Argument Essay

If you have ever taken an introductory psychology course, you were likely required to participate in one or more psychological “studies.” These are behavioral science experiments in which a researcher tests a hypothesis by performing some kind of “experiment” on a number of subjects. Most of these are not at all harmful—many are merely questionnaires. Yet they are an important way in which the discipline of psychology develops a more complete understanding of what makes people tick. In these studies, the information gathered is typically generalized to a larger group than just those college students who took part. Here’s an example.

Men have more confidence in their scientific abilities than women have in theirs. A recent study of nearly 400 students analyzed the attitudes toward science of students enrolled in entry-level general education courses at the University of Wisconsin–La Crosse in the areas of personal confidence, usefulness of the subject, perception of the subject as a male domain, and perception of the teacher’s attitude. Males in the study were found to have more confidence than females, and females perceived science as a male domain more than men. Expected achievement and attitude toward science were shown to be strongly related. —UW-L Journal of Undergraduate Research IX (2006)

From what you learned in Chapter 5, you should recognize this argument as an inductive generalization. Inductive generalizations, you may recall, conclude that some, most, or all members of a particular group have some feature based on evidence that a portion of that group has the feature. In this chapter, you will learn how to evaluate arguments like these according to their strength, from strong to weak. In order to determine the strength of an inductive argument, however, you must first learn to perform an additional analysis of the argument’s parts by identifying the sample, target, and feature in the argument. Once you can do that, you will learn to apply the proper criteria for consistent and accurate argument evaluation.

From the film Obedience © 1968 by Stanley Milgram. © Renewed 1993 by Alexandra Milgram and distributed by Alexander Street Press

One of the most famous and controversial psychological studies, the Milgram experiments, measured the willingness of study participants to obey an authority figure who instructed them to harm another person. Critics consider these experiments unethical and abusive. Today, many institutions and policies exist to ensure that human subject research is ethical and legal.

Learning Outcomes

In this chapter, you will learn how to evaluate inductive generalizations by:

· Identifying the sample, target, and feature of an inductive generalization;

· Determining whether the sample represents the target; and

· Writing a Critical Précis of an inductive generalization.

Analyzing Inductive Generalizations

Because inductive generalizations conclude that some, most, or all members of a group have a feature based on the evidence that some portion of that group has that feature, all inductive generalizations have the same basic structure. That is, even though you will encounter them in a wide variety of differing formats, all of them can be translated into a general form for inductive generalizations.

As you can readily see, inductive generalizations contain a sample (S), a target (T), and a feature (F). The sample is a portion of an entire class or group, and it appears in the single premise. The entire group or class represented by the sample is called the target; it appears in the conclusion. Thus, the sample is a part of the target. The argument works by citing the fact that the sample has a particular feature and making the inference that the target will have the feature, too. Thus, the feature appears in both the premise and the conclusion. When analyzing inductive generalizations, you must learn to properly isolate and identify the sample, target, and feature, each of which plays a critical role in evaluating the arguments.

Let’s work through a few examples showing how to find the sample, target, and feature in an inductive generalization, starting with this one.

Life on Earth can exist only when sufficient oxygen is present. Therefore, all life in the universe can exist only when sufficient oxygen is present.

Here’s the Formal Analysis of the argument, highlighting the sample, target, and feature.

Because the feature is given as an attribute of both the sample and the target, it must be stated in such a way that the feature makes sense when attributed to the sample and also makes sense when attributed to the target. Here are the three parts as you should identify them.

· Sample: life on Earth

· Target: all life in the universe

· Feature: exists only when sufficient oxygen is present

Not all inductive generalizations appear in this general form, but they all should be translated to fit the pattern. They often contain extra claims, as in the following example.

During the summer, the university does not follow the same schedule as during the school year. This summer the library, the dining halls, and the recreation center are all closed on Fridays. Therefore, most buildings on campus are closed on Fridays during the summer months.

First, diagram this argument.

