JMP Academic: Intro to JMP for Students

hello everyone and welcome to today's webinar introduction to jump for students you should notice that there's a poll asking a few questions it'd be great if you could answer those and get

a little bit of background of your experience in using jump and also the level of knowledge you have about statistics you'll know as you're familiar with

using Zoom that there's a q a panel and a chat panel if you could use the Q a panel for your questions it'll be a

little bit easier for us to manage and answer if you just wanted to make a general comment um maybe introducing yourself to the group or whatever you can do something in the chat but for questions keep it to

the Q a um now with with that uh Ross matusulum one of our statistical scientists of the academic team is going to be giving the

webinar today uh let me check in Ross are you uh there and ready to go yeah Kevin I'm here can you hear me okay I can um I'll turn it over to you and everybody

um you'll have a great session today all right well thanks everybody for coming to learn a little bit about jump Kevin thanks for hosting everybody as Kevin mentioned please use the Q a panel

uh for questions throughout and Kevin will be in there Fielding your questions live and then I'll pause a couple times to take some questions uh as we go along

but uh with that said let's move on over to uh jump and get started so uh

looking at our poll results and the poll is still open I appreciate people continuing to fill it out it looks like we have um people uh with a mix of levels of Statistics expertise even a couple

Advanced folks uh but mostly um none to novice and uh it looks like the majority of people who filled out the poll uh have no jump expertise

whatsoever and that's great um the today the goal is to take uh student users of jump particularly uh students who are either using jump in

your a course right now or we'll be using it soon um and give you the knowledge you need to be able to use it effectively essentially to to save you some time and

effort learning jump you know in real time while you're trying to do a class exercise or a homework assignment or a project so by the end of this I hope that you all feel comfortable taking

some of the the data that you have in your course and getting it into jump making some graphs performing some analyzes and saving your results

so let's talk about jump the software that we're looking at right now and before we get into any specifics for those of you who might be seeing jump for the very first time I want to just

boost your intuition for what jump even is now from the title of uh the webinar you know that it is about statistics software

um so let's actually just take a little look uh at some of the things that jump can do in that domain here we're going to be using this data set

that you are looking at and this is uh data from November 2022 so just a couple months ago on the most popular channels

on YouTube where we have the ranking of the channel where the ranking is based on the number of subscribers for that channel so here we have the number of subscribers in the Millions

and we can see that the most popular YouTube channel as of November 22 was a channel called T-Series which has

234 million subscribers in the category over here we can see that it is a music focused Channel and we can see some other things like it was started back in

2006 the channel currently has about 18 500 videos that have been viewed collectively about 212 billion times so

these values here in video views are in millions as well so jump is all about interactive data exploration and visualization and

Analysis all through a point-and-click interface so let's just take a look at a couple examples um to boost our intuitions here so let's say that we actually wanted to

see across the various categories of channels like music education entertainment and so forth which ones have the most subscribers out of this collection of popular channels

we might choose especially if we're just getting started with Statistics to answer that question with a graph so let's do that let's take a look at the different categories and their

subscriber numbers in fact you can see we've already gotten a simple graph just from a couple clicks here we can see that T Series is the most subscribed Channel again with about

234 million subscribers and let's go ahead maybe add some color to our graph so we're coloring it now by the category and maybe ask instead of

having all those points to just have bars that represent the sum total number of subscribers all right so what do we see well it looks like music and entertainment

dominate this group of popular YouTube channels in fact uh we have it looks like over 5 billion total subscribers uh about 5.2 billion subscribers in the music

category so jump has a lot of interactive Graphics this is just one example of making a graph with jump and we're going to take a look at this particular tool

in a little more uh detail shortly but uh jump also uh while it's very graphically focused um also performs a wide variety of statistical analyzes so let's start with

something simple let's say I actually just want the the numbers the numerical values behind this graph so the actual uh sum total subscriber numbers for each category

so jump has a tool that'll let us do that pretty easily so let's say that we want to look at the different categories and the subscriber

numbers and here we have the sum total so for that music category that we saw was the most subscribed we can see that 5.2 billion subscribers

the next most was entertainment and I can see that that's about 4.8 billion subscribers one thing you might be noticing is I'm clicking around in this table is that when I select a particular

category it's getting selected in that graph on the right and actually the individual channels in that category are getting selected back in our data table here's one more example if I click on

education we can see that education bar is selected and in our data table all of the education channels are being selected that's a core part of of jump

and its interactivity is that you can always just click directly on some of the data and have it selected everywhere which makes jump a pretty good tool for actually getting to know your data and

navigating it and exploring it so we've made a graph and done a basic numerical summary jump is also quite useful for a wide range of statistical analyzes so if you're just getting

started in statistics which I think a number of us here are you might start with analyzing just single variables and so here's an example of a statistical output where we're analyzing four of our

variables here so the subscriber number video views video count in the category and we can learn some things statistically for example it looks like the most popular category of these

popular YouTube channels is entertainment we have 238 entertainment channels which makes up nearly one quarter of our data set we might also notice in some of these

plots at the top that for each one of these measures here like subscriber number or video count we have some uh individual channels that are much higher than all the other ones so here we have

that uh T-Series and subscriber numbers again and we could even because jump has all these interactive Graphics go ahead and just select all of these subscribers and see where

they are in all of these other measures so we now have the most subscribed channels and I can see that they really dominate the total number of video views so that is all these High video views

channels tends to be the ones with the high subscribers but that's not necessarily the same with the video count so those little gray dots are channels that have a lot of videos that

they posted but aren't necessarily among the most subscribed channels in fact I could look at this one right here and see that this is GMA news with 342

