Bonus day: lab report writing tips
Oct 15, 2023
0:00 tips for using references and linked figures in Word
11:30 using EndNote Web Reference manager
23:25 Tips for preparing nice figures
34:10 digitizing data from figures using WebPlotDigitizer
42:28 nice microscopy images
44:30 Examples of nice figures
Okay, so I'm going to cover, I think, three things today.
The three things I wanted to cover, these are just really tips that I wish somebody
had shown me.
So if you've got somewhere to be, you can take off, but I think, I'll try and make this
worthwhile to be here.
The three things I'll cover is how to make hyperlinks to references and figures in Word
and how to use styles in Word.
So Word, there's easy and hard ways to type, and I'm going to try and show you an easy
way.
The second one is how to use a reference manager, and there's free ones that will make references
way, way, way better to use.
And the third one is how to make nice figures, right?
So let's start with the most important one, which is how to use Word.
So what I did, while you guys were off in the lab still, I came over here and I just
put together sort of like the bones of some of the stuff that goes in a lab report.
We've prepared a document, let me jump over to it.
We've prepared a document that describes what should be in a lab report in a little bit
greater detail.
Is this pulled up?
It's on Canvas.
If you click on files, and then you go to the MSC 2010 lab or whatever, oh, it's so
slow in here, and no internet.
We're going to have no internet today?
Here, I know what I'll do, I have the actual file, so I'll go to where I keep the file.
So you guys will see, let's see, MSC lab report writing guide.
This is in the file section, you should be able to find this.
So this talks about what should go in your lab report, things like, well, there should
be an abstract, there should be an introduction, right?
So I've only said a few words about it here, like almost nothing.
For the more complete treatment of what should go in there, check out this preparation guide.
I wrote this with Mike Scarpula, Jeff Bates, and what used to be the clear instructor,
Justin Whitney.
So these are all people who taught courses, so it's stuff that we actually care about,
so this is stuff we want to see in it.
We talk a little bit about the stuff I'm going to describe right now, like styles and using
internal references and stuff, but use this for a more detailed explanation of what I'm
going to cover today.
But for now, let's just start with our bare bones, right?
So let's say you've got a title, right?
So the very first thing I want to show you how to use is style.
So you're going to click on your title, and if you go to the different tabs up here, you've
got all these different tabs.
The first one under home, there's this style section.
Maybe you guys are unlike me, you already know how to use this, but go ahead and click
title, and it has one that's already set aside as title, and that's maybe not the type of
style that you want.
Maybe you want to tweak it a little bit.
Oh, I just open up Word and just type this.
Yeah, just pull up Word.
So here, maybe like, first off, they're using Calibri, they've got some size, sometimes
it's funny colors.
First thing I want to show you is that you can modify this, and it will stay that way
for the rest of the time that you're working this document.
So what you do is go ahead and highlight this text, and let's say you want your whole document
to be in Times New Roman.
You know, that's a pretty standard thing to do, but 28 is really pretty big for a title,
so you want 18, and maybe you want it bold or something, right?
So now you've got a title, and you also want it centered, right?
So there's your lab title.
What you can do is once you've got it how you like it, whatever you've done to it, you
can come up over here to the Styles under Title, where you were before.
If you right-click this, you can click this Update Title to Match Selection.
And now, if you were to have any other title in your document, it would assign that same
series of sort of style to it.
Now that's not very helpful here, because it's just your title, and you usually only
have one of those, so you can say, it's not saving me any time.
But there's other things where it will save you time, like under your section headings.
In a lab report, you typically have an abstract, you have an introduction, you have experimental
methods, you have results, you have discussion, you have conclusion, and you have references.
So it's like six or seven things.
Rather than formatting each one of those individually, I think it's worth your time just to set up
something.
So I come here, and I go to Heading 1, like the first type of heading.
I click it, and it's blue, and it's Calibri, and it's not stuff that I want, right?
So just like before, I can change this and make it times.
And maybe I want it size 14, but I do also want it bold, but I don't want that blue text.
I want it to be whatever color you want, black in this case.
Again, I'm going to come over here, and I'm going to update my headings to match this
selection.
Now, all I have to do is on all these other ones, I just come and click Heading 1, Heading
1, Heading 1, Heading 1, right?
You can do this to all of these, and it just makes it easier.
This also changes things like your line spacing, indenting preferences.
It'll change that all automatically to it.
You can do something for your normal text.
Right now, my normal text is Calibri.
You could do times, though.
Let me highlight this whole thing and do that, right?
And I could update all of my normal text to do that, and it changes it throughout the
document.
So this is the first thing I want to show you.
That will save you some time.
Right?
That's the first thing.
Next thing I want to show you is how to use built-in reference managers and hyperlinks
within Word.
This is critical for one thing, right?
So let's say you start out with your summary.
You put the stuff in the summary you want.
You talked about the abstract.
You described why you're solving your problem.
You tell people how you went about doing your research and basically the instruments
you used.
Finally, you get to your results.
And almost always in your results, this is where you start introducing figures, right?
You measured something in a...you looked at a picture under a microscope, right?
And you snapped a picture of it.
Or maybe you measured some property and you plotted that as a bunch of temperature or
load or whatever else.
All right.
So let's grab some data from the internet.
Now we have the internet again.
All right.
So grab some scientific data, right?
Something like this.
Looking very scientific.
So we're going to copy this.