① During the summer, the university does not follow the same schedule as during the school year. ② This summer the library, the dining halls, and the recreation center are all closed on Fridays. Therefore, ③, most buildings on campus are closed on Fridays during the summer months.

The first thing to notice is that claim ③, is the conclusion. Claim ② supports claim ③, but claim ① does not—it is an extra claim. The fact that the university does not follow the same schedule as during the school year is not evidence that most buildings on campus are closed on Fridays during the summer months. So in the diagram you would leave claim ① out.

Hint!

All inductive generalizations will contain only one premise and a conclusion. To make each claim in the argument make sense on its own, you often must rely on information presented in any extra claims that accompany the argument. But remember that you should not include the extra claims in your analysis, since they are neither premises nor conclusions.

The Formal Analysis, then, looks like this:

Hint!

Even though the conclusion is about most buildings on campus, as you can see, the target is still all buildings on campus. The target of all generalizations is always all of the members of that class. The claim is that most of them have the feature and the others don’t—hence, the claim addresses all members of the target.

Here’s another example of an inductive generalization, this one resulting from an opinion poll.

An online survey in the Journal of Computer-Mediated Communication suggests that people who post comments of a personal nature on their blogs are likely to have gotten into trouble for things they’ve written. A total of 492 bloggers filled out the online surveys. Of these, some 70% reported that they had gotten into trouble with family and friends for something they had posted. Researchers point out that additional research is needed to better understand privacy concerns inherent in the practice of blogging.

First, diagram the argument.

① An online survey in the Journal of Computer-Mediated Communication suggests that people who post comments of a personal nature on their blogs are likely to have gotten into trouble for things they’ve written. ② A total of 492 bloggers filled out the online surveys. ③, Of these, some 70% reported that they had gotten into trouble with family and friends for something they had posted. ③ Researchers point out that additional research is needed to better understand privacy concerns inherent in the practice of blogging.

This argument contains two extra claims—claim ② and claim ③.

Exercise 8.1.

· Your Turn! Complete the diagram of the inductive generalization above.

Here is the argument, presented in a Formal Analysis.

Notice that information from claim ② —the number of participants—must be included in the premise in order for that claim to make sense on its own. Many surveys include a number of extra claims whose information must be incorporated into the single premise and/or the conclusion.

Exercise 8.2.

Diagram each of the following inductive generalizations. Then provide a Formal Analysis using the general form, and state the sample, target, and feature.

1. I’ve eaten lasagna and spaghetti. I thought both were delicious. Thus, I’ll most likely find all Italian pasta dishes delicious.

2. Albuquerque is an arid region trying to create a stable economic base to fuel future growth. Restriction on urban sprawl helped Albuquerque become a more attractive place for business investment. So restrictions on urban sprawl will probably have the same results for every large community in the desert Southwest.

3. How do most Americans refer to a group of two or more people? They use the words you guys. This is the conclusion of an online survey of English usage conducted by Professor Bert Vaux of Harvard University that has had over 30,000 participants. When people were asked how they addressed a group of two or more people, the largest number (42.5%) said they used you guys. Interestingly, y’all was given by 14% of the respondents.

4. Since mice and rats are both rodents and they each make great pets, most rodents make great pets.

5. Hybrid cars are likely to get more than 40 miles per gallon. This is because both the Toyota Prius and the Ford Fusion get over 40 miles per gallon.

6. You sometimes hear it said that people who have good cognitive skills are just born with them. I disagree. Given that the cognitive skills required for computer programming, playing a musical instrument, and playing chess are acquired only after diligent and consistent practice, all cognitive skills must be acquired by the same means.

7. According to a Salary.com report, employees are dissatisfied with their jobs. The survey included over 7,000 employees, 65% of whom said they were actively looking for another job.

8. I have visited five small towns in southern Chile, and every one of them has had an old Spanish cathedral. It appears likely that the majority of small towns in southern Chile have old Spanish cathedrals.

