iSixSigma

problem with proportion tests

Six Sigma – iSixSigma Forums Old Forums General problem with proportion tests

Viewing 3 posts - 1 through 3 (of 3 total)
  • Author
    Posts
  • #47134

    Whitehurst
    Participant

    I am reading a study on the relationship between decision-making and different categories jobs. In the methods section the researcher wrote:
     
    Statistical analyses using proportion tests were conducted to determine the significance of the differences in the participants¡¦ decision-making in different categories of jobs. Since the results were presented as correctness percentages (i.e. proportions of correct decisions) rather than as different scores for different levels of correctness, proportion tests were regarded the most suitable statistical package for the study, as they are designed to determine the differences between two proportions with different sample sizes.
     
    I have a question for the researcher but I have no way to ask him about it. I would just like to know what a ‘proportion’ test is (the researcher¡¦s name for the type of statistical analysis) – is that an ANOVA or something else?  Or is the researcher using a series of Z-tests (and why these types of tests).
    Thank you for your help.

    0
    #156824

    Erik L
    Participant

    Joe,
    Proportions test can be used in scenarios where one is conducting a hypothesis of a population’s percentage relative to either a specified % or to assess a difference of one population’s percentage to another.  The simplest explanation that I can give is that these would be the discrete-data equivalents of a t-testing scenario that would have been appropriate for continuous data. 
    Regards,
    Erik

    0
    #156845

    Whitehurst
    Participant

    Thank you for your help Erik.

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

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