Here (finally) are the survey results:

You can see the full visualization here.  Click the ‘view in new window’ link to see the whole visualization.  bl.ocks.org is a convenient website to see D3 snippets and see how they work.

In terms of student makeup, we have 23 undergrads (38%) and 38 graduate students (62%).  Of the undergrads, 3are sophomores, 5 are juniors, and 15 are seniors.  For the graduate students, 20 are first-year, 7 are second-year, 4 are third-year, and 7 are more senior than that (4+).

Hopefully this will be a quick introduction to help you get used to how to use Tableau.  The application is very handy for quickly visualizing trends in data, connecting to a variety of data sources, and mapping cities/regions and their associated data.

If you’d like to discuss your use of Tableau, ask for help or share neat tidbits, please use the Tableau discussion on Canvas!

Getting Tableau

First thing is first; get Tableau Desktop by following the steps outlined in the Tableau discussion on Canvas.   By following this route, you will get a license good to let you use the program through early June.  If you go the ‘academic route‘, you can get a Tableau license for as long as you are a student (renewed in year increments).  You can have multiple keys active for your installation (check under Help > Manage Product Keys…).

Loading data

Once you have installed Tableau, you’ll be greeted with a splash screen like the below.

Click the orange ‘Connect to Data,’ and then click ‘Microsoft Excel‘ in the top-left.  It should immediate drop you into the default Tableau documents folder, but if not, the path is a variant of ‘Documents/My Tableau Repository/Datasources.’  Open the ‘Sample – Superstore Subset (Excel).xlsx‘ file.

Once the file is loaded, drag the ‘Orders‘ sheet to the light box with the ‘Drag sheets here‘ orange text.  This will determine the data that Tableau will pull into the workbook.  By clicking on header symbols, you can change the type of the data represented by the column (number, string, geographic representation).  Right-clicking on the header brings up the option to rename the fields.  At the top of the window, there are two options for ‘Connection’: Live and Extract.  A live connection will continuously update from the data source, so if the Excel sheet changes, so will the resulting data in the Tableau worksheet.  Extract will make a static copy of the data.

Once you are satisfied with how the data looks, hit the orange ‘Go to Worksheet’ button to enter the worksheet.  Let’s take a little time to understand the interface.

Much like in spreadsheet programs, we have the ability to make multiple sheets.  This is especially critical in Tableau when exploring data, because you can make multiple views in the data and keep track of your exploration history.

Data is sorted into measures and dimensions.  Dimensions help to narrow down the scope of the data, while measures are the values being compared.  If Tableau has guessed the type of a particular field incorrectly, you can simply drag the field into the other container.

The interface in the middle is where the magic happens.  You can drag dimensions and measures to the ‘Columns‘ and ‘Rows‘ gutters, where it will plot the data in the order given.  You can also drag data fields to the Marks container, and change the color, size, and the type of the mark based on a particular data dimension.  We’ll see how this works soon.

The last thing to note is the ‘Show Me‘ tab on the right-side of the window.  This is a quick way to change encoding types.  When multiple measures and dimensions are highlighted (via ctrl/⌘ click) in the Data pane, the Show Me tab will suggest visualizations based on the number of dimensions and measures selected.

Making a Quick Visualization

Since this is a commerce dataset, let’s try to figure out what products that we’re selling are doing well from a sales standpoint, and try to figure out how profitability depends on the type of product and the sales region.

To do so, let’s drag Sales and Profit from the Measures pane into the Columns gutter.  Let’s bring in Product Category and Region from the Dimensions pane into the Rows gutter.  The worksheet now looks like this:

It’s a good start, but nothing really popping out yet.  Let’s redundantly map profit to color.  Drag Profit from the Measures pane to the Color square.  Once you do that, all the bars will be colored based on their profit margin.  Knowing what we know about color, let’s break the color ramp into steps instead of being continuous.  Right-click the color ramp now on the bottom-left, and click the bolded ‘Edit Colors…‘.  You can change the color palette, but what we want to do is make the color ramp stepped.  Check the check box and select 6 steps. The worksheet now looks like the below:

Let’s sort our dimensions based on profit.  Right-click the Product Category field in the Columns gutter, and click ‘Sort…‘.  In the sort dialog, select ‘Descending,’ and click Okay.  The product categories are now sorted in terms of their profitability from left to right.

