Syllabus for ECON 0111: Economic Statistics

Author

Akhil Rao

Published

February 27, 2023

Spring 2023

Class meetings

Lecture

Tues/Thurs 9:30-10:45am (AXT 201)

Lab

Tues 1:30-2:20pm (V)

Thurs 3:20-4:10pm (W)

Problem/chitchat hours

Wednesday 1-2pm (Zoom), Thursday 1:30-3:30pm (Warner 301)

Contact information

Zoom room ID: 927 138 2648

Slack channel: ECON 0111 S23

Office: Warner 301 (mask required)

Email: akhilr@middlebury.edu

Course locations

Canvas site for submissions and material

All material is also hosted here

Executive summary

Format: Partly-flipped class with mostly-in-person meetings

Deliverables: Weekly problem sets; weekly lab reports; short quizzes linked to videos and class material; occasional challenge questions; optional exams; UpShot-style blog post. See schedule at the back for dates.

Consumables: lecture videos with slides/notes, class notes, lab notes, misc. notes as needed/requested

Grading: Points-based contract

Expectations: Come to meetings prepared (watched videos, worked through lab notes, done assignments)

Optional exams only. Take them to demonstrate knowledge & improve your grade; or don’t. Problem set solutions released upon submission

A particularly helpful question when feeling generally confused but you’re not quite sure what you’re confused by: “I’m not following, could you repeat that?”

Note: This is version 1.4 of a redesigned ECON 111 course. Professor Gong, Bea Lee (2020.5), and I did this redesign with the goal of making ECON 111 more engaging and inclusive. I would greatly appreciate your feedback on the course design, during and after this semester.

Course description and goals

This course is about asking and answering questions with data.

A lot of people think about research and statistical methods as something other people do. The goal of this course is for you to develop the capability to understand research other people have done, and to conduct independent research of your own. I want you to develop the confidence and knowledge to ask important questions and to generate scientific evidence. I strongly believe that regardless of your background, by the end of this course, you will be able to do quantitative research.

This course is also about building up your ability to read and write quantitatively. You will likely struggle at times. This is good! We often learn the most when we struggle a bit. Don’t shy away from trying, even if it doesn’t work out right away. Think of me as your statistics coach. I’ll challenge you a bit to help you learn and develop, but fundamentally my goal is to help you get the most out of this course. Don’t hesitate to ask for help.

Specific objectives:

By the end of this course, you will:

  • Be able to examine and describe data

  • Use two statistical programming languages (R and Stata)

  • Understand the connection between sample statistics and random variables

  • Be able to create compelling visualizations to describe data and convey insights

  • Understand the difference between correlation and causation

  • Appreciate the central limit theorem (the most important theorem in this class!)

  • Write models to frame a question

  • Estimate a multivariable linear regression

  • Test a hypothesis

  • Conduct empirical research

Course content

The major theme is statistics as a lens for understanding inequality. We will cover statistical content including descriptive analysis, visualization, hypothesis testing, inferential errors, and regression modeling. We will learn these topics by studying important and pressing questions relating to inequality, including:

  • Mass incarceration

  • Racial income gaps

  • Coronavirus case rates and income distributions

  • Income inequality and public sentiment

  • Climate change and public sentiment

  • Air pollution and environmental justice

Statistics offers one lens through which to view these topics. While it is a powerful lens, it is not the only one. The conclusions we can draw using statistics rely on our engagement with subject matter experts, particularly non-statistical accounts reflecting lived experiences. In econ-speak, quantitative analysis is a complement (not a substitute) to qualitative work. In addition to a useful statistical toolkit and the knowledge and confidence to apply it, I want you to take from this class a sense of humility about its limitations when used in isolation.

Positionality

I identify (among others) as an Indian-American man, an economist, a teacher, and a writer. I speak and write primarily in English and have spent my life financially comfortable. Most of my life has been in India and the US. My class, gender presentation, degrees, intellectual pursuits, and American accent are often more prominent in the spaces I inhabit than my ethnic identity, family background, or life experiences.

The topics and examples in this class reflect, among other things, my desire to engage more with issues of inequality and social justice. I recognize that some of these topics may have deep emotional resonances. I care deeply about not perpetrating harms by discussing issues inappropriately. While it is not your responsibility to educate me when I slip, I hope we can develop the mutual trust to support calling each other into more inclusive and empowered discussions. The topics and examples in this class also reflect my academic US-centric perspective. I recognize that many of you bring other perspectives to our discussions. I aim to support you if and when you choose to express your different perspectives.

