DS4UX (Spring 2016)

From CommunityData
(Redirected from DS4UX)
Community Data Science: Programming and Data Science for User Experience Research
HCDE598A - Department of Human Centered Design & Engineering
Instructor: Jonathan T. Morgan
Course Website: We will use Canvas for announcements, turning in assignments, and discussion. Everything else will be linked on this page.
Course Description:

Success in many UX related roles, particularly user research, require workers to possess an understanding of data science concepts and to have facility with the tools of data analysis. This course introduces students to widely-adopted programming and data science tools to give them the skills to use data to answer questions about the characteristics, behaviors, and needs of people who use a wide variety of products.

This course has students working with real data from real users. It is built around scenarios that are directly relevant to performing user research in industry, such as:

  • identifying user segments (e.g. power users) and popular content
  • manipulating very large datasets (too big for Excel!)
  • performing data visualization and statistical analysis using code (not GUIs)
  • implementing experimental designs such as A/B tests and funnel analysis



Overview and Learning Objectives[edit]

The goal of the course is to provide students with a basic grasp of programming and data science concepts using tools that they can reuse elsewhere. No previous programming experience is required, or even expected. While the course is framed around user research, the use-cases we will work with are relevant to a wide variety of non-engineering roles in software development and the broader technology industry.

Upon completion of this course, students should be able to:

  • Write or modify a program to collect a dataset from Wikipedia or the City of Seattle’s open data portal (Data.Seattle.gov)
  • Effectively read web API documentation and write Python software to parse and understand a new and unfamiliar JSON-based web API.
  • Understand database schemas and use MySQL to extract user data from relational databases.
  • Use web-based data to effectively answer a substantively interesting question and to present this data effectively in the context of both a formal presentation and a written report.



Where does this course come from?[edit]

This course is based on the curriculum of the Community Data Science Workshop series and Professor Benjamin Mako Hill's course COM597G: Community Data Science: Programming and Data Science for Social Media, offered first in Spring 2015, as well as in Spring 2016.

For the first three weeks, we'll be paralleling these previous courses quite closely. Once we have hammered down some of the basics, we'll start to explore some tools, skills, and use cases that are more directly aligned with HCDE backgrounds and UX professional contexts.

Note About This Syllabus[edit]

You should expect this syllabus to be a dynamic document and you will notice that there are a few places marked "To Be Determined." Although the core expectations for this class are fixed, the details of readings and assignments may shift based on how the class goes. As a result, there are three important things to keep in mind:

  1. Although details on this syllabus will change, I will not change readings or assignments less than one week before they are due. If I don't fill in a "To Be Determined" one week before it's due, it is dropped. If you plan to read more than one week ahead, contact me first.
  2. Closely monitor your email or the announcements section on the course website on Canvas. Because this a wiki, you will be able to track every change by clicking the history button on this page. I will also summarize these changes in an announcement on Canvas that will be emailed to everybody in the class.
  3. I will ask the class for voluntary anonymous feedback frequently — especially toward the beginning of the quarter. Please let me know what is working and what can be improved. In the past, I have made many adjustments based on this feedback.


Books[edit]

This class is going to be a studio and project based class. Although we will not rely very heavily on readings or discussing them in depth in class, Python for Informatics: Exploring Information by Charles Severance covers much of the material we cover in this course, and can serve as a reference work for review and self-study.

The book is available online for free but you can also buy a physical copy of the book from Amazon or get an electronic copy from the Kindle Store. According to the book's website: "The goal of this book is to provide an Informatics-oriented introduction to programming. The primary difference between a computer science approach and the Informatics approach taken in this book is a greater focus on using Python to solve data analysis problems common in the world of Informatics."

Some people find it's helpful to have a book to learn a new programming language: it tells you want you "don't know you don't know". Other people prefer to use adhoc resources. I'll point you to resources I find helpful throughout the semester.

General Notes[edit]

  • I expect you to come to class every day with your own laptop. Windows, Mac OS and Linux are all fine but an iPad or Android tablet is not going to cut it. We're going to install software during the class and you'll be working on projects for homework so please bring the same laptop each time. If for some reason your laptop dies mid-course, please contact me so we can get your new one up to speed.
  • Much of the class will be project-based and Ray and I will be available to help you through challenges you encounter in this work during class. If you have questions and need to reach to somebody outside of class, however, please reach out to me!



