Interactions of factorial components and testing occasions in the determination of scores on the a-state scale of the state-trait anxiety inventory

The factor structure of the A‐state scale of the State‐Trait Anxiety Inventory, interpreted as a state anxiety factor and a reported state of pleasantness factor, suggests the possibility of interactions between testing occasions and the two orthogonal factors. Ninety primigravid women completed the A‐State scale on three occasions: during the third trimester of pregnancy, in labour, and during the postpanum. Use of the full scale score, summed across both factors, would have led to misinterpretation of the occasions effect by indicating that labour provoked highest A‐State. Including factorial structure as a factor in the repeated measures analysis of variance revealed interactions between the structure and testing occasions. The interaction showed that the apparent increase in anxiety during labour was really a decrease in reported state of pleasantness. The use of part scores based on the factor structure of the A‐State scale is recommended.