I am committed to making this learning experience as fruitful as possible for you. Over my life I have struggled a lot with math. I recognize that my personal background shapes what I do and don’t immediately recognize as challenging. I hope you feel comfortable asking me to explain in more detail if something is confusing.

Etiquette and discourse

At Middlebury, we strive to make our campus a respectful, engaged community that embraces difference with all the complexity and individuality each person brings. Each student in this course is expected to contribute to an inclusive and respectful class environment. Students of all backgrounds including gender, sexual orientation, race, ethnicity, and religion are to be treated fairly and with honesty, integrity, and respect. Civil discourse without degrading, abusing, harassing, or silencing others is required of all students in this class.

Being intentional with our words and actions is always important, and particularly so when we discuss heavy topics like those in this class. Please, be mindful of how you frame points and arguments. Avoid pathologizing language in particular—it is often both harmful and a marker of a poorly-considered argument. Intentional and thoughtful discussions which do not perpetrate harms increase our collective freedom of speech and make it possible for us all to engage more fully. Similarly, when someone notifies us that we have perpetrated harms, robust intellectual discourse is best served when we listen, apologize, reflect, and adjust our behaviors moving forward.

To respect everyone’s right to a productive learning environment, please refrain from disruptive activities during class. Please set phones on vibrate mode. I understand that there are urgent and emergency situations. If a call or message comes through and you must respond to it, please do it discreetly or leave the room to address the situation.

Textbook and readings

We’ll be using Introduction to Statistics with Randomization and Simulation (ISRS) and The Effect (TE) by Nick Huntington-Klein.

ISRS and all associated materials (lab materials, code, videos) are free and open source; just go to the link and download/stream what you want. If you want a paper copy, you can pay a low fee for printing and shipping. This textbook is meant to be skimmed. Focus on topics covered in the videos – ignore concepts not covered in the videos/class (unless of course they interest you).

TE is also available for free online (though again, you can purchase a paper copy). You can find the relevant section to read by looking at the chapter titles. I recommend turning to this book for an explanation when you are confused by something in ISRS or when we discuss topics relating to causality or research design. Sometimes it helps to read the same idea written in two different ways.

ISRS has a bit more of the R-related content as well as some of material you may have seen in earlier statistics courses. TE is written with economics and causal inference in mind, and it has content (like identification and treatment effects) you probably haven’t seen before. It also has nice examples.

Neither textbook follows the exact order in which we will cover topics, but (nearly) all the content is there. Take good notes in class!

We will also read academic articles, essays, and book excerpts to complement our understanding of the substantive topics in the course. I will make these readings available on Canvas.

Statistical software

  • R and RStudio: We’ll be using R and RStudio for the first section of the course. R is a free open-source statistical software package, and RStudio is a popular integrated development environment which makes it easier to work in R. Given its open-source nature, R is very popular in the data science world.

  • Stata: You’ll be introduced to Stata in the second section of the course. It is not free and open source. Stata is the most-used statistical package in Economics and if you are continuing on to ECON 211 (required for the Econ major), you will be expected to know Stata. I recommend that you purchase a discounted student license (~$43) to install on your computer. Remote desktop access (currently via Apporto) will also be available for free, though many students have noted that it is less convenient than having a copy on your own computer.

Communication

I encourage you to direct all questions and communication to our Slack workspace. Slack is a popular communications platform that is more fluid than email. The major benefit is that your discussions are more organized and less cumbersome than email.

In general if your question is about specific code or math, please include a copy of the code (as a .Rmd/.qmd/.do file or code chunk) or the math (as a PDF, picture, screenshot, or text). I (try to) maintain work-life balance by not checking messages after 5pm Eastern time or during the weekends, but I promise that I will respond to any message you send me. If it’s been more than 72 hours and you haven’t heard from me, please follow up.

Zoom protocol and details

My Zoom room ID is 927 138 2648.

Think of Zoom as the equivalent of a classroom (during scheduled meeting times) or my office (all other times)—it’s rude to interrupt someone else’s scheduled time. Please use Slack or email if you want to ask a question not during scheduled drop-in hours or appointments.

I’ll be in my Zoom room during scheduled meeting times and problem/chitchat hours. Outside of those times, please contact me through Slack or email.

