Related Experiment Videos
Infant temperament measured by multiple observations and mother report
R Seifer1, A J Sameroff, L C Barrett
1E. P. Bradley Hospital, Brown University, Providence, RI 02915.
Child Development
|October 1, 1994
Summary
Maternal reports and direct observations of infant temperament showed low agreement, despite reliable individual ratings. Aggregated data improved reliability but not correspondence, highlighting the need for direct infant behavior observation in temperament studies.
Area of Science:
- Developmental Psychology
- Child Behavior Research
Background:
- Temperament assessment in infants often relies on maternal reports.
- Direct observation provides an alternative method for evaluating infant behavior.
Purpose of the Study:
- To compare maternal reports with direct observations of infant temperament.
- To assess the reliability and correspondence between these two assessment methods.
Main Methods:
- Fifty infants were observed weekly for 8 weeks in their homes by trained observers.
- Mothers also rated infant behavior using questionnaires aligned with observation scoring.
- Aggregated observations and maternal reports were statistically analyzed.
Main Results:
- Week-to-week correlations between observers and mothers were modest.
- Aggregated observations over 8 weeks showed high reliability for both mothers and observers.
- Little correspondence was found between direct observations and maternal reports, even with aggregated data.
Conclusions:
- While aggregated maternal reports and observations can be reliable, they do not correspond well.
- Direct observation of infant behavior is crucial for accurate temperament assessment.
- Findings challenge the interpretation of existing literature relying solely on maternal reports.