It’s starting to become clear that although furniture sales are high, they suffer from a profitability standpoint.  As an exercise to get you more comfortable with Tableau: see if you can determine why. (hint: furniture are heavier items to ship!)

Mapping data in Tableau

Let’s now make a quick map.  Make a new sheet by clicking on the New Sheet icon in the sheet tray at the bottom of the window.   Ctrl- or ⌘-click the following data fields: City and Sales.  In the ‘Show Me’ tab, click the Symbol Map (the map with tiny circles). The worksheet will think for awhile, then generate the following map:

Let’s make this easier to understand.  Boost the size of the marks by clicking on the ‘Size‘ square and pull the slider up to about halfway.  Add profit information by dragging the Profit fields into the ‘Color‘ square.  Edit the colors by clicking the ‘Color‘ square, decreasing the opacity (it’s misslabeled as ‘Transparency‘) to 35%, and adding borders by selecting black under the Border dropdown.  The map looks like this:

We can filter the data shown on the map to just be furniture.  Drag the Product Category field into the Filters container, then select ‘Furniture’.  If you want to easily switch between product categories, right-click the Product Category field within the Filters container and select ‘Show Quick Filter.’  A quick filter window appears in the top-right.

Other Handy Tricks

Making hierarchies — You can identify hierarchies in Tableau by dragging measures on top of one another.  In the example above, you can make a hierarchy by dragging Product Sub-Category onto Product Category and naming it “Products.”  You can then drag Product Name into that hierarchy.  Now, whenever you use the Product field, you can choose the level of granularity to show, and enable drill-down on any visualization you make.

Getting the raw data for the current view (crosstab) — When you get a view on your data that you like, you might want to get the data values back out.  Right-click the sheet’s tab at the bottom of the window, and select ‘Duplicate as Crosstab.‘  This will make another sheet with tabular data.  You can copy this data and paste directly in a spreadsheet program.

Getting the visualization to size ‘right’ — You can coax the visualization to fill the current window size by selecting ‘Entire View‘ from the size dropdown, in the middle of the toolbar.  Usually this dropdown says ‘Normal‘ by default.

Clearing your work/resetting the sheet — You can reset the current view on the sheet by clicking the icon in the toolbar that looks like a chart with a red X in the bottom-right corner.  You can also use the adjoining dropdown to clear any sorts, filters, and additional context.

Further Resources

Video tutorials: Tableau’s video training & tutorials are fairly comprehensive, but aren’t quite time-efficient for learning the software.  Unfortunately, a lot of information is hidden within these videos.  This document largely covers the same material of the ‘Getting Started‘ video (in less depth). If you’re interested in learning more, I’d suggest the Visual Analytics > Analyzing, Grouping, and Sets videos, as well as the Calculations and Statistics > Trend Lines, Residuals, and Forecasting video.

Quickstart: For a quick jump into particular features of Tableau, I’d suggest going to the Quickstart portion of the Tableau manual.  Find a topic of interest, and information and directions are presented as a single-page PDF to get you quickly oriented.  If you need more in-depth help, the videos usually pick up the slack (though the small idiosyncrasies of how to effectively use the program are sprinked around all the videos and aren’t well-organized).

I was asked again, so let me clarify more.

With the “smaller assignments” (reading discussions, design exercises, seek and finds, …) we’ve been having so far:

You cannot get an A on an individual assignment. Getting an A is about consistently doing well in all parts.

So far, we’ve been grading things nocheck/check/check plus (1,3,4). If you’re consistently getting 4s, that’s as well as you can do.
(The 5 is more a notation for ourselves that we want to come back to something, than a statement that its “better than a 4”)

We might change the scoring system in the future. We will probably want to have a richer rubric for larger assignments.

