How to Teach Programming (and Other Things)

Teaching Online


  • Learner can explain why expectations for massive online courses were unrealistic, and ways in which ad hoc use online learning is changing education.
  • Learners can explain several key features of successful online courses, and judge whether a particular course meets those criteria.
  • Learners can implement several logically different kinds of exercises using multiple-choice questions and/or write-and-run coding.

If you use robots to teach, you teach people to be robots.
– variously attributed

Five years ago, you couldn’t cross the street on a major university campus without hearing some talking about how teaching online was going to revolutionize education–or destroy it, or possibly both. Now that the hype has worn off, it’s clear that while almost all learning now has an online component, MOOCs (massive open online courses) aren’t as effective as their more enthusiastic proponents claimed they would be [Ubell2017]. As one specific example, [Koedinger2015] estimated “…the learning benefit from extra doing (1 SD increase) to be more than six times that of extra watching or reading.” “Doing”, in this case, refers to completing an interactive task with feedback, while “benefit” refers to both completion rates and overall performance.

One reason is that recorded content is ineffective for most novices learners because it cannot intervene to clear up specific learners’ misconceptions. Some people happen to already have the right conceptual categories for a subject, or happen to form them correctly early on; these are the ones who stick with most massive online courses, but many discussions of the effectiveness of such courses ignore this survivor bias.

Many enthusiasts also didn’t realize that MOOCs and other forms of online education were just the latest in a long series of proposals to use machines to teach. As Audrey Watters has chronicled, every innovation in communications–from the printing press through radio and television to desktop computers and now mobile devices–has spawned a wave of aggressive optimists who believe that education is broken (without actually knowing much about education) and that the latest technology is the solution (without knowing much about what’s been tried before, why it failed, or what success would actually look like) [Watters2014].

That said, technology has changed teaching and learning. Before blackboards were introduced into schools in the early 1800s, there was no way for a teacher to share an improvised example, diagram, or exercise with an entire class at once. Combining low cost, low maintenance, reliability, ease of use, and flexibility, blackboards enabled teachers to do things quickly and at scale that they had only been able to do slowly and piecemeal before. Similarly, the hand-held video camera revolutionized athletics training, just as the tape recorder revolutionized music instruction a decade earlier.

Today’s revolutionary technology is the Internet, whose key characteristics are:

  1. Students can access far more information, far more quickly, than ever before–provided, of course, that a search engine considers it worth indexing, that their internet service provider and government don’t block it, and that the truth isn’t drowned in a sea of attention-sapping disinformation.

  2. Students can access far more people than ever before as well–again, provided that they aren’t driven offline by harassment or marginalized because they don’t conform to the social norms of whichever group is talking loudest.

  3. Courses can reach far more people than before–but only if those students actually have access to the required technology, can afford to use it, and aren’t being used as a way to redistribute wealth from the have-nots to the haves [Cottom2017].

  4. Teachers can get far more detailed insight into how students work–so long as students are doing things that are amenable to large-scale automated analysis and aren’t in a position to object to the use of surveillance into the classroom.

The caveats in this list are a big part of why this book exists. Right now, most of the discussion about using technology in teaching is shaped by what tech can do rather than what students need, and by Silicon Valley’s craving for quick profits rather than by society’s need for competent, well-informed citizens. If people in tech know more about teaching, they’ll know more about what they should build to teach more effectively. And if they do that, they’ll be able to reach more people from more diverse backgrounds, so that there will be more voices in the room when key decisions about the future of education are being made.

General Guidance

MOOCs are useful for people who already a mental model and wish to remind themselves of things they already know or fill in gaps in their knowledge. Most of today’s online classes use some mix of recorded video presentations, automated exercise submission and grading, and web-based discussion, which can either be synchronous using chat tools and video calls or asynchronous using mailing lists and bulletin boards.

The two greatest strengths of this model are that learners can work when it’s convenient for them, and that they have access to a wider range of courses, both because the Internet brings them all next door and because online courses typically have lower direct and indirect costs than in-person courses. The disadvantages are that they have to shoulder much more of the burden of staying focused, and that the impersonality of working online can demotivate people and encourage uncivil behavior.

