Inductive Reasoning
I've noticed that every time I kick the ball up it comes back down. Therefore, I would argue that the next time I kick the ball up, it will come back down (to see how one might arrive at the same conclusion through deductive reasoning, click here)
Short explanation: Induction is the logical process of drawing general conclusions from observations of particulars.
Longer Explanation
Think back to the last time you left the CRMC campus to travel home. You likely stopped at the intersection of HWY 1 and Ocean View Drive. When the light turned green, you also likely proceeded into/through the intersection--either on your feet, on your bike, or in a car--and you likely did this before every single car coming from the North and South had come to a complete stop. Think about the risk you took: you put yourself directly in the path of oncoming traffic. You saw tons (literally) of heavy machinery rolling toward the intersection and you deliberately put yourself in its path. Why did you do such a thing?
Maybe you did so because you know that when your traffic light turns green, the traffic lights for the opposite traffic turn red, causing the drivers to stop their cars and making it safe for you to go. Probably true. But how did you know that the traffic lights for north- and southbound traffic were red in that specific intersection at that particular time? Did you check them both before you proceeded? Maybe you drove into the intersection because you could see that the oncoming cars were slowing down as they approached the intersection and looked as though they intended to stop before entering the intersection. Sensible enough. But how did you know those cars were going to come to a complete stop and not, instead, suddenly accelerate into the intersection and collide with your car or run you over?
You likely don't have much of an answer for doing what you did except this: In a thousand or ten-thousand previous instances when your light turned green, the traffic lights for the opposite traffic were red and those slowing cars did, in fact, come to a complete stop. You noticed a pattern, and so you made a kind of wager. You said to yourself, "I'll bet those lights and those cars do exactly what I've seen other lights and cars do before." And, hopefully, they do.
When you engage in this kind of thought process, you're using inductive reasoning. You are making observations (in the case above, it would be observations of a relatively small sample of cars and traffic lights in intersections in America), you are noticing patterns in your observations (i.e. when one light turns green, the others are red; and cars slowing for a red light stop), and you are drawing general conclusions (i.e. at intersections, when one traffic light is green, the lights for cars traveling in different directions are red and it is safe to proceed).
Think about it another way. Imagine that you travel to a foreign country and you're walking around downtown. There are road signs but you cannot read them. You are standing on a corner waiting to cross the street, and the lone car traveling toward you stops across the street, and the driver appears to wave you across the road. You cross safely and continue down the sidewalk until you come to another intersection. There is the same road sign at this intersection that was in the last one where the driver stopped. You see a car coming down the street at 30 mph or so. Here's the test: do you step out into the street to cross before the driver has come to complete stop and waves you to go? How many intersections would you need to cross (in other words, how many observations would you need to make) in order compare drivers' behaviors (in other words, to find patterns) and, ultimately, draw conclusions about what all or most drivers do in this country when they approach intersections with this signage?
As you can probably see, we use inductive reasoning many, many times each day--often without even being aware of it. We have to use induction to make choices in any situation that isn't governed by clear, already existing, relevant rules. However, it is important to emphasize here that when you reason inductively, you are drawing general conclusions based on just a sampling of evidence. In other words, when reasoning inductively, you are not looking at a complete body of evidence--just a part of it. In the example above, for example, you did not observe every car ever driven at every intersection in the US. You observed a sliver of a sliver of the evidence and drew a conclusion you'll apply to all or most intersections and cars.
The Elements of Inductive Reasoning
The Evidence (or "Sample"): The body of information or observations one is working with.
The Interpretation: The critical analysis of the evidence to find its meaning and/or to find patterns.
The Conclusion: The position or claim one arrives at through the process of interpreting the evidence.
Sample Inductive Process
Evidence or Sample: A polar bear you saw in the LA County Zoo; polar bears you saw on a National Geographic TV special; pictures of polar bears you've seen in various magazines; polar bears you've seen in movies.
Interpretation: All the polar bears in the sample are white.
Conclusion: Polar bears are white.
Induction and Probability
Conclusions arrived at by induction are not 100% guaranteed true because the sample does not consist of all possible evidence. It involves an incomplete sample. Because of this, conclusions arrived at by induction are only probable, and the degree of probability in an inductive conclusion is dependent on the quality of the sample and the logic of the interpretation.
The quality of a sample increases when it is
Known (or verifiable): If the sample has or can have verifiable existence it is known. Polar bears exist, as do automobiles and intersections. Therefore, it is possible to draw reliable conclusions about them. Space aliens, on the other hand, do not have verifiable existence (they have only speculative existence).
Sufficient: Generally, the more evidence one has, the more likely the conclusion will be reliable; the less evidence one has the greater the likelihood of prejudice.
