Generalization: White Men Can’t Jump

A fourth type of inductive reasoning is known as generalizing. When engaging in it, we use multiple instances, recurring patterns, or repeated observations to form inferences about an entire class or category. In short, we’re forming and/or applying general rules.

"What makes you think 1 wouldn't be up for sushi?”

Figure 1.2 Sign reasoning in action. Cartoon by Leo Cullum, The New Yorker, November 17, 2008.

© Leo Cullum/The New Yorker Collection/www.cartoonbank.com.

Women Are From Venus, Men Are From Mars

For instance, comedians often segue into a joke by saying, “Men and women are different. . ..” Generally speaking, they are right. For example, men are much more willing than women to engage in casual sex. In fact, according to one series of studies (Clark & Hatfield, 1989, 2003), when college students were approached by a stranger of the opposite sex who asked to engage in casual sex, males consented more than half the time. Females rarely consented. What’s more, college males and females hold different beliefs about romance. Males, for example, are more likely to believe that “bars are good places to meet a potential mate” (Abowitz, Knox, Zusman & McNeely, 2009, p. 276).

Of course, not all females and males think and act this way. Generalizing is a risky business. Yet, life would be impossible if all knowledge were fragmented and particularized and we could not form generalizations. How could you set your alarm clock if you couldn’t generalize about how long it took you to get ready in the morning? How could you get from home to work if you couldn’t reasonably estimate the time required for your commute? Once at work, how could you interact with co-workers if you couldn’t generalize about social norms governing communication? We hope you see our point. The key is to form accurate generalizations, qualify them appropriately, and articulate them to varying degrees, rather than seeing the world in black and white.

Universal Generalizations: All Unicorns Are White

Universal generalizations use terms such as “all,” “every,” “never,” and “always.” For example, one partner might tell the other “You never listen to me” or “You always have to get your way.” A friend once told one of the authors “Every year, an Indian kid (meaning a child whose family is from the Indian subcontinent) wins the national spelling bee.” “Every year?” the author asked. “What about Evan O’Domey?” (who won in 2007). As a matter of fact, in 2017 Ananya Vinay became the 13th consecutive South Asian-American (including co-champs from 2014—2016) to win the Scripps National Spelling Bee (Harper, 2017), which is a long way from “every year” yet still consistent with the generalization.

Whatever the case, as critical thinkers, we’d be well-advised to regard universal generalizations with suspicion. The saying “there’s an exception to every rule” acknowledges this guideline. Those who don’t run the risk of committing a fallacy known as a sweeping generalization, by lumping everyone together in the same group. Cultural and ethnic stereotypes, for example, often entail this fallacy, which we’ll return to in the next chapter. For now, however, it’s enough to know that, technically speaking, a single exception disproves a universal generalization. Imagine, for instance, that a business owner claimed, “We haven’t had a single case of sexual harassment in the workplace.” By documenting one instance of harassment, you would refute the claim. Practically speaking, however, a single exception may not be that damaging, hence the saying “the exception that proves the rule.” Suppose, for example, that a wife tells her husband, “We’ve been married 10 years, but you never remember our anniversary!” Would the husband be off the hook if he replied, “Not so, I remembered our 1st anniversary”?

Contingent Generalizations: It Depends

Considering this, generalizations that are qualified are a much safer bet. Specifically, by qualifying generalizations with words like “most,” “often,” and “many,” you reduce the odds that your claim will be contested. Thus, instead of saying, “We always spend Thanksgiving with your parents,” a spouse might say, “We’ve spent the last three Thanksgivings with your family.” And instead of telling a student “You never come to class on time,” a professor might say, “You’ve been tardy four of the last six class meetings.”

Another way to qualify generalizations is to make them contingent on a particular set of circumstances. For example, when spelling words, the letter “i” usually comes before the letter “e,” however the guideline, “‘i’ before ‘e,’ except after ‘c’,” specifies the conditions under which this is not the case. Even this general rule has exceptions, however. In the word “science,” for example, the rule is broken. As another example, a pediatrician might tell a parent, “Aspirin is generally safe for children over the age of four, unless they have a fever, virus, or flu-like symptoms, in which case they may develop Reye’s syndrome.” Clearly, specifying the conditions under which a generalization holds true makes for a stronger argument.

