Summary

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.

| iStudy | Module Descriptions | Getting Access | Contacts  | FAQ