Mentor Guidance¶

A rough guide on how to encourage learners.

Finding a level¶

It’s important to pitch your support at the right level for the person you’re helping, so that they’re able to absorb what you’re saying. One way to think about this is adding branches to a trunk of knowledge; if the trunk is a sapling then adding a large bough with many branches will crush the sapling. Conversely, presenting a bud of information can be frustrating for both parties if the person you’re helping is more experienced.

For complete novices, or when very first introducing a new topic (even to an experienced audience), this means that you should try to present an idea in its very simplest form. Try to think about the essence of what makes the idea important, rather than the details of how it works. A good example of this is the idea of “lies to children”.

As a concrete example, let’s consider the case of explaining variable assignment in Python:

foo = 42

A simple explanation here might be that we are “telling the computer to remember that ‘foo’ means ‘42’”. Note that there’s no mention here of “variable”, “identifier”, “assignment”, “integer” or even “number”. Depending on your audience, this might be a useful description.

For an audience which has some programming experience, though perhaps not in Python, you might instead describe this as “assigning 42 to the variable ‘foo’”.

Note that while there are some technical terms here, there’s still no mention of what identifiers are or what their naming rules are. It’s fine to leave explaining those until they become important.

A superb example of this in action is this video explaining quantum computing at five levels of depth.

Explaining the “how”¶

Of course after a while many students will want to understand the “how” of something as well as the “what”. The point where they start wanting to know this varies from person to person as does the impact it has on their learning.

As a result, while a hand-waved explanation (ignoring the man behind the curtain) will suffice when initially explaining a topic, you should be willing to explain in more detail if the student asks. Bear in mind that you still need to find the right level on this axis too -- if you switch to a very detailed explanation then it can still bewilder the learner.

For example, in the case of our variable assignment, you might be asked how the computer understands the idea of 42, or how it’s stored. A first response might be to talk about the differences between numbers (used for addition, etc.) and strings (used for text). Even if you do end up talking about the different in-memory representations (perhaps the learner has already heard about ASCII), you’d almost certainly want to avoid talking about the internals of CPython objects.

This line of questioning is sometimes related to the student trying to attach or relate the new topic so something they already know, but not being sure what the connections are. It’s reasonable to ask the student if this is the case, either by asking why they’re asking for more details (“what makes you ask that?”) or if they’ve seen something similar before.

Learn by doing¶

As much as possible, encourage the learner to experiment with what they’ve just learned. Learning is reinforced by putting it into practise and is often most enjoyable when discovering new things out for yourself.

For our learning Python example, this means getting the learner into a Python shell or similar environment where they can easily try new things. It doesn’t matter that they’re not writing a full program yet, the feedback loop from trying things and experiencing that there’s no downside to things not working as hoped is what you’re after.

Keep it interactive¶

Initially when explaining something you’ll need to mostly be telling the learner what things are and what they do. At this stage, it’s important to gauge their reaction to ensure you’re going at a pace that works for them. If there’s something you’re not sure that they’re understanding, do ask them!

Once the learner has formed enough of a basis of understanding of a topic you can start asking them to explain how they think things might work if there were slight variations. In our Python example, you might show a conditional block:

foo = 42
if foo > 25:
    print('greater')
print('nope')

And then ask the learner either what they think might happen if you changed the > to a < or to change the code to detect a different range of numbers. In either case you don’t need to tell them whether they’re right or not, but instead encourage them to try it themselves so that they can learn by doing.

For more advanced learners (much closer to your own level of understanding), then pair programming techniques can start to be used. You may have noticed that what’s described here is similar to the approach of having a driver who operates the machinery (changing the code) and a navigator who guides the overall direction (towards deeper understanding), albeit far less rigorous.

Let them do the thinking (provide options not answers)¶

It’s very tempting and very easy to just give a learner the answer to their question. While that can be suitable (for example if they’re clearly completely stuck), it doesn’t necessarily help them understand a topic. If you can break down the thing they’re trying to learn into small chunks and then ask them how they think they’d achieve each of the pieces, then this can both help them understand that topic better as well as how to apply the knowledge they already have.

While you need to be careful that it doesn’t turn a learning session into a never-ending series of brain teasers (which can be tiresome), this can be a useful approach. It’s reasonable to explain to the learner that the approach of breaking down a puzzle is a useful one and that that’s how you’re going to approach it.

Using our Python example again, you might start with a task of “print odd numbers up to 20”. It can seem that there’s not much to this, however there are a number of aspects which a novice might not know about -- at least: loops, mathematical operations, conditional blocks and printing variables. For each of these you can spend a little bit of time just explaining the basic concept and letting them explore how it works before getting them to bring the pieces together.

If you had lots of time, that might look something like this:

• Explain print("some text")

• Ask them to make it print “some other text”

• Explain foo = 42; print(foo)

• Ask them to make it print a different number

• Explain for foo in range(5): print(foo)

  note: I probably wouldn’t at this point explain the interaction between in and range being a generator; I’d probably just think about this like a loop over numbers

• Ask them to make it print up to a higher limit

• Explain foo = 42 % 3; print(foo)

• Ask them to make it detect numbers divisible by something else

• Explain foo = 42; if foo == 42: print("yes")

• Ask them to make it print if the number is divisible by three

• Ask them if they now think they could solve the original task, and what that might look like

Helping when they’re “stuck”¶

When a learner complains that they’re stuck it can be particularly tempting to give them the answer. However just as when giving a friend advice, it can be more useful to help them think through the problem themselves.

Get them to explain the problem they’re trying to solve and what they’ve done so far to fix it. This will help you understand their perspective, which you can then use to guide them towards a solution. It can also act as a rubber-duck device allowing them to realise why something doesn’t for themselves.

Answering their questions¶

Even the best teachers expect to be asked questions. You should ensure that you allow the learner opportunities to ask you questions and treat those questions with respect. It’s important that they feel that they can ask for help and know that they’ll be supported rather than scorned.

There are two extremes here, both of which deserve consideration. Firstly, if they ask a question which seems obvious to you, you should bear in mind that (for technical topics) they likely have less experience than you and thus don’t have the luxury of a more complete picture. It may help you to think about them as being one of the people who, at that moment, is lucky to be discovering something new and you as the person who gets to share the knowledge with them. (Some estimates suggest around 10,000 people find out about any given topic on a given day!)

Secondly, it’s possible that they may ask something which uncovers something that you don’t know. It’s completely fine (even encouraged) to admit that you don’t know something. If the question is on topic, you could then work together to find out the answer. (This may even be something you thought you knew, but their question yields new insights).

Example phrases¶

Here are some sentence starters you might find useful:

• Have you considered ...

• Why do you think X happens

• Tell me what you’re thinking

  (Useful if they get stuck and go silent)

• What happens if you change..

• Which bit of the code is responsible for…