# Hypothesis Testing

Six Sigma – iSixSigma › Forums › Old Forums › General › Hypothesis Testing

- This topic has 11 replies, 8 voices, and was last updated 14 years, 10 months ago by Mikel.

- AuthorPosts
- February 10, 2005 at 7:27 pm #38373
All:

I have data of volume of orders received on a daily basis and I am trying to see if there is variation by day. Is the receipt of my orders an X towards cycle time. I want to show if it is statistical or not. I can plot out by day my receipts, use an X-Bar chart or a zone chart but what kind of hypothesis test can I use? My data is normal data for that month (p-value > .05). I am assuming that my H0=No difference by receipt date and my H1= There is a statistical difference towards cycle time and volume receipt timing is a contributor towards cycle time. What type of hypothesis test could I run?

Thanks all

Julie0February 10, 2005 at 10:46 pm #114737

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

be notified via email.Julie,

It sounds interesting but I don’t understand what you are trying to test. Can you give more detail?0February 11, 2005 at 4:19 pm #114761Julie,The simplest answer would be multiple t-tests. For example, pick some time period you are interested in (for example, the past month or year-to-date). Take all of the data for Mondays and all of the data for Tuesdays and run a t-test on the data. Repeat for M vs W, etc and see if there are any significant differences.The problem is that as you get lots of categories (n=7 days in this case) you get lots of comparisons ( n(n-1)/2 = 21 in this case). This means that there is a pretty good chance of a false positive. The better technique here would be ANOVA. It is designed to say whether all the days are the same, or whether there is SOME difference in the various categories.Tim F

0February 11, 2005 at 4:46 pm #114763Doesn’t ANOVA deal with multiple categories? Multiple t tests result in a large cumulative error since each test has some assumed error. Julie has not provided enough info to give her any useful advice.

0February 11, 2005 at 7:42 pm #114772I think Tim’s answer has some validity. Hypothesis testing can be very confusing to those who have not used it in practical applications. Even though she has not given enough details to determine the best way to approach the problem, at least he has given her some possibile ways to categorize her data.

Sometimes people simply do not know enough about a subject to provide the experts with useful information, and a different approach can be helpful in generating a spark. :) I find different perspectives helpful when there is a subject I do not understand (and there are many!).

Lass0February 11, 2005 at 8:24 pm #114774Your generalizations have a couple of flaws. First, the poster was asking for a specific statistical approach to a specific data problem. If the poster does not know enough about the data then they have no business attempting an analysis until they understand their research question. Carnell gave the appropriate response. Tim’s answer was wrong. You do not do multiple t tests when you have multiple categories since each test has an error and by doing them in series, you compound that error. Anova is the tool for multiple Xs if you are looking for differences in categories. The confusing nature of the original post wasn’t even clear enough to determine whether he/she was looking to test differences or to establish relationships. The first case does call for some hypothesis test but the second might require regression.

0February 11, 2005 at 11:04 pm #114781

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

be notified via email.Thank you.

0February 17, 2005 at 12:46 am #114988Point well taken. I always find your responses well constructed and enlightening.

Lass0April 7, 2005 at 9:39 pm #117395I know what a hypothesis test is but what is the purpose?

0April 7, 2005 at 9:39 pm #117396I know what a hypothesis test is but what is the purpose?

0April 7, 2005 at 10:57 pm #117399

This site is a sweet deal…Member@This-site-is-a-sweet-deal...**Include @This-site-is-a-sweet-deal... in your post and this person will**

be notified via email.Hypothesis testing refers to the process of using statistical analysis to determine if the observed differences between two or more samples are due to random chance (as stated in the null hypothesis) or to true differences in the samples (as stated in the alternate hypothesis). A null hypothesis (H0) is a stated assumption that there is no difference in parameters (mean, variance, DPMO) for two or more populations. The alternate hypothesis (Ha) is a statement that the observed difference or relationship between two populations is real and not the result of chance or an error in sampling.

Hypothesis testing is the process of using a variety of statistical tools to analyze data and, ultimately, to fail to reject or reject the null hypothesis. From a practical point of view, finding statistical evidence that the null hypothesis is false allows you to reject the null hypothesis and accept the alternate hypothesis.

Courtesy of the isixsigma website, “New to Six Sigma” link, and look for Six Sigma terms and definitions. Now you try…0April 7, 2005 at 11:03 pm #117400My experience is that a hypothesis test’s main purpose is to thoroughly confuse students worldwide regardless of age. i have seen it serve this purpose more than any other.

0 - AuthorPosts

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