Assignments[edit]

The assignments in this class are designed to give you an opportunity to try your hand at using the technical skills that we're covering in the class. In addition to the four major assignments described below, there will be weekly assignments (tutorials and coding challenges) that I will ask you to complete and (sometimes) hand in, but these will only be graded as complete/incomplete. Like many technical subjects, Data Science tends to build on earlier ideas, so I strongly suggest that you devote time to assignments every single week.

Final Project Idea[edit]

Maximum Length: 600 words (~2 pages double spaced)
Due Date: May 2
Drop box: Turn in on Canvas

In this assignment, you should identify an area of interest, at least 2 datasets or APIs with relevant data, and at least 3 questions that you plan to explore. I am hoping that each of you will pick an area or domain that you are intellectually committed to and invested in (e.g., in your business or personal life). You will be successful if you describe the scope of the problem and explain why you think the data sources you've identified are relevant.

I will give you feedback on these write-ups and will let you each know if I think you have identified a questions that might be too ambitious, too trivial, too broad, too narrow, etc.

Final Project Proposal[edit]

Maximum Length: 1500 words (~4 pages)
Due Date: May 16th (at 6pm)

This proposal should focus on two questions:

  • Why are you planning to do this analysis? Make sure to introduce any background information about the topic, the community, your business, or anything else that will be required to properly contextualize your study.
  • What is your plan? Describe the data sources will you collect and how they will be collected. Are there any blind spots given the data you have available? Are there any visualizations or tables that you plan to build?

Your proposal should frame your final analysis, but it's also a chance to "sanity check" your plan. I will give you feedback on these proposals and suggest changes or modifications that are more likely to make them successful or compelling. I will also work with you to make sure that you have the resources and support necessary to carry out your project successfully.

Final Project Presentation[edit]

Presentation Date: June 3

Your presentation should do everything that your paper does and should provide me with a very clear idea of what to expect in your final report. I'm going to give you all at least a paragraph of feedback after your talk. This will be an opportunity for me to see a preview of your report and give you a sense for what I think you can improve. It's to your advantage to both give a compelling talk and to give me a sense for your project.

Timing
All presentations will need to be 'a maximum of 5 minutes long with additional 2-3 minutes for questions and answers. Timing is going to be tight and I'm going to set an alarm and stop presentations that go too long. Concisely communicating an idea in the time allotted is an important skill in it's own right.
Slides
You are encouraged to use slides for your talk but I will need your slides ahead of class.


Final Project Report[edit]

Due Date: June 8

For your final project report and presentation, I expect you to build on the first two assignments to describe what they have done and what you have found. Your final report should be no more than 3000 words and should include detailed information on:

  • The problem or area you have identified and enough background to understand the rest of your work and its importance or relevance.
  • Your research question(s) and/or hypotheses.
  • The methods, data, and approach that you used to collect the data plus information on why you think this was appropriate way to approach your question(s).
  • The results and findings including numbers, tables, graphics, and figures.
  • A discussion of limitations for your work and how you might improve them.

If you want inspiration for how people use data science to communicate this kinds of findings broadly and effectively, take a look at great sources of data journalism including Five Thirty Eight or The Upshot at the New York Times. Both of these publish a large amount of excellent examples of data analysis aimed at broader non-technical audiences like the ones you'll be communicating with and quite a bit of their work is actually done using Python and web APIs! A simple Five Thirty Eight story will include a clear question, a brief overview of the data sources and method, a figure or two plus several paragraphs walking through the results, followed by a nice conclusion. I'm asking you to try to produce something roughly like this.

Keep in mind that most stories on Five Thirty Eight are under 1000 words and I'm giving up to 3000 words to show me what you've learned. As a result, you should do more than FiveThirtyEight does in a single story. You can ask and answer more questions, you can provide more background, context, and justification, you can provide more details on your methods and data sources, you can show us more graphs, you can discuss the implications of your findings more. Use the space I've given you to show off what you've done and what you've learned!