802 videos posted to their Channel I believe that this is a a popular news channel in the Philippines if I'm not mistaken so it makes sense that they would have a lot of videos they probably

post videos for many many news stories each week so jump performs you know graphing uh summarization statistical analyzes we're going to keep it pretty basic here because we're assuming that for many of

us we might be starting out with statistics for the very first time but as we go along uh I I do want you to just keep in mind that jump has a really

wide range of capabilities of ways of analyzing data so if we just take a look at our analyze menu here you're going to notice things for basic statistics that tool we were

just using is one we'll see a little more in a moment it's called the distribution tool or I'll use the word platform which just describes one of these analysis tools that you find under the menus here

but it has a lot more as well so if you're familiar with linear regression there's general purpose linear regression tools from your simple regression to multiple regression generalized regression things of that

nature uh Suite of machine learning capabilities uh many of which in this case are available in Jump Pro uh your University might license jump Pro but if

you're using a standard jump you'll actually notice that some of the things that we might see in these menus actually aren't going to be present there but to highlight a couple other things

jump has tools for mining free text data or that is unstructured text things like um you know product review narratives for example

we have techniques for what we call multivariate statistics so analyzing many variables at the same time so some of you may have heard of things like principal components analysis or factor

analysis or structural equation modeling we have a lot of usage in engineering statistics and there's an entire Suite of tools relevant to that domain

and then looking up at these top menus we have a lot of other collections of tools as well for example designing experiments a lot of different graphing

tools and even tools for preparing data for analysis for example combining data for multiple data tables or working with the different rows in your data table to

figure out exactly how they should be analyzed and so forth so jump really is it's a pretty wide tool and we're just going to see a small bit of it today because the goal right now isn't

necessarily to teach you how to use the exact jump tools that you're going to be using in your course I'm sure you'll cover that as you go along it's to help you understand really at a very basic or

fundamental level how jump works so that regardless of the tools you're trying to use you have enough knowledge to get started so before we go on to talk about the

jump interface and making graphs performing analyzes and so forth uh let's talk just for a couple more minutes um in slightly more detail about exactly

what jump is so this is desktop data analysis software so it's not running you know in the cloud or on a server though there are options maybe your University could

deploy it that way it's primarily desktop software so you can download it and install it directly on your own desktop or laptop computer and it runs both Windows and Mac and as I mentioned

before they're actually there's standard jump and jump Pro so what we're looking at right now is the latest version of jump Pro that's uh version 17.

uh for you to get jump your University might have a license for it already many many universities do um if your University doesn't you can go to a website called on the Hub it's on

the hub.com and get what we call the student subscription of jump where you pay a small price to license jump for 12 months and your uh University or your

um instructor for your course may have you access it that way uh best bet if you're not sure how to get jump is to just ask the instructor for your course what the best way to get jump is

now jump is used widely across many different universities but it's also used in a wide range of Industries as well so just know that jump is a tool that you can actually make good use of

once you're out of University and and working so here's a page from our website that just summarizes some of the more prominent jump users out there and

how they use jump so if I scroll down you might notice some pretty prominent companies for example uh here's one that's kind of fun we have

um at Nike so the athletic apparel manufacturer they use jump in their uh Nike Air manufacturing Innovation Department um to actually help design some of those

Nike Air uh shoe Technologies or even right here you can see PNG that stands for Procter and Gamble which if you haven't heard of them you probably have one of their products in your house

they make a huge number and breadth of types of consumer goods for taking a look at another one here we have a story from a company called

regeneron which if you were following some of the pharmaceutical uh news and treatment news during the Copa 19 pandemic you probably came across their name as well

so there's just a really large number of even some of the world's most technologically uh Advanced and successful companies that use jump so this isn't just a teaching tool it's not just something for school it's something

that's actually used quite extensively out there in the real world and even though you have some really sophisticated you know scientists and researchers and Engineers who use jump

it is designed to be easy to use as you saw just for me making some of those graphs and Performing some analyzes it's no more than a couple clicks so jump is it's easy to use it's no coding so you

don't have to write one line of code if you don't want to and it's full of a lot of interactive Graphics to try to help you explore and understand your your data better so it's a powerful tool used a lot in

the real world but it's also particularly uh helpful and easy to use when you're first getting started and so um with that said let's actually take a

deeper look at jump and how it works to help you figure out um you know how to use it best in your course we're going to start by talking briefly about the jump interface so