And what you've probably done in the past in high school or wherever else you last wrote
a lab report, you just write this.
Type figure one, looks like climate stuff, right?
I don't know what it is.
But then let's say that you want to do another figure, right?
And it's going to be this awesome thing, right?
Clearly Poland's in a lab report.
So you're going to just paste this in here and go ahead and add a figure to wormhole.
So it's not a big deal.
But then you also have to consider that you're probably talking about these figures in your
text.
As we can see, the wormhole is blue.
See figure two, right?
So what happens, the challenge is, what happens if in your editing process that you're going
through and you're like, you know what?
I should have talked about the wormhole before I talked about the climate data.
And then you have to move it ahead.
Now this becomes figure one, which means everywhere in your text that you had see figure two,
you have to try and remember which figure was that now and which one did it move to.
So I promise if you only have three figures in your paper, you won't care about this.
When you write your senior thesis, you're going to have 20 figures, 25 figures, I don't
know, more.
You will definitely, it'll drive you crazy if you have to move things or something like
that.
So what you want, you want this to update automatically.
So there's an easy way to do that.
Here's how it works.
You go to your figure.
When you first pasted it there and you right click on it and you click insert caption,
right?
I apologize, this is tiny.
That says figure one up here and it gives you a label.
First off, it's asking, hey, is this a figure or is this something else?
Because sometimes people paste like equations as, as figures.
So you can tell, yeah, it's a figure, it's not a table or anything else.
It's a figure.
And what's the caption?
Well, you can make it say anything you want, but just leave it figure one is I think generally
the right way.
You can change your numbering.
If you want like Roman numerals or whatever else you can select whatever you want to label
it.
Just do figure one.
Yeah, let me show you one more time how to do it.
On the figure that you pasted on here, I right click and it said insert caption.
Hit that and it's going to pull this up.
We're just going to say, okay.
And it automatically creates this thing.
So now you say the climate stuff or whatever we were talking about there, right?
Now check out the difference.
When I highlight both of these, what's different about those?
One has a darker indent on one, which indicates it's like active.
Yeah.
It means it's hyperlinked to something, that it's referenced to something else.
Like it's not a regular number.
It's not like I just typed one there.
It's a word recognizes that is part of a series and therefore it keeps track of where you
move around in your document, which I'll show you, which is why that's cool.
By the way, the default setting for this is kind of funny looking, right?
The font there got small, it did italics.
If you wanted to change that, you could, right?
If you wanted this to be all bold for whatever reason, you can, and you just need to come
here to your styles and you notice that we're currently on caption.
It already knew that that's a caption, so it assigned it the style for caption, but
we can update that just like everything else, right?
If you want.
I think it's blue.
You could have changed the color, but you know, it's fine.
So that means we can get rid of this.
Now let's do the same thing over here for our wormhole.
You're going to right-click it, insert caption, say okay, and you're going to say wormhole
or whatever the caption is, right?
That's not a very good caption, but that's what we're using for right now, right?
It's using the same caption style as we just updated before, right?
Because it knows to use the right style, but now check this out.
This is where it gets really bad.
Oh, let me show you one more thing.
Instead of typing figure two here, now we want to call that figure, right?
We want to call the reference to it so that if we move it around, it stays with it.
So we're going to do this.
You're going to go C, and then you come up here to the reference tab, right?
Your word has this.
I don't know how far back it has, but you probably have this on your tabs.
Go to reference, and you're going to insert a cross-reference.
So we're going to go to reference, and then over here, it's right next to the insert caption,
but you're going to insert a cross-reference.
Click that, and it says, all right, well, what reference are we talking about?
Is this just a numbered item?
Like if you were talking about a series of sections in your paper, you could label them
like section one, 1.1.
That's the numbered item.
That's not what we're talking about.
We're talking about a figure.
So scroll down.
In this list, it's not an equation.
It's not an end node.
It is a figure.
Click figure, and by default, it lets you insert the reference as the entire caption.
I don't know why you'd ever do that.
See figure one, the climate stuff.
That doesn't make sense.
You just want it to say see figure one.
So change this to just the label and the number, and then pick the figure that you want it
to be.
I think we're doing figure two right now.
There it is, figure two.
So now if you highlight this stuff, you see that same darkened, highlighted text that
Brennan mentioned?
It's not only on two of figure two, but also the whole figure two that's in the text.
Go ahead.
That's how you know if it's a hyperlink as opposed to something you just typed, and then
what's great is you can take this.
I can cut this, and let's say your editor, Emily or Christian, they basically said like
why are you putting this at the end?
This should go in the beginning.
You say great, no problem.
At first, you say hey, it didn't fix it.
There's a problem.
Well, you can update it, and I can't remember where you find it, but the easy way I do it
is you just control A. That will select your entire document, and then hit function or
F9, and it will update those.
So now this is, because we moved it one, it is figure one.
This is now figure two.
Again, I promise if you're writing something short with two, three figures, you're like
what's the point?
When you guys get jobs, and you guys have to prepare much larger documents, and you
all will.
You'll have to prepare things with lots of figures.
This is the only way to do it.
It'll save you a ton of time.
It takes two seconds to set up in the beginning, and then it's way easier afterwards.
It's really easy to call back to like oh, as you expect from equation four, and it knows
the equation four is the specific one that you cited.
So that's the first trick I want to show you with Word.
Anything else I should show them here, guys, on Word, Christian, am I forgetting something?