I expect that your reports will include text from the first two assignments and reflect comprehensive documentation of your project. Each project should include: (a) the description of the question you have identified and information necessary to frame your question, (b) a description of the how you collected your data, (c) the results, (d) a description of the scope or limitations of your conclusion.

A successful project will tell a compelling story and will engage with, and improve upon, the course material to teach an audience that includes me, your classmates, and HCDE students taking classes like this in future years how to take advantage of community data science more effectively.

The very best reports will give us all a new understanding of some aspect of course material and change the way I teach some portion of this course in the future.

Code[edit]

Finally, you should also share with me the full Python source code you used to collect the data as well as the dataset itself. Your code alone will not form a large portion of your final grade. Rather, I will focus on the degree to which you have been successful at answering the substantive questions you have identified.

At least 25% of your grade for this project will be determined by the visualizations and tables in your report. Good visualizations should "stand alone" and motivate the core results in your paper all by themselves. A good question to keep in mind is "could I tell this story with the visualizations and a tweet?"

Weekly Tutorials and Coding Challenges[edit]

For the first few weeks of the course, I will assign you some short tutuorials as homework. These are intended to help you practice and remember basic programming concepts.

Each week I will also give you all a set of weekly coding challenges before the end of class. Most of these challenges will involve changing or adding to code that I've given you as part of the projects in the final parts of class to solve new problems.

These tutorials will not be turned in or graded. I may ask your to turn in some of the coding challenges--I'll be very clear in specifying which ones. Coding challenges I ask you to turn in need to be submitted, even if you only have a partial solution, however they will not be graded for correctness. You'll receive credit for attempting the coding challenge, not for whether your answer is correct.

I will share my solutions answers to each of the coding challenges by Monday morning of class. As you will see over the course of the quarter, there are many possible solutions to many programming problems and my own approaches will often be different than yours. That's completely fine! Coding is a creative act!

Please do not share answers to challenges before midnight on Sunday so that everybody has a chance to work through answers on their own. After midnight on Sunday, you are all welcome to share your solutions and/or to discuss different approaches. We will discuss the coding challenges for a short period of time at the beginning of each class.

Graded coding challenges only account for a small part of your grade, but if you repeatedly ignore them the numbers will add up. Worse: you're likely to fall behind in the course. Best practice is to always turn something in! Remember: you're not being graded on whether you answered the question correctly, only that you attempted to answer the question.

Participation[edit]

The course relies heavily on participation. The material we're going to be covering is difficult and we're going to be covering it quickly. It is going to be extremely difficult to make up any missed classes. Attendance will be the most important part of participation and missing more than 1 class is going to make it extremely difficult to excel in our class.

Nearly every week, we will begin by discussing challenges and problem sets that we'll define as a group at the end of the previous class. Please speak up and engage in this part of the class as well as asking questions anytime there is anything confusing. If you are feel confused about a new Python concept, it's highly unlikely that you are the only one.

In general, I prefer that students feel they can "politely interrupt" at any time to seek clarification or make a well-informed point.

Schedule[edit]

Week 1: March 28[edit]

Day 1 plan

Assignments due
  • fill out the pre-course survey
Agenda
  • Quick introductions — Be ready to introduce yourself and describe your interest and goals in the class.
  • Why Programming and Data Science for UX Research? — What this course is about
  • Class overview and expectations — We'll walk through this syllabus.
  • Group formation — We'll assemble in our peer programming groups for the first time.
  • Installation and setup — You'll install software including the Python programming language and run through a series of exercises.
  • Interactive lecture: programming concepts 1
  • Self-guided tutorial and exercises — You'll work through a self-guided tutorial to practice the basic concepts we introduced in the lecture.
Homework
Resources



Week 2: April 4[edit]

Day 2 plan

Agenda
Homework
Resources



Week 3: April 11[edit]

Day 3 plan

Agenda
Homework
Resources
  • go here



Week 4: April 18[edit]

Day 4 plan

Agenda
Homework
Resources
  • Two video lectures by Mako Hill, which covers most of the concepts from NEXT week's lecture (as well as some useful review):



Week 5: April 25[edit]

Day 5 plan

Assignments due
Agenda
Homework
Resources



Week 6: May 2[edit]

Day 6 plan

Assignments due
Agenda
  • An interactive lecture introducing the concept of user-defined functions
Homework
Resources