looking right here we have the jump data table and this is where you're going to start for probably most or maybe all of the things you do in your course so you may be looking at this and if

you've used something like let's say Microsoft Excel or Mac numbers thinking oh it looks kind of similar to a spreadsheet and I want to highlight that it's actually different from a

spreadsheet and one really key sense and that is that there's a particular structure to it each row in a data table corresponds to one observation in this case one YouTube

channel that we've observed here in each column stands for one variable or measurement so this one is just the name of the channel this is the number

of subscribers and so forth something like a spreadsheet you can enter any information anywhere you want a particular column in your spreadsheet can be a mix of different types of

values some of which are raw data some are summary statistics some are a graph and in jump and really all proper statistics software that's not the way

it works we have this one row per observation structure and one column per variable so that the software knows exactly what to do with all of the data

so for example let's take this subscriber number you can see at the top here we have 234 now in Excel I could type anything I wanted into here let's just type

um I'm going to type a Word in their text and you'll notice if I try to do that jump actually says Hey thought there were subscriber numbers in here and so putting characters or text in there doesn't make sense are you sure

you want to do that this is just one example of how you need to kind of obey this one column per variable structure here so I'm going to click revert to actually undo what I just did

now what I just did actually highlights another very important thing to know about the jump interface and that is that each column has what's called a modeling type associated with it and

really that's just telling jump what kind of data live in that column so jump knows what to do with it and they're best captured by these icons that you see in the notes on the left and then

that I've put a box around right here so we have at the basic level we have this blue triangle and red and green bars Let's uh stick with subscriber number

here you'll notice that that has a blue triangle and if I click on it you can see what's selected is continuous and this is telling jump well this is a continuous type of measurement or variable that is it's it's numbers and

you can do things like find the average of these numbers if we go to let's say category you'll notice that it's red bars and that

stands for a nominal variable where a nominal variable is just groupings or categories that don't have any natural ordering to them so that makes sense

here Music Entertainment gaming people and blogs and so forth and the types of analyzes and graphs you would do with a nominal variable are different than with a continuous variable and this is how we

tell jump which type of variable it is so jump knows what kinds of graphs to make and what kinds of analyzes to perform so when you first bring your data into jump you want to make sure that the

modeling types are set correctly so that jump actually behaves the way that you want to be want to behave so let's say that we have our data here and of course there's more as you can probably tell just looking at it more to

the data table than what I've just talked about but this is really at a fundamental level what you need to know to get started so let's say I have some data now I want to do something with it

make a graph perform an analysis what you do first is go to these menu items at the top so if I want to analyze the data I would go to the analyze menu if I want to make a graph I would go to the

graph menu if I want to actually manipulate the data table maybe I want to filter it or I want to combine multiple data tables into one I would go to the tables menu

if I wanted to let's say work with the columns in my data table for example I want to change things about the data types that are in there or how many decimal places should be displayed all kinds of Options under the columns menu

and so forth so you always begin at the top here in one of these menus and then once you get into a tool let's just take a quick look

here at a tool we'll see a little bit more of in a in a moment and let's say I'm analyzing a particular variable you will get some kind of default output but know that when you get these kinds

of outputs it's not all that jump can do you'll notice these little red triangle menus everywhere so you'll see one here and here some of them even live back in

the data table here's one over here here's one over here and over here these red triangles are all over jump and they are menus that actually contain most of

jumps features or capabilities so for example if I click here here are a number of different graphing and Analysis options for working with this category variable that I have and so when you're doing things in jump

you start with your data table you choose a particular tool from one of these menus and then once you get into that tool you can ask that tool to

perform additional actions for you using the red triangle options and that's the basic idea in kind of navigating at a really fundamental level here the jump

interface so now let's actually use that interface to do the the at a at least at an introductory level the basic things that you would do in one of your statistics

courses we're going to start with making graphs and to do that we're going to go back to the very first tool that we opened up in jump and that's called graph Builder

so uh let's go back to that graph that I made before uh very quickly and actually slow down and see how I did it so that we understand how this graph Builder tool works because no matter what domain

you're in you will have use for this tool so graph Builder lives at the top of the graph menu and when I open it you'll notice right

in the middle I just have this big blank area that says drag variables into Drop Zones so variables that refers to this list of columns in our data table over here and

the Drop Zones are all of these things that are lighting up blue as I start to drag one of these variables so when we started out I said let's take a look at the sum total subscriber

numbers across the different categories of YouTube channels so if I wanted to look at subscriber number by category I would I can start with either variable let's start with the subscriber number this time and

start to drag it and maybe I know that I want that on the y-axis of My Graph so I'm going to drag it over to the y-axis and you can see that before I even release it jump starts making a graph

for me so I'm going to go ahead and let go now because subscribers is a continuous variable jump started by making a simple Dot Plot so we just have one dot here

for each channel here again is that T-Series down here we actually have Coco melon which if any of you um spend any time around very very young children maybe you've heard of that

before you can see that I didn't have to tell jump right off the bat what graph to make it basically takes a best guess at

