Let’s get one thing straight, nobody wakes up one morning knowing how to tie data together. For some, it comes naturally and for others, it takes time to master. Understanding data of any kind takes some patience and an appreciation of some basic trends and rules. Firstly, data is like a puzzle. You just have to start with what you know, like starting with the outside pieces. Take something you know and build the frame. If you are working with Division of Interest data, identify your “Key ID”; which in most cases is a combination of Well Number and Deck/DOI Number. From the Well and DOI combination, you can identify pertinent information for that record; ownership, exemptions, burdens, etc. There are relationships that are built on this “Key ID” that help build the final picture. A Well Number alone will not tell you the entire or accurate picture on the ownership for that well. Secondly, not all data is black and white. It is repeated over and over that data is only as good as how it is entered and that could not be truer. It is my belief that one of the most important and pivotal roles is the data entry itself. Data integrity depends solely on how it was input. Keep in mind that the degree of system data cleanliness will not be able to be determined until a large data set is exported in mass. It is of good practice to scrub the data and normalize once exported. Data: You Can Do It!
By: Britanni Rutledge
Example Scenario: Report the number of wells in a particular State and County. If the State is both represented as abbreviated and long text (OK vs Oklahoma) then your summary is going to be a count on both OK and Oklahoma, not smart enough to understand that the two different records are actually one in the same.
Basically, normalizing data is the action of choosing only one way to say something. When working with large amounts of data, determine what fields are important, review them and consolidate to ensure consistency. Lastly, understand some basic rules in the system where the data manipulation will take place; most likely Excel so let’s start there. Here are a few tried and true rules to ensure accuracy in Excel. Rule #1: Filters must be across all data columns. Why? Try looking at a DOI download with hundreds of owners. Filters are put on only some of the column headers; row 1. If you filter the data, more than likely it’s not going to effect the relationships within the data, but
21
G r o w t h T h r o u g h E d u c a t i o n - J a n u a r y / F e b r u a r y / M a r c h
2 0 1 8
Made with FlippingBook - Online catalogs