Interpreting Data

by Don Boudreaux on January 6, 2006

in Data

Cornell University psychologist Tom Gilovich makes one of the most useful observations I’ve encountered in some time.  Here’s economist Robert Frank’s summary of Gilovich’s observation:

As the psychologist Tom Gilovich has suggested, someone who wants to believe a proposition tends to ask, “Can I believe it?”  In contrast, someone who wants to deny its truth tends to ask “Must I believe it?"

As much as I’d like to deny that Gilovich’s observation applies to me, I can’t.  It does apply.  But now being aware, I at least try to take a more critical stance toward propositions that correspond closely to my priors, and to take a more forgiving stance toward propositions that contradict my priors.

I thought of Gilovich’s observation when I saw this graph in today’s Econoblog discussion between Brian Wesbury and Barry Ritholtz.

Onaa793_econbi_20060105112803_3

When I first saw this graph, I thought exactly what Brian Wesbury wants me to think: "Yep, the tax cuts that kicked in May 2003 spurred investment spending in the U.S."  And perhaps they did.  The data, as they say, will bear that interpretation.

But as I look more closely and critically I see that the trend of business investment from the depth of the downturn through today might not have had anything to do with the tax cut.  Although still negative until the first quarter of 2003, business investment began to improve in the first quarter of 2002.  Furthermore, note that business investment was higher in late 1999 and early 2000 than it has been at anytime since the May 2003 cut in taxes.

None of the above is to disagree with Brian Wesbury.  I agree with him that these tax cuts likely explain much of the observed change in business investment over the past few years.  But it’s interesting how ambiguous even seemingly unambiguous pictures and charts and graphs and historical anecdotes become if you look at them critically, trying your best to see in them not what you want to see but what someone with a view very different from your own wants to see.

I’ll not here allow myself to get sucked into the black hole of methodology, save to express agreement with Deirdre McCloskey (and here I paraphrase) that no one was ever convinced by raw data of the truth of a proposition that he or she did not already hold to be true.  Data are important, but the theoretical filters in our minds are no less so.

Comments

24 comments    Share Share    Print    Email

Previous post:

Next post: