Sage: We probably need to review the basic statistical definitions. The first one is mean .
Heather: Mean – the average result of a test, survey, or experiment.
Sage: Thank you, Heather. OK. So what would the mean be if 7 people take a test (10 points possible). Their scores are: 4, 7, 7, 7, 8, 9, and 10. Anyone?
Heather: So, let me see, the average is the sum of the scores (52) divided by the number of people (7). The result is… 7.43. So that's the mean, right?
Sage: Correct. The next term is median .
Deena: That's the score that divides the results in half.
Heather: So, continuing with our example, 7 people take a test (10 points possible). Their scores are: 4, 7, 7, 7, 8, 9, and 10. The score 7 is the middle value when we put the scores in order. So we can say…
Brian: The median score on the test is 7.
Jose: Very good, Brian!
Brian: Hit me with another definition, Sage.
Sage: OK, Brian, how about mode ?
Brian: The most common result of a test, survey, or experiment.
Sage: You've been studying, haven't you? OK, Heather, using the same example, what would the mode be?
Heather: If we look at the scores, we see that 7 was the most common score on the test (3 people got a 7). So, we can say that the mode was 7, right?
Jose: Bravo! This isn't that hard. Even Brian's been doing well.
Sage: Well, things are going to get more difficult soon. Who wants to define significance ?
Heather: I can take that one. The measure of whether the results of research were due to chance.
Sage: So what about p-value ?
Brian: That would be the way in which significance is reported statistically. For example, p<.01 means that there is a less than 1% chance that the results of a study are due to random chance.
Sage: Correct! Are you trying impress us, Brian?
Heather: OK Brian, what do p-values need to be in order for a study to make any strong claims based on the results?
Brian: Usually the p-value needs to be pretty low - .01 and .05 are common.
Sage: How about correlation ?
Jose: The degree to which two factors appear to be related, I believe.
Brian: So, what is an r-value , Jose?
Jose: Ummmmm, I don't remember right now. Let's see, I used to know what an r-value was…
Brian: How about the way that correlation is reported statistically. It's a number between –1 and +1. Does that sound familiar?
Jose: Yeah, that's it. If r=0, there is little or no correlation between two things. When the number is higher, the correlation between two things is greater. Generally, r-values should be >.3 in order to report a significant correlation.
Sage: But, correlation should not be confused with causation. Just because two factors are reported as being correlated, we cannot say that one factor causes the other.
Heather: Well, that was fun. So that's it. Those were all of the terms and our time is up. See you later!
Return to the previous page by closing this window.