If you’re wondering, I am really pleased with the quality of things being submitted so far (admittedly based on sampling). The vast majority of what I’ve seen is good. I have no problem with a class where everyone does well. I have no problem giving lots of good grades.

See also: online discussions update; grading, discussions, participation, …

This class is an experiment. I’ve never done anything like it.

Here are some more thoughts on discussions. Hopefully, you’ve seen my discussion of discussion (in fact, read that first if you haven’t).

Some people have asked (to Alper, not to me directly), some variant of:

• How do I participate in a discussion?
• How do you evaluate a discussion?
• What if I have nothing to say?
• What do I need to get an A?

If you think you’re panicked about how you will be evaluated in this class, we’re even more panicked about how we will be doing the evaluation. Let’s try to make it easier on each other.

If you haven’t seen the course philosophy on grading, you might want to look at that.

It can be intimidating to join in a discussion. There is a diverse range of people in this class. There might be someone who knows more, who has thought about the topic more deeply, is a more eloquent writer, … You might have a great idea, but someone else beat you to it. You might have an idea that you’re unsure of, or that you don’t know how to express clearly. You might not understand what another person is talking about, or might misinterpret what they said.

And then, there’s this issue that these discussions are the primary data we have for evaluating your performance in class. So, you might not want to say things that are wrong.

Ideally, we can make the discussions be a “safe” place where people can express their ideas without being intimidated. Where people will support each other in learning. This requires all the participants to be polite, to help each other, to be supportive of each other – both giving and receiving. As far as we can see, people are doing a good job of that. We can’t police everything in real time, but it doesn’t seem like we need to. (if someone does something you think is inappropriate, please inform Alper and I immediately).

As far as grading: what we care about is that you participate effectively, and show some basic understanding. The discussions are set up (or we try to set them up) so that you can achieve the demonstration of basic understanding in your initial post. The discussion part can be hard: sometimes, there isn’t anything to add. In an in-person discussion, everyone can nod in agreement: on Canvas, a dozen people making “I agree” posts can get tiring, and dilute the content. Although, it’s a nice gesture to be supportive. There’s a fine balance, and hopefully it will work itself out (in some systems, they have mechanism for this).

So, here are some ground rules:

1. It’s OK to disagree, if you do it respectfully. If there was a question with a clear right answer, you wouldn’t need a discussion.
2. It’s good to ask questions if you don’t understand. This can be a great way to give people feedback on how they express their ideas. It’s best if you are specific, and critical that you are respectful. You can even ask questions about the content.
3. Be tolerant of others: the responses are often “off-the=cuff” – written hastily, sometimes in someone’s non-native language, without tons of time to think everything through.
4. If you don’t have anything to add, you don’t necessarily need to add it. Although, it can be nice to be supportive – especially if you can add a little, and be specific. Saying “I agree, and you explained it well” is a nice complement – but might not add too much to the conversation.
   Trying to add: “I agree, but I might have phrased it like X” might help everyone understand by giving a new perspective, and is a vote on the idea, but can be a subtle put down that you think you can express it better, and that you want to claim some “ownership” of the idea. For a class discussion, I think it’s more OK than in other settings. Especially since there is no way to enforce fairness in turn taking (some people will respond quicker)
5. For evaluation, in the original post, we are mainly looking for evidence that you did the required assignment (reading or design), and have given it some thought. It’s less about right answers (again, the interesting questions don’t have obvious right/wrong).
   Beyond the original post, we are mainly looking for evidence that you are thinking about things.
6. We understand that the opportunity to participate in discussion involves an element of luck. If no one says anything worth responding to, you can’t make a good response! If someone posted your idea before you, it’s hard to sound original. Hopefully, things are structured such that there will be enough opportunities.
7. If you say what you got out of a reading, or your opinion, it is correct (it is true that it is your opinion). We can try to (constructively) change your opinion, or help you get more out of a reading.

Trying to “grade” discussion is challenging, subjective, and possibly a fools errand. That’s why we’re more looking for consistency over the whole, and only “scoring” in broad buckets. And, we might need to replan if need be.

