Question: How many times have you, when reviewing IPEDS data, found basic data entry errors?
Better question: How many times have you missed such errors prior to submission?
The issues around data entry errors in IPEDS surveys are very real and addressing them isn’t always easy. But they are definitely correctable. You can reduce bad data entry during IPEDS reporting by following three general steps:
One of the first things you can do to alleviate errors is to simply define, “What is an error?” Some data entry errors are obvious, but some can be quite subtle. Think about addresses. Which of the following is correct?
- PO Box
- P.O. Box
- Post Office Box
How do you format names? All capital letters? Only the first letter capitalized? Or no capital letters?
Do you have any “intelligent” fields, such as program of study? Many institutions code the program name as ‘college-degree-major.’ If this is done incorrectly, programs may disappear at the most inconvenient of times.
The way to resolve these issues is to develop standards for data entry. This is not an easy task and involves most, if not all, functional areas. A project owner must first be determined. (This determination, alone, may be the topic of several meetings.) If a data governance process is in place at your institution, this task belongs to the data owners. If no governance process exists, the task can still be accomplished with ownership likely shifting to Institutional Research, as they are most affected by the issue.
The final output of this standards creation should be a list of fields used for data entry, broken out by functional area. Each of the fields should have the data requirements listed for it. A standard should cover the lifecycle of the data, and should include: who enters it, how is it changed (and by whom), and who uses this data. The standards should be a “living” document and must be reviewed periodically as the ERP, or other reporting sources, are updated over time.
An added benefit of creating data entry standards is that it won’t reduce bad data entry for just IPEDS reporting. It’ll result in more accurate and more consistent data across all campus reporting.
The second step to reduce bad data entry in IPEDS is to communicate the issue. Data entry people are not lazy, not forgetful, and not out to get you. They want to do a good job. If the issue is explained – and solutions are presented – to them, they are generally very happy to help make the data better. Remember, you are communicating with them – not accusing, yelling or berating.
Though not required, having standards in place obviously helps with the communication. The use of the data needs to be explained to those entering it, so that they have a good understanding of why data needs to be formatted in certain ways. Consistency is the main point to get across.
Realize that the people doing data entry are not automatons. They may have very good suggestions on ways the data can be formatted for easier – and correct – entry. They have as much skin in the game as the end users of the data.
Also, keep in mind that the lifecycle of the data may not be completely known. We think we know that this piece of data is entered by people in Department A, when it’s actually the folks in Department B. Convene a meeting to discuss where data originates, who enters it, and how it’s used. It will be better received, and less tense, than a meeting to discuss errors with the data entry.
Lastly, it’s important to check – and then double-check – that the data is correct. All the standards and communication in the world will not prevent data entry errors. The data is entered by humans and that means mistakes will be made. There must be a means of checking the data.
If creating a set of standards and establishing clear communication at your institution is not feasible, then you must have a method to check data! For instance, a list of data errors can be generated and sent to the appropriate group for correction without having standards in place or a discussion about data entry. This is not the most efficient process, but it can be done.
Checking the data means having a report (or several reports) that scans the data fields and identifies errors or anomalies in the data. The output is a list of data fields that need to be checked and possibly corrected. This process works better if data standards exist, as they would provide the criteria for those corrections.
Each functional area that performs data entry should have a routine or report that checks their work. Any errors detected should be provided to the functional area on a periodic – and timely – basis! (A list of errors from last term is not as helpful as a list of errors from last week or yesterday.)
By defining standards, communicating them to necessary personnel, and then double-checking your data, you can reduce data entry errors when doing your IPEDS reporting. Though each step can be implemented independently, it’s when they work together that you give yourself the best opportunity to provide clean data, not just for IPEDS, but for all institutional reporting.
Like this blog? Check out other blogs in this series and sign up for the webinar*:
*The March 28 webinar will expand on the content of these blogs.
Latest posts by Peter Wilbur (see all)
- How to Pull Data From the Cloud: Things to Consider - 04/24/2018
- IPEDS Data Entry: How to Reduce Basic Data Errors - 02/15/2018
- IPEDS Surveys Reporting: ‘Tis the Season (Part 2) - 12/05/2017