Tuesday, August 5, 2008

Star Plots


http://www.humble-inc.com/rof_app99-2.htm

Star Plots compare single data points with different variables. For example, point 1 is larger than 3 but smaller than 2. The figure above is a star plot created from gas chromatographic analysis of oil samples using selected peak ratios.

Correlation Matrix


http://www.livestockgenomics.csiro.au/Genome_to_Phenome/

A correlation matrix is another type of graphic tool that is used to show the correlation or linear relationship between two variables. The graph that I have selected above demonstrates the “heat map of the tissue to tissue correlation matrix. Thick lines separate cancerous from normal tissues. The spectrum goes from blue (correlation <= -0.45) to white (-0.05 < correlation <= 0.05) to red (correlation > 0.45).”

Similarity Matrix


http://www.fxpal.com/?p=eventDetector

A similarity matrix is a graphic representation of scores that demonstrates similarity between two different data points. The matrix above shows how a collection of 512 photographs relate temporal and content similarity.

Stem and Leaf Plot


http://www.highpointsmath.com/sitemap/Stem-and-LeafPlot.html

A stem and leaf plot is a useful tool for visualizing the distribution of quantitative data. This is a useful devise for showing exact scores. For example, if you are a teacher and you want to know how the class did on an exam, this would distribute all of the test scores.

Box Plot


http://math.youngzones.org/stat_graph.html

A box plot is also known by box-and-whisker diagram. Box plots are useful because they can show outliers, statistics that are removed from the concentration of data. Box plot can be horizontal or vertical, and can handle large data sets; they are not useful for exact values.

Histogram


http://math.youngzones.org/stat_graph.html

A histogram is another type of graph used to demonstrate values of different variables. A histogram uses the size of the bars rather than the height to demonstrate the value. The histogram above illustrates the rates of cigarette smokers by sex and age in 1992.

Parallel Coordinate Graph


http://www.curvaceous.com/six%20sigma%20application.htm

The parallel coordinate graph above comes from a company that designs software; therefore this graph is an example of their work; pretty good. Each line on the graph represents a number of values and is connected with a different variable. It would be almost impossible to find and study one variable. The intention of this type of visual aid is to do a multivariate analysis by studying the concentration of lines; you want to look for patterns and omissions.