If you do not have this background AND cannot do the following,
you SHOULD NOT consider taking these courses.
PSYC 225 -- Psychological Statistics
PSYC 480 -- Intermediate Statistics
PSYC 480 -- Intermediate Statistics Math and Stat Review (opens in a new window or tab)
Statistics Survival Guide
There are a small number of basic terms, concepts, and calculations that you must know if you're going to survive in statistics. You should not just commit them to memory. Rote memorization probably will not help you very much. Rather, you should make yourself so familiar with these ideas that you know what the terms mean without having to think about them. You should be so familiar with these concepts that they are second nature to you. This will not be enough to ensure an A in the course, but it will be a good start.
Population vs. Sample
Generally in the behavioral sciences, we are interested in how people behave and why they behave that way. To find this out, we watch them or do experiments using them as subjects, and we write down what we see or measure. We collect data on them, in other words. We cannot collect data on them all, however. There are over 6 billion of us after all! So we select a subset of people to study. This subset is the sample. We aren't really interested in what's true about the sample. Rather, we are interested in what the sample can tell us about all people, the population. The population does not have to be "all people." It can be all children under age 12, or all third graders, or all men, or all American Indians, for example, depending upon what our interests are. It can also be all mice, all olive baboons, all golden hamsters, etc. These groups--these populations--represent the general case. The subgroup--or sample--that we include in our studies represents a specific example of the general case. We use statistical analysis to try to determine if what we see in the specific example (sample) is also likely to be true in the general case (population).
Parameter vs. Statistic
A parameter is some number that tells us what is true in the population--i.e., in the general case. Parameters are usually denoted by Greek letters. An example is the population average or mean, symbolized by the lower case Greek letter mu. (It looks like the English letter u with a little tail on it: µ.) The values of these population parameters are usually not known for certain, but we often assume we know them or make hypotheses (educated guesses) about them. A statistic is some number that tells us what is true in a sample--i.e., in a specific example of the general case. Statistics are usually denoted by English letters. The sample average or mean is symbolized by the ordinary letter x with a bar drawn over it in most statistics texts. (My HTML editor won't let me show you this one. Look in your textbook.) It is sometimes also denoted by the capital letter M. Sample statistics are often used to estimate the value of a population parameter. Further statistical calculations then allow us to determine how accurate this estimation is likely to be.
Descriptive vs. Inferential Statistics
We are usually interested in what is true of a population--the general case. But we actually study samples--specific examples of the general case. Once we collect all of our data from a sample, we need some way to summarize it. One way is to calculate an average or sample mean. You'll learn other ways to summarize sample data during the course. These numbers that we calculate from sample data with the intention of summarizing the data are called descriptive statistics or summary statistics. The general method used to summarize sample data is also called descriptive statistics. Once we have the sample data summarized, we then want to know what it says about the population. We might want to test a hypothesis about the population, for example. The general method used to make inferences about the population from sample data is called inferential statistics.
Probability
Probability expresses the idea of chance. The idea is that the chance of something happening can be quantified, as least roughly. A probability is always a number between 0 and 1. A probability of 0 means the event in question has virtually no chance of happening. A probability of 1 means it's virtually a sure thing. The same idea is expressed by percent chance. A 0% chance means it won't happen, while a 100% chance means it's a sure bet. Although it's not always used this way in practice, it's best if you think of a probability statement as being a prediction about the future. There is not much sense in talking about the probability of something that has already happened.
Variability
Variability is the soul of statistics. Not everyone or everything is the same. Take an attribute like height, for example. If you look at a roomful of people, you'll find some who are tall, some who are short, and some who are in between. That's variability. Almost any attribute we are interested in will display variability, so it would be a good idea to have a way to talk about and quantify it. During the course you'll be introduced to terms such as variance, standard deviation, and standard error. Pay close attention to these ideas. They are critical to your success in statistics.
A Warning!
Statistics is not a spectator sport. It is hard work. You can't learn statistics by wishful thinking or by sleeping with the book under your pillow. If this is your plan, DROP THE COURSE NOW! You won't do well. Just like hitting a tennis ball, or playing the piano, or learning your lines in a play, statistics can only be learned through hard work and practice, practice, practice. Do all the homework problems in the problem sets, and do them on your own. It does no good to copy what someone else has done, and it is especially useless to wait until the professor does them on the board to copy them down. We have a word for this in the teaching profession. It's called LAZINESS. If you're a lazy student who doesn't want to do homework, I strongly suggest that you consider taking a different course. You've been warned!
Return to Dr. King's Home Page