When parents and 11-17 year olds are interviewed with the DAWBA, it is usually possible to shorten the interview by omitting some questions. Screening questions are used to predict when children are so unlikely to have a particular diagnosis that it is not worth asking about the relevant symptoms. Two sorts of screening questions are used:
| 1) | Start-of-section screening questions. After a brief introduction to the sorts of problems covered by the section, most sections begin with a single screening question. For example, the section on Activity and attention in the parent interview begins with: "Allowing for his/her age, do you think that [name of child] definitely has some problems with overactivity or poor concentration?". If the answer is "yes", then the parent is asked in detail about symptoms of hyperactivity, impulsiveness and inattention. |
| 2) | SDQ questions. Before they are given the DAWBA interview, parents and 11-17 year olds are asked to fill in the Strengths and Difficulties Questionnaire (SDQ). The SDQ includes five questions on emotional symptoms, five on conduct problems and five on hyperactivity-inattention, generating scores on each of these dimensions. DAWBA sections are not skipped when the score on the relevant SDQ scale is above the 80th centile. For example, when children score above the 80th centile on the SDQ hyperactivity-inattention scale, the subsequent DAWBA interview automatically asks all the questions about hyperactivity, impulsiveness and inattention. |
Most sections of the DAWBA can only be skipped if the start-of-section screening questions are negative and the relevant SDQ score is close to average (< 80th centile). This dual requirement is designed to prevent "respondent fatigue" leaading to too many sections being skipped. It is well recognised that respondents who get bored part of the way through an interview often switch to saying "No" to start-of-section screening questions to get the interview over with more rapidly. By contrast, the SDQ questions are asked at the start of the assessment (hopefully before boredom has set in), so it is much less likely that respondents will deliberately under-report problems at this stage.
The initial validation of the DAWBA examined the effectiveness of the skip rules. Overall, the skip rules functioned well, speeding up the interview considerably and only rarely leading to the omission of sections that would have been positive (Goodman, 2000).