Hybrid Vigor

The “pure MOOC” model described above is actually only a small subset of what could be done: [Brookfield2016] describes many other ways groups can discuss things, only a handful of which have ever been implemented online.) A hybrid approach that has proven quite successful is real-time remote instruction, in which the learners are co-located at one (or a few) sites, with helpers present, while the instructor(s) teaching via online video.

As an instructor teaching online, you should take advantages of the pros and do what you can to minimize or avoid the cons [Nilson2017,Miller2016]:

  1. Deadlines should be frequent, well-publicized, and enforced, so that learners will get into a work rhythm.

  2. Keep other all-class synchronous activities like live lectures to a minimum so that people don’t miss things because of scheduling conflicts.

  3. Encourage or require students to do some of their work in small groups (2-6 people) that do have synchronous activities such as a weekly online discussion. This will help students stay engaged and motivated without creating too many scheduling headaches.

  4. Create, publicize, and enforce a code of conduct so that everyone can actually (as opposed to theoretically) take part in online discussions.

  5. Remember that people learn best in small chunks, so use lots of small lesson episodes rather than a handful of lecture-length chunks. Also remember that, disabilities aside, they can read faster than you can talk, so use video to engage rather than instruct. The one exception to this is that video is actually the best way to teach people verbs (actions), so use short screencasts to show people how to use an editor, step through code in a debugger, and so on.

  6. Remember that the goal when teaching novices is to identify and clear up misconceptions. If early data shows that learners are struggling with some parts of a lesson, create extra alternative explanations of those points and extra exercises for them to practice on.

  7. Remember that you are not the first person to record educational videos: everything from short guides to entire books are there to help you.

Two-Way Video

Just as video lets an instructor show learners how she’s doing things (rather than just what she has done), learners can use it to show instructors how they are working. If you are teaching programming using desktop or laptop computers, have your learners record a 5-minute screencast to show how they solved a problem. You can then watch that video at 4X or faster to see how proficient they are with the tools they’re supposed to be using.

Freedom To and Freedom From

Isaiah Berlin’s 1958 essay “Two Concepts of Liberty” made a distinction between positive liberty, which is the ability to actually do something, and negative liberty, which is the absence of rules saying that you can’t do it. Unchecked, online discussions usually offer negative liberty (nobody’s stopping you from saying what you think) but not positive liberty (many people can’t actually be heard). One way to address this is to introduce some kind of throttling, such as a rule that says each learner is only allowed to contribute one message per discussion thread per day. Doing this allows those who have something to say to say it, while clearing space for others to say things as well.

When it comes to teaching platforms, you can either use an all-in-one learning management system (LMS) such as Moodle or assemble something ad hoc: Slack or Zulip for chat, Google Hangouts for video conversations, and WordPress, Google Docs, or any number of wikis for collaborative authoring. If you are just starting out, then use whatever requires the least installation and administration on your side, and the least extra learning effort on your learners’ side. (I once taught a half-day class using group text messages because that was the only tool everyone was already familiar with.)

The most important thing when choosing technology is to ask your learners what they are already using. Most people don’t use IRC, and find its arcane conventions and interface offputting. Similarly, while this book lives in a GitHub repository, requiring non-experts to submit exercises as pull requests proved to be an unmitigated disaster, even with its supposedly easy-to-use web editing tools. As an instructor, you’re asking people to learn a lot; the least you can do in return is learn how to use their preferred tools.

Points for Improvement

One way to demonstrate to learners that they are learning with you, not just from you, is to allow them to edit your course notes. In live courses, we recommend that you enable them to do this as you lecture; in online courses, you can put your nodes into a wiki, a GitHub repository, or anything else so long as it allows you to review and comment on their proposed changes, and the learner to make revisions based on your feedback, before those changes go live. Giving credit (or at least thanks) to people for fixing mistakes, clarifying explanations, adding new examples, and writing new exercises does increase the short-term load on the instructor, but reduces the lesson’s long-term maintenance costs.