the type of graph as soon as it receives some kind of information in this case it gets the Dot Plot is the best way to graph these data if all I provided was

this one continuous variable but you can change the default graph that's produced by using a number of different options

the first are the graph Elements which are depicted there in the notes but if we take a look up here you'll find them as a ribbon of options at the top you'll notice there's just one selected

and it's the dot element right here in fact if I turn that off we just end up with a blank graph so I could choose a different type of

graph for example I could choose a histogram or I could choose a box plot there are a number of different graphs I might make when I have just one

continuous variable now let's actually add categories we did before we wanted to look at these subscriber numbers across category so in this case I'm going to drag it to

the x-axis and you can see jump said okay well you gave me a nominal variable on the x-axis I have that continuous variable on the Y what I'm going to do is have a separate group of dots for

each category so it's just a Dot Plot by this categorical variable and what I did before was instead of looking at these dots I decided I wanted

to look at the bars so now we have bars representing the subscriber number and actually if I remember correctly I took this category variable and in

addition to putting it in the X role or the X Drop Zone I also dragged it over to this color role so that each bar would have a different color to help the US visually distinguish them a little

better so I'm pretty close to the graph we did before but what we have here is a graph of the uh mean number of subscribers so

the average uh per category and I can actually notice that by looking at something called the element properties so an element again is just the thing that I put in the graph in this case it's the bars

and here it's telling me which statistic is being plotted in this case the mean if I wanted to look at the sum what I could do is click on summary statistic

and then just choose the statistic that I want the bar to represent and we can see that now all the numbers have gotten quite a bit bigger I mean we're up to 5 billion some total

subscribers and the pattern has changed quite a bit so here if we go back to the mean you can see it's you know relatively even the mean number of subscribers per Channel but because we have more entertainment and music

channels than we have other channels when we add up the sum total of subscribers those two types of channels really dominate foreign

Builder here I've made a graph first by dragging and dropping a couple variables into the Drop Zone jump guest the best graph to make but I'm not stuck with

that graph I used the graph elements and the element properties in order to create the graph that I want now you may have even further things you want to do to your graph and you can't figure out well how do I get

a combination of an element and the element properties to do what I want odds are you can find what you're trying to do just by right clicking on the part of the graph you want to change for

example I'm not a big fan of this very light blue color for music so if I wanted to change that I might just right click directly on music in the in the legend here

and find a fill color option where I'll change it to let's say purple or if I wanted to change some aspect of

the axis here I might right click and find the axis settings where I can specify for example grid lines or the minimum maximum value of the axes and so forth

and eventually when I'm all done I can just click done and now I have my graph now that seems like probably a lot right I was talking for several minutes while I made this graph but again graph

Builder is pretty simple so if we were to just start from scratch here it's not a whole lot to make that graph again I just put category on the X

subscriber on the Y color by category and select the bars and I'm I'm basically there again the last thing we just did was in fact

change to the sum now what's really nice is uh graph Builder is great for just exploratory graphing and you can kind of just drag things around all over to see how

different things uh change and how the graph starts to maybe tell a story so let's say I'm going to start over here I'll put subscriber here but instead I'm going to go now to Maybe video count on

the X and I get a scatter plot here so now let's say I want to look at this by category well what should I do well I can just start dragging category to different places

to see when I might actually get a graph that looks sensible to me and if I end up let's say over here on wrap it looks like well now I can get a scatter plot

of this for each individual category and if I ever make a mistake I can just click undo so when you're in graph Builder and it's always good to have a good look at your

data using some graphs don't be afraid to just drag and drop different variables in different Drop Zones until you find a graph that seems to tell the story that you want you're not going to make any irreversible mistakes or

anything like that so that's a little bit about graph Builder I think probably the most generally applicable tool in all of jump now let's talk about uh performing analyzes actually now that I'm looking

at the clock though I'm going to go ahead and pause here Kevin to see now that we're at about the Midway Point um if we have any outstanding questions um that we could answer right now um going through them so I think I'm

keeping up with them but I'll let you know if I um if I can't answer them all but I'm typing in some answers as you speak okay great well thanks Kevin and thanks everybody I guess we will

um carry on here and talk about performing analyzes now again I just want to stress here uh the goal isn't to show you the actual statistical analysis

tools that you're going to need to use in your course that really depends on exactly what kind of Statistics course you're taking it's to give you an idea of just how performing analyzes uh Works in jumps so that you can get started

with whichever tool it is you're actually using oh Ross sorry to interrupt a question that came up that would be better for you to actually demonstrate than okay the answer so

going back to any of those graphs if you wanted to Omit a category you know what you want I've exclude some rows for a graph can you show how to do that sure let's see

okay so here's a graph that I'd saved I'll show you how to save and recall uh graphs and analyzes a little bit later but here's a uh you know uh box plots of

each of the uh category subscriber numbers and so I think the question is well let's say we just wanted to exclude one of these categories what would we do well I'm gonna go to the red triangle

and show you a tool that's all over jump so in all of the graphing and Analysis tools if you go to the red triangle and look near the bottom you're going to find something called the local data filter

