Modeling Prompt

The data provided is a spreadsheet of 200 schools, not all of which have information about each of the ten categories. Students will need access to technology such as Geogebra or Excel that will allow them to analyze and display the relationships that interest them.

If desired, the complete data set can be downloaded from the Department of Education website, and data can be curated in a different way. Different categories can be chosen based on the interests of the class, or the range of colleges can be narrowed down to those in the state. Historical data is also available. A few local colleges could be compared across time. If a customized data set is created, modify the launch accordingly.

1. Here are the characteristics of colleges that we have data about:

   Choose two that might be related. What do you predict the relationship between them is?

   • college ownership: public, private non-profit, or private for-profit
   • average SAT score of admitted students
   • acceptance rate
   • number of enrolled students
   • average cost of attendance per year
   • median earnings 10 years after graduation
   • highest degree awarded
   • completion rate
   • percent of students who self-identify as white
   • median ACT score of admitted students
2. Take a look at the data in the spreadsheet. Before calculating, do some estimation based on the data. Can you tell if your hypothesis seems reasonable?
3. Use an appropriate display to summarize the data.
4. Analyze your graph or table and do any other calculations needed. Do your results confirm your prediction? If not, what do you think explains the results?
5. Here are some headlines. What analysis do you think the article describes? Give some examples of evidence that would support the headline or contradict it.
   1. “Students from Selective Colleges Tend to Succeed”
   2. “Isolation at Larger Colleges Leads to Higher Dropout Rates”
   3. “High Cost of Private Colleges Not Outweighed by Lifetime Earnings”
       

The data provided is a spreadsheet of 100 schools for which there is information about each of the six categories. Students will need access to technology such as Geogebra or Excel that will allow them to analyze and display the relationships that interest them.

If desired, the complete data set can be downloaded from the Department of Education website, and data can be curated in a different way. Different categories can be chosen based on the interests of the class, or the range of colleges can be narrowed down to those in the state. Historical data is also available. A few local colleges could be compared across time. If a customized data set is created, modify the launch accordingly.

1. Here are the characteristics of colleges that we have data about:

   Choose two that might be related. What do you predict the relationship between them is?

   • college ownership: public, private non-profit, or private for-profit
   • average SAT score of admitted students
   • acceptance rate
   • number of enrolled students
   • average cost of attendance per year
   • median earnings 10 years after graduation
2. Take a look at the data in the spreadsheet. Before calculating, do some estimation based on the data. Can you tell if your hypothesis seems reasonable?
3. Use an appropriate display to summarize the data.
4. Analyze your graph or table and do any other calculations needed. Do your results confirm your prediction? If not, what do you think explains the results?
5. Here are some headlines. What analysis do you think the article describes? Give some examples of evidence that would support the headline or contradict it.
   1. “Students from Selective Colleges Tend to Succeed”
   2. “Isolation at Larger Colleges Leads to Higher Dropout Rates”
   3. “High Cost of Private Colleges Not Outweighed by Lifetime Earnings”