Technology Exercise 5: Learning About the World

Internet Delays (Question 1)

There is a command called traceroute that will show the route that information travels to get from one host on the Internet to another. It also shows the round trip time (in milliseconds) that it takes to get to each point along the route.

We're going to collect information about how quickly sites on the Internet in the United States can reach Richland's web server. Of more use to students might be the reverse condition, how quickly can we reach sites on the Internet. So, even though we're collecting the time it takes from the remote server to Richland, we'll use it as the time from Richland to them.

We will collect information from 12 different sites on the Internet. We will do this 5 times at different times. It's possible that the connections may be really good during one time of the day but slow at another time. That's why we're repeating the process and collecting at 5 different times. Those different times could be at different times of the day or on different days. I just ask that there be at least three hours between samplings.

You could collect information from home if you have an Internet connection and then bring it into school in an Excel file. Since the traceroute is from the remote website to Richland and not to your home machine, it won't be affected by dial-up or broadband connections.

Creating the Worksheet

  1. Label column 1 as "sample"
  2. Label column 2 as "site"
  3. Label column 3 as "time"
  4. Choose Calc / Make Patterned Data / Simple Set of Numbers
    1. Store the pattern in sample
    2. Start with the first value of 1
    3. End with the last value of 5
    4. List each value 36 times
    5. Click OK
  5. Choose Calc / Make Patterned Data / Simple Set of Numbers (** see note below if you want to be fancy!)
    1. Store the pattern in site
    2. Start with the first value of 1
    3. End with the last value of 12
    4. List each value 3 times
    5. Repeat the entire list 5 times
    6. Click OK
  6. Save your file in R:\01\tech5 or R:\02\tech5 depending on which section you're in. Use a filename that's unique to your group.

When you gather the information, there will be three times for each location. These should go in separate rows.

** If you want to be fancy when putting in the sites, you can put in the name of the sites rather than the number of the sites. That will make outputs look nicer and make it easier to see who is different. To do this, replace step 5 above with the following.

  1. Choose Calc / Make Patterned Data / Text Values
  2. Store the pattern in site
  3. Type in the names of the sites from below. You can abbreviate or leave off the top level domain (.com, .edu, .net) if you want to. For example, type in (or copy and paste from here) "above io his sdsc calweb getnet princeton opus1 socket wvi xmission playground".
  4. List each value 3 times
  5. Repeat the entire list 5 times
  6. Click OK

Gathering the Data

Click on each of the links that follow below. If the traceroute is successful, the last line should indicate www.richland.edu (64.107.104.12) and then have three times in milliseconds (ms or msec). Some sites will include an AS number, you can ignore that.

Sample Traceroute Output

traceroute to 64.107.104.12 (64.107.104.12), 30 hops max, 40 byte packets
1 inside.fw1.sjc2.mfnx.net (208.184.213.129) 0.317 ms 0.382 ms 0.214 ms
2 99.ge-5-1-1.er10b.sjc2.us.above.net (64.124.216.11) 0.519 ms 0.533 ms 0.500 ms
3 so-1-0-0.mpr3.sjc2.us.above.net (64.125.30.97) 0.512 ms 0.661 ms 0.512 ms
4 so-5-2-0.cr1.ord2.us.above.net (64.125.30.218) 51.915 ms 51.912 ms 51.889 ms
5 pos4-0.mpr1.ord1.us.above.net (208.185.0.198) 52.085 ms 52.030 ms 52.007 ms
6 206.220.243.106 (206.220.243.106) 52.265 ms 52.328 ms 52.359 ms
7 atm2-0-sub02-soib1-peoria-core.peoria.lincon.net (206.166.9.170) 58.997 ms 59.032 ms 58.847 ms
8 atm1-0-sub11-peoria-core-fat-elvis.springfield.lincon.net (206.166.9.186) 59.306 ms 59.209 ms 59.147 ms
9 206.166.9.254 (206.166.9.254) 59.716 ms 59.830 ms 59.056 ms
10 Richland.springfield.lincon.net (65.79.126.2) 60.999 ms 62.476 ms 60.782 ms
11 www.richland.edu (64.107.104.12) 61.648 ms 60.954 ms 61.564 ms

Enter the site number (these are already entered if you followed the steps above) and the three times into Minitab. Each site will take up three rows of data. Do not enter the units on the times. The data above would look like this.

row sample site time
1 1 1 61.648
2 1 1 60.954
3 1 1 61.564

Click on each link below to start the traceroute. Be patient, some of the traceroutes can take a while. Also, there will be a noticeable delay on most of the sites right before you reach Richland. In some cases, you may see a "* * *". This is okay, just be patient. After you have entered the information into Minitab, hit the back button on your browser to come back to here and visit the next site.

Collect all 12 sites' information at the same sitting so that we are comparing similar conditions. Wait at least three hours before repeating the process. If you want to do this at home, create an Excel spreadsheet with the two columns, then bring it into school and copy and paste the information into Minitab.