and what this is going to allow me to do is specify what essentially what subset of my data I actually want included in the plot so I'm going to select category here and click plus to add a filter by

category and now let's say I just wanted to look at entertainment music gaming and people and blogs I can select them and now those are the only ones that are included all the other

ones are excluded and so this local data filter tool lives all over jump in all kinds of different graphing and Analysis tools and you'll always find it under the red triangle in this bottom section here now

says removed since I had already added one so I can remove it and get back to normal all right uh thank you Eduardo for that question and uh thanks Kevin for catching it

so as I mentioned let's talk just a little bit about how you perform analyzes in Jump you'll start when you want to actually perform a statistical analysis of some kind by going to the analyze menu I think we kind of

mentioned that before we have these basic menu categories at the top and then all these individual tools that live under them let's say that I wanted to get a basic

summary table if you remember before uh when we made the graph of the sum total subscriber numbers by category we also made a little table that provided us the numerical values for those subscriber numbers as well

and to do that we use the tool called tabulate so I'm going to pop tabulate open right now this is a tool for creating summary tables you may also hear these referred

to sometimes as pivot tables or cross tabs or cross tabulations this tool works very much like graph Builder and that it's drag and drop so I

can grab category and drag it say into the drop zone for rows and immediately jump starts making a summary table in this case it just says the number of rows in my data table for

each category so I can see that entertainment is actually the most well represented at 238 popular channels music is next to 217 and so forth

if I wanted to look at subscriber numbers I could grab that and just drop it right over the data that's currently in the table and now it's given me the sum total subscriber number

for each one of these now you also notice here a lot of different statistics that I can use to actually summarize these data so for example the the mean or average the

standard deviation the minimum the maximum and so forth so if I wanted the mean instead of the sum I could take the mean and just replace the sum with it or if I undo that maybe I want the mean

and the sum I could take the mean and just put it oh if you notice that little blue drop zone right next to some and now I have the sum and the mean likewise I could add other variables

like video views and video count and I could either replace subscriber number or let me just put it just to the right of subscriber number and now I have the sum total

for all three of these variables maybe I'll switch it to the mean instead so this tabulate function or this tabulate platform is the place to go when you want to perform

just these quick numerical summaries uh no matter what different statistic it is you want to use to actually summarize your data now uh in addition to performing these

basic you know graphical summaries with a graph Builder or graphical exploration I should say and some of these tabular summaries there's a an analysis tool that will serve

um you in every domain you could possibly be in but also just as a great example of again how analysis Works in general in Jump and that's the distribution platform

which is uh your one platform for summarizing and analyzing uh single variables or just one variable at a time or if you're familiar with this term uh univariate statistics where uni just

means one and variant means variable this is the tool that lives at the top of the analyze menu now here for the first time we're really taking a look at a jump analysis launch window

so what this is is a list of all of our variables or columns on the left and then the different roles in the analysis that they might play on the right this is a simple one we just analyze each

variable independently so let's take just a couple variables here I'm going to click y columns and click OK note that if you wanted help at this point you could click this help button

down here and it would actually launch some help that will help you figure out how to use this tool and jump immediately produces an output for each variable so on the left with

subscribers a continuous variable we get a histogram with a box plot and down below we have quantiles and summary statistics and on the right we have for our

categorical variable or nominal variable category we have the counts and the proportions or labeled probabilities here of each one along with the bar chart and these as we mentioned before

are interactive so for example if I like the high subscriber numbers I can start seeing for example which categories we typically observe them in now there might be additional things I

want to do with these particular variables so taking subscriber number maybe a common thing that we do is assess the distribution or the shape of the distribution of a variable so if I'm

thinking that's what I want to do I should go to the red triangle right next to that variable and then I'll find the appropriate analysis option for example if I want to fit a distribution to these

variable or to this variable I can choose one here so I'll choose to fit let's say a normal distribution and I get some summary statistics down below

and then up top I actually get a visual picture of that normal distribution that I just fit now I did that kind of uh you know just to demonstrate obviously here it looks like a normal distribution is

not a particularly great fit this standard bell curve is not a good representation of these data the data actually look a little more like this but that's just to show you that again

when you're inside a jump tool if you want to perform an additional analysis just head to the red triangle and select it from the list here let's take a look at one more example of

this now let's actually look at analyzing pairs of variables so let's say the relationship between category and subscriber number

we're going to use the tool that lives right below the distribution platform the analyze menu It's called fit y by X and the name is meant to imply that you have One X variable and one y variable

and you just want to know the relationship between them so let's say that category is my X variable subscriber number is my y variable

uh this tool can take any combination of nominal ordinal and continuous variables as X and Y and it will automatically know which analyzes are appropriate again this is highlighting the

importance of getting those um those modeling types indicated by those little colored icons getting those modeling types correct when you first bring your data in

so when I click OK jump in this case first just gives me a Dot Plot this is kind of looking similar to what we were doing in graph builder at the start but if I want to let's say compare these groups to see if some have higher lower

