Switch course ▾
Session 3. Descriptive Observation: Seeing Without Judging
Objective. Record a short classroom visit in descriptive language and tell descriptive, inferential and evaluative statements apart.
Everyone who has visited a colleague's lesson has felt the urge to say 'good' or 'weak'. Those words feel like honesty, but they hide what you actually saw and give the host nothing to test. A network that can describe the same lesson in the same words can learn from it. A network that argues about whether the lesson was good learns nothing. This session trains the eye and the pen: what to watch, how to write it, and how to catch yourself.
Key ideas
Three kinds of statements
- Descriptive: what could be seen or heard. 'Eleven of 28 pupils wrote a sentence of their own.' Inference: a meaning you add. 'The pupils did not understand.' Evaluative: a verdict. 'A weak lesson.'
- Inferences are not forbidden. They belong in the debrief, labelled as guesses, and tested against the descriptions.
- Praise is evaluative too. 'Excellent questioning' tells the host nothing she can repeat.
What to write down
- Three streams: what the teacher says and does; what pupils say and do; what the task asks pupils to do. Add the minute so that you can place each note.
- Use exact words in quotation marks, counts ('7 of 28'), times ('4 minutes') and positions ('group at the window'). Numbers are the strongest friends of description.
- Write the task word for word. What pupils are asked to do shapes what they learn.
Few look-fors, not a long checklist
- A checklist of 30 'good teaching' items turns every visit into an evaluation. Choose one to three look-fors tied to the problem of practice.
- For the problem 'What do pupils do when asked to explain?', look for: the exact question or task, who speaks, how long, and what pupils write.
- You will miss things. That is fine: the network sees what no single observer can.
Traps and a quick test
- Warning words: good, weak, lively, boring, too, seemed, should, lazy, clever. They usually mark a verdict.
- Quick test: could another observer in the same room have checked this note? Would the host agree that it happened? If not, rewrite it.
- Never write pupils' names. Use seat numbers or 'group 3'. You study tasks and talk, not pupils.
From seeing to judging
Why even careful observers drift into verdicts
Writing is selective. You cannot record everything, and the moment you choose what to write, a theory of good teaching is already working inside you. If you believe good lessons are quiet, you will note noise; if you believe they are busy, you will note idle pupils. Most of this is not conscious.
The remedy is not to pretend you are a camera. It is to agree before the visit what to look for, to write exact words and numbers, and to check your notes with someone who sat in the same room. Disagreement between observers is useful information: it shows where a judgement entered.
Tools for describing: tallies, quotations and time lines
Three simple tools cover most problems of practice. A tally counts events, such as how many different pupils spoke. A quotation sheet copies exact teacher questions and pupil answers. A time line records minutes, so that you can say 'for six of the twelve minutes pupils copied'. Choose one tool for your look-for and use it for the whole visit.
Seat numbers or a sketch of the room let you refer to pupils without names. Add a margin for your own guesses, marked with a question mark, and leave them for the debrief. This keeps inference visible and separate, so that it can be tested instead of taken as fact.
Examples from the classroom
Counting the questions
Scene. In a Namangan school, a network wants to know how pupils take part in Grade 5 reading. Three observers each use a tally. In 13 minutes the teacher asks 16 questions; 12 can be answered in one word; five different pupils answer aloud. One observer's note said 'lively'. Nobody can say what 'lively' meant, but all three tallies agree on the five pupils.
Insight. Counts that observers share can be argued about usefully; adjectives cannot.
A kind word that taught nothing
Scene. In Bukhara, an observer writes 'Excellent questioning' about Dilnoza's lesson. In the debrief, Dilnoza asks, 'What did I do that I should do again?' Nobody can say. The facilitator asks the observer to read her notes. They show that Dilnoza waited four seconds after two 'why' questions. The group agrees that is the thing to repeat.
Insight. Praise without evidence cannot be repeated; the description beneath it can.
The quotation that changed the debrief
Scene. In Samarkand, an observer in a Grade 8 history lesson writes the pupil's exact words: 'I just copy what is in the book because the teacher said it is on the exam.' Read aloud, the sentence moves the conversation from pupils' effort to what the task and the exam signal. No one is blamed, and the group has something to work on.
Insight. An exact quotation carries information that a summary or a verdict loses.
Common pitfalls
- Using adjectives and adverbs ('very', 'too', 'clearly') and believing that they are facts.
- Writing about the teacher's manner instead of the task and what pupils did.
- Recording everything, so that nothing can be compared across rooms.
- Mixing guesses into the notes without marking them.
Script a Lesson Extract and Clean Your Notes
- Trios (8 min): the facilitator reads the SAMPLE lesson extract from the Kit aloud at a normal pace. Each person writes notes in three columns: teacher, pupils, task.
- Trios (10 min): compare notes. Count how many questions the teacher asked, how many different pupils answered and what pupils wrote. Mark every word in your notes that is a verdict (circle) or guess (underline).
- Whole group (7 min): share the counts. Differences in counts are normal; agree on what you can all stand behind.
- Pairs (20 min): in the Kit, sort the eight observer notes into D (descriptive), I (inference) or E (evaluative), then rewrite four evaluative lines as descriptive notes and say what you would have to count or quote to write them.
- Second short activity (10 min): each person writes one note from their own recent lesson in the three streams and swaps with a partner, who circles any word that is a verdict. Alone (5 min): write your school-based task.
Your facilitator will guide you through this activity in the session.
Questions to discuss
- Which of your first notes contained a verdict you did not notice?
- What changed when you counted instead of describing with adjectives?
- How would the host teacher feel reading your descriptive notes, compared with evaluative ones?
Reflect
- What is the most evaluative word you use about lessons in the staff room?
- What did the task, not the teacher, cause in the extract?
- What would you not want written about your own lesson, and why?
Try it tomorrow
- Visit a colleague's lesson (or your own with a partner) for 12 minutes and use only one tool: a tally, a quotation sheet or a time line.
- Read your own last lesson plan and rewrite one evaluative phrase in it as something that could be seen.
- After your next staff meeting, count how many comments about lessons were descriptive and how many were verdicts, and keep the numbers to yourself.
Further reading and links
Write what could be seen and counted, in three streams, and keep verdicts and guesses for the debrief where they can be tested.