Incomplete Evidence: Incomplete Comparison





This variant of Incomplete Evidence occurs when an argument supporting a comparative conclusion depends on ignoring or suppressing relevant evidence. It has the following general form:

Premise 1: A and B are compared.

Premise 2: Evidence relevant to the comparison is ignored, suppressed, or excluded.

Conclusion: Therefore, claim C about the comparison is true.

Like the standard Incomplete Evidence, this fallacy does not arise because the presented premises fail to logically support the conclusion. The fallacy persuades by conveying the impression that relevant information has not been ignored or excluded. While this fallacy can be committed in ignorance, it is often used in bad faith efforts. While there are far too many ways to make such incomplete comparisons to cover in this book, I will briefly discuss some common methods.

One method is to make a comparison using percentages while leaving out other numbers that would provide a context. This can be used to create the false impression of a significant difference when the actual difference is relatively small. For example, if you saw a headline reporting that cases of Squid Pox had increased 600% worldwide since last year, then you might be worried and accept the conclusion that it is a serious health threat. But if you learned that the increase was from one case to six, you would presumably be less inclined to accept that conclusion.

This method can also be used to downplay the seriousness or significance of something. For example, someone might point out that Black people made up only 27% of those shot by the police in 2021 and conclude that there is not a significant issue with racism and police violence.  But this reasoning ignores a relevant fact, that only 13% of Americans are Black. While there are those who argue that this disproportionate percentage is warranted, ignoring the population data would result in committing this fallacy.

Another method is to ignore differences in such things as standards, definitions, and reporting and recording practices when making the comparison. This can be used to create the appearance of a relative increase or decrease. For example, it might be claimed that efforts to combat a disease have been unsuccessful because the number of infected people has increased, but this increase in numbers is due to the change in the definition of the disease or due to more accurate recording and reporting of cases.

This method can be used to make a comparison seem favorable or unfavorable simply by ignoring such relevant differences. For example, an administration might claim that they have decreased unemployment relative to their predecessor but leave out the fact that they have redefined what counts as being unemployed.

A third method is to make a comparison while ignoring that the things being compared are not comparable. That is, relevant differences are simply being ignored.  For example, imagine that someone simply compares past marathon times with current times, and conclude that today’s athletes are better than past athletes (not just that they have better times). But if they ignore factors such as technological advances in running shoes, improvements in sports hydration and nutrition, then they risk committing this fallacy.

As would be suspected, this method can be an effective bad faith technique to mislead people. For example, a politician might argue that the minimum wage should not be increased because it is already much higher than when they were a kid mowing lawns in the 1980s. But they are leaving out a critical difference between then and now, namely the effects of inflation. While minimum wage is higher today in terms of the amount of money paid, it has less buying power.

As another example, imagine that someone wants to claim that the retail industry is being devastated because there has been an 88% increase in shoplifting. But they neglect to mention that this increase is relative to 2020, when the COVID-19 pandemic lockdowns and disruptions were in effect. They also neglect to mention that relative to 2019, recorded thefts have decreased.

A fourth method is leaving out key information when the comparison involves averages. While a useful tool, averages can be very misleading. For example, two cities might have the same average temperature, but this could be because one city has extreme lows and highs while the other city has a consistent temperature. As such, concluding that you would have the same experience in each city based on just the average would be an error. There are also different types of average which are calculated differently. Leaving out the type of mean being used in a comparison can result in this fallacy being committed. Leaving out the numbers used to calculate the mode can also result in this fallacy occurring.

When you think of an average, you probably think of the mean. The mean of a set of numbers is calculated by adding up the numbers in the set and dividing it by the number of members of that set.  A second type of average is the median, which is the number in the middle of a set of numbers. There are as many numbers of the set larger than the median as are smaller. A third type is the mode. In a set of numbers, the number that occurs most frequently is the mode.

Regardless of the type of mean used, extremely different sets of numbers can have the same (or similar) means. For example, the sets {12, 13, 70, 70, 250, 500} and {70, 70, 70, 70} would have the same mode. While this can lead to good faith errors, it can also be exploited in bad faith in this fallacy. For example, a professor could make a bad faith argument to address student complaints about bad grades by truthfully saying that the average of the class is 75 which is higher than the expected average of 70. The professor simply neglects to mention that they are using the mode and does not provide (anonymous) scores for the entire class.

Defense: To avoid committing or falling victim to this fallacy, the basic defense is to consider whether relevant information has been left out of an argument with a comparative conclusion. In the case of comparisons involving percentages, you would need to know the other numbers that would provide needed context. In cases in which standards, definitions, and reporting and recording practices matter when making the comparison, you would need to know what these methods are and whether they are the same for the things being compared. You should also consider whether the items are comparable and if such information has been left out. Finally, if the comparison involves an average, you will need to know the type of average being used, the numbers used to calculate it and any other relevant context.

If your argument is incomplete, you can fix it by adding the relevant information. If you encounter this fallacy, you should suspend judgment about the conclusion unless you can fill in the missing evidence. As always, even if this fallacy is committed it does not follow that the conclusion of a fallacious argument must be false. To think otherwise is to fall for the Fallacy Fallacy.

Example #1

News Anchor #1: “There was an 88% increase in shoplifting this year relative to last year. As you can see in this clip, a bold thief is just riding his bicycle down the aisle, grabbing merchandise, and mocking those trying to stop him.”

News Anchor #2: “This is why so many stores are closing in American cities. They simply cannot afford the losses they are suffering. This is why we cannot have nice things.”

Example #2

News Anchor #1: “In shocking news, health experts have reported a 600% increase in cases of Squid Pox. This terrible disease jumped species from squids to humans after a 300% increase in squid attacks last year.”

News Anchor #2: “How big of a threat is this disease?”

News Anchor #1: “Huge. As I said, cases increased 600% and squid attacks were up 300%.”

News Anchor #2: “I’m staying out of the ocean!”

News Anchor #1: “They can get you in the tub. Or shower.”

News Anchor #2: “Really?”

News Anchor #1: “As far as you know.”

Example #3

Student: “Professor, I talked to many of the other students, and they did badly in your course. They asked me to talk to you about this. If so many of us did badly, we should get a chance at extra credit or something.”

Professor Belial: “I appreciate you bringing this to my attention. I want to assure you that I review the course average regularly and compare it to past classes. Your class’s average is 77, which is significantly better than the standard average of 70. Based on this, the class is doing better than expected and no extra credit is needed.”

Student: “But we’re doing badly. If you aren’t going to offer any extra credit, what should we do?”

Professor Belial: “Study harder.”

Student: “That’s it?”

Professor Belial: “Yes.”

Example #4

Governor: “I am pleased to announce that because of my reform of unemployment and my Work Don’t Shirk plan, there are 33% fewer people on unemployment in our great state. This should silence my woke critics who complain that we aren’t doing enough to help the workers. Obviously we are doing a lot.”

Originally appeared on A Philosopher’s Blog Read More