nav[aria-label="Primary Navigation"] { padding: 0; & ul { list-style: none; width: 100%; display: flex; flex-direction: row; justify-content: start; align-items: start; gap: 30px; padding: 0; & li { margin: 0; } & ul li { list-style: none; } } }

JMP Glossary

Jeff_Perkinson

Below are some essential JMP-related terms that a user should know.

In the spirit of collaboration and wiki's everywhere this is an open document, editable by anyone. Please contribute your favorite terms or edit definitions for clarity.

Term	Definition
data table	Data organized in memory as rows and columns. Each column has a name and type, and each row is numbered. You can have as many data tables in memory during a JMP session as available memory allows.
table script	JMP Data tables contain more than just the data in the cells; JSL scripts can be named and saved with the table. The script can be edited or run with the red triangle menu.
table variables	In addition to the table scripts, a JMP data table can also have named values, separate from any column. Sometimes theses are used for documenting the table and sometimes as variables used by table scripts.
display box	rectangular elements glued together and nested in a hierarchical manner to contain the visual elements of a report
JSL, script	JMP has a language for saving and re-running platforms called JSL. JSL can also be used to create programs with Graphical User Interfaces.
Formula, Column Formula	JMP Data table Formula columns contain a JSL script that computes the value of each cell in the column using data from other columns. This script is referred to as a formula, especially when a platform saves a prediction formula to a new column.
Column properties	In addition to the very special formula property, a column has other properties such as notes for documentation of the column, Range Check and List Check for data entry, Value Labels for display, properties for specific platforms, and an Other property you can use.
missing value (•)	a cell without data is represented by a (•) for numeric data or null character strings. (missing value mark)
missing term	the empty red box in the formula editor is a place holder for an expression. It's waiting for something awesome.
modeling type	for a data table column, specifies how the values of the column should be treated in analyses. There are three modeling types: continuous, ordinal, and nominal.
continuous modeling type	numeric data that can take any of a range of values. Can be used in operations like sums and means.
ordinal modeling type	categorical data where the categories have an intrinsic order, for example, sizes: small, medium, large.
nominal modeling type	categorical data where the categories have no order, for example, gender: male, female.
platform	the collection of options and results for performing analyses. Each option under the Analyze and Graph menus launches a platform and presents appropriate options for continuing the analysis. In scripting the platform object is used to control the analysis or graph, while the report object contains the displayed results.
platform options	platforms can have multiple sub tools or specialized features, for instance Profiler has Surface, Contour, Mixture, and a few more.
report	the display resulting from an analysis. The display in a report is made up display boxes.

Comments

wilkap

Jeff -- Please let me know if I am on target before I add more.

Thank you.

Jeff_Perkinson

Thanks, Bill.

Your terms are more general than mine.I wonder if we should have two glossaries: a JMP-related glossary of terms that essential for JMP users; and a statistics glossary that references JMP where appropriate.

Anyone else have any thoughts?

-Jeff

michael_jmp

When I'm working on the Statistics Index that is inside JMP, I am often conflicted by how much effort/space I want to spend on general statistics terms. Especially considering that the internet (Wikipedia, etc) contains a lot of that information. I feel like creating a general statistics glossary on this site could be a high effort, low reward task.

stan_koprowski

I think both have value. It would be nice to be able to quickly search a JMP specific term and then link to additional statistical details as needed. Often I get asked by new JMP customers which platform in JMP do I need to use to perform a specific statistic technique. It might be interesting to have such a cross-reference to show similarities and differences.

wilkap

This request came when I widely asked my customers what they

would like to see in JMP so I could pass it on to you.

I think the problem some of my students are having is that terminology

is not always used consistently.

For example, JMP uses the term "Individual Confidence Limits" where

other sources use "Prediction Limits"

or "Confidence Limits on a Single Measurement."

Since these students are new not only to JMP but also to Statistics,

they aren't sure what JMP means by

these terms. They may try to look this up online, but not find the

exact terminology there.

Even something as simple as "Confidence Limits," which we know refers to

the mean, they are not sure about.

Another example -- in the Tolerance Limits the prompt asks for

"Confidence (1 - alpha)." On the web, alpha can have different meanings

and they don't feel qualified to judge. They don't know what alpha is

and so get nervous that they are missing something.

Whether the value is high enough for the effort required, I can't say.

I don't even know what percentage

of my students have difficulty. Since some are asking, it is likely

many who are confused just aren't asking (I think).

I answer there questions for them at no charge, but it seems they want

to avoid contacting me about this.

Do you have any idea how many of your customers are new to Statistics?

Bill.

Jeff_Perkinson

michael@jmp is right that if any of one of us tried to do this it would be high effort. However, if we crowd source this and make it open, wiki-style, it will be easier and have more impact.

Does anyone have opinions on whether two glossaries are better (easier to navigate?) than one?

wilkap

After thinking about this a lot, I think one Wiki is best. That way complex definitions can rely on simpler definitions in the same Wiki. Of course I am not stuck on this if any of you think otherwise.