01763 852 225

Tips on evaluating evidence in an investigation

Investigation 3rd July 2012

Tips on Evaluating Evidence by Tim Kingsbury, Head of Investigation

I find I need to analyse and evaluate evidence at different points during the investigation process.  The first phase begins once the Complainant interview is complete, and I analyse the record of interview to identify all the issues and incidents about which I will need to get evidence from the Respondent(s) and witnesses.  So the first questions I ask are ‘What are the issues?’ and ‘Who will have relevant evidence about these issues?’  The answers will form the basis for my investigation plan, and then for my preparation for each subsequent interview.

The way I go about this, after I have interviewed the Complainant, is to evaluate the record of interview.  I go through and highlight on the record of interview the individual allegations or issues that need to be addressed, and also highlight who was present at the time, to identify potential witnesses for interview.  I later use this highlighted record in preparing for those interviews and the questions I will ask.

The next phase comes once the interviews are complete.  The need then is to assess the evidence obtained so far.  The question I ask myself here is ‘Have I got all the evidence I need?’  There are three parts to this:

  • Have I interviewed everyone I need to?

If there are lots of witnesses to an incident it may not be necessary to interview them all, but to decide who were closest, or who will have had the best view or knowledge of the situation.  It is also necessary to be aware of potential bias in witnesses, and so take a cross-section so they are not all close friends of one party.  Have I seen enough people to be able to make an independent assessment of the truth of what happened?  And have I made a balanced selection from among the potential witnesses to an incident?  If I have decided not to interview someone, am I clear in my own mind why that is?  (It is worth checking your own potential bias in favour of one or other party in this.)

  • Have I got evidence on all the issues I need to address?

Here I check through the Complainant’s record of interview again, to make sure I haven’t missed anything.  Then I check whether I have evidence from all of the relevant Respondents and witnesses, i.e. did I forget to ask someone a vital question?

Sometimes there is an issue of clarity that needs to be addressed – ‘Looking at Witness X’s response now, I can see they misunderstood what I was asking’, or perhaps ‘The answer they gave is ambiguous and I need to understand what they meant’, or ‘They said they would forward me a copy of that e-mail.  I need it but I haven’t received it yet.’  If so, perhaps a telephone call is needed to re-interview the witness about that point, and make an addition to the record of interview.

However with complex cases I sometimes find I need to draw out the issues into a spreadsheet, listing issues (with the paragraph number from the Complainant’s record of interview) in a column, and Respondent’s and Witnesses’ evidence on each issue in separate columns (also cross-referencing back to the source paragraph number – this will help when writing the report.)  Once the interviews are complete I then check through this to make sure I have covered all the issues I meant to with each of the relevant witnesses.

With documentary evidence I create a Table and list all the documents in date order, showing the originator, date of document, and brief summary of what it is about or its importance.  This will later become an appendix in the report.  This table is sometimes helpful in making the chronology and development of a conflict clearer.

  • Have I sufficient evidence to reach a finding?  This question comes slightly later in the process (see below).

The next phase is to make a decision on the facts.  Where, on the balance of probabilities, does the truth lie about what actually happened?  Sometimes the Respondent will have agreed that the Complainant’s core allegation was factually correct.  In all the workplace investigations I have conducted that has happened to me just once!  More usually judgement is needed to decide between different accounts of what happened in order to decide, on the balance of probabilities, what actually did take place.  This is where the evidence of witnesses is crucial in weighing what corroboration for different accounts exists.  The test is for the investigator to have reached a genuine view based on the reasonable grounds (from the evidence available), after having conducted as much investigation into the matter as was reasonable in the circumstances (British Home Stores v Birchell [1978] IRLR 379 EAT).

It is worth bearing in mind here recent guidance from the Court of Appeal in the unfair dismissal case of Salford Royal NHS Foundation Trust v Roldan [2010] IRLR 721 CA.  In that case it was held that the more serious the consequences of dismissal for the employee, the more thorough the disciplinary investigation should be.  It also ruled that where a case turns on one employee’s word against another’s, it can be legitimate for the employer to give the accused the benefit of the doubt and to find the accusations ‘not proven’.

For the workplace investigator, who is seeking to establish the facts in advance of any disciplinary action, this means that the more serious the potential outcome of a case, the more careful and thorough the investigation needs to be.  So the question here is:

  • Have I sufficient evidence to reach a finding?

If not, and all of the available evidence has already been gathered, then it may be the outcome will be ‘Insufficient evidence to either corroborate or disprove the allegations’, in line with the Salford Royal NHS Foundation Trust v Roldan judgement.

The final phase is to reach a Finding.  Those of you who have attended CMP’s Qualified Workplace Investigator training course that I deliver will know that I show you how to create what I call Measures of Assessment.  I take the relevant policy definition, such as for harassment, break it into its constituent parts, and then phrase each part as a question.  For example, ‘Did the incident or behaviour occur as alleged by the Complainant?’ (Does the evidence establish the facts?)  This is always the first question.  Another might be, ‘Was the behaviour unwanted?’, if that is one of the elements of the definition.  I find this question is helpful in making me consider whether the evidence indicates the behaviour was unwanted by the recipient at the time it occurred.  It may be that the recipient did not complain at the time but nevertheless acquiesced to it only unwillingly.  However it sometimes happens that the behaviour was not unwelcome to the recipient at the time but the complaint has been prompted, for example, because of some other later unwelcome but justified management action or decision.

Having created the Measures of Assessment I can then ask those questions of the evidence available to me, and reach a decision about the answer on the balance of probabilities.  In that way I can assess the evidence in a consistent way against the whole policy definition in order to determine whether or not the evidence shows, on the balance of probabilities, that the standard of behaviour set in the policy definition has been breached.  I can then also demonstrate in a clear and logical way in the report exactly how I have reached my Finding, and what evidence I have relied upon to do so.