Analysis

CRITICALACTIONS

Guidance, not policy
This page is not policy and it does not set out a statutory requirement. It is supplementary guidance to support the learning and practice of everyone at Sparks Fostering.

Where the information becomes a judgement.

What this domain is for

Its purpose is to bring the evidence from every preceding domain together into a clear, reasoned safeguarding judgement. It is the point at which information becomes analysis, and analysis becomes a decision.

It holds the strengths and protective factors — assessed critically for whether they genuinely and reliably protect the person, not narrated as a strengths-based exercise; the vulnerabilities and aggravating factors, named explicitly including the ones that are uncomfortable, culturally sensitive or professionally difficult to raise; the unknowns and missing information, recorded so that uncertainty is visible and optimism or partial knowledge is not treated as evidence; how physical harm and emotional or mental harm are affecting the person now and what that indicates about future risk; and an explicit judgement about the current level of risk.

Vulnerabilities and unknowns are weighed more heavily than reassurance, and what is described is current risk rather than hoped-for improvement.

What an analysis actually is

Recording is what happened. Analysis is what it means, and what follows from it.

Analysis works in two ways. It is a way of thinking that runs through a whole piece of work: noticing, questioning, testing what you are being told, and changing your mind when the evidence changes. It is also a piece of writing, the part of a record where you pick out what matters most from everything you know, explain what you make of it, and say what should happen as a result.

There is a test a reader applies without realising: so what? If a paragraph states a fact and leaves the reader to work out for themselves why it matters, that is description. If it says what the fact means for this child, in this household, now, that is analysis.

A second test is harder and more useful. Could somebody who has never met this family read your analysis and arrive at the same conclusion you did, and see how you got there? If not, the reasoning is still in your head rather than on the page.

A list of strengths and vulnerabilities is not an analysis

This is the most common thing written in place of an analysis, so it is worth being plain about it.

Two lists, one of strengths and one of vulnerabilities, are an input to the analysis, not the analysis. They sort what you know into two columns. They do not tell the reader which items carry the most weight, how they act on each other, or what they add up to.

The analysis is what you do next.

  • Weigh them against each other. Three strengths and one vulnerability does not mean the strengths win. A single serious risk can outweigh a long list of positives, and one reliable protective factor can change the whole picture.
  • Say how they interact. A strength that holds only while one particular adult is in the house is a different strength once that adult is at work.
  • Say what it means for this child. The same circumstance can be a strength for one child and a risk for another. Age, history and what the child has already lived through all change the answer.
  • Reach a judgement and say what should follow. An analysis that stops before the conclusion hands the decision to whoever reads it next.

Five questions that anchor an analysis

These are the anchor principles set out by Brown, Moore and Turney for Research in Practice, and they work as well for a fostering household as for a child. Answer all five and you have an analysis; skip one and you usually have a gap somebody else will find.

1. What is this assessment for? What are you worried about, and what might the child or the carers be worried about? What do you need in order to do it properly, such as an interpreter or a chronology?

2. What is the story? Can you tell it from the child’s point of view? Do different people hold different versions of it? How does it make you feel, and how might your own background be shaping what you notice?

3. What does the story mean? What explanations have you come up with, and what are the alternatives? What effect is all of this having on the child? What have you not been able to find out, and what is disputed?

4. What needs to happen? What would have to change for you to step back? What do the child and the carers want? Are they able to make that change, and what support would it take?

5. How will we know we are making progress? What did you expect to have happened by now? What is working, what is not, and what will you do about the part that is not?

Work with more than one explanation

The quickest way to get an analysis wrong is to settle on the first explanation that fits and then collect evidence for it.

Write down at least two explanations for what you are seeing, including one you do not believe. A child who is withdrawn might be depressed, might be being bullied, might be worried about something at home, might be tired, or might be fifteen. Set them side by side and ask what evidence would tell them apart.

Then do the harder thing: ask what you would expect to see if you were wrong, and go and look for it. An explanation that survives a genuine attempt to knock it down is worth something. One that has only ever been agreed with is not.

Where you have landed on one explanation, say in the record why you preferred it over the others. That sentence is often the single most useful line in a file two years later.

What makes a protective factor real

Protective factors get written down far too easily. Before one goes in the analysis as something that reduces risk, it has to survive four questions.

  • Is it reliable? Has it actually held, over time, or is it something that was said in a meeting?
  • Has it held under pressure? A support network that works on a good week is not evidence about a bad one.
  • Does it depend on one person? If it does, what happens when that person is ill, working, or no longer there?
  • Does it protect this child, from this risk? A warm, experienced carer is a real strength, and still not a protection against a risk that happens outside the home.

A factor that fails those questions is not worthless, but it belongs in the record as something hopeful rather than something that lowers the risk. Writing it as a protection when it is not is how files come to read as reassuring right up until the point they do not.

The traps

These are well documented, and everybody falls into them. Knowing their names makes them easier to spot in your own writing.

Confirmation bias. Noticing what fits the view you already hold and passing over what does not.

The rule of optimism. Preferring the kinder reading when the evidence is mixed, usually because you like the people involved.

Anchoring. Letting the first account you heard set the frame for everything after it, including accounts that contradict it.

Start-again syndrome. Treating each new incident as the first one, so a pattern never accumulates. A chronology is the cure, which is why the repeated behaviours and recurring themes section exists.

Disguised compliance. Reading co-operation with the process as change in the situation. Appointments attended is not the same as anything being different for the child.

The useful habit is to re-read what you have written and ask which of these a stranger might accuse you of. If you can see it, fix it before somebody else has to.

Say what you do not know

An analysis that only contains what you found out reads as though you found out everything. Uncertainty has to be visible, because a gap is itself a finding.

Write down what you could not establish, who holds that information, what you did to try to get it, and what difference it would make to your judgement if the answer went one way rather than the other. Where information is disputed, record both accounts and say which you are working to and why, rather than quietly picking one.

Partial knowledge is not evidence of safety. Where you are relying on something unconfirmed, say so in the sentence where you rely on it.

Writing it so it holds up

An analysis is read by people who were not there, sometimes years later, and sometimes by the person it is about.

  • Separate fact, opinion and what somebody else told you. Each one carries different weight, and the reader cannot tell them apart unless you do.
  • Describe behaviour rather than labelling the person. What they did, when, and how often, not what kind of person that makes them.
  • Write it so the person could read it. That is not a reason to soften a concern. It is a reason to put the concern plainly and without contempt.
  • Keep it proportionate. Length is not thoroughness. Everything in the analysis should be there because it changes the conclusion.

The rules on what is recorded, how long it is kept and who can see it are on the records, confidentiality and data protection page rather than here.

Before you sign it off

Read it back and check you can answer yes to each of these.

  • Does it say what the information means, not only what it is?
  • Would a stranger reach my conclusion from what is written here?
  • Have I weighed the strengths and the vulnerabilities against each other rather than listed them?
  • Have I considered an explanation other than the one I settled on, and said why I preferred mine?
  • Is every protective factor one that has actually held?
  • Have I said what I do not know, and what difference it would make?
  • Is the child’s own view in here, in their words where possible?
  • Does it end with a judgement and what should happen next, rather than trailing off?

Where the judgement is that risk has reached a point needing immediate action, the crisis indicators and thresholds sections set out what follows.

You could be welcoming a child to your home soon

Or call 07920 097 834  ·  [email protected]