Cleaning, wrangling, munging, preparation, validation, ETL, transformation, ELT these are all different names for making data more useful. Whether youre doing one-off analysis, scheduled reports analytics, or integrating two different data systems or applications, when you work with data, data cleaning is an important part of almost all data processes. Data cleaning is taking data as-is and changing the content, structure or format to be more useful. Removing blank rows and deleting invalid addresses are examples of data cleaning.
Over the past several months, Inciter has been pleased to work on a project with a summer camp network. The goal of the network is to enhance accessibility and inclusion for campers and staff with disabilities by providing capital improvements, professional development, staff training, research, and evaluation to both participating camps and the field at-large.
In many ways the project is similar to the other program evaluations that we work on with organizations, foundations, and agencies - large and small.
Whether you are getting feedback from program participants, members, donors or other stakeholders, an online survey can be the easiest and most efficient way to do it. Of course, online surveys are not a one size fits all solution for collecting data to answer your burning questions, but they can be a cost effective method to gather certain data for your organization. No matter your field or audience, you have to follow some best practices if you want to make use of your data.
You want data at your fingertips, when you need it, in the right format. Who doesnt? But often when you go to get reports, the data is siloed and you cant bring it together, or some of its missing, or its not accurate, or you dont know what to make of it.
Maybe you think technology is the answer. Its understandable. Machine learning. Artificial intelligence. Big data. There is all this amazing technology out there that can make cleaning, analyzing and reporting on data so much easier.
Tired and unfocused during long meetings? Im going to help you with that. Last week, I had a seven hour Zoom conference call, with about 20 people in it. (I know. I know.) It was necessary, and all parties were actively engaged, and needed to be. This was an authors retreat for a large research project, where Inciter is serving as the data visualization team. It was important that I stay engaged, not space out, and listen to the findings, the nuances, the decisions people were making.
This is the eighth and final post in our Data Cleaning in Excel blog post series. As the series comes to a close, wed like to take an opportunity to organize and recap material weve covered. Well review some of the useful Excel functions weve used, and provide some guidance for incorporating the techniques into your data cleaning routine.
Every dataset has its own issues. Our posts in this series have been written to address ad-hoc common problems.
In this part of our data cleaning series, well be focusing on managing blank cells. There isnt a singular approach to handling blank cells in your dataset. Because there are numerous reasons why a cell might be blank, context is key when determining how to fill them. Sometimes, youll need to fill every blank cell in your data with the same constant. Other times, youll pick up clues as to what should be there from the surrounding data.
In this part of our data cleaning series, well help you find and remove duplicate entries in Excel. Repeats are a very common data entry mistake or error from a data pull. Duplicates in your data can create a variety of unfortunate consequences in administrative duties and analysis. Worst of all, they lead to a real misrepresentation of your results.
This is Part 6 in our Data Cleaning in Excel 101 series.
Ever try to do a calculation with numbers in Excel and get an error or the numbers don’t seem to be adding up? Ever fight with ZIP code formatting? Below you can find some methods of dealing with numbers that just aren’t acting like numbers. Well also review some instances in which youll actually want Excel to store numbers as text, and how to convert them.
This is Part 5 in our Data Cleaning in Excel 101 series.
Having the right data in the right columns to meet specific requirements for your analysis plays a major role in the data cleaning process. In Parts 2, 3, and 4 of our Data Cleaning in Excel series, well show you how to solve common issues by utilizing both standard and Excels powerful Flash Fill shortcut.
Part 2 focuses on splitting data from one cell to multiple cells. Part 3 covers the opposite: combining data from multiple columns into one column.