Assessment should include many areas of functioning (1), use multiple techniques (2) and use multiple sources (3). The integration of this information is often difficult due to the low rate of agreement between informants but this may reflect real situational variability in children’s behaviour. Informant disagreement is influenced by the type of behaviour being assessed and the measurement technique used across informants (i.e. assessment strategy).

The level of analysis should be taken into account when assessing a lack of agreement between informants as individual behaviours show less consistency across informants than broader dimensions of behaviours (e.g. diagnosis). To understand the different behavioural manifestations, it is important to understand the demands of different contexts.

Complex schemes refer to systems in which one source of information is weighed more heavily than others. Simple schemes (i.e. either/or approach) refer to systems in which information from all sources is weighed equally. Complex schemes are often based on clinical judgement rather than empirical evidence, meaning that simple schemes may be superior. However, this is only when informants are asked to provide information that they are expected to know. There may be a differential validity of various informants across behavioural domains.

The quality of information provided by different informants is influenced by the age of the child. The importance of parents and teachers as informants may decrease with age and the importance of children’s self-report may increase. Nonetheless, the reliability of parent report is still acceptable in adolescence. Marital conflict (1), parental adjustment (2), maternal alcoholism (3), marital difficulties (4), ethnicity (5), different motivation (6) and testing conditions (7) may influence the report of various informants and should be considered when interpreting discrepant information.      

There are several common cognitive strategies that can lead to errors in the problem-solving process:

1. Availability heuristic
   This refers to estimating the probability of an event based on the ease with which examples come to mind.
2. Representative heuristic
   This refers to when accessing a scheme by a given characteristic leads to the exclusion of other schemas (e.g. sadness lead the clinician to consider MDD and not other disorders).
3. Anchoring heuristic
   This refers to having predictions or decisions that are overly dependent on initial impressions and the discounting of later information.
4. Confirmatory search strategies
   This refers to using procedures that only seek to confirm initial impressions and failing to seek disconfirming evidence.

The use of these heuristics can be minimalized by using a systematic problem-solving approach. The problem orientation refers to the clinician’s overall theoretical orientation for viewing problem behaviour and defines the proper content and methods of assessment. An orientation of planned critical multiplism (i.e. clinical outcomes are brought about by multiple interacting factors) is a useful problem orientation to avoid the heuristics.

There are several steps for integrating information:

1. Document all clinical significant findings related to the child’s adjustment
   This includes reviewing all information and assessing what is significant.
2. Look for convergent findings across sources and methods
   This includes reviewing information and determining what is consistent across informants and methods.
3. Try to explain discrepancies
   This includes explaining discrepancies between sources and methods.
4. Develop a profile and hierarchy of strengths and weaknesses
   This includes developing a profile across the different domains of psychological functioning that have been assessed and includes prioritizing different areas of concern.
5. Determine critical information to place in the report
   This includes filtering the information to only include relevant information.

When prioritizing areas within a child’s profile, the degree of impairment in different settings (1), temporal sequencing of problem behaviours (2) and family history data (3), should be considered.