iSixSigma

One Tail Two Tails

Six Sigma – iSixSigma Forums Old Forums General One Tail Two Tails

This topic contains 2 replies, has 3 voices, and was last updated by  Bill Craig 13 years, 9 months ago.

Viewing 3 posts - 1 through 3 (of 3 total)
  • Author
    Posts
  • #40940

    AT
    Participant

    How to pick up the right P value for interpreting two sample or paired t-tests? How to know whether the one tail or the two tails value is to be considered?

    0
    #127970

    Robert Butler
    Participant

     If detecting a significant difference in either direction is important use the two tailed test. If your only concern is detection of a difference in one direction then use the one tailed test.
    Two Tailed – I have a drug and a proposed replacement. I care about knowing if there is an improvement but it is also important for me to know if there is a decrease in the effectiveness of the new drug vs. the old one.
    One Tailed – I”ve got a new machine and all I care about is will it do the job faster than the old one

    0
    #128529

    Bill Craig
    Participant

    Ramesh,
    Make sure you construct your null and alternate hypotheses before you do your experiment, and the selection of the correct p-value will be clear. Robert suggested some good examples.
    State Ho, Ha, and the decision rule up front. (ie reject Ho if t-calculated is greater than t-critical.)

    0
Viewing 3 posts - 1 through 3 (of 3 total)

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