Content validity of self-report measurement instruments: an illustration from the development of the Brain Tumor Module of the M.D. Anderson Symptom Inventory.

PURPOSE/OBJECTIVES: To illustrate one technique for establishing content validity of measurements using the initial development and testing of the M.D. Anderson Symptom Inventory Brain Tumor Module.

DATA SOURCES

Published articles, book chapters, and subjective judgments of experts.

DATA SYNTHESIS

Content validity is the essential first step in the development of items to be included in a measurement instrument. Content validity is a criterion-referenced process that is judged by how well each item in a newly developed instrument reflects its respective objective or content domain. The stages in addressing content validity include a developmental stage and a judgment-quantification stage. Steps involved in the developmental stage include domain identification, item generation, and instrument formation. The judgment-quantification stage is when experts review the items and either report validity of the items subjectively or with an empirically referenced method, such as calculation of the content validity index. The content validity of a set of questions designed to measure symptoms in a population of patients with primary brain tumors was ascertained by using the calculation of the content validity index.

CONCLUSIONS

The final version of the M.D. Anderson Symptom Inventory Brain Tumor Module consists of the 13 core items and 18 additional items designated as valid by a panel of experts. The instrument will be administered to a group of patients to determine construct validity and reliability of the items.

IMPLICATIONS FOR NURSING

Self-report instruments are used to measure various health outcomes in oncology. Oncology nurses are in a key position to develop such instruments to be used in clinical care and research of symptoms associated with cancer. Understanding the process of content validation is an essential first step in developing new instruments.