Data science? Yes, please. Data scientist? Meh.

I wrote a post a while ago about not being sure if I wanted to call myself a data scientist. The post was less about what title I wanted to ascribe myself and more about the many divergent ways “data science” seems to be defined. At the time, I wrote:

Bottom-up creation of data-driven capabilities: show don’t tell

I’ve been writing lately on what to do when people who make decisions in an organization say they want data-driven capabilities but then ignore or attack the results of data-driven analysis for not saying what they think the data ought to say. Some of the most productive things you can do in that situation include automating your work so you can devote more time and attention to more important (and labor-intensive) projects, as well as building support among the organization’s weak actors as a means of garnering positive attention from higher-power stakeholders.

Bottom-up creation of data-driven capabilities: weak supporters *10 = strong support

My previous post addressed the scenario of executives or managers saying they want data-driven capabilities but not accepting data-driven analyses when the findings are presented to them. As I discussed in that post, I think the best first step in such a situation is to regroup and streamline your workflow, automating as much of your more repetitive tasks as possible, in order to free up your time so you can devote more attention to bringing those managers on board. This post focuses on one of many possible next steps: building up support in a wide variety of seemingly trivial areas of your organization.

Bottom-up creation of data-driven capabilities: automate your work

My previous post on how to transform an organization into a more data-driven version of itself made a pretty big assumption that often doesn’t hold true. I assumed that people in the organization wanted their company or agency to become more data-driven. I think almost everyone says they want that if asked. I even think most people think they want that. But I don’t think most managers and executives actually want it (and Paul pointed out something similar in his recent comments). This is the first of what will probably be two or three posts on what to do when you want a more data-driven organization but the rest of your organization doesn’t.

Feeling Ineffective … Needing the “Haqqani network”

One of the things I dislike most about being in academia is the feeling of creeping complacency. I don’t feel it very often – that’s why it’s ‘creeping’ – but when I do it’s painful and soul-frustrating. Working for the military was where I felt least complacent (despite the numerous other downsides), so perhaps it’s not surprising that the things that make me feel most complacent typically have to do with the military and irregular warfare. The current situation in Afghanistan should be enough in itself to make anyone stop and think whether such a situation is really necessary and whether there isn’t anything that could be done about it. 326 members of the international coalition have died in Afghanistan so far this year; 3,021 Afghan civilians died in 2011; suicides among US troops have been averaging one a day in 2012; the Dept. of Veteran Affairs estimates 18 veterans are committing suicide every day. All this is an ongoing and overlooked tragedy. But it’s not the overall tragedy that makes me feel complacent. For me and for most people it’s about as outside the bounds of being influenced as is the weather. No, what makes me stop and feel terribly complacent are the errors in thinking about social phenomena in Afghanistan that policy-makers and military analysts continue to make, and that social scientists seem incapable of helping correct (…perhaps because we’re often not so immune to them ourselves).