Related Experiment Videos
Item reliability for the Griffiths scales of mental development
Child: Care, Health and Development
|May 1, 1982
Summary
This study assessed the reliability of the Griffiths Scales, finding that most items were reliably scored. However, some items require revision to improve the accuracy of developmental assessments for young children.
Area of Science:
- Developmental Psychology
- Psychometrics
- Child Assessment
Background:
- The Griffiths Scales are widely used for assessing child development.
- Preliminary revisions necessitate a thorough evaluation of the original test items' reliability.
Purpose of the Study:
- To evaluate inter-observer agreement and scoring reliability for the 498 items of the original Griffiths Scales.
- Identify specific items and subscales requiring revision for improved accuracy.
Main Methods:
- Analysis of videotaped test administrations to children aged 2 months to 7 years.
- Utilized large scoring panels to assess inter-observer agreement.
- Examined reliability across different subscales and age ranges.
Main Results:
- 57 items were inadequately administered or scored.
- 112 items showed unreliability in one-third of administrations; 23 items were unreliable in two-thirds.
- 88% of administrations for sufficiently scored items were reliable.
- Inter-observer agreement varied significantly with child's age.
Conclusions:
- While most items demonstrated acceptable reliability, a significant number require revision.
- Subscale performance varied, with no clear differences between locomotor, personal-social, hearing/speech, and hand-eye coordination scales.
- Age-specific weaknesses in subscales were identified, informing targeted revisions.