Are there ways to hyperlink text to other texts, like a certain page, or like to a certain
other paragraph?
I know you can do it with titles.
You can make these, you can make those part of it, because they become part of this numbered
item.
I haven't done it in a while.
I'd have to review it to see how you do that.
Let me show you the next valuable thing, though.
I know you can do that, though, because you can say as discussed in section two, and then
if you introduce a new section, it'll still know that it's section two as opposed to something
else.
It does track those.
The next thing I want to show you guys has to do with references, right?
You guys will have to write with references.
In fact, this is a big part of college, learning how to cite things properly.
And plagiarism is something that always comes up.
Every year there's always issues with plagiarism.
It's usually not intentional.
It's just that students sometimes haven't been taught what to cite, how much should
you cite, like what's the right way to do it.
So if you take anything that's original idea and contents that you came up with, you're
fine to just type it.
But if you're taking something that you took the results of somebody else, right, especially
if it's published results, you just need to cite it, right?
And I think some people think that they should just copy and paste it off of Wikipedia or
whatever, because it makes them look smart, because they found good content.
We like the content.
I like that to be in the live reports you guys write.
I just want you to cite it correctly, right?
So how do you cite things?
We're going to teach you a really helpful trick to do that.
If you guys go to just Chrome or whatever and type myendnoteweb or something, it should
pull up that stuff.
Yeah, myendnoteweb.com, all one word, myendnoteweb.com.
EndNote is a commercial software you can buy, and for some reason they made a free version
of their product, which works like perfectly great, like it's wonderful.
Maybe it like limits you to like 20,000 citations or something.
Like for what you guys are going to do, this works great.
There's other free ones if you'd like to use, there's Mendeley and Zotero, which are also
free ones.
I've used both of them.
They're fine.
I like EndNote Web.
Go ahead and click it.
When you go here, you have to sign up for an account.
Just like sign up with your Facebook or LinkedIn.
They have different options.
Do something easy to sign up.
So I've already got my account, but you could sign up using whatever, LinkedIn if you want
to be professional, right?
Go ahead, sign in, and the first thing you're going to see when you come here is the following.
We're all on it?
Yeah.
First thing you're going to see, if you haven't set it up, you're not going to have a lot
of this stuff.
I use this regularly, so I already see that I'm using this, so for example, I see here
under my references, I have a total of 965 references I've put in here, right?
Now they're not all just in a random pool.
I've set them aside to different projects.
So on one grant, like my porous thermoelectrics grant, we used a whole bunch, right?
Or fluoropolymers, we've got a few, and then some of these, if it has the icon of people
next to it, that means I've actually shared that with somebody else, so this is really
powerful.
You guys are going to work on teams, and if you're all doing this, you wonder, why are
we going to keep our references all put together?
I can invite Brennan, for example, to work with me on my ceramic algae filters, right?
My reference list, and he could contribute there as well, right?
So this is where it's going to go.
So then the question is, how do you get your references from wherever into end-to-end?
That's the next thing I want to show you.
So to do that, we use a tool called Google Scholar.
You guys are already using Google Scholar, I hope, at some point you've used it before.
It's genius.
When I was a student, it didn't exist.
We had to use something called SciFinder, which you had to pay for, which we had to
use it at a lab on campus.
It was just a pain in the butt.
Now it's just way, way, way better.
Google Scholar is as good as anything else out there, I feel like.
It's terrific.
It's what I use for all my research.
So here's how it works.
You're going to search for something.
So next week, you're learning about electrochromics, so you type in electrochromic films, right?
I have a question.
Yeah, go ahead.
When you're off-campus on research teams, you have to sign in, so if we're on campus,
you should have access to the articles if you're on campus, because it recognizes it
by IP address.
But you can get off-campus access, where do I have it?
Right here, let me open this.
You can get VPN access from home.
So oh, they just closed this?
Yeah, so just go to the library, and then research databases, and then it goes to Google
Scholar.
Ah, okay.
So from home, from your computer, you just log into the library page?
Yeah, and then you go to Google Scholar through the library page, and then you'll have access
to everything you would normally on-campus.
Good to know.
You don't have to go through the library to know about SciVal.
About SciVal?
That's the only way I know it.
Okay, we do.
So let's say you're looking through whatever, right?
Electrochromic films, something related to your research team.
You want to do your lit review right, meaning you're going to cite what people have done
in this field.
So read through these articles.
When you find ones that are relevant that you need to talk about, right?
The first person to do this was this guy, or this lady figured out the clever way to
do it with these materials, whatever it is.
Now you click on these, and the current way that you do references is how.
How would you do this?
If you wanted to cite this, what would you do?
Publisher and the author.
Yeah, it's this big, long process.
You look at it, and you're like, okay, we've got a title, we have our authors right here,
we've got the journal.
Or you steal the DOI and put it in.
Yeah, so here's what I suggest you do.
You can do that by hand, and sometimes these even have a little download our references
thing, but even that's worse.
Come over here, and under the heading, once you know that you do want to cite that article
and you know what you want to say about it, grab this little cite button right there,
click cite, and you could just copy it under, here's five different styles which are common
reference styles, but you can even do better.
You could just copy and paste that into your references at the end.
But you're still going to run into the same problems as before, right?
What if you use that sentence at the beginning of your paper, so you've got reference one.
And then you have to move it.
And then you've got to move it.
Because your things you write up should have a significant number of references.