Week 7: May 9[edit]

Day 7 plan

Assignments due
Agenda
  • Week 6 coding challenge solutions - Jonathan will review the solutions to the week 6 coding challenges and answer questions
  • Writing your own functions - Ray will give a lecture and lead us through a series of interactive exercises on creating custom functions to make our code simpler, clearer, and more flexible.
  • Working with location data - Jonathan will introduce some simple techniques for aggregating and visualizing datasets that have a location component, using a corpus of Seattle building permit data.
Coding challenges



Week 8: May 16[edit]

Day 8 plan

Assignments due
Agenda
  • Jupyter notebooks: intro and setup
  • Some new concepts: try/except, sleep(), dateutil.parser, datetime.datetime
  • Functions II: walk through examples in Jupyter and terminal
Coding challenges
  • No coding challenges this week!
Resources
  • Data Science from Scratch, Joel Grus (O'Reilly)



Week 9: May 23[edit]

Day 9 plan

Agenda
  • We will review the requirements for the Final Presentation and Final Project assignments
  • We will review the course as a whole, and what we accomplished
  • We will go through 1-2 more examples of how to organize a program into functions
  • We will have an opportunity to review key Python concepts as a class
  • We will have plenty of time to ask questions about and work on final projects
Resources
Free (mostly) Python 3 tutorials and reference works



Week 10: June 3 (DATE CHANGE)[edit]

Please note that this class we will meet from 6pm to 9pm on Friday evening, rather than Monday evening, because of the Memorial Day holiday.

Assignments due
Agenda
  • Final project presentations
Resources
  • one



Week 11: June 6[edit]

FINALS WEEK - NO CLASS

Assignments due



Administrative Notes[edit]

Attendance[edit]

Attendance in class is expected of all participants. This class is going to move very quickly and the things we learn will build on the things we've covered the week before. It will be extremely difficult to miss classes. If you need to miss class for any reason, please contact the instructor ahead of time (email is best). Multiple unexplained absences will likely result in a lower grade or (in extreme circumstances) a failing grade. In the event of an absence, you are responsible for obtaining class notes, handouts, assignments, etc.

Office Hours[edit]

Because this is an evening degree program and I understand you have busy schedules that keep us away from campus during the day, I will not hold regular office hours. In general, I will be available to meet in person for brief discussions during the hour before class.

Remote meetings, using Skype or Google Hangout, are always an option. Please contact me or Ray Hong on email to arrange a meeting.

Disability Accommodations Statement[edit]

To request academic accommodations due to a disability please contact Disability Resources for Students, 448 Schmitz, 206-543-8924/V, 206-5430-8925/TTY. If you have a letter from Disability Resources for Students indicating that you have a disability that requires academic accommodations, please present the letter to me so we can discuss the accommodations that you might need for the class. I am happy to work with you to maximize your learning experience.

Grades[edit]

Grades for individual components for this course (participation, coding challenges, final project deliverables) will be assigned based on the following criteria. For more information on grading policy, see the HCDE grading policy and the UW Academic Conduct policy for grading.

To learn more about how I will evaluate your overall performance for the course, please see Professor Benjamin Mako Hill's assessment rubric.

Assignment Percentage
Participation 30%
Required Coding Challenges 10%
Final Project Idea 10%
Final Project Plan 10%
Final Project Presentation 10%
Final Project Report 30%


Participation[edit]

You are not graded on whether you show up to class, or not. However, you are graded on your participation in class. Everyone misses class every once in a while. However, if you need to miss a class and you want to assure that your absence does not negatively impact your grade, follow the steps steps to demonstrate that you engaged with the material covered during that class session.

IMPORTANT: For full participation credit on a day you are absent, you must contact the instructor or TA before 10pm the day after class via email with this information:

  1. you have reviewed any materials we covered in the class, including instructor slides, links, videos, etc. (check the wiki)
  2. you have attempted to perform any in-class exercises that we performed during class (if they're listed on the wiki).
  3. you have turned in any assignments (graded coding challenges, project deliverables) due during the that week's class

In your email: please ask questions, what confused you, tell us how far you got on the coding challenges, and be honest about what you couldn't complete. I also highly recommend that you share your solutions or partial solutions in the email--sending me python files and/or pasting code into the body of an email is totally fine--this helps me and Ray understand your thought process and evaluate your effort and your progress.