Statistical Generalizations: Say It with Numbers

Another way of avoiding the pitfalls of universal generalizations is to provide statistical generalizations or averages. For example, an arguer might say “76% of felons are recidivists” (meaning that they commit crimes after being released from prison), rather than “all felons are recidivists.” Similarly, an arguer might say “America ranks ninth in obesity worldwide and first among industrialized nations,” rather than “America is a nation of fatsoes.”

Two Types of Generalizations

From the Whole to the Part

In the process of making an argument, there are two ways you can advance a generalization. In the first, you start from a general rule and then apply it to a specific case. For example, a father might advise his young son, “Being left-handed is an advantage in some sports. As a lefty, you might think about taking up baseball, fencing, soccer, or tennis.” The general rule, about lefthanders having an edge in some sports, is applied to a specific case, his son. Of course, the son may or may not excel in these sports, but the father is trying to stack the odds in his son’s favor by relying on a general rule.

From the Part to the Whole

In the second way to advance a generalization, you start with examples or a sample case and then form the generalization. Suppose Rita meets an LDS (Mormon) mom who loves to make casseroles, Jell-O desserts, and drink hot chocolate. Then she meets another LDS mother who also bakes casseroles, makesJell-O, and drinks hot chocolate. When she meets a third LDS mom who does the same, she may form the generalization that “LDS moms have a thing for casseroles, Jell-O and hot chocolate.” Of course, forming a generalization about a group based on a sample of three members could entail a fallacy known as a hasty generalization, which we discuss in the next chapter. That is why when pollsters survey people and conduct opinion polls they use a large sample of people, who are selected at random, and who represent a true cross-section of the larger class of people about whom they wish to generalize.

Example: Gimme a For Instance

A final type of inductive reasoning that we consider is argument by example, which involves offering anecdotal evidence about your own experience, or that of another, to prove a point. Arguments based on example are actually a sub-set of generalization, in which one or a few examples are offered as support for a claim about a larger class or category. Each example serves as proof for the larger generalization. Argument by example is so pervasive, however, that we believe it deserves its own place within the taxonomy of inductive reasoning.

When offering an example, the assumption is that the example is typical or representative of some larger class. If not, the arguer might be accused (or guilty) of “cherry picking” examples to suit his or her purpose. Testimonials, for instance, that accompany ads for weight loss products are subject to such bias. But how typical is their experience? How about the buff ripped models in fitness commercials? Do you really believe their physiques were ordinary before going to that gym or buying some exercise equipment?⁷ Nyet! Logically speaking, a single example doesn’t prove much. However, from a persuasive standpoint, a single example can be quite compelling. Statistics can be abstract, confusing, or daunting, but a single well-developed example can put a name or a face on a problem. Indeed, anecdotal evidence can be so powerful that people may give more weight to stories than they deserve (Kida, 2006). When using examples, then, a useful approach is to begin with an example, then provide statistics to show that it is not an atypical case (see Gass & Seiter, 2018). When challenging another arguer’s example, you can offer negative or counter-examples of your own. You also can show that the examples offered are neither typical nor representative.

Summary

Inductive reasoning is the stuff of which everyday arguments are made. The types of informal reasoning discussed in this chapter are part and parcel of ordinary arguments. Inductive arguments always require some sort of inferential leap and some leaps are more probable than others. A variety' of types of cause—effect reasoning were presented. Among these types, a necessary cause should not be confused with a sufficient cause and vice versa. Sole causes are somewhat rare in public policy controversies. Most social problems have complex causes. Analogies draw comparisons between a source domain and a target domain. They are useful for fostering critical and creative thinking and are highly persuasive. Analogies may be literal or figurative. The latter pose more difficulties for arguers because they compare different classes of things. When evaluating analogies, the focus should be on structural rather than superficial features. Sign reasoning is based on indicators whose presence predicts something else. Sign reasoning, however, cannot explain why one thing predicts another. Signs may be fallible or infallible. The latter are less common in ordinary arguments. Signs are also cumulative; the more signs the more likely that they mean something. Generalizing, while risky, is also inevitable. Contingent generalizations and statistical generalizations are preferred over universal ones. Generalizations may operate by applying a general rule to a specific case, or by forming a general rule based on a specific case or cases. Argument by example is a subset of generalization, with a sample of one. Although offering a single example is logically suspect, a well-developed example can resonate with people. People respond to stories and examples can be compelling.

