Help Pages » Tutorial: Custom stats
UUorld has a vast collection of statistics that are immediately available to you by clicking Browse Stats from within the main window, but if you want to import your own data, please read on!
First steps
Open the File menu and select Import Stats.
On the window that appears, choose the file you would like to import by clicking Browse for CSV Files. If you are using UUorld Pro, you can select an unlimited number of files to import at one time by pressing the Shift key while selecting your stats.
File Formats
UUorld understands CSV files in a simple format: regions on each row and date/times in columns. CSV stands for comma separated values. Most spreadsheet and data analysis programs support CSV files.
For example, at county level:
Or at country level, using ISO trigrams:
Different data sources use different naming conventions. Above are two examples, but the importer will handle many types of names for your regions and you can import new naming conventions as needed (details below).
Date & Time Formats
UUorld understands dates in a variety of formats. You can import data by year, month, day, hour, minute, or second.
UUorld attempts to match the dates in your file according to the following patterns. UUorld tries each of these patterns in order, and uses the first pattern that matches your data. (Note the 24 hour time format.)
- YYYY - e.g, 2001, 2002.
- MM/DD/YYYY - e.g. 01/01/2001, 02/01/2001.
- MMM-YYYY - e.g. Jan-2001, Feb-2001, Mar-2001.
- MMM-DD-YYYY - e.g. Jan-01-2001, Feb-01-2001.
- MM/DD/YYYY hh:mm:ss - e.g. 01/01/2001 15:30:00
- MMM-DD-YYYY hh:mm:ss - e.g. Jan-01-2001 15:30:00
Matching Names
After selecting your files, you will need to choose the border set which matches the data. We have included borders for Zip Codes (ZCTA actually, Zip Code Tabulation Areas, which are the Census-created equivalent), Counties, States, and Countries.If you want to bring new borders into the application, please see the border import tutorial.
After selecting the appropriate border set, you will be asked to choose the naming convention that your data uses. Your data must match a naming convention that already exists in the importer. We have included many naming conventions at each region level and you can bring in new names for regions as needed; please see the region names tutorial for details.
In the case below, we are importing county-level data and the Label1 listing is what is desired. (Notice that the names are not unique in the Label2 list.)
Error Handling
After selecting the appropriate label format, the importer will provide notices for any import errors. If your data is in the appropriate format there should not be any errors (although, that said, some errors may indicate syntax errors that do not disrupt the integrity of the data, for example if rows that do not matter are imported, the importer will try to parse them, fail, and give an error message). If less than 50% of the file contains errors, you will have the option to ignore errors and continue.
Below is an example of errors arising from importing a different file, one for Population Density by Zip Code (ZCTA).
This imported file was created by a spreadsheet program starting with two source columns and creating a ratio (population/area), and “ #DIV/0! “ is provided as a flag. The importer does not recognize this flag because it is not a number.
In addition, ZCTAs have 5 numbers in the selected naming convention, but the spreadsheet program has removed the first zero from “06001, 06002, 06003, etc.” and as a result, these names are not recognized. To include the data for these regions, either the spreadsheet formatting must be adjusted so that leading zeros are not eliminated, or a naming convention that accounts for dropped zeros must be selected.
All lines with errors will be discarded, so for the sake of data integrity, it is important to review these error messages and adjust data to reduce errors. In some cases, it may be appropriate to discard lines with errors, but you will have to use discretion as the application currently does not log these changes to the data.
Adding Metatdata
Once the data has been accepted, the importer provides an interface for logging metadata related to your data. By default the importer will use the title of your file as the title of the statistic, although you will have the opportunity to change the title later in the import process. For example, if your file is named "ZCTA-PopDensity.csv", then your statistic title will default to "ZCTA-PopDensity".
You have the option to skip the metadata process if you wish. In the current version of UUorld, metadata cannot be edited once a stat has been imported, so we recommend taking a few moments to input the metadata here.
Additional Note for Mac Users
If you are given the option, be sure when you are saving your CSV files for import into UUorld to save as a "Windows CSV." The importer has had a bit of trouble with other-formatted CSV files, but this will often immediately resolve the issue.
If you have questions beyond what is covered in here, don't hesitate to write to us. We're happy to help.