A major concern with any online community, learning or otherwise, is how to actually make it a community. Hundreds of books and presentations discuss this, but almost all are based on their authors’ personal experiences. [Kraut2012] is a welcome exception: while it predates the accelerating decline of Twitter and Facebook into weaponized abuse and misinformation, most of what was true then is true now. [Fogel2017] is also full of useful tips for the community of practice that students may hope to join.

One other concern people often have about teaching online is cheating. Day-to-day dishonesty is no more common in online classes than in face-to-face settings, but the temptation to have someone else write the final exam, and the difficulty of checking whether this happened, is one of the reasons educational institutions have been reluctant to offer credit for pure online classes. Remote exam proctoring is possible, usually by watching the student take the exam using a webcam. Before investing in this, read [Lang2013], which explores why and how students cheat, and how courses often give them incentives to do so.

Different Types of Exercises

Every mechanic has her favorite screwdrivers, and every good teacher has different kinds of exercises to check that her students are actually learning, let them practice their new skills, and keep them engaged. Some types of exercise are well known, but others aren’t as widely used as they should be.

The two key requirements are that an exercise has to be quick for learners to do, and it has to be possible to check the answer automatically. These requirements rule out some useful kinds of assessment, but many remain.

The first type of exercise that works well online is a multiple choice question that presents a question and asks the student to pick the correct answer from a list. Doing this might (in fact, should) require them to do more than just read and remember, and as a previous post discussed, multiple-choice questions are most effective when their wrong answers probe for specific misconceptions on the student’s part.

Example: You are in /home/marg. Use ls with an appropriate argument to get a listing of the files in the directory /home/marg/seasonal. Which of the following files is not in that directory?

  • autumn.csv
  • fall.csv
  • spring.csv
  • winter.csv

The second type of exercise is write and run, in which the student has to write code that produces a specified output. When the code is submitted, we check its structure and/or output and give feedback. Write and run exercises can be as simple or as complex as the instructor wants. For example, it’s often enough with novices to simply ask them to call a function or method: experienced instructors often forget how hard it can be to figure out which parameters go where.

Example: the matrix M contains data read from a file. Using one function or method call, create a matrix Z that has the same shape as M but contains only zeroes.

Write and run exercises help students practice the skills they most want to learn, but writing good automated checks is hard: students can find very creative ways to get the right answer, and it’s demoralizing to give them a false negative. One way to reduce how often this occurs is to give them a small test suite they can run their code against before they submit it (at which point it is run against a more comprehensive set of tests). Doing this helps to catch cases in which students have completely misunderstood the written spec of the exercise.

To help students realize just how hard it is to write good tests instructors can get them to do it themselves. Instead of writing code that satisfies some specification, they can be asked to write tests to determine whether a piece of code conforms to a spec.

Example: the function monotonic_sum calculates the sum of each section of a list of numbers in which the values are monotonically increasing. For example, given the input [1, 3, 3, 4, 5, 1], the output should be [4, 3, 9, 1]. Write and run unit tests to determine which of the following bugs the function contains:

  • Considers every negative number the start of a new sub-sequence.
  • Does not include the first value of each sub-sequence in the sub-sum.
  • Does not include the last value of each sub-sequence in the sub-sum.
  • Only re-starts the sum when values decrease rather than fail to increase.

Fill in the blanks is a refinement of write and run in which the student is given some starter code and asked to complete it. (In practice, many write and run exercises are actually fill in the blanks because the instructor will provide comments to remind the students of what steps they should take.) Novices often find fill in the blanks less intimidating than writing all the code from scratch, and since the instructor has provided most of the answer’s structure, submissions are much easier to check.

Example: fill in the blanks so that the code below prints the string ‘hat’.

text = 'all that it is'
slice = text[____:____]

As described earlier, a Parsons Problem is another kind of exercise that avoids the “blank screen of terror” problem: the student is given the lines of code needed to solve a problem, but has to put them in the right order. Research over the past few years has shown that Parsons Problems are effective because they allow students to concentrate on control flow separately from vocabulary [Ericons2017]. The same research shows that giving the student more lines than she needs, or asking her to rearrange some lines and add a few more, makes this kind of problem significantly harder [Harms2017]. Parsons Problems can be emulated (albeit somewhat clumsily) by asking students to rearrange code in an editor.

