text | sound | contrast | print
Creased Earlobe
COURSE TOPICS
PRACTICE PROBLEMS
COURSE SERVICES
[an error occurred while processing this directive]

since 11.15.04

 

Scientific Reasoning

Department of History and Philosophy
Learning Center M33
Lander University
Greenwood, SC 29649

Criteria for Analogical Arguments

  1. Analogical arguments are inductive arguments whose conclusion follows from the premisses with some degree of probability.

    1. The conclusion of an analogical argument does not follow with necessity. If the conclusion were to follow with certainty, the so-called analogy would not be analogous but would be a description equivalent to the original circumstances.

    2. There are several factors which affect the probability of analogical arguments as better or worse.

    3. The recognition of the presence or absence of these factors affect the appraisal of the argument in a qualitative rather than a quantitative manner.

  2. As an illustration for the following criteria consider the following example:

    • a Smith, b Jones, c Wilson, and d Johnson have P creased earlobes, Q depressed sternums, and R narrow hips.

    • a Smith, b Jones, and c Wilson all have the further property of S heart disease
    • _____________________________________________

    • Therefore, d Johnson probably has S heart disease as well.


    1. The number of entities, situations, or occasions (among which the circumstances are said to hold in the second premiss).

      1. The number of specific instances are the entities listed as what might be called the data base, i.e., in the example, a, b, and c listed in the second premiss.

      2. The number of entities or instances is not precisely related to the probability of the conclusion.

      3. This criterion is similar to the notion of induction by simple enumeration—the more examples enumerated, the higher the probability (ceteris paribus).

      4. Example: Specifically, the more the number of occasions of persons who have the further property of heart disease (in addition to the properties in the first premiss), the better the evidence. E.g. more persons, say White, Zindler, and McCall also were found to have creased earlobes, depressed sternum, narrow hips, and heart disease.

    2. The number of respects, properties, or attributes in which the entities are analogous (in the first premisss).

      1. The probability increases by the number of respects denoted by P, Q, and R

      2. The number of respects is not precisely related to the probability with which the conclusion follows.

      3. The respects listed in the first premiss are neither known to have an invariable connection, nor are they know to be unconnected.

      4. Example: The probability increases as the number of respects increase, such as the addition of experiencing chest pain and blackouts.

    3. Strength of the conclusion relative to the premisses.

      1. If the stated conclusion is  hedged, conservative, more cautious, or guarded relative to the premisses, the probability of the argument becomes stronger.

      2. "Playing it safe" or "under-rating" what could be said increases the expectation the conclusion is probable.

      3. Example: The probability of the conclusion increases with the following sequence of possible properties to be substituted in the conclusion: coronary heart disease ⇒ heart disease ⇒ heart trouble.

    4. Absence of the number of disanalogies or points of difference between the entities in data base with respect to the conclusion entity.

      1. The argument is weaker if there are many disanalogies between a, b, c, and with d in the above schema.

      2. In other words, similarities between entities in the data base and the conclusion strengthen the argument.

      3. Example: The probability increases if the person named in the conclusion (i.e. Johnson) is from the same culture and race as the persons named in the second premiss data base.

    5. The more dissimilar the entities in the data base, the stronger or more probable the argument.

      1. The conclusion is made more probable if the entities are varied because the data base becomes broader and more inclusive.

      2. Hence, disanalogies between entities in the data base and the entity in the conclusion are minimized by including more disimilar entities.

      3. Example: Smith, Jones, Wilson, and Johnson (the persons named in the first premiss) are from different cultures and races.

    6. The respects or attributes are relevant to the entities or situations.

      1. This criterion is the most important.

      2. The points of resemblance must be relevant to the conclusion drawn.

      3. Relevancy is often determined by a suspected causal or determining effect.

      4. Example: Depressed sternum and heart disease are thought to be associated with congenital heart defects.

Example adapted from

Elliott WJ, Powell LH. Diagonal earlobe creases and prognosis in patients with suspected coronary artery disease. Am J Med. 1996 Feb;100(2):205-11.

Shamberger RC, Welch KJ, Castaneda AR, Keane JF, Fyler DC. Anterior chest wall deformities and congenital heart disease. J Thorac Cardiovasc Surg. 1988 Sep;96(3):427-32.

Return to Scientific Reasoning Homepage

http://philosophy.lander.edu/scireas/FILE.html

Top of Page

[an error occurred while processing this directive]

Contact webmaster@philosophy.lander.edu
Please see the disclaimer concerning this page.
This page last updated 11.21.04
©2004 Licensed under the GFDL

 
Valid XHTML 1.1!    Valid CSS!