In this video, review a number of the potential sources of error that might creep into your own analyses. Once you know what they are, you can do your best to avoid them.
- [Instructor] Statistical analysis might seem like an exact science, but if you've ever tried to apply statistics in real life, you know that in fact it is not. In this movie I'd like to review a number of the potential sources of error that might creep into your own analyses. Once you know what they are, you can do your best to avoid them. One common source of error is to not use random samples. A famous example of this came from the 1936 United States presidential election where a telephone poll of subscribers to Literary Digest projected that Alfred Landon would beat Franklin Delano Roosevelt by a wide margin.
In fact, Roosevelt won nearly two-thirds of the popular vote. The error came from two sources. The first is that Literary Digest was a conservative publication, which biased the results, and also that the poll was conducted by telephone. In 1936, only the financially well-off had home telephones, so that biased the result as well. You can also fall prey to investigator bias. It's easy to anticipate what your data will tell you. That's normal, but you shouldn't let those expectations affect your judgment.
Many interesting discoveries come from the moment when you look at your data and think, "That's funny," because the results don't fit your preconceived notion. You could also have a problem of working with older data. In the digital age, it's hard to know what old data actually means. Even 10 years ago it might've been data that was six months to a year out of date. These days, if you change projects or if you update a website, old data might mean a couple of days ago. And finally, you should never base your policy on a survey or experiment with a small sample.
If you're testing a change to your website or perhaps a new product, you want to get as many opinions as you can so you have a better idea of how the public, or at least the audience you're trying to reach, will react to what you're offering. The more input you can get, assuming that you follow all of the other guidelines, the better your sample will be.
- Distinguish between the mean, median, and mode.
- Describe the relationship between variance and standard deviation.
- Identify a nondirectional hypothesis.
- Point out the difference between COVARIANCE.P and COVARIANCE.S.
- Explain correlation.
- Analyze Bayes’ rule.