Example: rearrange and indent these lines to calculate the sums of the positive and negative values in a list.

positive = 0
return negative, positive
if v > 0
positive += v
negative = 0
for v in values
negative += v

Tracing execution is the inverse of a Parsons Problem: given a few lines of code, the student has to trace the order in which those lines are executed. This is an essential debugging skill, and is a good way to solidify students’ understanding of loops, conditionals, and the evaluation order of function and method calls. Again, we don’t yet support this directly, but it can be emulated by having students type in a list of line labels.

Example: in what order are the labelled lines in this block of code executed?

A)     vals = [-1, 0, 1]
B)     inverse_sum = 0
           for v in vals:
C)             inverse_sum += 1/v
D)         pass

Tracing values is similar to tracing execution, but instead of spelling out the order in which code is executed, the student is asked to list the values that one or more variables take on as the program runs. Again, it can be implemented by having students type in their answers, but this quickly becomes impractical. In practice, the best approach is to give the student a table whose columns are labelled with variable names and whose rows are labelled with line numbers.

Example: what lines of text pass through the pipes and the final redirect when this file:


is run through this Unix shell command:

cut -d , -f 2 filename | sort | uniq > result.txt

Returning to debugging skills, another exercise that helps student develop them is minimal fixes. Given a few lines of code that contain a bug, the student must either make or identify the smallest change that will produce the correct output. Making the change can be done as using write and run, while identifying it can be done as a multiple choice question.

Example: this function is supposed to test whether a point (x, y) lies strictly within a rectangle defined by (x_min, y_min, x_max, y_max). Change one line to make it do so correctly.

def inside(point, rect):
    if (point.x <= rect.x_min): return false
    if (point.y <= rect.y_min): return false
    if (point.x >= rect.y_max): return false
    if (point.y >= rect.y_max): return false
    return true

Theme and variation exercises are similar, but instead of making a change to fix a bug, the student is asked to make a small alteration that changes the output in some specific way. These alterations can include:

  • replacing one function call with another
  • changing one variable’s initial value
  • swapping an inner and outer loop
  • changing the order of tests in a chain of conditionals
  • changing the nesting of function calls or the order in which methods are chained

Again, this gives students a chance to practice a useful real-world skill: the fastest way to produce a working program is often to tweak one that already does something useful.

Example: change the inner loop control in the function below so that it sets the upper left triangle of the matrix to zero.

def zeroTriangle(matrix):
    for c in range(matrix.cols):
        for r in range(matrix.rows):
            matrix[r, c] = 0

Matching problems are another entire family of exercises. One-to-one matching gives the student two lists of equal length and asks her to pair corresponding items, e.g., “match each piece of code with the output it produces”.

Example: match each function’s name with the operation it implements.

Function Operation
SGEMV triangular banded matrix-vector multiply
STBMV solve triangular matrix with multiple right-hand sides
STRSM matrix-vector multiply

Many-to-many matching is similar, but the lists aren’t the same length, so some items may be matched to several others. Both kinds require students to use higher-order thinking skills, but many-to-many are more difficult because students can’t do easy matches first to reduce their search space. (In technical terms, there is a higher cognitive load.)

Matching problems can be emulated by having students submit lists of pairs as text (such as “A3, B1, C2”), but that’s clumsy and error-prone. If available, the machinery built for Parsons Problems can be used let students drag and drop blocks of text to form matches.

Drag-and-drop opens many other doors: for example, tracing execution is easy to implement this way. So is labelling diagrams: rather than students typing in the labels, it is faster and more reliable for them to drag labels around to attach to the correct elements. The picture can be a complex data structure (“after this code is executed, which variables point to which parts of this structure?”), the graph that a program produces (“match each of these pieces of code with the part of the graph it generated”), the code itself (“match each term to an example of that program element”), or many other things.