Slack protocol and details

Use this invite link to join our class workspace (use your @middlebury.edu email). You can access the workspace using your browser or from the app.

I’ll be on Slack more frequently (basically whenever I’m at my desk but not shutting off all distractions to work). Use this for questions that don’t need a longer reply or discussions with each other. Please use the appropriate channel for your questions/discussions: problem set-specific questions should be in problem-sets, general course logistics questions should be in general, lab questions should be in labs, etc. Anything that doesn’t have a specific channel should go into random or general.

If you are asking me a question on Slack or would like me to weigh in on a discussion, please tag me (type @profrao once somewhere in your question). I’ll check Slack at least once per day during the week. I won’t be checking Slack over the weekend. If you ask me a question there and I don’t get to it right away, I’ll tag you in my reply.

You can also DM me with questions if you don’t want to post them in a public channel. I may post such questions (anonymized) to a public channel if I think it would help others in the class (unless of course you specify otherwise).

Email protocol and details

I will focus my attention on our Slack workspace and any emails I receive will have a longer response time (48-72 hours).

That said, email can be useful for longer questions. In the subject line, please start with “Econ 111:”. I use email filters to prioritize among the many emails I receive each day; not including the “Econ 111:” subject line will delay my response. Please allow me up to 48 hours to respond to emails. Don’t hesitate to follow up if I have not replied (you can just re-send the email).

If your question is about specific code or math, please include a copy of the code (as a .Rmd/.do file) or the math (as a PDF, picture, screenshot, or text description in the email).

Problem/chitchat hours

Just drop in. You don’t need to let me know in advance or ask or anything like that. Just stop by and ask questions. I enjoy getting to know students, and problem/chitchat hours is a nice place to do that. If there are multiple folks in problem/chitchat hours I’ll go through questions round-robin by entry order or try to triage questions so we address as many things as possible.

Seriously, just drop in. I’m just going to be at my desk—probably trying to do something productive, but more likely just scrolling. Stop by. Ask questions. Make chitchat. I’m looking forward to it.

Miscenalleneous notes

  • I love meeting with students. But! Please make an appointment if you want to meet outside of problem/chitchat hours. A quick email or Slack DM is fine. Friday is my “research day”, so I try not to schedule appointments then, and may be slow to respond.

  • This course may seem daunting—we’ll be covering complicated concepts, sometimes at a very abstract level. Don’t panic! With practice, repetition, and patience, it will come together. I’m here to help—don’t hesitate to ask questions.

  • On asking questions: sometimes a concept feels fuzzy or like it isn’t quite clicking, but it’s hard to frame a question precisely. Don’t worry! Ask anyway. We may not get to the bottom of it right away, but we can make progress. Questions about material also tend to be correlated across students, so by asking (especially in a public Slack channel) you often provide a public good to your classmates.

  • This material goes deep. I may at times limit our inquiry to keep us moving along. If you’re curious, I’m happy to talk in more detail in problem/chitchat hours. Think of statistics as a very big onion, and this course as pulling the first few layers off—you can chop those layers up and cook with them, plenty of recipes won’t need any more!—but there’s a lot of onion left.

  • Statistical programming proficiency is an important outcome of this class. I strongly encourage you to work through the code examples in the textbook and lab notes (actually type the commands in, don’t just copy-paste). You can look at the lab notes attached to the textbook for more examples. Stack Overflow is a particularly helpful online resource for programming questions in R and other languages. I encourage you to spend some time getting comfortable with using the site.

Course deliverables

There are a number of deliverables in this class. You do not have to do all of them to get an A (or whatever grade you’re after). See the grading contract below for details on what you need to do to achieve specific grades.

  • Video quizzes: I will post videos covering statistical and programming concepts. After you watch a video, you can take a short multiple-choice quiz testing recall and understanding. There will also be bonus quizzes on class content.

  • Lab reports: I will provide you with lab notes prior to each lab. These notes will contain exercises for you to do. The lab report for a week is an Rmarkdown file with your work on completing the exercises.

  • Problem sets: These are weekly collections of problems to solve. A completed submission is an Rmarkdown file with your work towards solutions. When you are unable to completely solve a question, sketch out your work so far, your thoughts on what the answer ought to look like, and what steps you think need to be taken to get there.