At least 10, I think, is the rule.
Right?
We want at least 10 references.
Like, learn how to read stuff.
So, that's a pain in the butt to keep track of.
So instead, we're going to use a reference manager.
So you're going to click End Note down here, right?
See these little links at the bottom?
Click the one that says End Note.
Here, I'll show you.
You don't have to highlight anything.
Let's pick on this one, right?
Let's say I wanted this compound, or this paper.
I'm going to click cite.
I don't have to highlight anything.
It did that automatically.
But I just come right down here to End Note.
And I click End Note, right?
And you see that it saved them.
It called one scholar one, scholar two, right?
And it saved as a .enw file.
End Note web file.
So we're going to take that, and we're going to throw that into our End Note web collector on the internet.
Here's how we do it.
You go to collect.
And then import references.
Right?
Collect.
Import references.
When you're there, you're going to go to choose your file.
I wish you could do these more than one at a time.
It's silly that you can't.
Last I checked, you couldn't.
Now you've got to click just one at a time.
So click this one.
You have to tell it what your import type is.
Because it can actually import lots of different styles.
Refworks, and you know, whatever else.
I've already told it that I'm only ever going to use End Note import by selecting your favorites here.
Just to make my life easier.
So there's only one option to click.
So I suggest doing that.
And then you have to tell it what group to send it to.
Right?
This isn't a real project, so I'm just going to send it to this cryptically named Alex group.
I don't really know what that is.
Right?
Alex is going to have a very weird reference.
He's getting all sorts of new things.
He's like, what is this?
And then hit import.
And it'll think for a minute.
And it'll say, I'm importing records.
And then right here it'll say one reference was imported to the Alex group.
You can do the same thing with your second reference.
Actually, we imported that same one yesterday.
So we're going to have two of it in there when I go to look at it.
Right?
Import it.
It's going to say one reference import.
So now you can click on the Alex group or whatever group you're sending these to.
And when you look in here, now it's got all your references you brought in.
Yeah.
So we did this on Tuesday.
So some of these are repeats.
It doesn't know when you have a repeat.
Right?
So you can delete it.
If you see that you've got a repeat, you can delete it.
Where is it?
Delete.
Right.
You can kick it out like that.
But now that I've got them here, let's say I've collected some data and I'm feeling pretty
good about that.
Now I want to bring it into Word.
So how do you do that?
You have to go to format, cite while you write plugin.
This is a free plugin that you download.
And then the next time you open Word, what you'll find is there's a new tab that wasn't
there before that works.
It just makes it easy to interact all these references into Word.
So I've already done it.
So when you get a chance, go ahead and upload it.
But you'll see this EndNote tab appears up here that wasn't there before.
Go ahead and click that.
It has to log in.
Tell it to save your information so you don't have to type it every time.
It's doing it automatically for me right now.
It takes just a second.
It's accessing my EndNote web account.
Now it's open.
When I want to cite something, like if I'm here, tell us what's been done in the field.
If I want to cite something right there at the end of that sentence, all I have to do
is click insert citation.
And I have to type something.
I think these titles said something like W03 in the title.
Yeah, there we go.
I can cite one of them.
I can cite both of them.
I can highlight both.
Let's say we want to cite both of those.
I click it.
It's automatically applying a certain reference style, whatever the style I had for the references,
which is academic medicine, for whatever reason that's the last one I was using.
And if you look at the bottom of the report, it creates a reference bibliography right
here automatically.
It's in that same dark highlighted text, meaning this is something that Word is doing for you
that you don't have to mess with.
And it will move with you.
That is...
It's super powerful.
They should teach this in English classes too.
They should teach this.
I agree.
But they don't.
I had to rearrange 40 sites.
And it's awful.
It's awful.
It's the absolute worst.
It's garbage.
Three hours of English.
And then again, if you want to add something else, let's say I want to cite something else
I've written or a couple things.
Let's say I want to cite these three papers.
I throw them in.
It's doing the same thing.
And then, for whatever reason, this section after you edit your paper, you're like, you
know, I really need to move that or something.
You move it here.
You say...
I think you might have to highlight the whole thing and then hit update citations.
And now it's 1, 2, 3, and 4, 5.
Just super, super slick.
Let's say that you're writing this because we're telling you, hey, for this lab report
this week, we want you to pretend that you're writing an article that would be submitted
to Journal of the Electrochemical Society, right?
Easy.
You just search up here.
You find Journal of the Electrochemical Society.
If you don't see it, come up here to select another style.
And there's like a gazillion more.
You could scroll down.
Just go to J.
Well, that's not going to help.
There's a million journals, right?
You would scroll till you find Journal of the Electrochemical Society.
It's going to take forever.
I'm not going to go there.
Journal of Colloid Interface Science, right?
You pick the one that you care about.
You click it.
And it will automatically change how these things are written.
What do I mean?
It will change whether or not they have italicized text, whether they have bold.
Whether you just use the first initial of the name, like J for Jeff, as opposed to writing
out Jeff.
Let's pick another one where it does the whole thing.
Let's do Academic Medicine.
I think it actually has the whole names, right?
It should say, yeah.
No, I thought it didn't.
Let's go to ACS.
Oh, they're all doing it.
But you see how they're changing?
Some things are bold.
Some things are italicized.
Because every journal has its own persnickety way of wanting their references.
I have no idea why.
Why they can't just use some standard style.
But the great thing is, we don't care.