As you have noticed, we are asking you to upload a lot of images – in time there will be more (since we’ll be asking  you to make pictures as well).

Two things to keep in mind:

1. With Canvas, there is a 50MB upload limit per student. We don’t know how to get around it. This should be enough – unless you upload massive pictures. If you start to run against the limit, contact Alper and we’ll work out another mechanism.

2. For some assignments coming up, we’ll ask you to do things on paper and upload them. 

If you don’t have access to a scanner, a cell phone or tablet (iPhone or Android) can serve well, given appropriate software. I like CamScanner (http://www.intsig.com/en/camscanner.html) – its one of many scanning apps. It has the advantage that it is cross platform (iPhone and Android), and they give the “full” version free to academics (look on their support page – its hard to find). It does a nice job of keystone correction, image cleanup, …

No matter what you use, watch the file size – its easy to make gigantic files with a scanner! 

If you’re totally stuck, you can ask Alper for help with scanning.

Note that in some cases, we will ask you to scan things and bring the original to class.

Link: Discussions on Canvas

When we originally designed the online discussion assignment type, there was a lot of uncertainty about how the class would be run. (both technically, and in terms of how people would respond). With 2 weeks of experience we can now be more specific and less confusing.

With the kinds of topics we’re covering in this class, discussion is important. Whether its talking about different viewpoints, or ways the ideas might be applied in practice, or critiquing designs, or even just clarifying ideas.

We will use online discussions as the primary mechanism for assignments in this class. Even for things that are primarily “handing things in”, there is value in conversation and critique beyond the initial submission.

Discussion assignments (either reading or design) will always be due the day before class. All deadlines are aligned with the “day” (so if its after midnight, it’s the next day).

So there will be two discussion assignments per week. If it’s a Monday or a Wednesday, there’s a discussion assignment initial posting due. Unless we explicitly tell you otherwise.

For discussion assignments, there is a deadline for your initial posting. Because participating in the conversation is important (and graded), we will not do evaluation for at least 1 week after the initial submission date. Effectively, the discussion is “open” for 1 week after the initial submissions for conversation where everyone has submitted their initial assignment. (of course, you can discuss before the deadline too).

So, at any given moment, you’ll have one discussion assignment that’s starting, one that you’re in the middle of discussing, and one that’s coming to an end.

Note: in addition to discussion assignments, there are also “seek and find” assignments, and 838 Bonus assignments. And there might be “non-assignment” discussions (either student initiated discussions, or non-required discussions, or …)

Grading Discussion Assignments

Discussion Assignments are graded. For this we are still working out the details – it is hard, since it is quite subjective, and we want to have enough bins to get differentiation, but not so many that the noise of subjectivity dominates. As of now, the scoring system for discussions is (sorry, we’re stuck with numbers because of Canvas):

• 0 = did not turn anything in
• 1 = minimal – not a complete assignment
• 2 = late, but otherwise good assignment (minimal late assignments get 0)
• 3 = complete
• 4 = good / great
• 5 = rare event – something notable

Note: we don’t expect that 5s are that common. For a “top grade” 4 is sufficient, a 5 is more like a 4 with a notation to ourselves that we want to remember it for some reason.

For a reading assignment…

• to get a 3, you must really answer the given questions in a way that convinces the grader that you’ve done the readings and thought about it.
• to get a 4, you must show depth of thought in your initial answer, and contribute meaningfully to the conversation.

For design assignments, we might need a different grading scale, and we might need to have separate grading for the initial posting and discussion.

Other Notes

We will divide the class into discussion groups (since having a discussion with 65 people is too hard). We will periodically shuffle the groups – not too often (since getting to know people is valuable). Canvas makes elements of this convenient, but there are issues when you are “in your group” you can’t see things outside your group.

Please be respectful of others in online discussions. There are lots of discussions of online etiquite (some nice ones are from CSU, Madison College, generic core rules, etc.) But the biggest thing: be a reasonable person and respect others. Remember that in this class, there is a wide range of students: people with different backgrounds, different levels, different abilities to write English, etc.). We reserve the right to remove inappropriate postings.

