Lab 11

lab #8 statistical associations/Digital Mapping Lab #8.docx

Lab #8

Learning Outcomes for Lab:

1. Interpret summary tables

2. Spatially analyze data sets

3. Map quantitative variables

4. Analyze and interpret spatial patterns

5. Compare density

6. Create choropleth map

7. Explore choropleth classification

8. Sort/Organize spatial data for analysis

Learning Outcomes for Quantitative Literacy:

· Understand quantitative models that describe real world phenomena and recognize limitations of those models;

· Perform simple mathematical computations associated with a quantitative model and make conclusions based on the results;

· Recognize, use, and appreciate mathematical thinking for solving problems that are part of everyday life;

· Understand the various sources of uncertainty and error in empirical data;

· Retrieve, organize, and analyze data associated with a quantitative model; and

· Communicate logical arguments and their conclusions.

Part I – Understand and Analyze Summary Tables

The data below are summarized from the Excel spreadsheet we have been using for the previous 2 labs.

1. What do you think the count variable stands for? (a few words)

2. Describe the sum column? What do the numbers represent? (a few words)

3. Why might you want to look at a summary table instead of the whole data set? (2 sentences)

4. Answer the following the questions about the data table below

a) Tell a two sentence story about Phillies and Miles based on this summary table.

b) Which data category (1,2,3,4, or 5) had the highest Sum_Miles? (one number)

c) Why do you think that this category (1,2 ,3 ,4, or 5) had the highest Sum_Miles? (one sentence)

d) What is the total number of miles distance from where folks grew up that answered the Phillies question 2?

5. Answer the following questions about the data table below?

a) Is there any relationship between wealth (how important is to you that you are wealthy) and the average amount of time it takes our study participants to get to work? In other words as wealth increases does commute decrease? Do people who really care about wealth have shorter commutes? Is there something else going on? (2 sentences)

b) Which category (1,2,3,4,5,6) has the highest count? (one word)

c) Uh-oh – there is no 6 category for wealth in the dictionary (see the Excel file). What do you think happened? (one sentence)

d) What is the average number of minutes it takes for folks to get to work who answered the wealth question with a 3? (one number)

Part II - Create Your Own Summary Tables in Excel

1. Find two variables (from our Excel database) and copy and paste them next to each other in Excel.

I am using Growup and Minutes. Please choose a different pairing of variables for your example!

One should be a nominal or ordinal type – the other should be a ratio – you did something similar in the previous lab when you were calculating average and standard deviation

Here is the example I will use.

2. Please sort the data by the variable on the left (the categorical/ordinal) data type

3. Highlight both columns and choose the Data Ribbon/Menu at the top of Excel.

Choose ‘subtotal’

The categorical/ordinal type should be in ‘at each change in’

Use the function average and make sure ‘replace current subtotals’ and ‘summary data below’ are checked.

You should get something like this (below) where at each break between Growup values Excel takes (in this case) the average number of minutes it takes folks to get to work from each category of Growup.

Deliverable - Create a table that has each category (in this case 1,2,3) and the average number of (minutes in this case). (Submit a table like the one below)

It should look like this:

Deliverable - Finally – create a chart or graph that shows your averages for each category. Think about the histogram we created a couple of labs ago. (Submit a chart, graph, or histogram)

Part III - Visualize Points in Map Shaper

Navigate here

Map shaper visualizes and converts geographic data.

Extract the contents of the density zip file. There should be 7 files with extentions like .dbf, .prj, etc. These are all files needed to create a map. Please double check to make sure that you have extracted these files – just opening them in the zipped folder isn’t correct. Once these files are extracted into a separate folder – drag and drop them into mapshaper.

Here is a link to help with unzipping if you need it.

Then press import

Deliverable: Take a screenshot of your map and paste it here. (Screenshot of the map you created by importing files into mapshaper)

Answer this question – What part of Philadelphia are the interviews (each dot represents an interview for the survey) located in? Use directions like north/south/east/west or locations like center/periphery. (one sentence)

Answer this question – How does mapshaper know where to plot the points? What information is in the files you are feeding it that gives it location information?

Part IV – Experimenting with Density and Creating a Choropleth Map

Density is a great measure that allows us to take into account the size of a geographic area like a neighborhoods, states, or countries when we analyze our data.

We might expect that a large neighborhood has many interviews and small neighborhood would only a few or that a large country has a lot of people while a small country has less people.

In the case of population this isn’t always true. Metropolitan Tokyo actually has more people than all of Canada!

Geographic data isn’t distributed evenly so sometimes data points in small geographic areas are very heavily concentrated and data points in large areas are lightly concentrated. See below.

Image credit: Science World

In the following exercise I want you create what’s called a choropleth map based on the density of points/interviews in each neighborhood. You will use the map you created in part III to assess which areas are least and most dense in regards to the survey data we collected.

Using the blank map below use different shades of the same color (pick light for less, medium for medium, and dark for more to show what areas on the map are most dense.

Most dense = darkest shade of the color…

If you go the draw tab and choose the highlighter. If you click on the drop down menu you can change the color. Click on more colors.

(To get rid of the highlighter and go back to the word cursor click on draw to left side of the ‘draw’ ribbon/menu

Deliverable (using blank map below) - Based on the map you made in map shaper choose a light color for areas in which there aren’t many interviews, a medium color for areas that have some neighborhoods, and a dark color for areas that are highly concentrated. These three colors should fill the map AND THEY SHOULD BE IN THE SAME COLOR SCHEME (blue or green or red etc). You can be pretty broad/general with your assessment, you don’t have to keep switch back and forth between colors. This isn’t correct – but it might look something like this:

Please make sure you take a snip before you do anything else on your word document as your map may change position without the colors attached to it!!!

Finally – once you have taken a snip of your color shaded map

Visit this website

Last Question: What are the neighborhoods in Philly that have the highest density of interviews? (this doesn’t have to be exact – answer these questions based on the map you created) (List 2 -3 neighborhoods)