subscriber numbers I would go to the red triangle in this case the analysis I might want to perform is called an anova an analysis of variance and so I can select that from the list to perform that

analysis and get these statistical outputs below if you're familiar with hypothesis testing you might notice this p-value right here is less than a typical one of 0.05 we might say that we

have evidence that at least one of these uh YouTube channel categories is different from at least one other one

but again that the statistical analysis is not really the point here it's again just highlighting in a different tool now that you specify at the outset which variables you want to analyze and what

role they should play and then you go to these red triangles to select uh the actual analyzes that you want to do foreign okay well what is it going to look like

in some of the analyzes I need to perform uh when we cover finding help at the very end I'm going to show you where you can find video demonstrations much like the very brief ones I just did for the specific analyzes that you need to

do hey rosk I'm just sorry to interrupt I know you mentioned this at the end and show people the location of where this recording will be but there's two people

that need to leave early and so we're asking if you could just share that link now of where it's going to be sure we can do that right now let's find out I actually meant to have this on my

clipboard and I forgot so Excuse me while I find our way to it well here I'll actually just show you the easier way that'll bring you to a recording for this when it's posted as

well as a lot of other ones so if I Google jump academic webinar Library the first hit will be all of the recordings of the webinars that we do

now some are aimed towards teachers that is your instructors might come here to learn about how to use uh jump from the teaching perspective and also some Advanced research topics but this one

will be posted alongside there let me grab this and I'm going to drop this link into the chat here and so later this week we will

post the recording of This webinar to this page um and you will also find once you click through to it some helpful links like to the quick guides that I mentioned and

will be showing shortly here as well as this journal so that you can actually reproduce some of the things that we did okay so uh we've seen how to make graphs and perform analyzes I think we know the

basics of the jump interface let's talk about getting our data in to Jump now the easiest way would be if you end up with Native jump files that is dot

JMP files in which case you just double click them for example let me just close that down and I'm going to open this folder here and here is a DOT JMP file

that is for our YouTube channels and when I double click it jump just opens it automatically and uh if you're using it exclusively for classroom purposes there's a decent chance that your

instructor or the textbook that you're using will provide dot JMP files and that's all you have to do but but maybe not a lot of data comes from a lot of different sources in a lot of different

formats know that jump can open a lot of other file types directly as well just by going to file open for example

I have here a DOT CSV file that stands for comma separated values it's just a text file where each row is on a line

and each value is separated by a comma that's kind of like this so jump can read this very common file type

we just go file open select the CSV file and when I click open you can see jump launches this tool to help me import the

data and that's because maybe it's the case that there are column headers in the data uh in that CSV file but maybe there aren't um maybe I only want to grab some of the

columns of data and not all of them maybe I just want to see a preview to even know what's in there and how jump would bring it in which is what I have at the top here so jump provides these

tools import Wizards we call them to help you bring your data in from a number of different file types now I'm not going to get into the specifics of how to use this particular import wizard the idea is just to show you that you

can point to a non-jump file and get a tool that helps you bring it in and then once I import the data here we are ready to go

now you may get your data as Excel files so here we have our Excel file that contains these data

now we could of course go to file open point to the Excel file and now we have another import wizard this one built specifically to handle

Excel files that we could use to bring the data in another easy way if you have a spreadsheet that has your data and it's in this nice format already one row per

observation one column per variable and maybe you even have column headers at the top you can always just copy paste so let me

copy these data then go to file new new data table and that's just going to be a blank jump data table and then under edit I can paste the data

in or because I had column names that are copied as well I can select paste with column names so jump nodes to take the first row from that Excel sheet and call it the column headers or

column labels here and again I'm in and now what I would want to do here is actually clean my data up a little bit to get it ready for analysis we said get those modeling types right

and they look pretty good rank I might actually want to call an ordinal variable instead of a nominal so I'll change that there may be some other things I want to clean up for example right here at number two we have YouTube

movies claiming a video count of zero uh in fact I see a few with zero video counts so I think I'm not sure how that got in there but I'm I don't think I want to analyze it because I don't think

it's a accurate reflection of a real Channel and so I might let's say select all of the zeros and then find a red triangle in this case by my rows and say hey why don't

you just delete all those rows or you might notice I have a lot of different categories more than we started with initially for example I have pets and animals I have non-profits and activism and some of them they

really only have a few channels associated with them so I might right click here and actually recode this variable let's take Auto and

vehicles and some of these other smaller categories and let's just combine all of them together and call them other

now click recode and now I have a cleaned up version that I've called just category two here where I've collapsed a lot of these together that's just a couple quick examples of

of how you can clean up your data get it ready for analysis and jump if it's not already clean for you now for classroom use it probably will be pretty clean most of the time but uh especially if

you go out there for a project or just to find some data online to analyze you may need to do some data cleanup up front and in the Learning Materials we're going to look at you'll see a

variety of of materials to help you actually learn how to do that but it's a little bit beyond the scope of what I intend to cover in detail here so I'm going to close that table down

