The Craft of Research

So I’m finally done reviewing all 22 beat memos in both sections.  Phew!  I spent at least an hour on each one.  You do the math.

First of all, special congratulations to the 7 who merited a check-plus.  For those who didn’t, take heart that this will be your least significant research grade.

Kudos also to those who had the tenacity to navigate the user-UNfriendly Census Bureau Factfinder to get any 2007 ACS PUMA data.  I was perfectly willing to accept the equivalent 2006 data available on the more easily-navigable NYC.gov, so that was impressive.

It was actually one of a handful of things that impressed me.

A few of you discovered good sources that I hadn’t listed on the blog or in handouts.  One was the 2007 Community Snapshots in the Children’s Services section of NYC.gov.  Mind you, these snapshots weren’t great in all the cases people cited, but they were particularly good for CD income and unemployment data (although there may be some issues with unemployment…more on that later).  Another was the Association of Religion Data Archives site, a perfectly acceptable alternative to Social Explorer for 2000 religion breakdowns by county.

What else stood out?

Brian Winkowski used Infoshare.org to track immigration trends for his CD (through 2002), which I cited in all my successive memo reviews.

Jeanmarie Evelly did something so mind-numbingly impressive to compile CD voter data I’m not even going to bother describing it here.  Ask her.  It was above and beyond the call.

A few of you used the web sites of your Community Boards to get information.  Others talked to board members.  Either was smart.  (Talking to people “in the know” is always encouraged.)

I was also happy to see at least some of you understood the value of a ReferenceUSA SIC code search.  My in-class demonstration of how to find churches by zip code was not wasted.

What disappointed me?

Blank answers, first and foremost.  I at least expect an explanation when you can’t find an answer.  Not far behind on the disappointment list was the lack of a source for answers.  As I posted on the blog October 1st, your research grade will be based primarily on your sourcing of answers.  (Remember…”Says who?”)  Keep that in mind for the final quiz, when I might be inclined to give partial credit for a wrong answer that cites a good source.  A wrong answer with absent or weak sourcing will get nothing.

On the subject of weak sourcing, try to stick to the sources we’ve reviewed in class or that I’ve listed in handouts or on the blog.  Occasionally, you might discover some other good sources (e.g. Community Snapshots or ARDA), but the chances you’ll find something worthwhile that I haven’t already provided for you are not great.  Several of you cited secondary sources in your beat memo, such as Wikipedia, Trulia.com, Rentometer.com, BestPlaces.net, and even NYU and NYTimes.com.  Generally, these are sources reporting data that other entities compile.  That’s not what you want.  When you’re reporting data, it should be second hand, not third hand (or worse).

Another source many of you used that should have been good in theory was James Brown of the Labor Department.  I’m always good with your contacting experts, but in this case, some of his unemployment numbers varied so widely from what I found in the ACS PUMA tables that I don’t know if he was using different calculations or definitions or geographies or what.  Had any of you investigated the sources I provided first, you could have asked him why his numbers were so different.  Maybe they’re “better” than ACS, but without an explanation, I can’t assume that.

Lastly, I always prefer the most up-to-date data, as long as the sourcing is solid.  Quite a few of you reported 2000 numbers when much more recent ones were available from equally good sources.

We move on.  See you Wednesday.