Representative: The sample is representative when it is typical of the whole class of things being studied.
Common Mistakes in Inductive Reasoning
Insufficient Sample: How large a sample needs to be varies. If the stakes are high, the sample should be big. For example, if we are going to generalize about races or sexes or cultures, we should work with a very large sample (otherwise, we run the risk of creating hurtful racist and sexist stereotypes). One the other hand, if the stakes are low, a smaller sample may suffice. If, for example, we are looking for a place grab a bite after work one day, perhaps one good experience at a restaurant is enough to warrant going back. However, if we're looking for a good restaurant to take our boss to lunch, we may hesitate to pick a restaurant based on just one piece of evidence; the importance of the lunch may cause us to rely on a "sure thing"--i.e.. a restaurant with a larger collection of evaluations (and these evaluations may be our own or others we are aware of).
Sample insufficiency is often the weakness of conclusions drawn from polling data. The purpose of polls is to discover, through the questioning of a relatively small group of people, something about a mass. A conclusion like "44% of Americans think the President is doing a good job" or "86% of the people in Fort Bragg want a Wal-Mart to locate in their town" has been drawn from an interpretation of only part of the population. Not every American or every resident of Fort Bragg was questioned; only a small percentage were, and the bigger the number, the better the conclusion can be. Polling 100,000 Americans instead of 10,000 is going to allow you to draw more reliable conclusions about attitudes toward the president.
Unrepresentative Sample: For a sample to be "typical," it has to reflect the qualities of the whole. Think of the president approval polling example above. If the pollsters questioned 1,000,000 people, that might be a very good (i.e. sufficient) sample size; however, if a disproportionate number of that 1,000,000 were Black or Asian or members of the upper-middle class or residents of the Midwest, etc. the sample would be unrepresentative and the conclusions would be unreliable. In other words, one cannot draw reliable conclusions about a diverse group of people if that diversity is not represented in the sample. For our logic to work, we would have to either alter our conclusion (to something like "44% of Asian Americans on the West Coast think the president is doing a good job") or make the sample more representative.
Another example: if biologist Jane wants to draw conclusions about the spawning habits of Coho salmon, she would have to collect a sample (she cannot, after all, observe every Coho). As indicated above, the bigger the sample the better for interpretation. So let's say Jane, working with a research team, collects data on 500,000 Coho. Super good sample size. But what if the whole sample was made up of only fish spawning in Northern California rivers? Could she draw conclusions about Coho in general? Can California Coho represent Coho in British Colombia's rivers? What if Jane's samples are only from the last two years? Does the last two years of Coho represent typical Coho behavior?
Illogical Interpretation: One may have a sufficient, representative sample and still blow it. One of the most common ways to blow it is by assuming connections or cause/effect relationships that don't actually exist. So, for example, say I have a big sample of data on convicted murderers, and I notice a pattern: 85% of them indicate that they did not finish high school. Can I conclude from this that dropping out of high school leads to murder? How about this one: My friends all own cars that get better than average gas mileage. All their cars are silver. Can I conclude from this that silver cars get better than average gas mileage?
Take Home Quiz: Type your responses to each of the following.
1. You're traveling through a town you've never been in before, and you're in the mood for Mexican food. You see four restaurants in a plaza: "El Patio," "Eats," "Seabreeze Cafe," and "Sheshiang Garden." Which do you choose? What is your rationale?
2. Someone wants to bet you $50 on the following: A person who knows nothing about CR or its employees comes into the CR main office and sees Charlie Claybrook's nameplate. Will this person assume Charlie is male or female? There is no rule about what names have to be given to which sex, so why does it seem like a lock to bet that this person will assume Charlie is male? What pattern are you counting on this person to have recognized in his/her previous experiences with people named Charlie?
3. An English instructor at CR asked 80 math students in Todd Olsen's evening class whether or not CR needs to offer more evening library hours. 45 of these students said more hours are needed. Is it safe to draw a fairly general conclusion like "A majority of CR students believe the college should offer more evening library hours" from this data? Why or why not?
4. Let's say I take a philosophy class at CR. I do poorly on all the exams, turn in every assignment late, and never participate in class. Is the instructor on safe ground when she concludes that I'm not college material?
5. I've seen many of Quentin Tarantino's films: Reservoir Dogs, Pulp Fiction, Jackie Brown, and Kill Bill Vol. I. I haven't seen Kill Bill Vol. II and some others he's made. I've noticed quite a bit of graphic violence in the films I've seen. Am I safe in drawing the following conclusion: Many of Tarantino's film's depict graphic violence? How about this one: Quentin Tarantino likes violence. How about this one: Modern Hollywood films often contain graphic violence.
Example of Inductive Argument