Although nonprofit leaders recognize the importance of measurement, they often find the process of data analysis and decision-making to be daunting. To the contrary, our 7 Steps to Data Driven Decision Making mirror the scientific process that we all learned in middle school. The key is to set up the “experiment” appropriately: clearly define the issue you are investigating, develop clear hypotheses about what may be going on, test the hypotheses and then draw conclusions from the analysis.

This second installment in our 7 Steps to Data Driven Decision Making series details the second step in the process, Hypothesis Development. Click here to read our last installment, Step 1: Framing the Issue.

Hypothesis development

Hypothesis development is a key step that many organizations skip entirely. Once they identify an issue, they immediately begin to collect data without taking the time to develop a clear understanding of the data they need or the questions they are trying to answer. As a result, it is very difficult for them to draw clear conclusions that inform decision making from the data they gather.

Think of hypotheses as statements that you will accept or reject once you collect and analyze data. We encourage the nonprofits we work with to generate as many hypotheses as they can and then prioritize those that they would like to test.

The best hypotheses are specific and measurable. Start by identifying the cause and effect of the issue you framed in step 1. We found two approaches organizations can take to develop clear hypotheses: using a logic model and brainstorming with colleagues.

Use a Logic Model

Your logic model outlines connections between inputs (staff, time, grant dollars), activities (training, programs), outputs (engaged volunteers) and short- and long-term outcomes (improved volunteer retention, better living conditions for target population). To generate a hypothesis, pick two points in the logic model and generate an “if…then” statement.

Let’s return to Sarah, Program Manager for a capacity-building foundation who wants to increase volunteer retention.

Sarah’s logic model connects grant money, volunteer training, volunteer engagement and volunteer retention. Based on this, she develops the following hypothesis:

"If a nonprofit that receives grant money uses it to train its volunteers (input), then the volunteers for that organization will feel more committed to the organization and will stay there longer (output). "

Brainstorm with Colleagues

A brainstorm will help you to think of all possible hypotheses. These hypotheses are often derived from anecdotal evidence, gut feelings or emotional responses. Although our experience shows that over 80% of such hypotheses are rejected by the data, they are still quite helpful as organizations set strategies and focus limited resources where they are needed the most.

After using her logic model, Sarah decided to brainstorm with her team to develop additional hypotheses. Her first instinct was to skip this process and dive into her data. However, as she engaged in the brainstorm, Sarah realized that she would have missed several potential drivers of retention rates if she had skipped this step.

Sarah’s team drew on their experience working with the nonprofits in their grant portfolio to develop a list of hypotheses about what might be influencing volunteer retention rates. The list they developed contained over 15 hypotheses.

Prioritize Hypotheses to Test

Now that you have developed as many hypotheses as possible, it is time to choose which of these you would like to test. Select those hypotheses that most closely relate to the mission of your organization or those that impact the most people. Remember that hypotheses based on variables that are out of your control will not help you make actionable decisions.

Sarah’s team developed more than 15 hypotheses through their logic model and brainstorming session. Of those, she chose the following three to prioritize and test, based on her mission and potential impact:

• If the nonprofits in the grant portfolio invest more time and money into training their volunteers (input), their volunteer retention rates will be higher (short term outcome).
• Nonprofits that have more impact will have higher retention rates because volunteers who see the impact their service has on the organization feel more committed.
• Nonprofit organizations that have multi-year relationships with the foundation will have higher volunteer retention rates.

With clear, measurable hypotheses in hand you are prepared to collect the right data, which will ultimately inform effective, data-driven decisions.

Check out our next installment: Step 3: Data Collection.