Notes

• 1 Constant conjunction is also referred to as concomitant variation or cause-effect contingency (Buehner & McGregor, 2009).
• 2 There may be real physical cause—effect relationships, but humans still must infer their existence. With respect to physical phenomena, cause—effect relationships may be so direct that inferences are easy to draw. For example, if a person turns a light switch on and off, the person is closing and opening an electrical circuit.The flow of electricity (cause) makes the bulb glow (effect), while an interruption in the circuit makes the bulb dim (effect).This seemingly obvious causal connection must still be inferred, however. Suppose a light switch was not working. Instead, the circuit is connected to a motion detector. If a person enters a room and flips the switch on, the person would likely infer that flipping the switch turned the light on. In actuality; it is the person’s movement that activated the motion detector and turned on the light.
• 3 For you vampire aficionados, we recognize that the ways a vampire can be killed depend on whether one is reading Bram Stoker, Anne Rice, Stephanie Meyer, or some other source. We’re not sure Count Chocula can be killed!

• 4 There are actually three types of Down Syndrome; Trisomy 21, Mosaic Down Syndrome, and Translocation Down Syndrome, but all three are the result of abnormal cell division in chromosome 21.
• 5 For convenience sake, we’ve grouped all arguments based on comparisons, including metaphors, similes, parables, and precedents, under the rubric of analogy.
• 6 Some critics (e.g., Rudy, 2014) have suggested that, although the results of the hurricane study are interesting and may be correct, additional research is necessary. Specifically, because almost all hurricanes before 1979 had female names, its possible that the study’s results are due to improvements in forecasting hurricanes rather than the masculinity/femininity of names. If this is the case, the author’s conclusions are a good example of faulty causal reasoning.
• 7 As an example, Bowflex holds casting calls for fitness models. For example, Gregg Pitt, one of the Bowflex models, has graced more than 100 magazine covers. Another, Andrew Brown, also endorses numerous other fitness products. Kristia Knowles, one of the females in the Bowflex commercials, is also a spokesperson for other fitness products.

References

Abowitz, D.A., Knox, D., Zusman, M. & McNeely, A. (2009). Beliefs about romantic relationships: Gender differences among undergraduates. College Student Journal, 45(2/Part A), 276—284.

Adler, R.B. & Proctor, R.F. (2011). Looking out, looking in (13th Ed.). Boston, MA: Wadworth.

Alter, A. (2013). Drunk tank pink: And other unexpected forces that shape how we think, feel, and behave. New York: Penguin Books.

AmericanRhetoric.com (2017). John G. Roberts, opening statement to the Senate Judiciary Committee. Retrieved on November 1, 2010 from: www.americanrhetoric.com/speeches/johnrobertsenatejudiciaryaddress.htm.

Barnes, R. & Eggen, D. (2010, June 29). Supreme Court expands gun rights. Washington Post, p. A-l.

Barnum, C. & Perfetti, R.L. (2010). Race-sensitive choices by police officers in traffic stop encounters. Police Quarterly, 13(2), 180-208, doi: 10.1177/1098611110365690.

Blanchette, 1. & Dunbar, K. (2001). Analogy use in naturalistic settings: The influence of audience, emotion, and goals. Memory and Cognition, 29(5), 730—735, doi: 10.3758/BF03200475.

Buehner, M.J. (2005). Contiguity and covariation in human causal inference. Learning & Behavior, 33(2), 230-238, doi.org/10.3758/BF03196065.

Buehner, M.J. & McGregor, SJ. (2009). Contingency and contiguity trade-offs in causal induction. International Journal of Comparative Psychology, 22(1), 19—22, doi: 10.1037/a0020976.

Busloft', S.E. (1993). Comment: Can your eyes be used against you? The use of the horizontal gaze nystagmus test in the courtroom. Journal of Criminal Law & Criminology, 84(1), 203-238, doi: 10.2307/1143891.