  • Problem set corrections: When you submit your problem set, you will receive access to the solutions. You can review the solutions and submit a “redo” file explaining what you got wrong. If you got everything right, you can just say so.

  • Challenge questions: These are more challenging/involved questions intended to test your understanding and stretch your skills. You will have two weeks per challenge question.

  • Optional exams: These are take-home exams. You will have 48 hours to complete them. You may use any of the course materials, including your notes. You may not discuss the exam with anyone other than me (Prof Rao) and you may not consult the broader internet (e.g. no StackOverflow).

  • UpShot-style blog post: These are the capstone research project for this class, blending statistical analysis and communication.

Please feel free to ask me questions about any assignments. However, you will probably get the most out of questions if you ask them after working on them a bit and running into an issue.

More details on deliverables

Video quizzes

After each video, you can take a short multiple-choice quiz to test your recall and understanding of the video you just watched. The quizzes are accessible on Canvas. Video quizzes have a small point value attached to reward the effort of watching the video.

Some quizzes will be directly based on videos, while others will be based on class material. Class material quizzes will typically have a higher point value to reward the effort of coming to class and paying attention, but will also be a bit stricter in terms of grading.

Lab reports

The lab report notes will guide you through the mechanics of the new code we’re learning that week. The exercises are designed to give you some practice and (sometimes) lead you to new insights. You will automatically get credit for submitting a completed lab report whether the answers are correct or not. You’ll submit lab reports on Canvas.

I recommend not spending more than 2 hours on a single lab report. There are diminishing marginal returns to staring at the same thing. Eventually we all hit a wall of “I’m clearly not getting something but I don’t know what and it’s not getting any clearer.” You won’t be penalized for wrong answers, so you might as well submit when you hit the wall.

Problem sets and corrections

I want to incentivize you to put effort into the problem sets even if you aren’t sure you have the right answer. My criteria for “effort” is “the question was attempted and completed, or attempted and incomplete with a brief explanation of where and why the student got stuck”. “Brief” means “enough to convey understanding and the issue”, not “everything the student knows about the subject”.

After you submit, you will get access to the solutions. You can then submit problem set corrections: revise your solutions to each question you got wrong and resubmit for credit. If you got the question right, just say so. This is to give you an incentive to review the solutions, even if you got everything right. The idea here is that the effort of reviewing solutions is rewarded, whether you got the answer right or not.

I recommend not spending more than 3 hours on completing a single problem set.

Challenge questions

These questions are meant to stretch your skills a bit more and maybe teach you something new and useful. Every so often (roughly every 3-4 weeks), I’ll post a new challenge question. You can discuss the questions and the issues you face with others, but I think you’ll get the most out of these questions if you first make a serious attempt on your own. They’ll often deal with real issues researchers study; if so, I’ll post some underlying research the question is based on after the challenge question is due, and either we’ll discuss it a bit in class or I’ll make a video (or both, I’m not sure yet). You’ll submit challenge questions on Canvas as Rmarkdown files detailing your work.

I recommend not spending more than 4 hours on a single challenge question.

Optional exams

I used to give in-class exams. Then COVID happened and I moved to take-home exams. I haven’t looked back.

There will be two exams in this class, one roughly in the middle of the semester and one near the end. They are both take-home open-book (notes and any materials I have shared are ok; general googling and working with others is not) exams. You will have roughly 48 hours to complete them. Late submissions will be penalized an amount proportional to the number of hours late, with a penalty of 100% if the exam is 24 hours late or later.

These exams are optional in the sense that every assignment in this class is optional: you don’t need it to pass or get a decent grade (not necessary for a B), you probably need it to get a really good grade (generally necessary for an A), and I’m giving you the choice of whether to take the exam or not.

These exams serve two purposes for those who take them:

  1. They test your knowledge / understanding of facts, concepts, and tools taught in this class.
  2. They force you to confront what you do and don’t know and teach you new things.

This means they will be fairly comprehensive and probably kinda tough. Prepare accordingly.

Exam eligibility

Grading exams takes a lot of time, and in the past the optionality in these exams has encouraged students to take a “roll the dice” approach to them – “take the exam with low preparation because why not?”

This is not ideal for many reasons, e.g. it wastes everyone’s time. So, this semester the final exam will be “locked” to begin: you must unlock the final exam by earning 200 points before the Monday of finals week (12/12).