Learning Objectives

For reading assignments, the discussions are mainly to make you think about the concepts you are learning, and to build upon them by conversation with your peers.

For design discussion assignments, the goals are to give you practice at the skills of doing visualization (e.g. making designs, identifying tasks, critiquing, …) and also to re-enforce class concepts by having you reflect on others’ work and  getting feedback (and engaging in discussion) about your work

I always thought a “syllabus” was a description of the topics covered in class.

But, it is actually a more general discussion of official information about the class. (see https://kb.wisc.edu/vesta/page.php?id=43346 for the university’s definition).

This brief syllabus has the basic info. Really the course web is the syllabus, and most topics here are discussed in more detail somewhere (links provided from this document – although this document is written first, so that the links may appear later). In general, you should look at the “Course Policies” topic (which has lots of articles), the individual postings about readings, and the “Class Schedule.”

Basic Info:

Course Name and Number:

CS638: Data Visualization
CS838: Data Visualization

There are, technically two separate classes that meet together and have overlapping assignments, readings, and activities. Grading will be done separately for each class. Unless otherwise specified, all information applies to both classes.

The course numbers (638, 838) are the “Topics in Computing” – the generic numbers we use for new classes that haven’t gotten their own numbers yet.

Instructional Staff:

Instructor:
Michael Gleicher, Professor, Computer Sciences
Office: 6385 Computer Sciences
Office Hour: Tuesday 2-2:45, Wednesday 11-11:45, or by appointment
Preferred contact is by email: gleicher[at]cs.wisc.edu

TA:
Alper Sarikaya, Graduate Student, Computer Sciences
Lab: 1347 Computer Sciences (Graphics Lab)
Office Hour: Monday 11a-12p, Thursday 1p-2p, or by appointment
Preferred contact is by email: sarikaya[at]cs.wisc.edu

Class Meetings:

Tuesday and Thursdays, 11:00 am–12:15 pm
Room 1221, Computer Sciences

Note: this room has tight aisles, and we will be filling most chairs. If you come late, you will be climbing over other students to get to a chair. So please come on time!

Also note: students are required to come to class.

Exam Period:

Monday, May 11, 2015, 5:05 pm–7:05 pm.
We probably won’t have a final exam, but I reserve the right to schedule something in that timeslot.

Main Topics: (see the detailed schedule for more information)

• Motivations for visualization and the types of visualization
• Data and Task Abstractions
• Evaluation, Validation, and Design Process
• Human Perception, and its impact on Design
• Encodings and Layout
• Graphs, Networks, and Multi-Variate Data
• Clutter and Dimensionality Reduction
• Experiments and statistics (for visualization, and visualization for)
• Animation and Motion
• Scientific Visualization
• Depicting 3D

Learning Outcomes:

• Students will understand the potential of effective data visualization.
• Students will understand the key principles for the design of effective visualizations.
• Students will be able to design and evaluate data visualizations for a variety of tasks.
• Students will understand the relevant basics of visual perception and its role in designing.
• Students will understand some standard visualization methods and their applicability.
• (838 only) Students will gain exposure and practice with some of the skills required to be a researcher and practitioner inthe field of Visualization.

Things that we will not do (Non- Learning Outcomes):

We will not teach students about implementation of visualization systems.
We will not teach students to use specific visualization systems.

Text and Readings (summary, see the other postings)

There are two textbooks: Munzner’s Visualization Design and Analysis and Ware’s Visual Thinking for Design. Both are available online via the library.

There will be many other readings distributed via the web. There will be a combination of book chapters (provided under terms of academic fair use), research articles, and web pages.

