I was recently intrigued by this passage regarding time-series data from The Grammar of Graphics, section 3.2:

Among the many prevailing views of the role of empirical data in modern science, there are two opposing extremes. On the one hand, the realist assumes data are manifestations of latent phenomena. In this view, data are pointers to universal, underlying truths. On the other hand, the nominalist assumes data are what they describe.


The last line is a little hard to interpret, I concede that and we should all bear that in mind. The definitions of realists needed some clarification, but the original phrasing of the book sounded very philosophical, so I will omit the exact wording, but suffice to say, the spirit of the distinction as per the author is as follows: nominalists view data as self-contained, exogenous entities whereas realists view data as pointers to underlying truths.

[End Edit]

I'm not 100% clear on how these perspectives clash in practice. From the passage, it would seem they are diametrically opposed

With a given data set, let's just say commodity prices (anything, like steel, copper, wheat), the realist narrative would sound something like: "Behold, this data is consistent with supply and demand as we theorized. The invisible hand is everywhere!" (ok maybe I'm exaggerating)

I'm not quite sure what the nominalist would say here. Maybe: "It's all in the eye of the beholder. The data are the data. You realists are overthinking it."

But I still wonder, maybe there are certain situations where a state of ambiguity is possible. The book later raised the work of Skinner (1969) and that further provoked my curiosity.

Question: Assuming my characterization of realists and nominalists are fair and/or accurate (please tell me if they are not), are there any significant inferences/insights in my example (commodity prices) that would change based solely on one's realist/nominalist association?

  • You may support/clarify your answer with statistical assumptions
  • Try to be specific about what inferences or insights would be contingent upon the realist/nominalist divide (i.e. population size, correlation/causation, ect)

Optional: And, as a bonus question are there any occasion where the two views agree?

migrated from stats.stackexchange.com May 1 '18 at 18:07

This question came from our site for people interested in statistics, machine learning, data analysis, data mining, and data visualization.

  • This sounds rather like the distinction between the pursuits of modeling and testing on the one hand and, on the other hand, exploration and description. It's difficult to say, though, because the final sentence in the quotation, taken literally, is a laughable position: nobody confuses a description of something with the thing itself. It must mean something else that is revealed further on. – whuber May 1 '18 at 16:13
  • @whuber I take your point. That is where it leaves off, oddly enough. That probably added to my confusion. I will see if I can paraphrase the later parts after I am sure of the position. – Arash Howaida May 1 '18 at 16:15
  • You might want to read the relevant entries in Stanford's Encyclopaedia of Philosophy: scientific realism, realism (in metaphysics), nominalism in metaphysics, the medieval problem of universals. – gung May 1 '18 at 16:55
  • @gung I tried adding a few stipulations to the question in hopes that a quantifiable answer can emerge. I asked the answerers to explain what inferences/insights are at stake. Depending on your view of data, what aspects of statistical inference would be most affected. The philosophical overtone is still there, but at least there is a baseline for answers now. I think the expertise is best suited for stats SE, but that's just me. – Arash Howaida May 1 '18 at 17:30
  • They don't make the answers statistical. Philosophers of science have some familiarity with science, statistics, etc., to understand how their theories connect to issues in practice. Someone with the chops to provide a good answer to the philosophical issues that are the core of this question should be able to provide a concrete example on the level asked, but not vice-versa, IMHO. I'm going to migrate this. – gung May 1 '18 at 17:42

In sciences, such situation has been encountered where the set of experimental data leads to a variety of conclusions and the analyst has to sieve through the data with background disposition as to his seemingly 'subjective choice' but a choice based on the history of such investigations.

In selecting/choosing commodity prices - such biases do work. When a fresh crop is arriving the price of the 'produce' is lowest, when the seasonal demand period is coming to the prices shoot up.

I'm not quite sure what the nominalist would say here. Maybe: "It's all in the eye of the beholder. The data are the data. You realists are overthinking it."

the following quotes from Millikan's measurement of charge of an electron can throw some light on the issue-

From the feature article "In Defense of Robert Andrews Millikan" by David Goodstein (American Scientist, January-February 2001):

Awkwardly, an examination of Millikan's private laboratory notebooks indicates that he did not, in fact, include every droplet for which he recorded data. He published the results of measurements on just 58 drops, whereas the notebooks reveal that he studied some 175 drops in the period between November 11th, 1911 and April 16th, 1912.