Blog posts

In previous versions of this class, I’ve assigned a research project. This was typically a two-person project where students produced an academic-style paper showing off their statistical and programming skills on a question they found interesting. I’ve come to think this isn’t the best assignment for a couple reasons. First, though free riding was rare, each instance led to significant discomfort for those picking up the slack, less learning for those exerting less effort, and a lower-quality product for me to assess. Second, the academic writing style for statistical research will probably not be relevant to most of you later on, and students spent a lot of energy mimicking that particular writing style. Nobody who hasn’t spent years stewing in academia likes to read a typical academic research paper dealing with statistics; even the best-written ones generally have low readership.

The UpShot, on the other hand, is widely read by academics and non-academics. Each post is generally well-written, has an interesting question and a key point, and gets to the point fast. It features statistical analysis and visualizations as needed to drive the key point home and address likely reader questions. A non-specialist could read an UpShot post and come away smarter. Whatever you do later on, it’s (probably) the kind of output you’ll need or want to produce to convey quantitative insights to a broad audience.

In the honor code section I say you can collaborate on anything “except the blog post”, which is worth clarifying. You are free to discuss your blog post with others, to review each others work and code, and even solicit suggestions. But you must submit your own work, reflecting your own effort and interests. It’s fine if you end up asking the same question as someone else in the class (this happens surprisingly frequently in academic research), but you must have your own analysis and interpretation. This doesn’t mean you need to disagree with others working in similar areas.

Where you use someone else’s ideas, you must cite them. This applies to any source you use: popular press articles, academic sources, etc. Use inline citations (not footnotes) in the Chicago Author-Date style used in Economics or hyperlinks. Your bibliography should sit between the main text and statistical appendix.

You’ll submit the blog posts as html or PDF files (your choice) on Canvas.

I recommend starting the blog post early.

Effort tiers

The effort tiers here are cumulative, i.e. to get a “complete” you must satisfy the “basic” and “complete” requirements. I will assess whether a post meets the criteria for a given tier or not with feedback on suggested improvements. If you are unsatisfied with my assessment, you can redo the blog post up until the final posting date for blog posts. There is no limit on the number of redos. Points in a higher tier for which criteria have not been met will be discounted by the fraction of the preceding tier you have completed, e.g. if you have done 40% of the “complete” tier requirements but only 50% of the “basic” tier requirements, you will receive only 50% of the “complete” tier points you have done the work for.

This assignment is inspired by Professor Wolcott, and many of the criteria below are borrowed from her rubric. Blog posts at any level must include a bibliography.

  • Basic: At a minimum, a blog post should have a clearly defined and stated purpose/question. It should be obvious to the intelligent layperson reader why they would be investing time in your post. There should be at least one data visualization/graphic. The visualization should satisfy the following criteria:
    • It should be helpful, and aid the reader’s understanding of the article.
    • It should be necessary, and without it the article would be less effective.
    • It should be easy to understand (e.g. cleanly labelled, minimal chartjunk), so that the reader gets the point of the visualization in seconds.
  • Complete: To be a more complete representation of your learning from this class, a blog post should include some data analysis. The post should contain some basic data description, robustness checks, a regression with appropriate interpretation/presentation of the results, and a “Statistical appendix” explaining the messy details for the interested reader. The data description should be enough that your reader understands what’s in the data and where it comes from (words are fine in the main text). The robustness checks should be enough to answer some questions the reader may have about the generalizability/quality of the finding(s) (one or two in the main text is enough). If you make any causal claims, you should explain the necessary assumptions clearly in words; the reader should be able to understand these assumptions and when they might be violated. The results should be presented in a way that a reader who hasn’t taken ECON 210 can follow along (e.g. don’t just drop a table of R/Stata output in there, explain in words and maybe make a picture).
    • The main text of the post should not be too heavy on the statistical analysis. Weave it in with the narrative, and don’t bore your reader.
    • In the process of writing the post, you will likely find that you have more data description, robustness checks, or theory than would be appropriate for an UpShot-style blog post. This is what the statistical appendix is for. It’s a good place to show tables of summary statistics, details on how you merged data (if you merged anything), additional robustness checks (if any), and causal diagrams (if you’re making causal claims).
    • To earn the full “complete” mark, you must meet the project milestones (e.g. project proposal, data summary) throughout the semester.
  • Extensive: To receive an “extensive” appraisal, you must (1) incorporate evidence outside the data, (2) write well, (3) show sustained commitment to the project by meeting milestones throughout the semester, and (4) provide a replication package for your results. (1) involves choosing an appropriate set of peer-reviewed papers and seamlessly integrating them into the arguments, summarizing key points or issues in the sources cited to critically analyze those ideas and relate them to the post’s purpose. (2) involves controlling pace, rhythm, and variety; words chosen should be apt and precise; sentences should flow smoothly together and clearly open, develop, and close topics. Use the active voice. I recognize that “writing well” is a subjective thing. I encourage you to go to the writing center for help with writing. (3) is pretty straightforward, same as for the complete; On (4): A replication package is a zip file with
    • All the scripts (.r files / .do files / .Rmd files) you used for your analysis, with comments inside explaining what each line/block of code does
    • A readme file (a text/markdown file named README.txt or README.md) explaining how the scripts should be run and in what order, and what datasets are necessary
    • All the datasets you used. Someone who knows R and Stata but doesn’t really know what you did in your project should be able to take your replication packages, follow the instructions in your readme, and replicate every result/figure you used in the blog post and statistical appendix without modifying the code scripts at all (it should be a turnkey experience; push the buttons to run the scripts and they just work). Replication packages are an important part of open science, and are increasingly a requirement for publication in academic journals and prestigious non-academic outlets.