  1. above.net
  2. io.com
  3. his.com
  4. sdsc.edu
  5. calweb.com
  6. getnet.com
  7. princeton.edu
  8. opus1.com
  9. socket.net
  10. wvi.com (rounds to nearest ms)
  11. xmission.com (generates entire page before displaying, be patient; rounds to nearest ms)
  12. playground.net

Conducting the Hypothesis Test

Check the assumptions. If they aren't satisified, address any concerns that may have in interpreting the data.

To make a probability plot, do the following. You can also use the graphical summary under Descriptive Statistics to get a histogram.

  1. Choose Stat / Basic Statistics / Normality Test
  2. Select the time variable
  3. Click OK

To conduct the hypothesis test

  1. Choose Stat / Basic Statistics / 1 Sample t
  2. Select the variable time
  3. Enter the claimed value for the mean in the test mean box.
  4. Go into options and make sure the values are set properly.
  5. Click OK

Designated Hitter (Question 2)

Entering the Data

  1. Choose File / New / Minitab Worksheet to create a new worksheet.
  2. Label three columns as Team, League, and OBP.
  3. Go to the Major League Baseball site at http://mlb.mlb.com/NASApp/mlb/mlb/stats_historical/mlb_historical_team_stats.jsp and enter these options on the left side in the "Historical Team Stats" section. Note these values may be the default choices and the results you want already displayed, but double check to make sure. You should do step f to sort the data even if you don't have to do the rest (having the teams in alphabetical order will help with question 3).
    1. Choose Major League
    2. Choose Hitting Stats
    3. Split by the Entire Season
    4. The Timeframe is 2003 Season
    5. Click GO.
    6. Click on Team at the top of the data.
  4. For each team, enter the name of the team, the League (AL or NL), and the On Base Percentage (OBP)

Performing the Hypothesis Test

  1. Choose Stat / Basic Statistics / 2 Sample t
  2. The samples are in one column
  3. The samples are in the OBP column
  4. The subscripts are in the League column
  5. Check the Options and make sure they're set properly
  6. Click OK

Better Pitcher at Home? (Question 3)

Entering the Data

Just use the same worksheet you started for question 2.

  1. Label two more columns, one as Home and one as Away.
  2. Go to the Major League Baseball site at http://mlb.mlb.com/NASApp/mlb/mlb/stats_historical/mlb_historical_team_stats.jsp and enter these options on the left side in the "Historical Team Stats" section.
    1. Choose Major League
    2. Choose Pitching Stats
    3. Split by Home
    4. The Timeframe is 2003 Season
    5. Click GO. Note that this is extremely slow. You may need to wait 2-3 minutes for the results to come up. Be patient.
    6. Click on Team at the top of the data.
  3. Enter the Earned Run Average (ERA) values into the Home column for the proper teams. If you clicked on Team like step f says, then the order of the teams should be the same as what you previously entered. If not, be careful to match up the ERA values with the proper team.
  4. Repeat step 2, except this time, Split by Away.
  5. Enter the ERA values for each team in the Away column.

Performing the Hypothesis Test

  1. Choose Stat / Basic Statistics / Paired t
  2. The first sample is in Home
  3. The second sample is in Away
  4. Go into options and make sure they are set properly. Note that Minitab compares sample 1 to sample 2, so make sure your alternative is set up properly with Home on the left and Away on the right.
  5. Click OK

Show the T Approaches the Normal (Question 4)

Start this in a new worksheet. You do not need to supply a context with this problem, but you should describe what we're trying to show here.

Creating the Graph

  1. Label the first column as x, the second as z, the third as t1, and the fourth as t5.
  2. Go to Calc / Make Patterned Data / Simple Set of Numbers. Start at -3 and go to 3 with a step size of 0.01. Store the results into x
  3. Go to Calc / Probability Distributions / Normal. Select Probability Density, set the input column to x and the optional storage to z.
  4. Go to Calc / Probability Distributions / t. Select Probability Density, set the degrees of freedom to 1, the input column to x and the optional storage to t1
  5. Go to Calc / Probability Distributions / t. Select Probability Density, set the degrees of freedom to 5, the input column to x and the optional storage to t5
  6. Go to Graph / ScatterPlot / With Connect Line
    1. Create three plots. For the y-variables, use z, t1, and t5. Use x for each of the x variables.
    2. Go into Labels and add a title
    3. Go into Data View and uncheck the symbols box in the data display window.
    4. Go into Multiple Graphs and check the Overlaid on same graph radio button.
    5. Click OK

Cleaning up the Graph

  1. Click on the "Y-Data" label on the vertical axis and delete it.
  2. Click on the "X" label on the horizontal axis and delete it.
  3. Double click on the curves
    1. Change the lines to be custom
    2. Change the size to be 2 so they show up better
    3. Click OK

Copy the graph into Word and add an explanation of what we're looking at.

Here's an example of what the graph might look like.

Graph showing how student's t curves approach normal curve as degrees of freedom increase