Quality Tools－Scatter Diagram.ppt

Constructing the Scatter Diagram Step 1 Collect Paired Data Step 2 Draw the Axes Step 3 Plot the Data Step 4 Interpret the Data Scatter Diagram Essentials Key Success Behaviors Use the tool only when credible data exist Test first for the most likely relationships Don’t reject unexpected correlations Speak in terms of relationship strength, NOT cause and effect Steps at a Glance: Scatter Step-by-Step Construction Step 1 Collect Paired Data Step-by-Step Construction Step 2 Draw the Axes Step-by-Step Construction Step 3 Plot the Data Step-by-Step Construction Scatter Example Scatter Example * ? 1995 GOAL /QPC Scatter 1 Scatter 2 CG p. 191 Scatter 3 Start Develop hypothesis between two variables Data available? Can data be collected? Assemble paired data Collect raw data Construct data table Construct correlation table Draw x and y axes Axes cover range of data? Plot all paired data Duplicate pairs of data? Need precise measure of correlation? Using software? Use median/quadrant analysis Estimate correlation “by eye” Run correlation tests Determine strength of relationship Act on data Plot circled data points MJII p. 145 Scatter 4 Course Average Session Rating (on a 1-5 scale) Average Experience of Training Team (days) Theory: There is a possible relationship between the number of days of experience the training team has received and the ratings of course sessions. 1 2 3 - - - 40 4.2 3.7 4.3 - - - 3.9 220 270 270 - - - 625 Scatter 5 MJII p. 146 4.5 4.4 4.3 4.2 4.1 4.0 3.9 3.8 3.7 3.6 3.5 150 200 250 300 350 400 450 500 550 600 650 Dependent variable (“effect”) Independent variable (“cause”) Looking for relationships, not cause and effect MJII p. 146/CG p. 197 Scatter 6 Information provided courtesy of Hamilton Standard 4.5 4.4 4.3 4.2 4.1 4.0 3.9 3.8 3.7 3.6 3.5 150 200 250 300 350 400 450 500 550 600 650 Average Experience of Training Team (in days) 4.0 y axis 430 x axis Overall Rating of the Session Step 4 Interpret