In this module I learned how to use different data classification method to show how populations of seniors are distributed uses percentages and counts.
The data classifications I used were the following:
1. Equal Interval-this uses the highest and lowest amount of 65 years old's and breaks that up into 5 categories. This showed the left census tract having more 65 year old's than any other method. Equal intervals do not take into account outliers so this can inaccurately display data
2. Quantile-this method breaks the total number of classes into equal categories. This helps eliminate outliers skewing the data. This and the natural break class ended up having very similar looking maps.
3. Standard Deviation- Standard deviation method is using the standard deviation to add or subtract from the mean of the data. Showing how far data deviates from the mean can cause outliers to stand out more. Using standard deviation is best with “normal” datasets. this map has the least variation than any other map, having the largest amount of the map be in the <-1.5 Std. Dev. category.
4. Natural Break-Natural breaks uses the natural grouping over data within the dataset. This method used the Jenks optimal method. This method does consider the outliers when creating the classifications. This map ended up looking very similar to the quantile method.
I believe the population count normalized by area using natural breaks gives the best idea of where most of the seniors are distributed in Miami Dade County. This map shows where the seniors actually are, and not showing misleading data from percentages living in sparse areas. If one area has a small population of people, but most of those people are seniors, the percentage map would show this area as an area of interest, even though across the map, only a very small number of seniors actually live there. The natural break method gives the most accurate representation of the population. This map takes the outliers into consideration.
No comments:
Post a Comment