and bring our original one back and talk about saving and sharing results okay so here's our data table before

so let's say that I am working on a graph let's see I'm gonna just for fun let's make a a new one here um let's take a look in this case I'm going to go category

by subscriber and maybe I'll also just to show some flexibility of graph Builder I'm going to have several different y variables like video views and video count

let's actually we'll still color by category all right so on the X5 category and then I have basically a separate pain in my plot for each of these three continuous variables

so now let's say I want to save this I may need to come back to this graph later your instincts might be to go to file save

it can and it'll save a DOT JRP file and that'll be linked to your data table and when you try to open this.jrp file

it's going to ask you to point to the data or maybe you can actually just embed the data in that JRP file I recommend something else instead under the red triangle and here let me pull up

the notes so that we can map to that all over jump you're going to find this category in the red triangle it says save script and I'm going to choose to

save it to my data table and now I'm going to call it test capital t e s t just so we can keep track of it and then keep your eyes up here in the background in our data

table while I click ok you'll notice that that script that says test has been added to the data table underlyingly this is the the code the

jump scripting language or JSL code that actually creates this graph and so when I click this green arrow next to it it creates the graph again for me

and now what's really nice about this is that it actually re-runs the graph on the data table in real time so that if you update the data

uh it will be reflected in the graph for example let's just say I mean I don't let's say I've just take out all the entertainment entries here so I've selected them in

this little header graph I'm going to right click and go to hide and exclude which basically tells jump hey don't show this these rows to me in any graphs and don't include them in any analyzes so when I rerun this we

shouldn't see the entertainment category anymore and we don't if you go through this quick list here you'll notice that entertainment is not present so this is a great way to just have

reproducible graphs and analyzes all bundled inside your data table so you have one file with everything you're working on and again that's going to save script save to data table

now there are some other options for example if you want to get this graph into a document or slides you can use a

simple copy paste to copy something go to the jump tools they're just this toolbar at the top of your jump window or you can actually click the tools

right here and go to the selection tools this uh plus sign looking thing select everything you want right click and go to copy and then

let's bring up a Word document let's say so maybe I'm doing a report for class and I want this to go in my report

I'm going to instead of going to paste I'm going to go to an option that's known as paste special and that lets me choose the format that

I'm going to paste in on Mac PDF is a good format on Windows you'll see something called EMF it stands for enhanced metafile and that'll be a good option for you on Windows

and so now I have this copied into my document and the nice thing about using PDF or EMF is that you can resize it without ever losing quality so things don't get blurry or pixelated so as you

can see as I zoom in all the way it got pixelated at first but then the graph was essentially redrawn and that's a benefit of some of these uh particular image types like PDF and EMF is that

they can just be redrawn however big they need to be on your screen or on your sheet of paper so I've gotten this into a document I could do the same for slides I may also just want to get a standalone file like

an image file like a JPEG or something you can always from a graph or an analysis just go to file export meaning save it as something other than

a jump file and then I could select image and maybe the type of image that I want to save it as and and so forth for the sake of time I won't do that but just know that you can so these are the I think for at least

for purposes of classroom use the three main ways you'll want to save your work either saving scripts back to your data table to be able to reproduce What You Did copying pasting into your slides or

your documents or if you need a standalone file going to file export now with a little bit of time we have left I want to shift gears from focusing

um on you know using jump for data analysis and graphing to learning with jump and learning about jump first you may notice actually I'm just

going to close this down so that we can maintain our focus on what we're going to talk about you may notice that I have a menu item called student if you download the

student subscription jump you'll have this installed automatically and you'll notice there are a number of different options and we're going to talk about those in just a second but uh first know that if you don't have

this student menu maybe because your University licenses jump and you download it from your University software portal you can always add it to your version of jump um so when you go to the recording and

get this jump file here this journal with these notes you'll have a link to where I'm about to go and also show you how to find it otherwise but it is a jump and as we call them that will add

the student menu to your version of jump so you would go to this page and you would click this button right here to download this dot jump add-in file and then double click it and it will put the

student menu in your version of jump and if you just need to find this maybe via Google I'm just going to type jump Student Admin and it's the very first hit that you'll

find here now just know if you go this route uh this add-in is available only in English I presume if you're here and following us along then um your English is

probably quite good and you will have no problem using the add-in but I do know that we have people from around the world joining us here today so I just wanted to make sure and point that out so let's see what's actually in a

student menu and what it is it's a collection of calculators and teaching and learning demonstrations that are particularly helpful for student use let's say for example I'm doing some exercises in at the end of a chapter in

my book and they ask you to do some calculations and oftentimes they don't provide you with raw data they provide you with some summary statistics and say ask you to calculate a confidence interval and so here we have a calculator that

allow me to do that either from a raw data table or let's say summary statistics so maybe they tell me to calculate a confidence interval when the mean was