9. Recently a team of archaeologists found a previously undiscovered Native American campsite in Michigan that contained hundreds of hunting implements. Five of the several hundred arrowheads and other hunting implements found at the site were carbon-dated. They were shown to be over 2,000 years old. So it’s likely that the vast majority of the tools that were discovered at the site are a couple of thousand years old.

10. Salt marshes on the East Coast are among the most productive ecosystems in the country. The majority of them are in South Carolina. Unfortunately, these marshes are home to fewer birds and fish than in the recent past. It’s a good bet that most salt marshes on the East Coast are having similar problems.

Evaluating Inductive Generalizations

In Chapter 5, you learned that the first step in evaluating inductive arguments is to examine the structure of the argument. That is, you must ask yourself, If the premises were true, would the conclusion likely be true? To do this

, you must determine how well the sample represents the target. As the sample gets more representative, the argument gets stronger. Inductive generalizations lacking that kind of support from the premises are weak arguments.

Hint!

Recall from Chapter 5 that strong and weak are relative terms ranging from very strong to very weak. For this reason, the exercises in this section will instruct you to compare pairs of inductive generalizations to determine which argument is stronger.

The size of the sample is the most important factor in determining how well the sample represents the target in an inductive generalization. As a rule, the larger the sample is, the stronger the argument is. This is because larger samples tend to be more representative of the target. Of course, if an argument had the largest possible sample—that is, if all members of the target were in the sample—the premise would be identical to the conclusion. So in an inductive generalization, the sample will always be smaller than the target.

Exercise 8.3.

· Your Turn! Why must the sample always be smaller than the target in an inductive generalization?

At the other extreme, when the sample is much too small to offer even minimal support for the conclusion, the argument commits a fallacy called hasty generalization. For example, if I argue that because my cousin, her husband, and I all think Proposition 22 should be defeated, a majority of voters believe the proposition should be defeated, then my argument is a hasty generalization. This is because the sample consists of only three voters (my cousin, her husband, and me). A sample of three people does not offer enough possible differences of viewpoint and opinion to provide evidence regarding all voters in any state.

In order to see how the size of the sample can be evaluated in an argument, we will consider a pair of generalizations. Examine the first one.

The student newspaper conducted a survey of its readers to determine which video games were most popular. Over 70% of the 175 who responded to the random survey said that World of Warcraft was their favorite. Thus, World of Warcraft must be the most popular video game among all readers.

First, use a Formal Analysis to identify the sample, target, and feature.

Now, let’s consider the second argument.

The student newspaper conducted a survey of its readers to determine which video games were most popular. Over 70% of the 545 who responded to the random survey said that World of Warcraft was their favorite. Thus, World of Warcraft must be the most popular among all readers.

Again, use a Formal Analysis to identify the sample, target, and feature.

Both arguments have the same target (all readers of the student newspaper) and the same feature (said World of Warcraft was their favorite video game). However, the size of the sample is different in each argument. The first contains a sample of 175 readers, whereas the second contains 545. A larger sample provides more evidence for the conclusion than a smaller one. So, if the respondents were chosen in the same manner, the second survey would be the stronger argument.

Exercise 8.4.

· Your Turn! How do you use your identification of the sample, target, and feature to evaluate the strength of an inductive generalization?

Exercise 8.5.

For each of the following pairs of inductive generalizations, determine which is stronger and justify your evaluation.

1.

· A. I have visited 10 amusement parks, all of which had a wide selection of roller coasters. So I bet all amusement parks I visit will have a wide selection of roller coasters.

· B. I have visited two amusement parks, both of which had a wide selection of roller coasters. So I bet all amusement parks I visit will have a wide selection of roller coasters.

2.

· A. All professionals earn much higher than average salaries. This is because lawyers, doctors, engineers, and high school teachers all have advanced degrees, and all of them earn a much higher than average salary.

· B. All professionals earn much higher than average salaries. This is because lawyers and high school teachers have advanced degrees, and all of them earn a much higher than average salary.

3.

· A. I tasted three of the oranges from my tree, and all were sour. This tree must have very poor fruit.

· B. I tasted 10 of the oranges from my tree, and all were sour. This tree must have very poor fruit.