In a classic case of cooking, the accusation goes, he reported results that supported his own hypothesis of the smallest unit of charge and discarded those contrary results that would have supported Ehrenhaft's position. And, to make matters very much worse, he lied about it.

Millikan's 1913 paper contains this explicit assertion: "It is to be remarked, too, that this is not a selected group of drops, but represents all the drops experimented upon during 60 consecutive days, during which time the apparatus was taken down several times and set up anew." (Emphasis in the original). Thus, Millikan is accused of cheating and then compounding his cheating by lying about it in one of the most important scientific papers of the 20th century.

The author defends some of Millikan's actions.

[...] More than one of the entries in his notebooks show the result of a computation and then the comment "very low something wrong," perhaps with an indication of what Millikan thought might have disturbed the measurement. Needless to say, such entries were not included in the 58 drops Millikan published.

At first glance, this procedure certainly appears questionable. But one needs to dig deeper. The notebooks also contain a calculation with the comment "This is almost exactly right, the best one I ever had!!!" And yet Millikan did not include this drop either in his crucial 1913 paper. These discarded measurements, the good and the bad, were all part of a warm-up period during which Millikan gradually refined his apparatus and technique, in order to make the best determination possible of the unit of electric charge. The first observation that passed muster and made it into print was taken on February 13th, 1912, and all of the published data were taken between then and April 16th. This period of roughly two months is what Millikan refers to when he talks about "60 consecutive days," although the interval was actually a bit longer (63 days), in part because 1912 was a leap year.

During these nine weeks, Millikan recorded in his notebooks measurements on roughly 100 separate drops. Of these, about 25 series are obviously aborted during the run, and so cannot be counted as complete data sets. Of the remaining 75 or so, he chose 58 for publication. Millikan's standards for acceptability were exacting. If a drop was too small, it was excessively affected by Brownian motion, or at least by inaccuracy in Stokes's law for the viscous force of air. If it was too large, it would fall too rapidly for accurate measurement. He also preferred to have a drop capture an ion a number of times in the course of observation, so that he could investigate changes as well as total charge, which had to be an integer multiple of the fundamental unit, e.

[...] He had no special bias in choosing which drops to discard: Allan Franklin of the University of Colorado reanalyzed Millikan's raw data in 1981 and discovered that his final value for e and for its margin of error would barely have changed had he made use of all the data he had, rather than just the 58 drops he selected.

the details given above challenges the 'realist' view and boils down to nominalist approach. surprisingly in 21 st. century the 'fudging' of data has become fact of life and established scientific norms and ethics are being flouted.

rehttps://hsm.stackexchange.com/questions/2756/is-millikans-famous-oil-drop-experiment-a-fraud/2759#2759f.- >


According to Wikipedia, it would be helpful to frame the conflict between nominalism and realism according to the problem of universals, which might be stated thusly:

Are properties (such as bigness, color, superiority, and quality) and other abstract objects real? Or do they only exist as names in thought and speech?

A realist would say the properties are real. Plato was a realist. A nominalist would say that the properties are not real, however much they appear in speech and thinking. Another stance opposed to realism is anti-realism, as enunciated by Michael Dummett.

Back to your question: how would this debate play out over commodity prices? On the face of it, seemingly a realist would believe that the price---a consensual monetary value applied to a lump of wheat or lead or empty cattle wagons---is a real object that truly exists; this real object can be causally related to other objects.

Seemingly a nominalist would concede that the price is there on paper, but it is not itself real: it is the name people give to the dollar value of some future economic transaction; it has no causal connection to anything in and of itself.

I suppose that today we would find realism to be a bit of a strange theory. With realism, there exists some real standard of "big", "green", "best", etc. against which every thing may be measured objectively. Where would this standard exist? What is its origin? Realists would need to tangle with all of these objections.

Your Answer

By clicking “Post Your Answer”, you agree to our terms of service, privacy policy and cookie policy