Introductory Project: Temperatures

Your group is tasked with collecting temperatures at random from the Decatur, IL, airport and seeing if there is a difference in the temperatures between the different years. Information is available for January 2003 or later. This means that you will not be able to gather information for Winter 2003 or Summer or Fall 2006. Consider December as part of the Winter for the next calendar year. That is, December 2004 is part of Winter 2005.

See the section at the end for Saving Your Work. You may do that at some point along this process. You probably should not wait until the end to save it.

Random Selection of Dates

Use Minitab to randomly select dates from the time periods you have been given. Select 10 dates from each year.

This example assumes you are working with the Spring 2003 season. Adjust the values for your season and year.

  1. Label the first column in Minitab as dates. Then label three columns for each year of data you are gathering. Labels the sets of columns as date##, temp##, and norm## where the ## is the year (03, 04, 05, 06). If you are doing Spring, then you will have thirteen columns labeled when you are done: dates, date03, temp03, norm03, date04, temp04, norm04, date05, temp05, norm05, date06, temp06, and norm06.
  2. Choose Calc / Make Patterned Data / Simple Set of Date/Time Values. Store the patterned data in dates, set the first date to 03/01/2003 and the last date to 05/31/2003. Click OK.
  3. Choose Calc / Random Data / Sample From Columns. You want to sample 10 rows from the dates column and store the samples in the date03 column. Do not sample with replacement. Click OK.
  4. Repeat steps 2 and 3 for the other years. Change the dates for step 2 and the location where you store the samples for step 3.

Gather the Data

Recent Temperatures (2003 - 2006)

Your information will come from the National Weather Service Forecast Office for Central Illinois at

  1. The Product is the Preliminary Climatology Data (CF6)
  2. The Location is the Decatur Airport
  3. The Timeframe is Archived Data although the month will vary.
  4. When you click GO, the data will open in a new window and you will collect the information from the appropriate column (2 for MIN, 3 for MAX, and 4 for AVG).

For each date that you have randomly selected, gather the temperature and record it next to that date in the table. For example, if the third entry in the date06 column is 04/15/2006, then the third entry in the temp06 column should be 71, which was the temperature on April 15, 2006. This process will go quicker if you have the team look up information and give it to the person using Minitab to enter.

Normal Temperatures (1971 - 2000)

The thirty year normal daily temperatures for Decatur can be found at

Scroll down to the appropriate table and record the normal daily temperature in the norm## columns next to the date## and temp## values for the same date. The data from this table goes into all of the norm## columns in Minitab; the ## is to help match it with the temp columns, which are year specific.

Manipulate the Data

Stacking the Data

Some of the ways that we will look at the data require all of the information to be in one column, instead of three (or four) different columns.

  1. Label three blank columns in Minitab as date, temp, and norm.
  2. Choose Data / Stack / Blocks of Columns. There are a series of rows where you can enter variable names.
    1. Click in the first row and then highlight date03, temp03, and norm03 from the variable list on the left. Click on Select to transfer the list to the right.
    2. Click in the second row and then highlight date04, temp04, and norm04 from the variable list. Click on Select to transfer the list to the right.
    3. Fill in the rest of the table by repeating steps 2a or 2b for any remaining years.
    4. Put the information in a column of the current worksheet. The columns should be date, temp, and norm.
    5. Do not use variable names in a subscript column.

Computing the Difference Between Observed and Expected Temperatures

  1. Label a column as diff.
  2. Choose Calc / Calculator. Store the results into diff. The expression is temp - norm.

Changing the Difference into Categories

Sometimes it is useful to display information based on categorical variables instead of numerical ones. In this case, we're going to determine whether each day was colder than normal, normal, or warmer than normal. As a group, decide how many degrees away from the normal expected temperature before you would consider it cold or warm. In my example, I'll use 5 degrees, but use your numbers in the problem. In Minitab, a range of values can be expressed as min:max, so 5:16 would be the same as saying 5 ≤ x ≤ 16.

  1. Label a column as feels.
  2. Choose Data / Code / Numeric to Text.
    1. Code the values from the diff column into the feels column.
    2. Code the original values -100:-5 to Cold. Again, that -5 should be the value you decide upon as being too far away from normal. The -100 could actually be the most negative difference you have, but I'm pretty sure -100 will catch all of them.
    3. In the second row, change the original values -4:4 into Normal.
    4. In the third row, change the original values 5:100 into Warm.

Saving your Work

Wow, that's the end of the first part, but you'll want to make sure that you save it.

There are two ways to save a file in Minitab: a project or a worksheet.

A worksheet contains only the data. You can have multiple worksheets in a project, but none of the output or graphs are saved. That's okay for right now since you don't have any of those anyway. When worksheets are opened, a copy is saved within your project and the original is left alone. When I give you data as a class, it will be in the form of a worksheet from which you will create a project. Use File / Save Current Worksheet to save it. Use File / Open Worksheet to open a worksheet.

The second way to save a file is as a project. A project may contain multiple worksheets as well as statistical output and graphs. The graphs will be saved as long as you don't close them before saving the project. You can minimize the graphs and keep them in the project, but when you close them, they are gone. Unless what you're working on is so trivial that it doesn't need saved or you only have data, you will probably want to save your files as projects. Use File / Save Project to save your project. Use File / Open Project to open a project.

When you are logged into the network at Richland, there are two places where you can save your files. You may save them under My Documents. The My Documents folder is actually on the networked drive U:, which is the other place you can save files. Drive U: is networked, meaning that you can have access to it from anywhere on campus. We do not have remote access configured at this point.

Sharing Your Data with Your Group

The My Documents folder or the U: are yours. Other students in your group do not have access to them. That means that it isn't really easy to share files with other students. One way to do this is to email copies from your account to theirs. Another way is to bring in a USB Jump Drive or a floppy disk and transfer files between computers manually.

Written Paper

Go into Microsoft Word and open the file P:/james/math113/introproject.doc. Save this file under My Documents

Fill in the information asked for in the document. Most of the material will be explained in class.

When you are done with the paper, email it to the instructor at