Method effects as a source of construct invalidity were evaluated for parent, teacher, and child ratings of problem behaviors among children and adolescents with serious emotional disturbances. A multitrait-multimethod (MTMM) matrix consisting of 4 types of problems (i.e., internalizing, thought, attention, and externalizing), rated by 3 sources, was analyzed using confirmatory factor analysis. Subjects were 489 youngsters, ages 8 to 18 years (M = 13.86), who were classified as having serious emotional disturbances. Results indicated large method effects for each rater. These effects may represent either measurement artifact, true situational differences, or a combination of both. Modeling method effects is necessary to obtain unbiased estimates of the relationships between children's multiple problems and other variables
No takes yet. Share an insight, caveat, or question.
Greenbaum et al. (1994) studied this question.