Session 4. Tools, Records, Confidentiality and Evidence of Impact

A minimal toolkit: fewer papers, each with a reader

• Six tools are usually enough: the working agreement, a one-page goal sheet, a three-line meeting note, an evidence note from a visit, a programme register (who is paired with whom, meetings held) and an end-of-cycle review. These come from courses 11.1 and 11.2; here you decide which ones the programme requires.
• Test every form with two questions: who will read this, and what will they do with it? If nobody reads it or nothing follows, delete it. Paperwork that feels like inspection is the fastest way to lose trust.
• Paper is enough. A folder per coachee held by the coachee, plus one register held by the coordinator. A group chat is not a record system; do not put observation notes or personal details into messaging apps.

Records and confidentiality: what to keep, share, destroy

• Keep facts and agreements: the goal, what was seen with a time, the next step and date. Do not keep judgements ('weak', 'lazy'), health or family information, or pupils' names. If you would be embarrassed for the coachee to read it, do not write it.
• Agree three levels of sharing at the start. Private: the coach's and coachee's own notes. Programme: counts and themes without names. Escalated: only when a pupil or person is at risk, or the law or school procedure requires. The ICF Code of Ethics for coaches follows a similar logic: confidentiality kept, with exceptions for legal requirement, risk of harm or consent, agreed in advance.
• Decide how long records are kept (for example, destroyed at the end of the school year unless the coachee asks to keep them), where they are stored (a locked cupboard) and who may see them. Follow school policy and national rules on personal data; if unsure, ask the head or the district.

Evaluate at five levels, and plan it before you start

• Thomas Guskey describes five levels of evidence for professional development: 1 participants' reactions; 2 participants' learning; 3 organisation support and change; 4 participants' use of new knowledge or skills; 5 student learning outcomes. Kirkpatrick's four levels (reaction, learning, behaviour, results) are a similar older model.
• For coaching, translate them: Was it useful? Can coachees state their goal and technique? Were time, coaches and clinics really protected (level 3, where most programmes fail)? Is the practice seen in lessons? Is there any sign in pupils' work or exit tickets?
• Plan the evaluation before launch: for each level write the question, the evidence, who collects it, when and how much time it costs. A baseline (the same measure before coaching) is worth more than a long report afterwards.

Honest evidence for a board

• Small programmes have small numbers and no comparison group. Say 'we saw', 'it contributed' and 'we are not certain', not 'it proved'. Look for at least two sources that point the same way, and report the awkward ones too.
• Expect practice to change before pupil results do. The Kraft, Blazar and Hogan review found larger effects on teaching than on achievement, so a first-year report that shows changed practice and early pupil signs is a healthy sign, not a weak one.
• Never turn programme data into a ranking of teachers or coaches. A good one-page board report has: purpose, who took part (counts), three indicators, one short quotation used with permission, what did not work, and the decision requested.

The front-page test for any record

Before you write a line, imagine that the coachee, the head and the coachee's colleagues will read it on the school notice board. If you would not write it there, do not write it anywhere. Use a three-part formula: Goal in the teacher's words; Seen, with the time; Next step with a date. Everything else is memory, and memory belongs to the people in the room.

Decide who holds what. The coachee keeps their own folder and decides whether to bring it. The coach keeps only a minimal log of dates. The coordinator keeps the register of pairs and counts. Agree when records are destroyed, for example at the end of the school year, and follow school and national rules on personal data; ask the head or district if you are unsure.

A cheap method for each of the five levels

Level 1, reactions: a three-question note at mid-term (useful? one thing to keep? one thing to change?). Level 2, learning: each coachee states their goal and technique in one sentence. Level 3, organisation: the count of meetings held and the timetable check for protected time. Level 4, use of practice: the coach counts the focus practice in three lessons before the cycle and three after, with times.

Level 5, pupils: an exit ticket or a small piece of work from the same class in week two and week nine, reported for the class and never as a score for the teacher. Look for two sources that point the same way, report the awkward ones as well, and write 'contributed' rather than 'proved'. Level 3 is usually where programmes fail, so start your evaluation there.

Try it tomorrow

• Rewrite one of your own meeting notes in the three-part formula and delete everything else.
• Count one practice in three lessons this month, with times, as a simple baseline.
• Write your programme's three levels of sharing on a card and read it to the next colleague you coach.