Example: label the following diagram to show which structures the variables x, y, and z refer to after these three lines of code are executed.

x = 3
y = [x, x]
z = [x, y]


Drawing diagrams of things like data structures is also straightforward to do on paper but very difficult to grade automatically. One way to make solutions gradable may be to constrain the drawing in the same way that Parsons Problems constrain code construction, i.e., give students the pieces of the diagram and ask them to arrange them correctly, but this is a long way off.

We mentioned earlier that matching problems require students to use higher-order thinking skills. Summarization also does this, and gives them a chance to practice a skill that is very useful when reporting bugs rather than fixing them. For example, students can be asked, “Which sentence best describes how the output of f changes as x varies from 0 to 10?” and then given several options as a multiple choice question. Similarly, ranking problems present the student with several choices and ask them to order them from fastest to slowest, most robust to most brittle, and so on. (Ranking is more manageable when implemented with drag and drop than as a multiple choice question.)

One other kind of exercise that can be implemented as a multiple choice question is fault mapping: given a piece of buggy code and an error message, the student has to identify the line on which the error occurred. In simple cases this will be the line mentioned in the error message, but in more subtle cases, the student will have to trace execution forward and backward to figure out where things first went wrong.

Other kinds of exercises are hard for any automated platform to provide. Refactoring exercises are the complement of theme and variation exercises: given a working piece of code, the student has to modify it in some way without changing its output. For example, the student could be asked to replace loops with vectorized expressions, to simplify the condition in a while loop, etc. The challenge here is that there are often so many ways to refactor a piece of code that grading requires human intervention.

Example: write a single list comprehension that has the same effect as this loop.

result = []
for v in values:
    if len(v) > threshold:

Code review is hard to grade automatically in the general case, but can be tackled if the student is given a rubric (i.e., a list of faults to look for) and asked to match particular comments against particular lines of code. For example, the student can be told that there are two indentation errors and one bad variable name, and asked to point them out; if she is more advanced, she could be given half a dozen kinds of remarks she could make about the code without guidance as to how many of each she should find. As with tracing values, this is easiest for students to do when presented as a table, which we currently don’t support.

Example: using the rubric provided, mark each line of the code below.

01)  def addem(f):
02)      x1 = open(f).readlines()
03)      x2 = [x for x in x1 if x.strip()]
04)      changes = 0
05)      for v in x2:
06)          print('total', total)
07)          tot = tot + int(v)
08)      print('total')
  1. poor variable name
  2. unused variable
  3. use of undefined variable
  4. missing values
  5. fossil code

All of this discussion has assumed that grading must be fully automatic in order to scale to large classes, but that is not necessarily true. [Paré2008] and [Kulkarni2013] report experiments in which learners grade each other’s work, and the grades they assign are then compared with grades given by graduate-level teaching assistants or other experts. Both found that student-assigned grades agreed with expert-assigned grades as often as the experts’ grades agreed with each other, and that a few simple steps (such as filtering out obviously unconsidered responses or structuring rubrics) decreased disagreement even further. Much more research needs to be done, but given that critical reading is an effective way to learn, this result may point to a future in which learners use technology to make judgments, rather than being judged by technology.


Give Feedback (20 minutes)

  1. Watch [Wilson2017] as a group and give feedback on it. Organize feedback along two axes: positive vs. negative and content vs. presentation.

  2. Have each person in the class add one point to a 2x2 grid on a whiteboard (or in the shared notes) without duplicating any points that are already up there.

What did other people see that you missed? What did they think that you strongly agree or disagree with? (You can compare your answers with these lists.)

Adapting Multiple Choices Questions (30 minutes)

Pick one of the examples given in this chapter of using multiple choice questions to implement some other kind of online programming exercise, create an example, and swap with one of your fellow learners.

Adapting Write and Run Exercises (30 minutes)

Pick one of the examples given in this chapter of using write and runs exericses to implement some other kind of online programming exercise, create an example, and swap with one of your fellow learners.