The Grading Contract

The concept

Grading can be problematic. It induces anxiety and stress, fosters competition amongst students, induces proclivities towards honor code violations, and focuses student time and attention on grades instead of learning. Many professors also think it is the worst part of our jobs. Grading takes up a lot of time and typically doesn’t result in timely or productive feedback. Finally, grading can create an adversarial relationship between professors and students; office hours discussions around grades (e.g. “What can I do to get a better grade?”, “Can I do extra credit?”) are generally unpleasant for all involved.

All that said, I still have to assign you a final grade in the course.

What to do?

Instead of receiving a final letter grade based only on mastery of course materials, I will reward you based on the effort that you put into the class and the mastery you demonstrate. In a traditional grading approach, your scores on assignments are determined by how correct your answers were, and the scores determine your grade (grades are based on knowledge). In this grading contract, assignments are scored on a mix of completeness and correctness (grades based on labor and knowledge).

The underlying theory here is that exerting effort has benefits even when it feels like you’re not getting much out of it. I want to rewire the usual effort-output-reward loop a bit so that you’re more directly incentivized to do things that will (on average) help you learn, even if in the moment it feels like a struggle and you’re not getting anywhere. From what I’ve seen, all else equal,

  • an effortful attempt almost always teaches you more than no attempt,
  • an effortful-but-incorrect attempt often teaches you about as much as an effortful-but-correct attempt, and
  • the average effortful-but-incorrect attempt teaches you more than the average effortless-but-correct attempt.

Why do this?

I start from the premise that if you’re in this class, you want to (a) learn statistics and statistical programming and (b) achieve a certain grade. All too often, course (and other) incentives students guide students to focus on more on getting good grades than on actually learning. With this system, I hope to remove some uncertainty from your final grade (you can achieve the grade you want with a relatively well-understood effort allocation) and make it easier for you to learn (less anxiety about assessments and more focus on process). Think about this question: “If I just wanted to learn this material and wasn’t getting a grade for this assignment, would I still do it?” If the answer is “No”, then maybe the assignment isn’t the best use of your time.

I want to take the emphasis away from grades, and put it back on where it belongs: the process of learning.

Finally, grading contracts can create a more (but not totally) equitable environment. Using a traditional grading system, students that have prior knowledge of the topic have a distinct advantage. With this grading contract, all students must put in time and effort; you cannot coast based on what you already know.

What does this look like?

There will be a set of assignments that you can choose to do. When you submit most assignments, it is either accepted (receive credit) or declined (no credit). Thus, it is possible to get up to a B+ purely on the basis of your effort.

Three types of assignments will be graded for correctness as well as completeness: the challenge questions, the optional exams, and the blog post. These assignments are meant to showcase your mastery of the material, so you must be able to demonstrate not only effort but understanding. Doing well on these assignments will help you get into the A- / A range.

In case it wasn’t obvious, I am not Emma Stone

I want to be very clear: the grading contract does not mean that this class is “an easy A”. In all likelihood, getting an A in this class will require more work than getting an A in your other classes. An A signals a high degree of competence and comfort with course material. If you get an A in this class, you can be assured that your understanding of the material is superlative.

