This is first part of 7 part series discussing ideas presented in “How Learning Works: 7 Research-Based Principles for Smart Teaching”.

In the first chapter of “How Learning Works” authors describe how Students’ prior knowledge affects their learning. Its main message is:

Students’ Prior knowledge can help or hinder learning.

Different Aspects of Knowledge

When prior knowledge is activated, sufficient, appropriate and accurate it will facilitate learning. When it’s inactive, insufficient, inappropriate or inaccurate it will have a negative impact on learning.

We need to understand what those terms mean before we dive into practical ways of improving the learning experience.

If a student cannot make the connection between prior knowledge and new concept, this knowledge is inactive.

Sometimes there is some information lacking and the learner does not have enough knowledge to grasp new concepts. It is important to understand here that there are different ways of “knowing something”. There is declarative knowledge — knowing what and why and there is procedural knowledge — knowing how and when. For example, a student may know what the unit tests are but have never written one.

In some cases, the knowledge we have seems to fit the new problem, but in fact, it cannot be used in the new context, it is inappropriate. For example writing an essay and technical documentation both require writing skills but they apply different rules. Another example will be when students proficient in one programming language try to apply same approaches in newly learned language (eg. loops vs vectorised code).

When prior knowledge is inaccurate or wrong students may have problems understanding new concepts because they will be in contrast with what they already know. This can be very straight forward and easy to fix. If experienced Python programmer indexes from 0 while working in R, he or she need to be corrected and with time will make fewer mistakes. It can be much harder to revise deeply held misconceptions. It can also be hard to spot them since the student will be convinced she or he understands the concept well.


Before the instructor can start teaching she has to assess both the amount and nature of students’ prior knowledge. How can we gauge the extent and of students’ prior knowledge? I will try to present some solutions based on what the authors say and how I think it fits with Workshops.

Talk to the Hosts of the Workshop. Learn about the background of the students. What domain are they coming from? What are the datasets they usually work with? Do they collaborate on the code they write? This will help you get a beter grasp of students’ background.

Have Students Assess Their Own Prior Knowledge. Self-assessment designed well will allow you to tell if students have declarative or procedural knowledge. Software Carpentry uses pre-workshops surveys that students fill in before the workshop. The instructor should read it carefully before the workshop.

Administer a Diagnostic Assessment. Having few short exercises with different levels of complexity before starting the lesson will help you gauge learners expertise.

Use brainstorming or Assign a Concept Map Activity. I found those especially useful since they integrate well with the rest of the lesson. We use those techniques in Data Carpentry spreadsheet lesson when students come up with different problems with spreadsheets by themselves and the result of this is the great start for the discussion.

Look for Patterns of Error in Students Work. Paying attention to what kind of mistakes learners make in exercises will help you grasp better the state of understanding.

What to do now?

Once we know what is the status of prior knowledge in the group of learners we need to address those issues.

How to Activate Prior Knowledge? One way is to emphasise the connection to previous or other lessons. When teaching Python and showing joins in Pandas connect this to what students learned in SQL lesson. Other is to use analogies from everyday life. An example could be cooking recipe and Makefiles. You have to be careful though to see and tell students when analogy breaks.

Lack of Necessary Prior Knowledge means you have to adjust workshop content to get students up to speed. Since the experience may vary across the group, it helps to pair less advanced students with savvier ones.

To help students Recognise Inappropriate Prior Knowledge highlight the conditions of applicability. For example, not all data fit tabular format or bar plot is good for presenting categorical data but not relationship beteen two variables. Again show learners where the analogies break down.

How to Correct Inaccurate Knowledge? The best approach is to let student explain their reasoning. One of the methods we use during the Workshops is to pair students for exercises and let them explain their choice to a partner. Also allowing students to try out their newly acquired knowledge through different exercises will help with reinforcing new accurate knowledge.


There are many good and bad ways students’ prior knowledge may affect learning. Above I have explained what are the culprits and solutions with some practical tips that can be applied id Carpentries Workshops.

I hope you found above post interesting and learned something new.