Have you ever been in a presentation or a meeting where someone was sharing important updates for stakeholders or a new report and they either:
- Butchered the explanations of the methods underpinning the findings and conclusions (or provided no explanation at all), or
- Immediately directed any follow up questions to someone else for a response, lacking ease and confidence about what they just said?
It’s not a very comfortable situation. (Similar scenarios can play out in writing, too.)
The thing is, we can’t talk about our impact and outcomes in all of their glory if we don’t understand how we’ve come to the conclusions we’re highlighting.
What’s more, we can’t trust the conclusions we’ve come to if we don’t understand how the analysis was done in the first place. This is not me saying don’t trust your analysts’ findings. I am making the point that people trust you based on what you can explain with confidence, and it’s nearly impossible to explain something you don’t understand.
So when we’re talking to our stakeholders, communities, and donors anyone speaking about your work needs to have a solid understanding of the WHY and HOW we’ve come to our stated conclusions.
Understanding the fundamentals behind the methods gives the speaker, writer or presenter much more confidence in what they’re saying, which also gives more confidence to the people in the room that this person and their organization know what they’re talking about, and this builds trust.
So, does every middle manager or director needs to go out and take several statistics courses or learn how to run regression analyses? No. Definitely not. There are dedicated people for things like that, be they external or internal. (Who’s your person?)
Here are some things to try:
- Hand your analysts the mic. Presenting is certainly not everyone’s cup of tea, but if your data person/team can distill the analyses and conclusions into plain language, convey the main points and explain the reasoning behind the methodology in basic but clear terms, let them co-present or write about the findings.
- Create a half to full page cheat sheet of the methods used, assumptions behind them, limitations, and why these methods were chosen for anyone who will be discussing the work. It doesn’t have to be a textbook chapter, but it should cover the basics. Everyone should understand why certain methods was chosen for an evaluation, analysis, or QI initiative. Tap your analysts or researchers to help with this.
- Make sure the whole team has a good grasp on important concepts.These are the ones I see people either confuse or forget often: “controlled for”, “multivariate”, “significant”, “bias”, “correlation” and “causation”. Understanding the difference between correlation and causation is a particularly strong need.
- Use this newfound understanding to enhance the clarity of your communications. Make statements like this: “We found that our intervention reduced the level of local emissions by X%, a statistically significant reduction from 3 years ago when we started”…instead of this: “we have significantly reduced local emissions through the intervention put in place 3 years ago”. The last one uses “significant” as editorial language and not in the statistical sense. Also, consider communicating visually as much as possible using high-quality graphs, charts, or infographics. Humans digest information faster from visuals than they do words.
- Share your methods and assumptions in bullet form, for yourself and the audience. It’s clear, short, and to the point. Information without overload, which is especially important to avoid when discussing something bulky like methodologies.
These relatively simple and straightforward practices can help you gain and keep the interest and trust of the people in the room, head off skeptics (because there are always SOME limitations and now the whole team will be able to discuss them), and you keep the focus on your data story.
Has statistical literacy ever been an impediment to adequately sharing the progress and impact of your work?