4.

· A. Grocery prices at a locally owned market are not much higher than those at one of the national supermarket chains. We bought an identical shopping basket of foods from the local market and from the chain market. The price of the basket of foods from the local market was higher by only 4%.

· B. Grocery prices at a locally owned market are not much higher than those at one of the national supermarket chains. We bought an identical shopping basket of foods once a week for a month from the local market and from the chain market. The price of the basket of foods from the local market was higher by only 4%.

5.

· A. The vast majority of students who eat in the dining commons think the food is quite good. A group of graduate students in sociology surveyed over 100 students who eat at least one meal a day in the dining commons, and 65% of them had no complaints about the quality of the food.

1. B. The vast majority of students who eat in the dining commons think the food is quite good. A group of graduate students in sociology surveyed 30 students who eat at least one meal a day in the dining commons

· , and 65% of them had no complaints about the quality of the food.

Although the size of the sample is our first concern, in order for a sample to be representative of the target, it also must contain a similar degree of whatever relevant diversity exists in the target. So, if, for example, you want to draw a generalized conclusion about some group you belong to (such as a sports club, local political group, or college major) and the group has both male and female members, the sample needs to include both some men and some women in proportion to the make-up of the group in order to be representative of the entire group. Thus, when you choose your sample, you need to make sure that both men and women are among the ones chosen.

One way to evaluate whether an argument contains representative diversity is by considering whether the make-up of the sample is likely to emphasize or diminish the viewpoints favoring one kind of answer. For instance, if the sample at the sports club was created by choosing every fifth or tenth name on the membership list, we can assume that the position of a person’s name on the list is unlikely to indicate how they would respond to your survey. So such a process should result in a strong argument. Another possibility for a strong argument like this would be to randomly interview people in a common meeting place. Of course, better and worse ways exist to create a random sample, but in each case, the aim is to ensure that the diversity of the target is reflected by the sample. What would not be a random sample is one that excludes part of the target. For example, a sample of sports club members chosen from those in the men’s locker room would not include any women members.

Hint!

Professional pollsters ensure a representative sample by making the sample random. Interestingly, sample size does not always have to be huge if the sample is random. For instance, most nationwide surveys done by reputable polling companies use samples of only 1,000 to 1,200 respondents.

By comparing two similar examples, you can see the difference between an argument with a representative sample and one with a sample that is not representative of the target. Let’s consider the first argument.

As part of a class project, I surveyed owners of a variety of businesses to see whether they expected to hire additional workers next year. Of the 36 owners, 20 said they had no plans to increase their workforce, 10 said they weren’t sure, and 6 planned to add employees. Thus, the majority of local business owners probably will not be hiring more workers next year.

First, use a Formal Analysis to identify the sample, target, and feature.

Now, consider the second argument.

As part of a class project, I surveyed business owners who had filed for bankruptcy to see whether they expected to hire additional workers next year. Of the 36 owners, 20 said they had no plans to increase their workforce, 10 said they weren’t sure, and 6 planned to add employees. Thus, the majority of local business owners will not be hiring more workers next year.

Again, use a Formal Analysis to identify the sample, target, and feature.

In these two arguments, the target is the same (all local business owners) and the feature is the same (do not plan to hire more workers next year). The same number of business owners was sampled in each argument. Even the conclusion is the same. What is different is the sample itself: The business owners in the first argument come from a variety of businesses, so the sample is representative. However, the business owners chosen for the second sample had all filed for bankruptcy and are, presumably, less likely to hire new workers. That means the sample is not representative. When an inductive generalization’s sample misrepresents the target, the argument commits a fallacy called a biased generalization. So this second version of the argument is weak because it is a biased generalization.

Hint!

Because biased generalizations and hasty generalizations are common mistakes in reasoning that are often rhetorically persuasive, they are referred to as fallacies. You will learn about other common fallacies in Chapter 11.

