The process of using your descriptions and evidence to classify what you have observed is known as ‘coding’. Coding can be done to an existing classification structure or you can develop codes from the data. Or to put it another way, you can impose a coding structure or let the coding immerge from the data as you analyse it. It is important to note or code only what you observe, not what you infer.

Allow time to write up your notes soon after the observation. Maintain a distinction between what you observed and how you reacted to it.

• In a structured form, this means leaving room for written comments.
• In a less structured ‘blank piece of paper’ approach, a line down the middle of the page might suffice (observations on the left, reflections on the right).

Notes from a 40 minute classroom observation could range from one A4 page to up to 10 pages or more, depending on what is observed and how much detail you record. Drawings and sketches are useful for laying out who was where and when. The evaluation team might also like to develop some shorthand for words or situations to enable more detail to be captured with less effort.

Data collected during observation can be subject to bias and it is important to keep this in mind as you collect the data. Possible biases to be mindful of include:

• confirmation bias – looking for and seeing things that confirm our thoughts
• observation bias – the streetlight effect, where we look in places where we think we will find what we are looking for
• the halo effect – initial impressions of a person or group of people that influence subsequent thoughts about this person’s or the group’s capacity or ability.