100 the standard deviation was 15 and we measured 40 individuals and so right here with that information without ever having the raw data I've calculated the confidence interval for

us which is listed here and then depicted in the graph below there's a wide variety of these calculators in the student menu that are particularly useful for those situations

when you're performing calculations using summary statistics now under the applets you're going to find a number of very helpful things

that will allow you to explore statistical Concepts in an interactive way while you are learning let's say that I'm learning linear regression so I'm going to go to this demonstrate regression

now this is a tool that will allow us to see kind of how uh various aspects of the data affect a linear regression line so by a linear regression line I mean

just this uh best fitting line through this cloud of points for example I could grab one of these points here say okay you know I'm trying to understand how the value of an individual Point

actually affects the slope of this line and I can do that interactively in this tool just by clicking and dragging and I can see that the line moves quite a bit maybe I want to see well what happens

though if the sample size is a lot bigger let's say now instead of 25 it's a hundred well I can see that moving a point around like that doesn't move the line

nearly as much so the influence of one point starts to become diminished the more and more points that we have now that's not something I'm necessarily trying to teach right now it's just showing you how you can use some of

these interactive tools to help maybe you know test and build your own intuitive understandings of how different data analysis techniques work so if we look under the applets we have

a number of different applets for things like sampling distributions or confidence intervals or hypothesis tests these core um statistical techniques that I think

at some point we all learn when we're learning statistics at the outset here so finally just the few minutes we have left let's answer possibly the most pressing question at this point which is

going forward when you need help trying to figure out how to get jumped to do what you need to do what should you do so we bring our data back let's talk about different ways of

getting help first thing you might let's say you want to do an anova that's before when we looked at the different categories and compared their subscriber numbers and you don't know where and jump to go

know that Under The Help menu you can search jump for the tool that you need so if I go to search jump I get this little search dialog and I can type Nova into it

and this first hit tells me exactly where to go it says location analyze fit y by X then it has this little symbol which basically means hey there's an extra note you should read in

this step and says click go to launch and then you'll find it under the one-way option that's called means in Anova and if I want to launch that tool

I can just click go to get started and then I follow that path that it just specified which again it said go to means Anova and there we go

so use that uh help search jump if you're not sure where and jump to go to perform the analysis that you need now once you're in an analysis let's say we're looking at the distribution of

subscriber numbers and you're wondering you know I could use some more information on one of the things I'm looking at like let's say that box plot at the top there's a tool and jump It's called The

Help tool it's a little question mark here where you can directly on something in a jump window like a jumper analysis report and get help on that thing you can see this took

me right to the page for the outlier box plot with these bullet points that tell me all the different ways of interpreting the individual components and this nice um

graph at the top with the labeling to tell me what the different components of that are so when you're in jump and you want to just get right to the right page of help

that you need uh the most direct route is to again choose that help tool shaped like a question mark and then just click

directly on the thing you want help with so those are a couple of the tools that you're going to find Under The Help menu

or ways of getting into the help menu in the case of the um the help tool uh basically just tools built into jump to give you help but let's talk about

some online resources now uh first as I have mentioned multiple times we have a collection of quick tutorials that show you how to do things in jump and once you're actually performing specific

analyzes for your course I highly recommend checking these out and it's called this collection is called our jump learning library and if I pull up a browser window I'll take us to it you can see in the notes

right there it says it's at jump.com learn this is a collection of approximately 100 uh entries for doing things in Jump

For example just opening jump for the first time getting started with it or making box plots or assessing the normality of a distribution just all these different quick things

you might want to know how to do in Jump let's say I want to make a box plot and I don't know how so I would click on box plots I get a quick tutorial video here

that's uh in this case it's a minute and 41 seconds long and it shows us a number of different ways of actually making box plots in Jump just walking us through step by step

and if I click view guide it gives me a one-page PDF that I can keep for a reference so if I'm making box plots later I can just consult the steps

so this learning library has again about a a hundred I think just over 100 entries for all different kinds of things from basic to more advanced things that you might need to do and jump

this is the place to go if you're wondering well how do I do X go to jump.com learn um note that uh we're currently rolling out some updates to our worldwide site

so the most up-to-date version is the English site at uh en underscore us you can also click on worldwide sites to go to the US

site uh if you are in another geography and when you do you can always just then search for learning library in the little search box right next door and

then find it that way as well now finally um if you're wondering about statistical um Concepts that you're learning so apart from jump itself and just want another explanation of something like

correlation or regression I recommend our statistics knowledge portal so if I go to jump.com skp or skip as we like to call it uh it will take us to a collection of pages

that provide a professionally written high quality explanations of basic statistical concepts for example I'll select correlation on the right and we

have a relatively brief easy to digest explanation of what a correlation is and how to visualize it effectively so you can see we have coverage of some

of the core statistical Concepts and techniques that are covered in most you know introductory uh statistics courses across a number of different domains so if you want another explanation the

stats concept I highly recommend again going to jump.com skp or skip so that's a little bit about finding help and that's we're going to wrap it up

here now that we've just hit the top of the hour I do hope that in this past hour for those of you who've never used jump before which was I think a pretty strong majority of us that you feel

comfortable now both um you know using jump at least opening a data set from your course and getting started and know a little bit about where to find help when you need it