Data can be like chemistry: the elements interact, sometimes to explosive effect. Ignore it at your peril – and use it to your benefit. This month’s Datassist posts dig into data’s role in accurately parsing out issues of race, sexual diversity, and gender gaps. Data from these different areas work in tandem. The correlation between such variables is called the Interaction Effect – and it’s a beautiful, useful thing.
One of Datassist’s upcoming projects has got me reflecting on this recently. It studies how many parents use available child care programs, and why others don’t. Initial findings seem clear: “Focus on Moms!” This study found that female parents use ~38% of available programs, while males use ~44%. Other factors seem to matter less: self-identified East Asian parents use ~41%, while self-identified Europeans use ~40%.
Interactions of gender and ethnicity, however, tell a more useful story. East Asian moms use ~32% of available programs – way below the average – and European moms use ~45% – far above. East Asian moms seem to need the most outreach. Why are they not coming? In general, moms say, they stay away due to price (45%), difficult registration (42%), or inconvenient times (41%). Ok, now we’ve got it, right?
But these numbers – you guessed it! – alter by ethnicity. Price only dissuades 30% of those East Asian moms not attending. 49% don’t come because classes are at bad times for them. We even flat-out missed some barriers: it’s not on our list, but 42% of East Asian moms (and 35% of Europeans) say they don’t have time for classes at all. And 22% of European moms would get involved but have trouble finding out how.
It’s not enough to design programs and outreach policies for “Moms”. Considering the needs of different kinds of moms through Interaction Effects can tell us more about our communities’ unique needs and challenges. Do these effects still do some generalizing? Sure they do – but thinking about Interaction Effects get us a lot further ahead than viewing one category at a time.