Chandler, J., Griffin, T.M. & Sorensen, N. (2008). In the “I” of the storm: Shared initials increase disaster donations. Judgment and Decision Making, 3(5), 404—410, doi: 2008-09227-005.

Clark, R.D. & Hatfield, E. (1989). Gender differences in receptivity to sexual offers. Journal of Psychology and Human Sexuality, 2(1), 39—55, doi: 10.1300/J056v02n01_04.

Clark, R.D. & Hatfield, E. (2003). Love in the afternoon. Psychological Inquiry, 14(3), 227-231, doi: 10.1207/S15327965PLI1403&4_8.

Ducheyne, S. (2008). J.S. Mill’s canons of induction: From true causes to provisional ones. History and Philosophy of Logic, 29(4), 361-376, doi: org/10.1080/01445340802164377.

Dunbar, K. (2001). The analogical paradox. Why analogies are so easy in naturalistic settings, yet so difficult in the psycholog)' laboratory? In D. Gentner, J.K. Holyoak, & B.N. Kokinov (Eds), The analogical mind perspectives from cognitive psychology (pp. 313—334). Cambridge, MA: MIT Press.

Gass, R.H. & Seiter, J.S. (2018). Persuasion, social influence, and compliance gaining (6th Ed.). New York: Routledge.

Gentner, D. (2003). Analogical reasoning, psychology of. In L. Nadel (Ed.), Encyclopedia of cognitive science (pp. 106—111). London: Nature Publishing Group.

Gentner, D., Rattermann, J. & Forbus, K. (1993). The roles of similarity in transfer: Separating retrievability from inferential soundness. Cognitive Psychology, 25(4), 524-575, doi: org/10.1006/cogp.1993.1013.

Glassbrenner, D. (2003). Estimating the lives saved by safety belts and air bags. National Center for Statistics and Analysis, National Highway Traffic Safety Administration, Paper no. 500. Retrieved April 2, 2005 from: www-nrd.nhtsa.dot.gov/pdf/nrd 01/esv/esvl8/CD/Files/ 18ESV-000500.pdf.

Glick, M. & Holyoak, KJ. (1980). Analogical problem solving. Cognitive Psychology, 12(3), 306—355, doi: org/10.1016/0010-0285(80)90013-4. ”

Hannigan, S.L. & Reinitz, M.T. (2001). A demonstration and comparison of two types of inference based memory errors. Journal of Experimental Psychology: Learning, Memory, and Cognition, 27(4), 931—940, doi: 10.1037/0278-7393.27.4.931.

Harper, B. (2017, June 2). Why have Indian-Americans won the last 10 national spelling bees? Fatherly.

Retrieved on December 14, 2017 from: www.fatherly.com/love-money/indian-americans-dominate-scripps-spelling-bee.

Hiltzik, M. (2010, October 31). Why big-time CEOs make terrible politicians. Los Angeles Times, p. B-12.

Holyoak, KJ. (2005). Analogy. In KJ. Holyoak & R.G. Morrison (Eds), The Cambridge handbook of thinking and reasoning (pp. 117-142). New York: Cambridge University’ Press.

Jung, K., Shavitt, S., Viswanathan, M. & Hilbe, J.M. (2014). Female hurricanes are deadlier than male hurricanes [Abstract]. Proceedings of the National Academy of Sciences of the United States of America, 111, 8782-8787, doi: 10.1073/pnas/140278611.

Karpinski, A.C. & Duberstein, A. (2009, April 16). A description of Facebook use and academic performance among undergraduate and graduate students. In poster presented at the meeting of the American Educational Research Association, San Diego, CA.

Kida, T. (2006). Don’t believe everything you think: The 6 basic mistakes we make in thinking. Amherst, NY: Prometheus Books.

Koenig, C.S. & Griggs, R.A. (2004). Analogical transfer in the THOG task. Quarterly Journal of Experimental Psychology, 57/1(3), 557-570, doi: 10.1080/13546780342000070.

Krawczyk, D., McClelland, M.M., Donovan, C.M., Tillman, G.D. & Maguire, MJ. (2010). An fMRI investigation of cognitive stages in reasoning by analogy. Brain Research, 1342,67-73, doi: org/10.1016/j. brainres.2010.04.