The mechanics

Each grade requires you to obtain a number of points in the course. You can get points by completing assignments and receiving credit for them. For the UpShot-style blog posts only, I will track variation in output quality. The tiers of blog post quality are discussed in the “Blog posts” section of “More details on deliverables”. Every assignment except the blog posts, optional exams, and challenge questions will be assessed on an accept/decline basis.

The table below shows one (recommended) approach to securing specific grades. The point requirements for each grade are listed alongside the recommended quantity of assignments to complete to get there. The points attached to a marginal assignment of each type are in the subsequent table.

Note: You must complete the Stata problem sets to earn an A! Even if you have the points necessary for an A, you will not receive one without having completed all the Stata problem sets.

Assignment point values
Assignment Point value
Video quizzes 1
Lab reports 6
Problem sets 3
Problem set corrections 3
Challenge questions 12
Optional exam 1 24
Optional exam 2 48
Blog post (basic) 20
Blog post (complete) 32
Blog post (extensive) 48

The point values are meant to be loosely proportional to how helpful I think each assignment is, at the margin, in furthering your learning. Credit is assigned on an accept/decline basis for all assignments except the challenge questions, blog post, and optional exams. They also provide a convenient “exchange rate” to use between assignments, so that you have options in case you’re unable to meet the recommended requirements for the grade you want.

Some quizzes may be worth more than 1 point. I reserve the right to modify grade thresholds and assignment point values at my discretion.

Engagement

Being engaged in a class can improve the learning environment for everyone. “Being engaged” doesn’t mean talking all the time, though it can mean asking questions. It means making efforts to improve the class learning environment – by being present, by asking questions, by helping others during activities, and by showing up having done the work. Frequently missing class, not participating in activities, watching sports during class time – these are all examples of behaviors which will harm your engagement score.

At the end of the semester I will award up to 16 points of credit for engagement.

The Secret to Doing Well

The course is front-end loaded and the course builds on itself. I strongly recommend to keep up with course deadlines, especially in the first half of the semester.

Your attention is one of your scarcest and most valuable resources – and everything is vying for it. Here’s are some tips to put you in a position to succeed:

  • Each week, dedicate days and times when you work solely on Econ 111 in a distraction free environment. For example, on your calendar, block out Mondays and Wednesdays from 8:30-10:30 am to work on Econ 111.
  • When you work on Econ 111: Focus on this class and don’t multitask (i.e. checking email, talking with friends, working on another course at the same time, …). Give your phone to a friend or leave it in your room and go to the library.
  • Use Freedom, Cold Turkey Blocker, SelfControl or another app to block web and app access on all your devices. When you work on this course, you want to eliminate access to anything that can distract you.

If you are interested in discovering more learning techniques and strategies, read up on Cal Newport. He’s a successful computer science professor at Georgetown who has interviewed a number of top undergraduate students.

I’ll add a couple more here based on my experience teaching this class:

  • Take notes. Students who regularly took notes in class have tended to perform better. It also gives you a resource for optional exams. I strongly recommend actually writing things down – research has shown that writing engages the brain more fully than typing, facilitating better memory formation and recall.

  • Practice programming regularly. It’s like learning any other skill (instruments, math, throwing trash into the bin on the first try) – you need to practice to get good. Doing assignments is a good form of practice, and doing them consistently over the semester will develop the mental muscles much better than trying to cram it all in at the end.

College resources

The college has many resources to help you when you are struggling. You can find a list of resources here.

I’d like to make a particular mention of the Anderson Freeman Resource Center. The AFC has some great programming, including a peer mentorship program, peer writing tutoring, wellness support, counseling, and social events (granted we’re in a pandemic, but even online events can be nice). The AFC is physically located at Carr Hall and can be reached at afc@middlebury.edu or 802-443-2214.

Accommodations

Students who have Letters of Accommodation in this class are encouraged to contact me as early in the semester as possible to ensure that such accommodations are implemented in a timely fashion. For those without Letters of Accommodation, assistance is available to eligible students through Student Accessibility Services. Please contact Jodi Litchfield or Peter Ploegman, the ADA Coordinators, for more information: Peter Ploegman can be reached at pploegman@middlebury.edu or 802-443-2382 and Jodi Litchfield can be reached at litchfie@middlebury.edu or 802-443-5936. All discussions will remain confidential.I will make best efforts to provide accommodations but, given the current circumstances, I can’t guarantee the requirements of the letter will be met.

