• US
  • China
  • France
  • DACH
  • Italy
  • Japan
  • UK
  • JMP
  • JMP Software
    • JMP
    • JMP Graph Builder for the iPad
    • JMP Pro
    • JMP Clinical
    • JMP Genomics
    • Free Trial
    • Buy JMP
    • About Us
  • Using JMP
    • Customer Stories
    • Application Areas
    • White Papers
    • Industries
    • Capabilities
    • Academic
  • News and Events
    • Seminars
    • Webcasts
    • Users Groups
    • Conferences
    • JMP in the News
    • JMP Blog
  • Resources for Users
    • JMP User Community
    • Support
    • Training
    • Learning Library
    • JMP Blog
    • File Exchange
    • Discussion Forum
    • JMP for Life Sciences Resource Center
    • Users Groups
    • Newsletters
  • Knowledge Base & FAQs
  • System Requirements
  • JMP® 10 Online Documentation
  • Alphabetical List of Features
  • Downloads
    • JMP Software Updates
    • JMP Genomics Software Updates
    • JMP Documentation
    • Sample Six Sigma Menu
      (JMP 6.x)
      *
    • Sample Six Sigma Menu
      (JMP 5.x)
      *
    • List of JMP Downloads (Requiring SAS Profile Registration)*
  • Books and JMP
  • Report a Problem
  • JMP Administrators
  • Find Your Serial Number or Site ID
  • Why Register JMP?
  • Register JMP

Protected LSD

Protected LSD  
Download Details
Download
License: Freeware
Size: 2 KB
Type: ZIP Archive

Description: This script oversees an ANOVA and subsequent Means Comparison procedure. The Means comparison is only performed if the F ratio from the ANOVA is significant. Otherwise a message is issued that it is not appropriate. It assumes that the data table is already open and ready for analysis.

This script was developed for a large manufacturing company. They originally implemented a scheme in SAS that set up multiple comparisons in two stages, following the method known as 'Fisher's Protected Least Significance Difference.' The scheme sets up a conventional one-way ANOVA in the first stage. If this analysis is significant (p-value less than alpha), then the second stage is applied. The second stage uses the pair-wise Student's t-test for each pair. If the p-value for the F ratio is not less than alpha, then the script stops and sends a user message to the Log.

An alternative method would be to use Tukey-Kramer Honest Significant Difference which effectively adjusts alpha for multiple comparisons. This alternative method is also available in the JMP Oneway platform under Compare Means options.

Instructions: Open the data table for the analysis. Open and run this script. Select the numeric, continuous response variable. Select the nominal or ordinal predictor variable. Select the significance level (alpha), the default value is 0.05 alpha. Select OK.

Requirements: An open data table.

Note: No support shall be provided for any JMP scripting language code furnished on this site.

SAS | JMP is a business unit of SAS.

  • Terms of Use
  • Privacy Policy
  • Site Map
  • Contact