My son: I can’t find my sweater anywhere!
Me: Have you checked in your bag?
My son: I know it’s not in there.
Me: But have you checked?
My son: No, but I know it’s not there!
Five minutes go by until I go into his room, open his schoolbag, and pull out his sweater. This conversation happens frequently at my house, and I doubt we’re alone.
Some of us who are older and wiser may feel this is the sort of exchange one only has with the young and foolish. The fact is, we are all guilty of this crime. We fail to differentiate between what we know and what we think we know. And this isn’t just about where we left the car keys or if we paid the phone bill — what do we know about data? (And what do we just think we know?)
Policymakers and Gender Equality Data
A group called Equal Measures 2030 recently published a report on policymakers and gender equality that examined how closely decision makers’ perception of gender equality aligned with actual data on equality in their countries. They surveyed policymakers (including government officials at various levels as well as NGO and private sector leaders who might influence policy) in an effort to answer these questions:
- How do policymakers perceive progress on gender equality in their countries?
- What most needs to change in order to improve gender equality?
- What data and evidence do they rely on to make their decisions?
- How confident are they in their understanding of the major challenges affecting girls and women in their countries?
What Do We Think We Know?
The report is fascinating, and I encourage you to read it. But while I believe the issue of gender equality is a critical one, it’s not all I want to focus on today. To demonstrate how much we like to think we know about data, I’d like to highlight some of the responses the survey elicited:
One of the survey questions asked respondents what percentage of girls in their country were married before the age of 18. In Colombia, policymakers’ estimates ranged between 4 and 80%. (The most current data puts the figure at 23%.)
Another question asked what percentage of the country’s parliamentary seats were held by women. Kenyan responses spanned almost the full range of possible answers — from 6 to 90%. (The actual number, according to the most recent data, is 21%.)
When asked what percentage of the country’s labour force was female, Indian respondents estimated anywhere from 20 to 70%. (The most recently available data says 27%.)
A serious problem arises when people who are making decisions think they know about the relevant data but don’t actually know about it at all.
“If policymakers actually think that 70% of the labor force is constituted of women when it’s only 27%, (it is) no surprise the issue is not prioritized.”
-AB Albrectsen, CEO of Plan International
Smart vs. Informed
The team at Equal Measures 2030 wasn’t trying to disparage the leaders they surveyed; neither am I. Being ill-informed doesn’t make those policymakers stupid or ill-intentioned. But it does affect their ability to make decisions that will benefit the people they serve. Recognizing the difference between what we know about data and what we only think we know is an important step towards making the world a better place — for all of us.
David Ruttka wrote a brilliant blog post about the difference between knowing and thinking you know. While his points came from a summary of a conversation with his five-year-old, many of them can be applied to what we know about data.
When You Don’t Know About Data
- You didn’t know. You thought. There’s a difference.
And that difference can make a huge difference in the decisions you make and the stories you tell with data. Maybe you’ve got the numbers totally wrong. Or maybe you think you know what your data means, but are being tripped up by ecological fallacy or complicated data relationships.
- Not knowing is OK. It’s just one more thing you can learn.
Learning isn’t just for children or people training to acquire a new skill. I regularly attend workshops and conferences where I learn all kinds of things I wouldn’t have otherwise known. And we’ve made it our goal here at Datassist to keep offering learning opportunities for people from all walks of life. Make what you know about data something you actually know.
- Just because someone says they know doesn’t mean that they know. Trust, but verify.
Verifying that your data is accurate, complete and free from bias is critical. In every introductory data analysis course I’ve ever taught, I’ve tried to highlight the importance of ensuring that what you think you know about data is actually true. Just because your data comes from a reliable source, doesn’t mean it merits blind trust. Create a data biography to help uncover the secrets your data might be hiding.
- Again, always, always remember that it’s ok to be wrong.
Everybody makes mistakes. You don’t have to hide yours or cover them up. In fact, sharing your failures is a great way to help others learn. (So they don’t have to make the mistakes you’ve already made.) The best way to increase what we know about data is to share what we don’t know.
Let’s Share What We Know About Data!
At Datassist, we want to help everyone know more about data. Stay tuned for weekly blog posts and announcements about upcoming data courses, or subscribe to our monthly newsletter for data tips, tricks and tools. Got a data project you need help with? Get in touch with us today.