|While the Internet provides huge databases which could fill spreadsheets as large as a football field, it does little good to visit such sites unless they know how to "crunch" the numbers, interpreting the huge databases with mathematical tools and reasoning. We must teach them to recognize the questions that reside within such data collections along with the skills to pose and explore such questions.
It is not enough to gather information or interpret the information. We would hope that our students will learn to take their thinking about numbers to the prediction level. Based on their understanding of correlations, older students should be able to test the strength of relationships between variables.
At the same time, we can see problems with statistical inference - as when we can prove an association without proving a cause.
You and your partner have been hired by the governor of a Southern state in the USA to study ways to bring down the infant mortality rate. You will analyze data from the Annie B. Casey Foundation to decide what strategies are most likely to make a difference.
Open Inspiration and create a diagram with Infant Mortality in the center and at least 10 variables such as nutrition that might have an influence on the infant mortality rate.
To help you decide which of these matter the most, you and your partner will look through the state rankings and profiles at KIDS COUNT. KIDS COUNT is a project of the Annie E. Casey Foundation.
This site will not supply all the data you would need to do a complete analysis but it will give you a start.
You will have 10 minutes to look at the following data:
Option One: For those with strong spreadsheet and statistical skills, you may use this table to calculate 2-3 correlations. Modify your Inspiration diagram to include new variables you did not think of at first. Enter your calculations as notes in the diagram.
Click here to download these data in Excel format. You must then save it and re-open in Excel.
Option Two: If statistics are not part of your skill set, try scanning the data above to see if you can "eyeball" relationships. Which factors seem most closely associated with lower infant mortality rates? Which would be worthy of investment?
In your word processor, summarize for the governor your predictions of where new funding might do the most good.
© 2000, Jamie McKenzie,
Conditions of use: These materials may be used without charge by individual visitors to this Web site, but schools, districts, consultants, companies and any other organizations may not use these pages for group sessions without explicit permission and without purchase of rights. Click here for information on site licenses and costs. In no case may they be downloaded or whacked to be served or distributed or used on any network without explicit written permission from the author and copyright holder.