Overview
The Social Communication Questionnaire (SCQ; Rutter, Bailey, & Lord, 2003) is an autism spectrum disorder screening instrument for individuals aged 4 years and up, with a mental age over 2 years. The SCQ contains 40 yes/no items, which can be completed in less than 10 minutes by a parent or other caregiver. It is based on the Autism Diagnostic Interview-Revised (ADI-R), and agreement between the two instruments is high at the total score level (Rutter, LeCouteur, & Lord, 2003). The SCQ has two forms: the Lifetime Form, which focuses on behavior throughout development; and the Current Form, which focuses on behavior during the most recent three months. The instrument yields a Total Score for comparison to defined cutoff points, which identifies individuals who require evaluation because of elevated likelihood of ASD.
Summary
Age: 4 years to 99 years
Time to Administer: Less than 10 minutes
Method of Administration: 40-item parent questionnaire; Current and Lifetime forms
Yields total score that is compared with cutoff
Age range is over 4 years, with a mental age over 2 years
Subscales: Overall Composite Score: Total Score
Subscale Scores: Reciprocal Social Interaction; Language and Communication; Stereotyped Patterns of Behavior
Screening/Diagnosis: S
Autism Related Research
Rosenberg et al. (2018)
Age Range: 48-63 months
Sample Size: 2,557
Topics Addressed:
Effect of demographic factors on SCQ
Outcome:Rosenberg et al. (2018)
Child sex, household income, race, and maternal education were associated with SCQ scores. A cut-point of 11 was optimal for most of the demographic factors. This demonstrated higher sensitivity for ASD (0.88) among children seen for developmental concerns than among children in the general population (0.78). Specificity was lower for children served in clinical or educational settings (0.74) than for children in the general population (0.95), suggesting that the target source population in which the screener will be applied is important to consider. It is especially important to note that lower maternal education and minority status may be more prone to inaccurate screenings (i.e., false positive results).
Conclusion: although generally shown to be a good choice across demographic subgroups in this study, the SCQ cut-point of 11 was associated with poor specificity for families with lower income and education.
Barnard-Brak et al. (2016)
Age Range: 1.58-25.92 years
Sample Size: 344
Topics Addressed:
Comparison of SCQ sensitivity vs. specificity
Outcome:Barnard-Brak et al. (2016)
Sensitivity and specificity of the SCQ at the recommended cutoff of 15 are not in agreement with results obtained by test creators. Results suggested a cutoff score of 10 may be more appropriate. AUC values improved when participants were 7.5 years old and older individuals had better AUC values, suggesting a lack of age neutrality of the SCQ in screening for ASD (i.e., 4-40 is not likely an appropriate age range for the SCQ). ROC curve analyses suggested use of SCQ for those under 4 was significantly worse at accounting for ASD than a sample that had only those 4+.
Conclusions: SCQ sensitivity at cutoff 15 is not as high as originally thought; SCQ is not age-neutral (i.e., modified scores or items may be needed for different ages).
Brooks & Benson (2013)
Age Range: 18–40 years
Sample Size: 69
Topics Addressed:
Utility of SCQ in adults with ID
Outcome:Brooks & Benson (2013)
At test authors’ recommended cutoff score of 15, sensitivity of 0.71 and specificity of 0.77 in sample. When sensitivity prioritized over specificity, the optimal cutoff score was 12, yielding sensitivity = 0.86 and specificity = 0.60. Internal consistencies (Cronbach alpha) of total (0.87), social interaction (0.83), and restricted repetitive behavior (0.81) were good; communication subscale internal consistency poor (alpha 0.48).
Conclusion: SCQ is a promising screener for adults with ID but lower cutoff score than 15 is recommended for use with this population.
Oosterling, Rommelse, deJonge, Van der Gag, Swinkels, Roos, Visser, & Buitelaar (2010)
Age Range: 20–40 months
Sample Size: 208
Topics Addressed:
Utility of SCQ as with different cut-off scores, for toddlers
Outcome:Oosterling, Rommelse, deJonge, Van der Gag, Swinkels, Roos, Visser, & Buitelaar (2010)
The SCQ was not satisfactory at accurately identifying high-risk toddlers and resulted in a number of false positives for toddlers with symptoms of ASD.
Cutoff > 11 Sensitivity = .92; Specificity = .26 Cutoff > 12 Sensitivity = .88; Specificity = .35
Cutoff > 15 Sensitivity = .76; Specificity = .58
Cutoff > 22 Sensitivity = .29; Specificity = .86
Conclusion: SCQ is not optimal as a second-level screener in toddlers at high risk of ASD. When used in diagnostic evaluation, the SCQ does not appear to add anything over and above the ADOS. However, when used in combination with the ADOS and for distinguishing between ASD/non-ASD, it is a better and cheaper alternative than the expensive and time-consuming combination of ADI-R and ADOS, which has shown superior diagnostic validity in older children.
Snow & Lecavalier (2008)
Age Range: 18-70 months
Sample Size: 82
Topics Addressed:
Comparison of M-CHAT and SCQ
Outcome:Snow & Lecavalier (2008)
A cutoff of 13 may be optimal when the SCQ is used with younger children. Sensitivity was higher than specificity. Negative associations found between level of functioning and total scores for the PDD group, suggesting that level of functioning should be considered in interpretation of screening scores.
Chandler, et al. (2007)
Age Range: 9.8–14.5 years
Sample Size: 913
Topics Addressed:
Examine SCQ properties in ASD cohort and in general population
Outcome:Chandler, et al. (2007)
SCQ showed strong discrimination between ASD and non-ASD cases (sensitivity 0.88, specificity 0.72) and between autism and non-autism cases (sensitivity 0.90, specificity 0.86). Findings not affected by child IQ or parental education.
Conclusion: the SCQ is useful as a first-level screen for ASD in school-age children.
Wiggins, Bakeman, Adamson, & Robins (2007)
Age Range: 17-45 months
Sample Size: 37
Topics Addressed:
Discriminative validity of ASD in very young children
Outcome:Wiggins, Bakeman, Adamson, & Robins (2007)
Discriminative validity of ASD in very young children
t cutoff 15: sensitivity 0.47, specificity 0.89); at cutoff 11: sensitivity 0.89, specificity 0.89.
Conclusion: Consistent with other studies, the recommended cutoff score of 15 with very young children results in low sensitivity, but this is increased (without compromise to specificity) at a cut-off score of 11.