- Cafe Hayek - http://cafehayek.com -

Data Are Unlikely to Settle This Dispute

Posted By Don Boudreaux On October 17, 2013 @ 10:55 am In Data,Science,Seen and Unseen,Truth-seeking & ideology,Work | Comments Disabled

Commenting [1] on this graph posted earlier this year by Stanford’s John Taylor [2], Jamie Newman says

Chicken? Egg? Can’t tell from looking at that lovely chart, can you? Neither can Mr. Taylor, who admits as much at the end of his post.

Mr. Newman is correct.  What Mr. Newman highlights (whatever was his intention) is the indispensability of theory – the plain fact that data never speak for themselves.  Theory is always necessary to make sense of data, and the same data can yield very different tales depending on the theory used to interpret the data.

My point here is obvious and often-made.  Yet it seems as though it cannot be made often enough, given the many accusations, in public and private discussions, that people are “unscientific” or are “willfully ignoring the data” if the lessons they draw from the data differ from the lessons drawn by those who hurl accusations of “science denial” at those with whom they disagree.

Of course, there are indeed people who are genuinely unscientific.  There are indeed people who do ignore the data or who study or ponder the data lazily or with undue prejudice.  But the existence of such unscientific people and practices does not mean that all scientific disagreement is the result of one side or another failing to examine the data with sufficient care and objectivity.  Differences in theory explain many differences in conclusions – and resolving these differences in conclusions can seldom be done simply with more and better data.

Theory is not only indispensable, it is of at least equal importance as are ‘the data’ in helping us to better understand reality.

Therefore, if a theoretical proposition is firmly established and unquestioned in its application to a certain class of phenomena, peeling out some phenomenon from the class and claiming that the established proposition does not apply in its widely understood manner to that phenomenon requires a powerful and compelling theory to explain why that particular phenomenon happens not to follow the same theoretical rules that are accepted as governing other similar phenomena.

Finding data that show that government-mandated higher wages for low-skilled workers does not result in reductions in the employment opportunities for low-skilled workers is child’s play.  So, too, is it easy to find data that show the opposite effect.  The first effect has been found by some studies; the second effect has been found by other (I believe larger-in-number) studies.  Some of these studies are more careful than others.  Some of these studies (on both sides) are products of serious scholarship; some other of these studies (also on both sides) are products, not of genuine scholarship, but of unscientific prejudice or jejune analysis.  How to choose among these studies?  Theory is a good way to make this choice – indeed, I believe it is the only way to make this choice.

As Sam Grove frequently asks in the comments section [3]:

When you decide to buy something, do you shop for a higher price or a lower price?

This question, straightforward as it is, signals the strength of the law of demand.  I’ve seen no remotely plausible theoretical reason to suppose that this law somehow doesn’t apply to the demand for the services of low-skilled workers.  So, yes, my theory plays a huge role in prompting me to dismiss as false the claim that raising employers’ cost of employing low-skilled workers will not reduce the employment prospects of low-skilled workers.

I am no more (or less) willing to believe that the law of demand doesn’t apply to low-skilled workers – or that that law is, in the case of low-skilled workers, so routinely overwhelmed by exceptional circumstances that its effect is nullified – than any respectable biologist is to believe that natural selection doesn’t explain the phenotype and behavioral patterns of human beings.

Be Sociable, Share!
  • [4]
  • [5]
  • [6]
  • [7]
  • [8]

Comments

comments


Article printed from Cafe Hayek: http://cafehayek.com

URL to article: http://cafehayek.com/2013/10/data-is-unlikely-to-settle-this-dispute.html

URLs in this post:

[1] Commenting: http://cafehayek.com/2013/10/quotation-of-the-day-783.html

[2] this graph posted earlier this year by Stanford’s John Taylor: http://economicsone.com/2013/02/05/investment-unemployment-link-still-on-track/

[3] Sam Grove frequently asks in the comments section: http://cafehayek.com/2013/10/minimum-wage-legislation-lowers-the-cost-of-racial-discrimination-in-hiring.html

[4]

  • : http://twitter.com/intent/tweet?text=Data%20Are%20Unlikely%20to%20Settle%20This%20Dispute%20-%20http%3A%2F%2Fcafehayek.com%2F2013%2F10%2Fdata-is-unlikely-to-settle-this-dispute.html%20

    [5]

  • : https://mail.google.com/mail/?view=cm&fs=1&to&su=Data%20Are%20Unlikely%20to%20Settle%20This%20Dispute&body=http%3A%2F%2Fcafehayek.com%2F2013%2F10%2Fdata-is-unlikely-to-settle-this-dispute.html&ui=2&tf=1&shva=1

    [6]

  • : http://www.stumbleupon.com/submit?url=http%3A%2F%2Fcafehayek.com%2F2013%2F10%2Fdata-is-unlikely-to-settle-this-dispute.html&title=Data%20Are%20Unlikely%20to%20Settle%20This%20Dispute

    [7]

  • : http://www.google.com/reader/link?url=http%3A%2F%2Fcafehayek.com%2F2013%2F10%2Fdata-is-unlikely-to-settle-this-dispute.html&title=Data%20Are%20Unlikely%20to%20Settle%20This%20Dispute&srcURL=http%3A%2F%2Fcafehayek.com%2F2013%2F10%2Fdata-is-unlikely-to-settle-this-dispute.html&srcTitle=Cafe+Hayek+where+orders+emerge

    [8] Image: http://www.blinklist.com/index.php?Action=Blink/addblink.php&Url=http%3A%2F%2Fcafehayek.com%2F2013%2F10%2Fdata-is-unlikely-to-settle-this-dispute.html&Title=Data%20Are%20Unlikely%20to%20Settle%20This%20Dispute

    [9]

  • : http://www.myspace.com/Modules/PostTo/Pages/?u=http%3A%2F%2Fcafehayek.com%2F2013%2F10%2Fdata-is-unlikely-to-settle-this-dispute.html&t=Data%20Are%20Unlikely%20to%20Settle%20This%20Dispute

    [10]

  • : http://reddit.com/submit?url=http%3A%2F%2Fcafehayek.com%2F2013%2F10%2Fdata-is-unlikely-to-settle-this-dispute.html&title=Data%20Are%20Unlikely%20to%20Settle%20This%20Dispute

    [11]

  • : http://news.ycombinator.com/submitlink?u=http%3A%2F%2Fcafehayek.com%2F2013%2F10%2Fdata-is-unlikely-to-settle-this-dispute.html&t=Data%20Are%20Unlikely%20to%20Settle%20This%20Dispute

    [12]

  • : http://sphinn.com/index.php?c=post&m=submit&link=http%3A%2F%2Fcafehayek.com%2F2013%2F10%2Fdata-is-unlikely-to-settle-this-dispute.html

    [13] Tweet: https://twitter.com/share

  • Copyright © 2011 CafeHayek.com. All rights reserved.