Popular in Course
Popular in Geography
verified elite notetaker
This 4 page Class Notes was uploaded by Cortney Leuschke on Saturday September 12, 2015. The Class Notes belongs to GEOG 494M at West Virginia University taught by Staff in Fall. Since its upload, it has received 22 views. For similar materials see /class/202691/geog-494m-west-virginia-university in Geography at West Virginia University.
Reviews for GEOG 494M
Report this Material
What is Karma?
Karma is the currency of StudySoup.
You can buy or earn more Karma at anytime and redeem it for class notes, study guides, flashcards, and more!
Date Created: 09/12/15
Geog 494m Geodemographics Spring 2009 Example 7 Regression Obiectives 1 To analyze and describe the relationship between two variables 2 To understand the components of the regression equation Data WVA ttributes spreadsheet WVElectionshp shapefile Step 1 Examine the relationship between Obama percentages and McCain percentages I Open the WViA ttributes spreadsheet in Excel I Highlight the column where you calculated the percentage of total votes that Obama received Hold down the control key and also highlight the column where you calculated the percentage of total votes that McCain received I Leaving these two columns highlighted go to insert 7 chart7 scatter Use the first option for how to display the scatterplot I Now you ll see some new tabs along the top Go to design 7 chart layout and choose layout 9 with the equation included I Now you have the plot trendline and regression equation included on your chart Examine the regression equation What does each element in the equation mean in terms of the relationship between the two variables How would you describe this relationship How would you describe the correlation Is this relationship positive or negative Step 2 Use regression to describe the relationship between per capita and rent I We want to know to what degree per capita income explains the average rent price in each county I Highlight the rent column and the pcincome column I Create the scatterplot in the same way you have been doing Describe the relationship between the two variables From this analysis can you say that high pcincome causes an increase in rent How would describe the correlation between these two variables Is this relationship positive or negative Step 3 Go back and examine some ofthe relationships that we have looked for during previous exercises Use regression to gain a more detailed understanding the relationship between some of these relationships I What information can you get from a regression analysis that you cannot get from a chisquare analysis Step 4 Open the WVElectionshp le in GeoDa I Use the explore 7 scatterplot property to look at some of the variables you were exploring above in Excel I In GeoDa you can highlight individual points You can also rightclick in the plot and choose to exclude the highlighted points Try this and see what happens to the plot particularly the trend line Geog 494m Geodemographics Step 5 Spring 2009 In the visualization exercise we looked brie y at the meaning of the slope value in GeoDa Now that we have covered regression how would you describe this value Mapping residuals Open Arch and load your WViEZECl l39OI thp We are going to map the residuals for our first regression analysis that examined rent and per capita income Open the attribute table for WViElection and add a new eld Name the field RentResid and classify it as oat Now highlight this new eld rightclick on the top and choose field calculator We are going to use our regression equation to create a column of residuals What exactly do these residuals represent How does each component of the regression equation relate to the residuals Why do you think we want to map them What information might this give us Enter the following formula into the eld calculator rent 7 00099pcincome 1202 Hit ok and watch as ArcMap populates your new eld with the residual values Now that we have a field of residuals we just need to symbolize them in our map Choose to categorize by quantities and select an appropriate color scheme and number of classes Notice that you have some positive and some negative residuals What do the negative and positive values tell you Do you see any patterns in the data What can you learn from mapping the residuals Geog 494m Geodemographics Spring 2009 Example 12 KMeans Clustering Obiectives 1 Create clusters of counties in West Virginia based on five variables 2 Analyze statistics for each cluster in order to characterize its properties 3 Create a choropleth map of the clustered variables 4 Compare the results of the K means clustering with the previous results from hierarchical Data cl ustering WVElectionXls spreadsh eet WVElectionshp shapefile Step 1 Step 2 Step 3 Open JMP and open the WVelectionsxls le Open JMP from the desktop From the opening menu choose to open a new data le You will have to specify that the new data le as an Excel xls le When you open the le you will notice that it is a reduced version of the data that we have been using for most of our other exercises this semester There should be a listing of each county in West Virginia accompanied by ve different variables We will use these ve variables to cluster the counties in West Virginia into distinct groups Create k means clusters Once you are in IMP go to Analysis iMultivariate 7 Cluster To create the cluster you will select all ve variables pcturban pcincome pctcoleal unemploy and homevalue and add them to the Y columns section This basically de nes your cluster analysis on those variables Now make sure that kmeans clustering is chosen and de ne 3 clusters Keep the standardize alata box checked Click ok and examine the results of the analysis They should quite different from the hierarchical clustering results For one thing there is no dendrogram Instead you can choose the red triangle and display a biplot and a 3D biplot What do these visualizations say about the clusters How would you interpret them Now go back to Analyze iMultivariate 7 Cluster Repeat the process except select 6 clusters instead of 3 Again examine the biplot and 3D biplot How do you interpret them now Now repeat the analysis one more time choosing 10 as the number of clusters Examine the differences How does the number effect the arrangement of the clusters Where do you notice the biggest changes We will now use the 6 cluster set in the rest of our analysis Luckily we do not have to calculate averages and standard deviations in Excel because JMP does this for us Examine the values for the 6cluster analysis and characterize these clusters in the same way you did with the hierarchical clustering exercise How are your characterizations the same or different from the ones in hierarchical clustering Go back to the little red triangle on the 6cluster analysis and choose to save clusters Now go to save as and save your table as a new Excel le called WViElection76kmeansxls Map the hierarchical clusters Now open Arch and load WViElectionshp You can also load the xls le you just created WViElectioni6kmeansxls Geog 494m Geodemographics Spring 2009 Join the WViElectionizikmeansxls to the WViEZECl l39OI thp using Name as the eld from both layers Now symbolize the WViEZECl l39OI thp layer by unique categories using the cluster eld This will show you counties that correspond to each cluster Consider your characterizations of each cluster Do these counties t these characterizations Why or why not What can you learn about these particular counties based on this information
Are you sure you want to buy this material for
You're already Subscribed!
Looks like you've already subscribed to StudySoup, you won't need to purchase another subscription to get this material. To access this material simply click 'View Full Document'