Because it's as easy as selecting it from the drop-down list, right?
So those are the things I wanted to show you about this.
That's half of our time.
We covered how to use Reference Manager and how to do some stuff in Word.
Can I answer any questions before we move on to figures?
You said you're going to...
This is a screenshot.
I'm screenshotting this.
I'll post it to YouTube.
Will you post the link to your YouTube?
I'll share that again.
Any other questions with this?
I promise this will make your lives a lot easier.
So the only tricky thing is downloading that site while you write plugin.
And even that's not hard.
Again, format.
Site while you write plugin.
It's like one file.
I think it's painless.
I hope it is for Mac people.
You Mac people that are out there.
I hope it's also painless for you.
You have shaky grips with shit.
I love shit.
That's my favorite band.
I love it too.
My absolute favorite band right now.
What was I going to say?
Okay, let's move on to figures.
Having talked about Word and how to...
We didn't have time because I've got to run to group meeting at 5.
But please, please do read this document.
The MSC Lab Report Writing Guide.
This is under files on Canvas.
This really just talks in more detail about what should go in different sections.
Should you use present tense or past tense?
Should you use the word we versus I?
It talks about some of that stuff.
What's the right length?
It covers a lot of stuff that I think that you'll like.
So it's worth reading through this.
Faculty who you'll be writing lab reports for.
We came up with this.
Now let's talk about figures.
I am passionate about nice figures.
To the point where I'm quirky about it.
Poor Emily and Christian who have worked for me before
know that it's a pain in the butt
because I like my figures to look really nice.
And the reason why is because I did a post-doc
with this guy named Ram Seshadri at UCSB
who converted me to believing that
if you're going to collect data
and you're going to spend the time to do the research
make sure that people can understand what you're talking about.
Make sure you support data mindless of whether people are listening
or they can interpret it or they care.
Make sure you answer those three things first
that they can understand it, that they're actually listening
and that they care about it before you ever tell them
what it is you did.
And part of that is making nice figures.
So let me skip some of his philosophy on that.
And he's got opinions on everything on fonts.
I don't know.
I don't feel quite as strong on this.
But there is some science behind this
on whether you use different fonts.
And there's really two classes of fonts
Serif and Sans Serif.
Serif means it has little feet
and Serif means feet.
Do you spend time on their feet?
Well, I just feel they're kind of like the buttons.
Okay.
So these letters have those things
and because they do, it makes it easier for your eye
to read long blocks of text that are in paragraphs
where it might be easier to shift around if it didn't.
So it's not a big difference
but it is a little bit nicer.
So if you're going to write a block of text
as in a research article, a Serif font is nice.
I don't care if you do this on your lab reports.
Personally, pick whatever looks nice to you.
I've seen it done both ways.
At least that has some rationale behind it.
Now if you're doing a PowerPoint presentation
or something where it's a small block of text,
a short blurb, you don't need the feet
and some of these actually look a little nicer.
Like the Sans Serif.
And he has a lot of opinions about those.
I'm going to skip some of this.
Obviously, there is some font which we just shouldn't use
because it looks childish.
Please don't do it.
Just last year,
one of the two,
they did a review meeting
that we were getting together.
Christian and I and our team
were showing what our results were
and they were showing their results
and theirs was in this old-timey font
with the bullets shot through the letters.
It was bad.
So don't do that.
Let's now get to figures.
Oh yeah, Logan was there.
Now let's talk about figures.
And figures, it's sort of one of those things
like you know a good figure when you see it
and you know a bad figure when you see it
but there's aspects that we can actually identify, right?
A good figure has things about it that we can actually talk about.
For example,
in this figure, an example of a good figure
you can feel free to dip it, it's fine.
But one of the things I like about this
is that it's very easy to see
that there's two different data sets we're talking about
and it's very easy to tell them apart in this figure.
That's one thing I like about it.
Second thing I like about this is that
you should be able to see more or less
what these values correspond to.
So if you wanted to read a value off
it's a little bit easier to do so, right?
They've done things like made tick marks
which are intelligently spaced.
They're actually the same spacing on the X and the Y axis
which makes it look more uniform and nice.
The font here, everywhere throughout
legend, labels, tick labels
are all the same size and it's all large enough
that you can see it from far away.
They've done a broken line here
as opposed to a solid line and made the stroke rather large
So if you print this, if you're a cheap grad student
and have a color printer and you print this in grayscale
you'll still be able to see these two apart.
On Tuesday I had a student who was color blind
and red and green are very hard to tell apart.
There's different types of color blinds
and that's one that's really hard to tell apart.
So they've taken care to think about this stuff.
Let's show some more.
By the way, this is huge.
If this was an 8.5 by 11 sheet of paper
that's a 5 inch figure.
So on your lab reports, imagine how big 5 inches is.
It's like that big sitting in the middle of your sheet.
You probably don't want it that big.
You're going to shrink it down.
In fact, many journal articles
they require you to shrink it to a two column format.
I'm going to ask Emily and Christian
to have you guys turn in
in addition to using different references
I'm going to have them also occasionally
ask you to turn in a two column format
research report for that week.
And by the time you do that
that's going to make your figure shrink to 3 inches.
If you had small text
maybe that text seemed big for you up here
but if it was any bit smaller
by the time you shrink that to 3 inches
it's getting difficult to read this.
Same with the stroke.
If the stroke wasn't very thick
these things can get vanishingly smaller.