How Students will be evaluated (class activities – see specific postings about types)

• Participation / In-Class Activities: We will not grade participation in the usual sense (speaking and interacting in class). Participation will be assessed via in-class activities.
• Reading: There will be a substantial amount of reading for this class. Most readings will have small online assignments that will give you a chance to demonstrate that you have done the readings and learned something from them.
• Online Discussion: There will be assignments where your job is to participate in an online discussion. Generally, we will break the class into small groups and each group will have a separate discussion.
• Seek and Find: You will be asked to find an example of a visualization (generally on the web) that satisfied some criteria, and to answer a question or two about it. Generally, the critiques will happen seperately.
• Design / Re-Design Challenges: You will be asked to make a visualization that solve a problem: either from scratch, or based or to improve an existing visualization.
• Peer Evaluation: (838 only) students will be required to evaluate their peers (and assessed on their ability to do so).
• Summative Assessments: I do not plan on having a traditional exam in this class (but I reserve the right to have a take home final). However, we will have an activity that requires you to review what you’ve learned. (the benefit of an exam is that it forces you to review and internalize the material – but hopefully, we can find less painful ways to make that happen).

All activities will be considered in determining the final class grade.

Grading:

The two sections (638/838) are graded independently.

Grading Standards:

Students who successfully complete most assignments will earn a B. Earning a better grade will requiring doing better than just “succeeding” at the assignments.

We will use a curve at the end of the class. If fewer than 10% of the class does not meet our absolute standard, we will shift grades accordingly. Of course, more than 10% of a class can earn As. If too few people earn As, its the staffs mistake (we didn’t teach well enough). If the entire class earns As, we have an exceptional class (don’t laugh – it has happened!).

Course Infrastructure (summary – see infrastructure posting)

We will use email to the students @wisc.edu email address for individual communications.

We will use the university provided class mailing list (to students’ @wisc.edu mailing lists) for announcements. However, the class web will be the primary mechanism for announcements.

The class web will be the primary source of information on the class, and students are required for its content. The website provides a facility to send email notifications of new information. You must sign up for this if you prefer this mechanism.

We will use Canvas as the course management system. Students will be required to submit assignments there, as well as to participate in online discussions.

We will provide readings online through a protected reader (web access only to students in the class) to comply with academic fair use standards.

Course Policies (summary – see postings):

• Class attendance is required, in-class assignments cannot be made up. There is some leniency.
• Late assignment policy will be specified for each assignment.
• You are required to be aware of content on the web site.

Software and Computing:

Students will be expected to be able to use basic computing software (a spreadsheet, a presentation preparation program, …). Using more advanced tools (for software development, data analysis, statstical analysis, etc.) is optional.

For students interested, Tableau has made their software available to the class through the Tableau for Teaching program. Tableau’s data visualization software is provided through the Tableau for Teaching program. Use of this software is optional.

You can use computers in the CS computing labs if you need to. Contact course staff to make arrangements. It’s probably best if you use your own computer (if you have one), or at least one you have more convenient access to.

As you probably observed, this class is actually two classes: CS638 – Undergraduate Topics in Computings and CS838 – Graduate Topics in Computing. These two classes will meet together, and share (almost) all of the material.

You should ask yourself: which one should I take, and sign up for the correct one. Note: if you want to switch sections, we will make it possible for you to do so during the 1st week (and before). Don’t worry about the other class being “full”.

How you should decide:

1. If you are an undergraduate, sign up for 638. Period.
2. If you are a CS graduate student and want “core credit” sign up for 838.
3. If you aren’t willing to program, don’t sign up for 838 (we’ll expect you to implement things – and assume that you’ll be able to figure out tools to do so).
4. Otherwise, read the description of what the differences will be below and choose. Selfishly, I would prefer that more people sign up for 638 (so that the numbers look good), but its more important that people choose what’s right for them.

The requirements for 838 are a superset of those for 638 (that is, an 838 student will be asked to do all the work a 638 student does – and more). The two groups will be evaluated independently (separate curves).

The differences in the classes stem from the fact that I view a graduate class as being a way to train students to be researchers, educators, and advanced practitioners. It’s not just about more advanced material.

Note: none of the extra work for 838 is “busywork.” I actually think that by doing the additional things, you will get more out of the class – both in terms of the amount you learn about the topic, but also in terms of general “skills you should be developing as a grad student” development. But you will also put more in.

Normally, I would expect grad class students to work on presentation skills and to have an independent project as part of the class. Given the large size of the class this semester, that may not be practical.