02 1.34 1.16 1.02 0.92 To fix this, we can combine some of the columns together: X (First Digit) 1 2-3 4-9 Probability 0.301 0.301 0.398 6 8 Expected (rounded) 6 Do the same for your data from Project 2. Part 1: 2) Do you have evidence at the 5% level of significance that Benford's Law does not apply to the populations of countries? Include an in-context conclusion about the result of your test. Part 2: Now let's try again: this time, I used flightaware.com to take a random sample of the first digits of the flight numbers of 20 flights in the air at the time: 1 2 3 4 5 6 7 8 9 X (First Digit) Frequency X (First Digit) 2 3 1 3 3 1 2 4 1 1 2-3 4-9 Frequency 2 4 14 3) Do you have evidence at the 5% level of significance that Benford's Law does not apply to flight numbers? Include an in-context conclusion about the result of your test. Concluding note: One of the requirements of Benford's Law is that the numbers need to span several orders of magnitude, but flight numbers for a single airlines are usually all three-or all four-digit numbers.