Keeping a data agenda
Previously, I discussed the importance and benefits of keeping a data dictionary. But as your program grows, you’re going to want to get familiar with the data dictionary’s close cousin – the data agenda.
What Is a Data Agenda, and Why Should I Keep One?
While a data dictionary is the inventory of every piece of data that your program collects, a data agenda represents the next step in the process: a list of data points that would be valuable to your program but are not currently being measured. These data points typically:
- are important indicators internally that your program is meeting its goals
- are important for managing your program, i.e. they are connected to real decisions and action steps
- would communicate to the outside world that your program is meeting its goals
- would be relatively easy to collect (eg. administrative data or a simple survey).
As your program grows, your goal should be to keep flexing and developing your ability to make new data in a way that supports how you want to manage your programs. Having a data agenda allows you to plan for the measures that will help you understand your program’s effectiveness as it develops and changes.
What Does a Data Agenda Look Like?
Your data agenda will look much like your data dictionary; but while your data dictionary shows what you’re doing, your data agenda will show what you want to do.
To create your data agenda, simply build an excel workbook, and create one tab per program with the key elements defined:
- Measure name
- Owner
- Data source
- Calculation
- Cadence (how often you look at it)
- Report (where you’ll look at it)
The goal is to eventually be successfully gathering and managing the data listed in your agenda, which may seem a bit overwhelming at first. But the key is to tackle one item at a time, moving the datapoint from the data agenda to the data dictionary each time you implement a new measure. Over time, you will see your success as your dictionary grows.
I Don’t Have a Data Agenda Yet. What do I do?
Organizations are often working on an issue before they are measuring the data they need, so don’t use the fact that you don’t have the data yet to avoid the problem. In the meantime, you can impute data. Use your memory. Use an estimate to fill in historical data. You can work to address the issue at the same time that you work on creating better measures to track your progress.
Still need to develop some measures? Need help including them in your logic models? Contact me below to talk about hosting a workshop at your site.
{{cta(’41ba56f8-8a80-44cd-95a2-dbc1e2e96b15′,’justifycenter’)}}