They're hard to read.
Notice that they're using nice colors
that look well together.
They're using complementary colors, orange and blue.
If you want a trick for doing this
you can just go to the color wheel.
If you want colors in general that look nice
there's tricks you can use
blue and orange look great together
purple and yellow look beautiful together
so picking things opposite each other
on the color wheel is one way to do that
but pick colors that look nice
and avoid things that are hard to see
like look at this green and this yellow
that yellow you can't see
or some stuff really stands out
and maybe that's ok
maybe you want it to stand out
because you're trying to draw attention to a data point
or one of your data lines
just be cognizant that it destroys consistency
Now let's take that exact same data
the exact same data you just saw
in the other figure plotted
this is the default output of a plot called Maple
Anybody use Maple before? It's a math software.
If you say plot these two data sets
this is what it spits out
Look how much worse this is than the other one.
What things do you see that you don't like about this?
Anybody notice something?
The numbers are tiny. You can't read any of these.
Green is really faint.
It's thin stroke lines
are too thin
Also this is red and green
You can't see these.
That looks the same to you.
What about this?
That's your separation between major tick marks
minor tick marks
Here that's your major tick marks
This is not consistent
This is not a square plot
I'm a big fan of square plots
for the reason that when you shrink it
if you put it in a two column format
when you shrink this thing down
compare these two data sets
by the time you shrink this down to 3 inches
because you have this in a square format
it's taller
you can just see this a little bit better
These are completely illegible
You've got these giant labels and you can't read any of the numbers
Most of this is common sense
I'm not telling you anything that should be surprising
I think
but I think what we often do as scientists
because this is common
you'll see stuff like this all the time
We just don't think about our audience
We just write it up
We know that I'm writing for TAs
or I'm writing for a professor
or whoever
but I don't think about who's going to interpret this later on
A little bit of care to make sure that it's easy to interpret
goes a long ways
There's other things you can do
There's things like
nice insets
You guys have all seen figures that has an inset in it
The normal way that people make an inset
is they make two figures
They shrink it down and plop it on top
So the text that was on this
usually gets shrunk super duper small
Here they've made a point to make it the same size
as all your text throughout
It's all the same font size so it's easy to read
even though you have taken up a small portion of the plot with this
Now insets work
but I think a better way to do insets
is actually to do multi-panel plots
and Excel cannot do multi-panel plots
that I know of
Maybe they're changing
They can't do it as far as I know
I'll show you briefly how to do it if we have time
Multi-panel plots are just a better way to show things
I'll show you some examples of why
in just a moment
Again, same exact data
In literature you'll see this all the time
where they'll just take another figure, shrink it and throw it on there
You can't read any of this
Why are you bothering to show me this if I can't interpret the data there?
By the time you shrink that down to a two column format
it's just gone
Another one
is sometimes instruments we use are so old
that it's not, or for whatever reason
it's easier to get the XY data off
It just creates a figure
I'll see this all the time where students will take
from the x-ray diffractometer
they'll just take the image and they'll just copy that
and throw it in their lab report
and it looks cheesy
it doesn't actually focus on the information you care about
I just don't like it
So what would be great
is if we had a way to take that data
and turn it into something that you could plot in Excel
or whatever else
So how would you do that?
How would you do it?
Well, the way that I did it
is we would get the file
emailed to us
Like a figure
Well, it wasn't a figure
it was numbers that you turned into a figure
and then you put them
you copy the two columns
that you want and you put it in Excel
and then you make a graph
out of that
Absolutely, that's what we'd like to do
We'd like to, in Excel, have a bunch of numbers
So let's put
equal random between
0 and 100
We're going to generate some numbers here
Once we have the numbers
it's easy to plot it in Excel
We just go to insert
and you do this little scatter chart
or whatever type of chart makes sense for you to plot
and you can plot it
Excel, by default, gives you stupid things
like chart titles, which you should never use
It makes no sense to have a chart title
You should change those
Justify it
Just put your caption
underneath
I think you can make these look kind of ok
Make it squarish
I would make my font larger
on both of these axes
You can't pick both at the same time
It's obnoxious
You can make an Excel plot look
kind of nice
Just take some care
because the default setting is not always the best way to plot it
You can't plot data in the first place from a JPEG
Right here
I can screen capture this
How do I do that?
Conrad knows all about this
because he spent a summer doing this
If you ever need to do a web plot
You go to web plot digitizer
Web plot digitizer
So if you just google
web plot digitizer
I think it's the best tool out there
Here's why I like it
You don't have to download anything
It will collect your data automatically
Here's how
First off you have to grab your data
Previous to you guys being here
I just did a screen capture
I've got that saved
as a JPEG or something
Right here
The first thing I'm going to do is load an image
This is just like the practice
You want to learn about their software
We want to do our own data
I'm going to load an image
I'm going to choose our file
I have to tell it that it's actually
a 2D plot with XY points
It's not like a bar plot or a radial plot
Apparently you can read those
I've never done anything but a 2D plot
The next step is we need to align the axes
You need to give it certain data points
so you can read the axes
You have to give it 1, 2, 3, 4 points on the axes
and tell it what those correspond to
After you do that, you can read any points you want
The problem is
the barbarians that made this figure
did not even give us tick marks
There's a lot of problems with this figure
The biggest one is that they didn't give you tick marks
and this is a spectroscopy file
The key thing you want
from a spectroscopy file is to figure out
the wave number, the position
at which one of these events occurs
Often times the intensity is not even that important
You just need to know where it's at
You can't even read that
First off, it's getting bigger that way
and then they have no tick marks
So we're going to do our best
The fact that they didn't include it
We're going to do our best
We're going to put one of these
on the 3500 right in the middle
If you look in the top right corner
it zooms in
You can actually move it with your arrow
Oh genius, I didn't know that
I don't know how to do that
Try and get there
Can I move it once it's selected?