Coding Challenges[edit]

Only about half of the coding challenges will be graded (and I will be very clear which ones those are, so there is no confusion). Graded coding challenges are evaluated as complete/incomplete. You gain a 'complete' for coding challenges by turning in the assignment (whether or not your answers are correct, or your code runs) via the submission channel I specified for that assignment.

Final project deliverables[edit]

Grades for all final project deliverables (idea, proposal, presentation, and report) for this class are based on a rating scale.

Rating-scale grades are based on the faculty member's assessment of each assignment as opposed to a calculation from earned and possible points. The broad criteria for the ratings are given below. The ratings for some assignments may be multiplied by a constant (e.g. 2 or 3) so as to count more toward the final grade. The final grade is calculated as the average of all ratings.

4.0 - 3.9
Excellent and exceptional work for a graduate student. Work at this level is extraordinarily thorough, well reasoned, methodologically sophisticated, and well written. Work is of good professional quality, shows an incisive understanding of data science-related issues and demonstrates clear recognition of appropriate analytical approaches to data science challenges and opportunities. Clients who received a deliverable of this quality would likely develop loyalty toward the vendor to the exclusion of other vendors.
3.8 - 3.7
Strong work for a graduate student. Work at this level shows some signs of creativity, is thorough and well-reasoned, indicates strong understanding of appropriate methodological or analytical approaches, and demonstrates clear recognition and good understanding of salient data science-related challenges and opportunities. Clients who received a deliverable of this quality would likely recommend this vendor to others and consider a longer-term engagement.
3.6 - 3.5
Competent and sound work for a graduate student; well reasoned and thorough, methodologically sound, but not especially creative or insightful or technically sophisticated; shows adequate understanding of data science-related challenges and opportunities, although that understanding may be somewhat incomplete. This is the graduate student grade that indicates neither unusual strength nor exceptional weakness. Clients who received a deliverable of this quality would likely agree to repeat business with this vendor.
3.3 - 3.4
Adequate work for a graduate student even though some weaknesses are evident. Moderately thorough and well reasoned, but some indication that understanding of the important issues is less than complete and perhaps inadequate in other respects as well. Methodological or analytical approaches used are generally adequate but have one or more weaknesses or limitations. Clients who received a deliverable of this quality would likely entertain competitor vendors.
3.0 - 3.2
Fair work for a graduate student; meets the minimal expectations for a graduate student in the course; understanding of salient issues is incomplete, methodological or analytical work performed in the course is minimally adequate. Overall performance, if consistent in graduate courses, would be in jeopardy of sustaining graduate status in "good standing." Clients who received a deliverable of this quality would likely pay the vendor in full but not seek further engagement.
2.7 - 2.9
Borderline work for a graduate student; barely meets the minimal expectations for a graduate student in the course. Work is inadequately developed, important issues are misunderstood, and in many cases assignments are late or incomplete. This is the minimum grade needed to pass the course. Clients who received a deliverable of this quality would likely delay payment until one or more criteria were met.



Plagiarism[edit]

Writing

Please don't use other people's writing as your own. It's easy to spot plagiarism of this type, and the consequences can be serious (see link below).

Code

Programmers routinely copy chunks of code from other people's programs into their own programs. This in itself does not constitute plagiarism--though when in doubt, it is best to inform your instructor that you copied/adapted some part of your code from another source. I encourage you to adapt code from our in-class demonstrations, exercises, and coding challenges to your final project if you think that code will work for you. However, please do NOT copy whole scripts or substantial parts of scripts from other sources and try to pass it off as your own. Again, this is easy to spot--experienced programmers have their own 'styles' and conventions and it generally contrasts with the style of less experienced programmers. Also, if you're copying code because you don't know how to reproduce it yourself, you're probably not learning very much, and that will come back to bite you later in the course, and maybe even in your professional life!

For more information, see the UW policy on plagiarism and academic conduct.

Retrieved from "https://wiki.communitydata.science/index.php?title=DS4UX_(Spring_2016)&oldid=184674"
Attribution-Share Alike 3.0 Unported
Powered by MediaWiki