A scatter plot is when a variable exists under the control of the experimenter. There are two different types of variables: Independent and dependent variables. Independent parameter exists systematically incremented and/or decremented by another parameter and is plotted on the horizontal axis. The dependent variable is customarily plotter along the vertical axis. If a dependent variable does not exist, the type of variable can be either on the axis and a scatter plot. This will only illustrate the degree of correlations between the two variables. A scatter plot can also suggest different kinds of correlations between the variables with a confidence interval. The correlations can be positive (rising), negative (falling), or null (uncorrelated). When the patter slopes from upper left to lower right, it is a negative correlation, and when the opposite happens it is a positive correlation. Then a line of best fit, trendline, will be drawn to study the correlation between the variables. Therefore and equation fro the correlation can be determined by the established trendline. The trendline for a linear correlation is known as a linear regression and is guaranteed to generate a correct solution in a finite time. The most powerful aspect of a scatter plot is the ability to show nonlinear relationships between the variables. The data is represented by a mixture model of simple relationships (visually evident as superimposed patterns.)
Some basic tools of quality control used for a scatter diagram are: histogram, Pareto chart, check sheet, control chart, Ishikawa diagram, and a flowchart
Example:
Calculator: Diagram:
No comments:
Post a Comment