Okay
So do your best to figure out
where these are
Hopefully they have tick marks
In this case the y-axis doesn't really matter
They're arbitrary units
but they didn't tell us that
because they're barbarians
Anyways, we've got this
So now that we're done with those four points
we hit okay
and it asks what are those points
So we say the first x point is at 3500
The second x point is at 500
The first y point is at 0
The second one is at 1
And neither of them are on a log scale
They're both linear scales
We hit okay
And now at any given point
Let's say this point up here
That should be about 1700 and 1
And you see it's doing that
I know that's science implementation
But 1700 and 1
Pretty great
But it gets even better
So now what you can do
Right now it's set to manual mode
And I'm set to add a point
If I click anywhere on here
It'll start grabbing those points
That works
And sometimes you have to do that
But it'd be even better if we could automate this process
And have it just grab our points
So let's go do that
Let's click automatic
We're going to clear our points
Let's click automatic mode
In automatic mode
It will collect everything of a certain color
So let's say this is a black line
So let's just tell it to click
Click here and then take everything that's black
And collect that data
And it's going to freak out
Because I'll show you
It's looking at the whole screen
It sees everything that's black on my screen
And it says data
And it's grabbing it
So we don't want all that data
But there's an easier way to limit it
Here's what we're going to do
We're going to clear the data points
We're going to now go to the pen
And with this pen
I'm going to draw on this thing
And you can actually scribble
Because as long as you only get the black stuff
If I scribble over here it won't matter
I just need to collect all this black stuff
So I would take a minute and I would do that
But I'm going to hit run
And now it only collects data points
Where you had that line on there
Which is pretty sweet
You still might get a few random ones
Especially if you have data points that are overlapping
Let's say you have one line here and one line here
Where they overlap it might collect some on the other line
It might get a little bit wrong there
But you can always edit that in Excel afterwards
You can throw those individual ones out
Generally speaking this is a pretty easy way to collect it
Now this default setting is that every 10 pixels
It takes a data point
But there's some really fun details here
We could probably get even more accurate
If we change it to collect more often
So here under this delta x and delta y
Let's clear our points again
And let's do
Set this to 1 and 1
So every single pixel it's going to collect another data point
So it's doing something funny
But there's a ton of data there
So whenever you're ready
You're ready to export your data
Then you just go to view data
It's all right here
You can copy and paste it from here
Or you can download the CSV
If you don't want to do the CSV
In Excel you can just change the
Format while you're saving it
The other important thing to know
About digitizer is up top
If you screw up your actions
You can always recalibrate
There's a data tab
You can make multiple data sets
If you have one graph
If you go to manage data sets
You have one graph that has five different lines on it
You manage data sets
And you can add data sets
You can name each of those data sets
And then you can
When you are selecting data right here
You choose this
You go to the next data set
So you just made a data set
And then when you're exporting
The data sets
When you click view data
You just also have a drop down
And that way you don't have to reload
The same image multiple times
Or redo the axis
You can just really easily
Off one graph make four or five
Different plots
That's terrific
Once you have it
You download this file
It's going to open it up in Excel
It already separates it for you
And then you can go ahead and create
You have to invert the axis
If you want it to look like the original
Any questions on how to do this?
So you can pull up any image
From any website
And then just save it
And create your own data
Obviously if you're using somebody else's data
You should cite it
And you should say data adapted
Figure adapted from such and such a reference
Then you're totally clear to do that
Nothing unethical about it
If you're grabbing data from a table
It's called Tabula
Rather than going through the table
And having to try to copy paste everything
It recognizes tables
And will automatically export it
To a CSV for you
It's usually pretty good
Giving you a really nice
So you can text with it
It takes tables from PDF
So you just download the PDF
As long as you're able to download the PDF
You can quickly go through it
But it is a software
It's not the same way
Okay let me say a few more words
I've got to split in five minutes
So let's just say a few more words about this
You guys will generate a lot of
Microscopy images this semester
You'll take optical micrograph images
As well as scanning electron micrograph images
Anybody used a SEM before?
Just today
Just today?
You did already today?
Oh well done, fantastic
So scanning electron microscopes are fantastic
And honestly it's one of the more important skills
That you'll get from this degree
People hiring MSc grads
They want you to do characterization a lot of times
They want to hand you a part that broke
And they want to say figure out why it broke
And often times that means looking at it under a microscope
Identifying flaws, identifying what went wrong
So you're going to generate images like this
This is a fine image, this is okay
It can be better though
One of the reasons I think it can be better
Is because if you look at these
Most SEMs will automatically generate a list of data down here
It's associated with how you collected the image
Things like the accelerating voltage, the spot size
The working distance
It's not that those things aren't important
It's just that they don't belong in this figure
You're not reporting on how you collected that when you show the figure
If it was important you would have described it in your experimental methods section
If it was important
Sometimes it's not
You don't need to report working distance and spot size
In your methods unless that's a key part of the research
So it would be good to get rid of this
Because it's small and hard to see
And so it's a distraction
You want people to draw attention to the signal
But you care about them learning
And minimize things that are going to distract them
So what you can do is
You can take this and you can make it a square image
Just by cropping that whole part out
And then the one piece of information that you did want
Re-plotting it yourself
Such that it's easy to see
The piece of information you wanted is the scale bar
Are we looking at the surface of the moon?
