Analyzing the FACT-G Questionnaire for Cancer Patients
Author Information
Author(s): Adam B. Smith, Penny Wright, Peter J. Selby, Galina Velikova
Primary Institution: Cancer Research UK Clinical Centre, St. James's University Hospital, Leeds, UK
Hypothesis
The study aims to investigate the dimensionality and measurement properties of the FACT-G using Rasch Models and Factor analysis.
Conclusion
The study found that while most FACT-G scales were unidimensional, the Social & Family Well-being scale showed multidimensionality and contained misfitting items.
Supporting Evidence
- The factor analysis revealed a four-factor structure corresponding to the FACT-G subscales.
- Internal consistency for the scales was very good with Cronbach's alpha ranging from 0.72 to 0.85.
- Rasch analysis showed that most subscales were unidimensional except for the Social & Family Well-being scale.
Takeaway
Researchers looked at a questionnaire used for cancer patients to see if it accurately measures their quality of life, and found some parts of it need more work.
Methodology
The study used factor analysis and Rasch analysis on the FACT-G completed by cancer patients.
Potential Biases
No significant item invariance or bias was found for the items from the two samples.
Limitations
The sample was not fully representative of all cancer types, and demographic details were missing for some patients.
Participant Demographics
461 patients with 323 females (average age 55.7 years) and 138 males (average age 60.8 years).
Statistical Information
P-Value
0.006
Statistical Significance
p<0.05
Digital Object Identifier (DOI)
Want to read the original?
Access the complete publication on the publisher's website