This paper aims to introduce the Schmid–Leiman solution (SLS) as a useful tool to interpret the results of higher-order factor analyses in marketing research irrespective of the type of higher-order factor structure used (formative or reflective).
Two studies, one with retail shoppers in India and another with undergraduate students in Hong Kong, are used to compare different types of higher-order factor structures to test the utility of SLS.
The authors show that whether a reflective or a formative model is used to operationalize a higher-order construct, using SLS as an additional analysis gives useful insights into the factor structure at different levels and helps isolate their unique contributions to the explained variance.
The authors test higher-order models for store environment and consumer impulsiveness with data from retail shoppers and undergraduate students in two Asian countries, which may restrict the generalizability of the study findings. Future research may try to replicate our findings with other higher-order constructs and consumers in other countries.
The authors offer a checklist that can be used by future researchers to evaluate alternate higher-order factor structures and choose the appropriate one for their research context.
The authors show that using SLS is especially useful when there is a lack of clarity on the nature of relationships between the factors at different levels or about the independent contribution of the factors at different levels, in a higher-order factor structure.
