# P-Value

Six Sigma – iSixSigma › Forums › Old Forums › General › P-Value

- This topic has 8 replies, 5 voices, and was last updated 18 years ago by Mike Carnell.

- AuthorPosts
- November 20, 2001 at 9:09 pm #28255

Reynol DiazParticipant@Reynol-Diaz**Include @Reynol-Diaz in your post and this person will**

be notified via email.P-value is a kind of hypothesis test. Also it means the number of samples whose intervals will not catch the parameter.

That is, if you have 100 samples and your confidence coeficient of 95%,you will have 100 intervals, but only 95 of them (samples) will have the paramater and 5 of them do not. It is the meaning of P-value.0November 21, 2001 at 3:29 pm #70101Im not sure if this message was from a previous thread or just a declaration

A p-value is a characteristic of a statistic that defines a hypothesis test. It is not a kind of test.

For example, the test statistic for a t-test is (ybar-mu0)/(s/sqrt(n)), where mu0 defines the hypothesis. This test has a p-value for a given set of data based on this statistic. Its not like there is some different statistic that would yield a t-tests p-value.

Your interpretation is interesting. A common interpretation of a p-value is “the probability that we would get a more extreme test statistic if the null hypothesis is true”. So, if a p-value is 0.05, you could say that the chances are 1 in 20 that you would get more extreme results if you repeated the experiment.

I think your interpretation is for the confidence coefficient of confidence intervals

I highly recommend the Cartoon Guide to Statistics by Gonick and Smith (in all seriousness). I think that is does a great job of bringing these concepts to a common sense level0November 23, 2001 at 10:26 pm #70152

Mike CarnellParticipant@Mike-Carnell**Include @Mike-Carnell in your post and this person will**

be notified via email.I am not sure I comprehended your explanation. Let me give you a reference. “Basic Statistics for Business and Economics” by Lind and Mason. ISBN 0-256–19408-4. P values are defined on pages 268 – 269.

A p value is also a hypothesis test called proportion testing. We also have p values in attribute control charts. I don’t think from your description that these are what you are interested in.0November 24, 2001 at 4:56 am #70165This question has already been answered – see the earlier postings. Incidentally, you’re confusing proportions (the p’s in a proportion chart) with the lowest attainable significance level – which is what the p-value is (formally).

0November 24, 2001 at 4:01 pm #70166

MikeCarnellParticipant@MikeCarnell**Include @MikeCarnell in your post and this person will**

be notified via email.I gave him a reference for p value.

I am not confusing anything. Try understanding that p is used in more than one way.0November 24, 2001 at 4:45 pm #70167

Mike CarnellParticipant@Mike-Carnell**Include @Mike-Carnell in your post and this person will**

be notified via email.I went back and read the answer and you are incorrect. I had not been answered. The original question had to do with someone who was attempting to comprehend and internalize a concept. The response was a definition. They are two different things. In the original question the guy was already at a higher level on the adult learning model (Comprehension) that the person who responded with a definition (knowledge) or at least the response was on a lower level.

Since I didn’t feel I adaquately understood the original message I provided a reference. Next time you hip shoot a response that something has been answered you might want to think it through.

Now as far a p value. The original question mentioned hypothesis testing. The author of the question is obviously working hard at comprehending the statistical side and getting to the point it becomes useful to him. P is used in many places and in different ways. He seems to have the inclination to work through that. People who have only the ability to regurgitate “formal” definitions typically are of marginal use to someone who is working on comprehension.

Before you assume I am confused you might want to consider if what you believe is help really is. You may be confused. It is the difference between mentoring some one and being “shove it to them” instructor.0November 26, 2001 at 3:29 am #70178Kindly note that my definitions are entirely in keeping with the Encyclopedia of Statistics and proper statistical practice in quality control. Feel free to refer to textbooks by Montgomery, Box, etc. Correct definitions are imperative for statisticians to avoid confusion, and I feel that as such should be encouraged, not dissuaded, especially on a public site such as this.

Happy Thanskgiving.0November 26, 2001 at 3:32 am #70179or even Thanksgiving.

0November 26, 2001 at 5:53 am #70180

Mike CarnellParticipant@Mike-Carnell**Include @Mike-Carnell in your post and this person will**

be notified via email.Kindly note it doesn’t have anything to do with looking something up in a book. The person asking the question had translated the idea of P value into his own words. He has gone from Knowledge (reciting the book definition) into comprehension the next level UP in the adult learning model. You aren’t offering help because he has moved past the point where you are operating.

As far as how you feel about what people do or do not promote, you are entitled to your opinion. Interesting but not really relavant to what I do.

Happy Thanksgiving.0 - AuthorPosts

The forum ‘General’ is closed to new topics and replies.