Samples may also be biased when surveys require participants to initiate contact rather than using a survey taker to actively solicit responses. For example, surveys requiring that participants respond by sending a text message, going online, or phoning in their response are likely to get unrepresentative results, since the respondents are self-selected. Only people who are particularly interested in the issue are likely to respond to the survey. To make matters worse, unless surveys prevent respondents from contributing their answers more than once, the data are likely to be skewed by unscrupulous repeat respondents who are trying to influence the outcome.

Putting it All Together

Writing a Critical Précis of an Inductive Generalization

In Chapters 2, 3, and 4, you learned how to recognize and analyze arguments. In Chapter 5, you also learned how to distinguish among different kinds of arguments. Now that you can evaluate inductive generalizations, you can put all of these skills together by adding inductive generalization evaluation to your Critical Précis assignments.

Directions for a Critical Précis of an Inductive Generalization

In paragraph form, use complete sentences and proper English grammar and spelling to do the following:

Step 1: Identify the passage.

· Completely and correctly identify the author and the source (whenever such information is given).

· If the passage contains an argument, identify it as such. Then move to Step 2.

· If the passage does not contain an argument, identify it as a nonargument. Then specify which feature of an argument is lacking.

Step 2: Analyze the argument.

· Clearly and completely identify the issue, conclusion, and premise or premises in that order.

· If the passage is a multiple argument, write separate paragraphs to analyze each separate argument.

Step 3: Diagram the argument.

· Verify that the diagram is consistent with your Critical Précis.

Step 4: Identify the kind of argument.

· If the argument is deductive, identify it as a categorical argument or a truth-functional argument.

· If the argument is inductive, identify it as an inductive generalization, an analogical argument, or a causal argument.

Step 5: Evaluate the argument.

· If the argument is an inductive generalization, evaluate its strength by considering how well the sample represents the target.

Let’s walk through a few examples of a Critical Précis before you try some on your own.

It appears that most of the teachers at East High School are very unhappy with the school district administration. We surveyed a dozen teachers who were recently laid off, and they complained about administrators.—The Telegraph

This passage is an argument, since it contains at least two claims, one of which offers reasons as evidence for the truth of one of the other claims. So you should provide a Critical Précis and an argument diagram.

It appears that ① most of the teachers at East High School are very unhappy with the school district administration. ② We surveyed a dozen teachers who were recently laid off, and ③, they complained about administrators .

This passage from The Telegraph contains an argument. The issue is whether most of the teachers at East High School are happy with the school district administration. The conclusion is that most of the teachers at East High School are very unhappy with the school district administration. The premise is that the dozen laid-off teachers who were surveyed complained about administrators.

Notice that the extra claim is left out of the analysis and diagram. Because the passage is an inductive generalization, you must next evaluate the argument.

This argument is an inductive generalization. It is weak because the sample is biased and small. The sample is not random—it consists only of teachers who were recently laid off. Since they are more likely to be unhappy, they do not represent all of the teachers. The sample size is only a dozen teachers, so it is too small to make a strong argument.

Here’s another example, this one a survey of the kind that is readily available in newspapers, magazines, and websites.

One Step Further

Stereotypes are generalizations about a group of people whereby we attribute a defined set of characteristics to this group. While stereotyping enables us to respond rapidly to situations, it often leads us make judgments about people that might not be true. For example, in the summer of 2011, a gunman killed 69 young people at a summer camp near Oslo, Norway. Terrorism experts initially proclaimed that this attack was the work of Al-Qaeda or another Muslim terrorist group. Who else, they initially thought, would be killing blond-haired, blue-eyed Norwegian children?

The reasoning of law enforcement officials seemed to follow something along these lines:

The most recent attacks on civilians in Western countries have been the work of Muslim terrorists. Thus, it is most likely that all attacks on civilians in Western countries are the work of Muslim terrorists.

1. Evaluate this argument. Is it strong or weak? Justify your evaluation.

2. The gunman, who later confessed, was a right-wing nationalist from Norway named Anders Breivik. Should knowing that fact change your evaluation of the argument’s strength? Why or why not?