This add-in provides an interface for importing MTX files to JMP Data Tables. To use the add-in, select Add-ins > Import MTX . From the launch window, you can browse for the MTX file you wish to import by pressing the Select File button. Once a file is selected, you will have the option to remove the file by pressing the Remove File button and then searching for another file. There are two options for the format of the resulting data table. Coordinate Table will produce three columns, A, B, and Distance. Adjacency Matrix will produce a square matrix where every A value is a separate row and every B value is a separate column. There is also the option to explicitly add a Rows column. Additionally, the Importer reads any notes within the file, and produces a symmetric matrix if the keyword symmetric is found. Pressing OK will initiate the importing process. See Figure 1.1 below. Example Data Original MTX % Symmetric 2 1 0.5 3 1 0.6 4 3 0.8 Resulting Data Table - Coordinate Table Resulting Data Table - Adjacency Matrix Resulting Data Table - Adjacency Matrix with explicit Row column ... View more

This add-in provides an interface for importing XML files to JMP Data Tables. To use the add-in, select Add-ins > Import XML. From the launch window, you can browse for the XML file you wish to import by pressing the Select File button. Once a file is selected, you will have the option to remove the file by pressing the Remove File button and then searching for another file. Pressing OK will initiate the importing process. Figure 1.1 Launch Window Example Data Original XML <header>   <title>Title Text</title>   <content>     <array>Text1</array>     <array>Text2</array>     <array>Text3</array>   </content>   <title2>Title Text 2</title2>   <content>     <array>Text4</array>     <array>Text5</array>     <array>Text6</array>   </content> </header> Resulting Data Table header.content.array._text header.title._text header.title2._text Text1 Title Text Title Text 2 Text2 Title Text Title Text 2 Text3 Title Text Title Text 2 Text4 Title Text Title Text 2 Text5 Title Text Title Text 2 Text6 Title Text Title Text 2 Note that tags that are not at the most nested depth get repeated and in some instances the order of tags at a given depth are not preserved. It becomes alphabetical in JMP. ... View more

A funnel plot is a scatter plot that displays proportions of measurements (Y Axis) collected from various sample sizes (X Axis). The purpose of a funnel plot is to help distinguish random from non-random variations in the data. Points inside the limit curves are more likely to differ based on randomness, while points outside the curves are more likely to be the result of some special situation not caused by randomness. This example shows you how to use the Funnel Plot add-in to create a funnel plot with a z-score distribution based on a Sample Size column and an Occurrences column. The columns should be set up so that when any given value in the Occurrence column gets divided by its corresponding value in the Sample Size column, the resulting plotted proportion value is between zero and one inclusively. First, open a data table with occurrences and sample sizes that you want to use a funnel plot to analyze. With the data table open, select Add-ins > Funnel Plot Generator. From the launch window, assign Sample Size and Occurrences to their respective roles. See the figure below: Figure 1.1 Launch Window Use the Show Legend and Show Histogram options to control what appears on the plot. After you click OK, your graph looks similar to the figure 1.2. Figure 1.2 Proportions vs. Sample Size Points that fall outside the 95% limit curves represent data that can only be attributed to randomness 5% of the time. Similarly, points that fall outside the 99.8% limit curves represent data that can only be attributed to randomness .2% of the time (which is extremely rare). Figure 1.3 If you look at the circled points in the funnel plot, you might assume the point circled in red has a higher significance than the point circled in green; however, this is not the case. In fact, the point circled in green represents a much more extreme value. Since the sample size of the value circled in red is so small, it is much more likely to be the result of randomness. Whereas for the data of the point circled in green, the proportion is much more significant in relation to the rest of the values, and therefore cannot be due to randomness. Instead, the proportion’s extreme value is much more likely to be the result of some significant situation or special cause. (See the links below for more information.)   The histogram on the right is a distribution of z-scores calculated from the data. Z-scores are used to normalize the data given that the original data samples do not come from the same distribution. By using z-scores, the points now have a mean of 0 and a standard deviation of 1, following a normal distribution. Statistical Details Potential issues Ensure that no occurrences are greater than their corresponding sample size, as this will disrupt the calculations used to generate the plot and the distribution. For more information on funnel plots and their uses check out these links: http://www.perceptualedge.com/articles/visual_business_intelligence/variation_and_its_discontents.pdf http://understandinguncertainty.org/fertility Graph Makeover: Where same-sex couples live in the US ... View more