Research and the library

You may need to find data for your blog post. I encourage you to go to the library and work with the librarians to find more resources, but be sure to cite what you use and provide your own insights. I highly recommend you reach out to Ryan Clement, the Economics reference librarian, as soon as you start settling on a topic. Ryan is a wealth of knowledge about economic data and literature. His email is rclement@middlebury.edu, and his website is at go.middlebury.edu/ryan.

Honor code

I expect you to adhere to the Middlebury Honor Code. The Middlebury Honor Code is described at go/honorcode. You can collaborate on all assignments except the exams or blog post (see the “Blog post” section under “More details on deliverables” for more details). If you collaborate with someone, you must acknowledge those you worked with and submit your own final writeup. For the blog post, an acknowledgements section at the back is a good way to do this.

Deadlines and extensions

I expect you to submit work by the deadlines. Why? It really helps facilitate our group discussions. It also keeps students on track for the course. We all know that students/professors prioritize what they need to do by deadlines.

Late submissions will be accepted for reduced credit with a penalty proportional to their lateness, reflecting their diminished learning value when completed late. There is a small grace period immediately after the due date before the penalty kicks in. An assignment submitted 14 calendar days (or more) later than the original submission deadline will receive no credit.

There is an element of externality pricing to this: the learning environment for all is diminished when some are unprepared. The reduced credit reflects this, too. A submission is considered late if it is submitted after the next submission deadline (5 pm EST on a day with an assignment due). So:

  • an assignment due Friday is considered late if it is submitted after the following Monday;

  • an assignment due Monday is considered late if it is submitted after the following Thursday;

  • an assignment due Thursday is considered late if it is submitted after the following Friday.

Practically, this means you have very little room to delay lab report submission; more room for problem sets; and maybe a little less room for challenge questions. This distribution reflects how important I think it is for your learning that you do each assignment type on time (i.e. very important for lab reports).

It’s not a problem if you need a small extension (small = not enough to be considered late), just send me a Slack DM letting me know that you’ll be submitting it a little later. If you need a longer extension, reach out and we can discuss.

Since you won’t be penalized for effortful-but-incorrect answers, please don’t take extensions if you’re at the wall and unsure if your answers are correct. I don’t mind extensions, but it won’t do much for your learning beyond maybe getting you frustrated. You’re almost surely better off just submitting your effortful attempts and reviewing the solutions carefully.

Names and Pronouns

If you go by a different name or pronouns than what is listed on the roster, please let me know. I expect us to refer to each other by our preferred names and pronouns.

Basic needs

Sometimes people get sick, or are otherwise unable to come to class/submit assignments for whatever reason. That’s ok. If you need to take time off for illness, a mental health break, or some other reason, please do so. You don’t need to let me know in advance, though advance notice is always appreciated. I’m always happy to discuss course content you’re struggling with, just send me an email/slack DM and we’ll go from there.

A brief note letting me know where you’re at and what I can do to support you is fine, even if it’s after the fact. I don’t need doctor’s letters or detailed explanations. My working assumption is that you want to be in class and learn and that if you’re not there it’s because something came up and you were either constrained. My goal to make sure you’re able to take time when you need it without seeking approvals or permissions and with the support you need to catch up. If you anticipate or experience a longer absence (e.g. two weeks due to COVID), please do reach out so I’m aware you’re ok (or not) and I can help you with any material you’re struggling with.

Please take care of yourself. College can be a demanding (and rewarding) experience. If you or one of your friends, peers, classmates is facing mental health problems, please consider seeking help. Resources include your Common’s Dean, Resident Life Staff, Middlebury College Counseling Center, and ULifeLine website.

Class schedule/semester at a glance

The schedule below shows major assignments and due dates for the course. I will revise the schedule as/when things change. Unless specifically instructed otherwise, assume all assignments are due at 5 pm Eastern US time of the day they are listed as due.

The statistical concepts are listed in loose clusters by chapter/week and are meant to give you some guidance on what you can expect to learn. For example, the topics listed next to the first week will likely cover the first two weeks. Within each cluster they aren’t necessarily in order.

Key:

  • LR: Lab report
  • PS: Problem set
  • CQ: Challenge question
  • VQ: Video quiz (due at the end of the week)