Or is this microscopic?
You don't know without the scale bar
Now that you've got the scale bar
And you know how big they are
You know that the particles are roughly
About a micron in size
The pores are about a micron in size
So this is now easy to see
I have my students do this all the time
Take your SM image
Crop it square
Put your own scale bar on there
So it's easy to read
There's a bunch of good examples here
We'll just talk about a few of them briefly
So here's an example of a really nice multi-panel plot
We use multi-panel plots
Most of the time when you have a common x-axis
It makes it easier to interpret sometimes data
If you have a common x-axis
In this case, they did three different
Experimental techniques
They did the top
This is a magnetic measurement
Here they did a specific heat measurement
So it's a thermal measurement
And here they did diffraction
Which measures the structure of something
Here we have Jake
Always
So the point
The reason that they showed this figure
Is they were trying to argue in their paper
Is that at a certain temperature
There's an event occurring
That is coupled magnetic, thermal, and structure
All those things
Something happens that relates to all three of those
So by plotting them all in the same x-axis
And making it easy to see
They've got the x-axis here
The same one here
You guys can cut across
They were able to show it really nicely
Now heatmaps are a great way to show a ton of information
If you've got 3D data, heatmaps are nice
But it's hard to tell with your human eye
Exactly what's going on here
So they were nice and they added as add-ons
On the wings of this heatmap
What the actual data looked like for one slice of this
They took that 3D data and made one slice of 2D
And they plotted it
And they showed this is what the actual data is here
Versus up here
So you can clearly see that there's an obvious difference here
I think this is a really nice way to present data
You should show your data points
As points
Even if they're lying on top of each other
Because this helps you know
How small of a step size did they take
With respect to temperature or whatever else
If you just show a line here
I'm left to wonder
Is this from 5 or 6 data points
But by showing the actual data
It tells you something about the carefulness of the experiment
So whenever you collect data
I'm a big fan of showing the actual data points
Even if they lie on top of each other
Okay, let's do a couple more
We're almost done
Again, even if you shrink that down
You can still read it nicely in a two-column format
Other things you can do
So he suggests that you avoid
Dark backgrounds
And even if you have amazing data
A dark background can make it look kind of stupid
Here's the discovery of superconductivity
By all means, a pretty awesome thing
And it looks stupid on a dark background
So I would suggest that we also not do that
Here's an example of a micrograph image
Which I think was really well done
What they're trying to show you here
They wanted to show that as you had different length scales
Of your material
You had different things happening
So if you zoom way out
You've got a big scale bar indicating that it's zoomed out
Here they show the microstructure
And you can see that there's this heterogeneous interesting structure
So then they zoom in down to 50 microns
And they label, oh these are actually two different phases
The half-Hoyzer and the whole-Hoyzer
That would be in the caption
You have to tell your recruiters what that is
And then, see, they actually do a selected area diffraction pattern
Which is proving that they know what one of these phases are
So, again
It's only taking up one figure spot
You could have made three separate figures
But this just looks nicer
It's telling you the key information
That as a function of hierarchy in terms of length scales
You see something happening here
One more key thing I want to show you
Which is how you make figures like this
I won't be able to show you how to do it today
But I'll at least say that it's possible and it's not hard to do
If you're talking about crystal structures
We're material scientists
A key part of what we do is we relate a property
To a processing to a structure
The property you got is due to the structure that you started with
Or if you change the structure
You're going to get a different property
So it's critical oftentimes for us
To describe changes in the structure
And that can even be crystal structure
Let's say you doped it or you mixed a different mixture of elements
And you go from having SR2 Ti1
To having SR4 Ti3
Now you can try and explain
What's happening in the crystal structure
What's happening in the crystal structure with words
You can say alright there's these
It's layered structure and there's different blocks
This is called a Perovskite block
And there's different numbers of those
And you're going to lose your audience if you try and do that
I usually have my students try and do this like interpret one
Where they don't use figures and they explain it
It's just so hard
Picture like a thousand words
This easily tells you what's going on here
You've got three different structures
And in the text you can talk about the number of these repeating units
But here you can easily show
One to two to three
As you change the composition
So you can make figures like this using a software
Called Vesta
It's free
It's actually very easy to use
At some point in the semester I'll make it
Required on assignment
So you'll get a chance to try it out
But just realize it's not hard to do
And you can make really nice looking high quality figures
I think we're almost done
Let me see if there's anything else I definitely want to cover
Before we wrap it up
Some more resources if you want to read more
There's a guy named Jean-Luc Dumont
And I'll share a link to him
He's really the master on communicating science
It is well well well worth your time
To take an hour over your lunch break
While you're eating your sandwich
And just watch a YouTube video of what he has to say about communicating science
Because it forever changed how I do it
I think it will be in your best interest
Okay any questions I can answer?
I did, this whole thing is screen capped
I did, I've got all those
Okay so you should post that
So you did an entire hour and a half talk on this slideshow
And went into detail about the software he uses
To make the graphs look really good