Lander, E.S. & Ellis, JJ. (1998, November 5). Founding father. Nature, 396, 13-14, doi: 10.1038/23802.

Retrieved February 23, 2019 from: www.nature.com/nature/journal/v396/n6706/full/396013a0. htmL

Lewis, J. (2004). Thomas Jefferson and Sally Hemings redux. William and Mary Quarterly, 57(1), 121 — 124, doi: hdl.handle.net/10822/504513.

Lorenz, E.N. (1972). Predictability: Does the flap of a butterfly’s wings in Brazil set off a tornado in Texas? Presentation to the American Association for the Advancement of Science, Washington, DC.

Mankiewicz, J. (writer) & Medak, P. (producer) (2004, December 21). The Socratic method. House, M.D., Fox TV.

Miller, K. (2007). Racial profiling and postmodern society: Police responsiveness, image maintenance, and the left flank of police legitimacy. Journal of Contemporary Criminal Justice, 23(3), 248—262, doi. org/10.1177/1043986207306868.

National Safety Council (2010). Odds of dying from. Itasca, IL: National Safety Council. Retrieved September 26, 2010 from: www.nsc.org/news_resources/Documents/nscInjuryFacts201 l_037.pdf.

Osborne, H. (2003). It’s like what you already know: Using analogies to help patients understand. The Boston Globe’s On Call Magazine. Retrieved December 14, 2017 from: http://healthliteracy, com/2003/01/01/ analogies.

Porter, T. (2017, September 10). Don’t fire guns at Hurricane Irma, Florida police warn. Newsweek.com. Retrieved on October 30, 2017 from: www.newsweek.com/dont-fire-guns-hurricane-irma-florida-police-warn-662494.

Prusher, I.R. (2001, May 29). Some unexpected relief for endangered species. Christian Science Monitor, p. 7.

Rice, D. (2014, June 2). Ladykillers: Hurricanes with female names deadlier. L7S/1 Today. Retrieved on December 27, 2017 from: www.usatoday.com/story/weather/2014/06/02/hurricane-female-names-deadly/9868413.

Rudy, K. (2014, June 6). Do the data really say female-named hurricanes are more deadly? [blog post]. The Minitab Blog. Retrieved on December 27, 2017 from: http://blog.minitab.com/blog/ the-statistics-game/do-the-data-really-say-female-named-hurricanes-are-more-deadly.

Shanks, D.R., Pearson, S.M. & Dickinson, A. (1989). Temporal contiguity and the judgment of causality by human subjects. Quarterly Journal of Experimental Psychology Section B: Comparative and Physiological Psychology, 41(2), 139-159, doi: 10.1080/14640748908401189.

Sharp, P.M. & Hahn, B.H. (2011). Origins of HIV and the AIDS pandemic. Cold Spring Harbor Perspectives in Medicine, 1(1), a006841, doi: 10.1101/cshperspect.a006841. Retrieved February 23, 2019 from: http://perspectivesinmedicine.cship.org.

The Economist (1998, May 30). Why rhinos recommend Viagra. The Economist, 347(8070), p. 76.

von Hippel, W., von Hippel, F.A., Chan, N. Sc Cheng, C. (2005). Exploring the use of Viagra in place of animal and plant potency products in traditional Chinese medicine. Environmental Conservation, 32(3), 235-238, doi: 10.1017/S0376892905002353.

Warren, P.Y. (2010). The continuing significance of race: An analysis across two levels of policing. Social Science Quarterly, 91(4), 1025—1042, doi.org/lO.llll/j.1540-6237.2010.00747.x.

White, E.B. & White, K.S. (Eds) (1941). A subtreasury of American humor. New York: Coward-McCann.

Young, J.R. (2009, April 24). Facebook, grades, and media hype. Chronicle of Higher Education, 55(33), p. A-13.

Young, M.E. (1995). On the origin of personal causal theories. Psychonomic Bulletin & Review, 2(1), 83-104, doi: 10.3758/BF03214413.

Chapter 8

Fallacies in Reasoning, Part I