Is there a way to test the reliability of a child's witness statement years after it was taken? Maybe, as a child psychologist explains

Analysing child interview evidence

Child psychologists have consistently emphasised the importance of specialised methods when conducting an interview with a child witnesses (for example, Ceci & Bruck, 1995(1); Westcott et al, 2003(2); Lamb et al, 2008(3)). Based on this, the Scottish Government compiled very detailed practice guidelines for social workers and police officers who conduct joint interviews with a child witness (Scottish Government, 2011(4)).

Covering core practical issues, these guidelines clearly set out how a reliable interview with a child witness (under 16 years at the commencement of the initial interview) should be conducted, pointing out the common pitfalls and how to avoid them. Such texts and research are extremely valuable in specifying procedures and techniques that ensure child witnesses are treated fairly, sensitively and appropriately when interviewed by police officers and social workers, and that their evidence during the interview is treated as reliable when presented in court.

Where there are any doubts about the reliability of a child’s original interview evidence – perhaps due to the quality of information the child provided, or the method or style of questions used by the interviewers – the professionals involved have an opportunity to conduct as many follow-up interviews with the child as required, until they are satisfied no further clarification can be gained. It is that follow-up interview option which ensures a thorough assessment of the child’s testimony. By controlling and extending the interview process in this way, professionals can make reasonably confident judgements about the reliability of the child’s original comments.

Analysing “cold case” child testimony

But what happens when there can’t be further interviews? “Cold case” analysis of existing child evidence using the original interview transcript alone – that is, when existing child evidence is re-examined several years after the original interview – typically does not have the follow-up option. Here are some real-world scenarios where a cold case analysis is sought:

  • A young adult learns that when he was a pre-schooler, he accused one of his relatives of physical abuse, though he remembers nothing of the incident. This deeply troubles him and he now feels responsible for the family breakup that subsequently occurred. As a result, he asks for his original interview to be re-analysed for reliability.

  • A four-year-old alleged that his father beat him, but this was not proved. Five years later, when the father cohabits with a woman who has children of her own, the social work department objects because of the previous allegation. The father contests the reliability of the original evidence, but the whereabouts of his ex-wife and son are unknown.

  • A five-year-old alleged her father abused her and her mother, and the father is refused further contact. At the time, he claimed the mother coached the child to lie. Years later, the father asks for the original interview in which the child made the allegation to be re-analysed, as he wants to resume contact with her. The mother won’t allow the child to be re-interviewed.

In each of these scenarios, cold case analysis of the transcript of earlier child evidence is required, but the crucial option of re-interviewing the child is either inappropriate, impractical or impossible. Therefore, analysis of an original child interview years after it was conducted is much more challenging than a contemporary interview analysis. In effect, the absence of follow-up interviews could significantly weaken any re-assessment of reliability.

Despite this, cold case analysis of the reliability of child evidence given in an interview is not automatically de facto less valid than a contemporary analysis. The difficulty created by absence of the follow-up interview option can be countered by conducting the cold case analysis using a systematic contextual model that is based on the psychology of child development. In the remainder of this article, I will propose such a psychological model, and I will also suggest that although this form of reliability assessment is by no means an exact science, a systematic contextual approach using the model outlined in this article can lead to informed opinion.

The contextual model

This model involves asking a series of questions, progressing from stages 1-8, as follows:

Stage 1: Did the original interview methods follow the best-practice guidelines?

This involves a thorough analysis of the interview structure and the style of questioning to verify that the requirements of the best-practice guidelines were met. Key issues are covered such as whether questions put to the child were leading, whether suggestions were made to the child that encouraged allegations, whether the child was familiar with one of the interviewers, and so on.

Stage 2: Did the child make the allegation more than once during the original interview?

An allegation of, say, abuse made by a child several different times during an interview is likely to be more reliable and more convincing than an allegation made only once, particularly if the reaction of the interviewers to the allegation is neutral each time and therefore does not reinforce the salience of the child’s comment.

Stage 3: Was the child’s understanding age-appropriate at the time of the original interview?

Using previous records (from nursery, school, or social work reports), an assessment can made of the child’s cognitive development. Opinions from parents and professionals who knew the child at the time of original interview can confirm that the child's understanding was age-appropriate, which in turn would add weight to the reliability of the interview evidence.

Stage 4: Did the child use language that was commensurate with their level of development?

Three-year-olds typically use three-year-old language, five-year-olds typically use five-year-old language, and so on. Any mismatch between the words and language structure used by the child in their original interview and the words and language structure expected of a child that age could cast doubt on the reliability of the evidence.

Stage 5: Did the child appear to make the allegation spontaneously?

Even when a child’s language during the interview is consistent with their developmental stage, the style of that language can suggest that the child might have been “coached” by one of the adults in the child's life. Key signs to look for are exact repetitions of the same phrase or sentence, or allegations made without any accompanying emotional responses.

Stage 6: Was the child’s psychological development within normal limits at that time?

While some childhood psychological problems are unlikely to impair a child’s ability to provide a reliable account of their experiences (for example, anxiety, sleep problems, fussy eating), others could have an adverse effect on the child's judgments (for example, autistic spectrum disorder, oppositional disorder). An examination of nursery or school reports will confirm this.

Stage 7: Was the child known to be reliable outwith the interview context?

Examination of school reports and, preferably, discussions with professionals who worked with the child at the time the original interview was conducted (nursery staff, teachers, for example) will confirm if the child had a history of making false accusations against, say, their peers, or if the child had a propensity for providing fanciful accounts of their experiences to their teachers and carers.

Stage 8: Is there any objective evidence that gives any credence to the original allegation?

Even in cases where a child’s allegation during interview is not fully substantiated by external evidence, there can be some unexplained information to support the child's comments. For example, an allegation of physical abuse may be unproved, yet there might be a history of hospital visits due to childhood injuries which were not entirely consistent with the parents’ explanation.

A tool to assist

Cold-case analysis of a child’s original evidence using this stage model, which sets the original interview comments in the context of the child's broader psychological development as well as in the context of the interview style itself, will lead to a more accurate assessment of the reliability of the evidence. The more “yes” answers to the stage 1-8 questions, the stronger the likelihood that the child’s original evidence is reliable. As such, this contextual model can be a useful tool for lawyers, psychologists, social workers and other professionals involved in cold case analysis where follow-up interviews are not possible.

No method of assessing the reliability of a child’s interview evidence is 100% foolproof – it is always a matter of degree. Cold-case assessment of the reliability of child evidence, however, is even more hazardous. The model proposed in this article provides a useful tool that can help a child’s voice be genuinely heard in legal proceedings.

Dr Richard Woolfson is a child psychologist, Fellow of the British Psychological Society, and is registered with the Health & Care Professions Council. He provides expert witness reports in cases involving children and families.



(1) Ceci, S and Bruck, M (1995): Jeopardy in the courtroom: A scientific analysis of children’s testimony. Washington, USA: American Psychological Association.

(2) Westcott, H, Davies, G and Bull, R (2003): Children’s testimony: A handbook of psychological research and forensic practice. UK: John Wiley & Sons.

(3) Lamb, M, Hershkovitz, I, Orbach, Y and Esplin, P (2008): Tell Me What Happened: Structured Investigative Interviews of Child Witnesses and Victims. UK: Wiley-Blackwell.

(4) Scottish Government (2011): Guidance on Joint Investigative Interviewing of Child Witnesses in Scotland. Scotland: Scottish Government.

Share this article
Add To Favorites