Pied Beauty

Gerard Manley Hopkins (1844-1889)

Glory be to God for dappled things—
For skies of couple-colour as a brinded cow;
For rose-moles all in stipple upon trout that swim;
Fresh-firecoal chestnut-falls; finches' wings;
Landscape plotted and pieced—fold, fallow, and plough;
And all trades, their gear and tackle and trim.

All things counter, original, spare, strange;
Whatever is fickle, freckled (who knows how?)
With swift, slow; sweet, sour; adazzle, dim;
He fathers-forth whose beauty is past change:

Praise him.

Thursday, November 11, 2010

On Trees and Statistics

Did you see the publicity on the recent study, conducted in Portland, Oregon, that concluded that large street trees actually reduce crime?  Now I love trees.  But when I heard about this study, soon to be published in the journal Environment and Behavior, I felt a curmudgeonly skepticism rear its ugly head.

I love trees.  Even Bradford pears.
The study, performed by the US Forest Service, looked at several years of crime data, cross-referenced with aerial photographs, site visits and tax data.  In part because one of the authors lived there, the study was confined to a single police precinct area in Portland consisting mainly of single family homes.

I love trees.  Especially cedar elms.
Perhaps not surprisingly, the authors find that the presence of large trees is negatively correlated with crime.  Smaller trees, however, are positively correlated with crime.  Their conclusion?  Big trees prevent crime, while small trees cause crime.  I'm oversimplifying a tiny bit; the authors do explain in the final paragraph of a 29-page report that correlation is not causation, and that other variables are more strongly associated with crime and that unknown variables may have a large impact on crime prevention. 

I love Baldcypresses.
You wouldn't know that from reading the rest of the report, though.  Nor could you tell that from the press the study has received so far. For heaven's sake, don't put up burglar bars.  They're positively correlated with crime.  Make sure you have a tree that's at least 42 feet tall instead.

I love Weeping Yaupon.
I'm pretty sure that the large-tree factor is negatively correlated with crime.  I'm just not convinced trees are preventing the crime.  The same characteristics that cause people to care for trees is perhaps responsible for a (very slightly) lower crime rate.  How about, for starters, home ownership versus renting?  Income levels?  Education levels?  Previous history of crime? Economic conditions? Ethnic makeup?

I love conifers, too.
Oh well.  Perhaps I'm just experiencing a "Bah humbug!" moment.  May it soon pass!

You can read the entire study here.


  1. In architecture school, landscaping classes to be exact, we were taught the relationship and correlation. But living in a city, on a block with 100 foot tall trees, decent incomes and education, ethically diverse, and homes worth considerably more than the surrounding areas; we have had a couple of murders, drug busts, identity theft arrests, a peeping tom also arrested, a car theft ring, and numerous lesser offenses. All on one block, the block I described above. So, unless they want to expand their study, I am not so sure I am on board with the findings. Crime can happen anywhere.

    On Monday's GBBD post, I have a story of tree theft. So even the trees get stolen.

  2. Sounds rather like the "Chanticleer Effect" to me!