Reliability of Internet vs. Telephone Questionnaires for Smokers
Author Information
Author(s): Amanda L Graham, George D Papandonatos
Primary Institution: Georgetown University Medical Center / Lombardi Comprehensive Cancer Center
Hypothesis
Do Internet-administered questionnaires have comparable reliability to telephone-administered questionnaires among diverse smokers?
Conclusion
The psychometric properties of most Internet-administered measures are consistent across different racial/ethnic and income groups.
Supporting Evidence
- Test-retest reliability was satisfactory to excellent for most measures across all strata.
- 9 of 12 continuous variables had intraclass correlation coefficients ≥ 0.70.
- 10 of 18 binary variables had kappa coefficients ≥ 0.70.
Takeaway
This study shows that asking smokers questions online works just as well as asking them over the phone, even for people from different backgrounds.
Methodology
Participants were enrolled in a randomized trial and completed both telephone and Internet surveys to assess the reliability of various measures.
Potential Biases
Potential bias due to the self-reported nature of race and income data.
Limitations
The study's findings may be influenced by learning effects and the short time frame between measurements.
Participant Demographics
52.4% non-Hispanic White, 22.9% Black, 11.6% Hispanic, 7.8% Asian, 4.4% American Indian/Alaska Native, 1% Native Hawaiian/Other Pacific Islander; 49.4% had an income of $40,000 or less.
Statistical Information
Statistical Significance
p<0.05
Digital Object Identifier (DOI)
Want to read the original?
Access the complete publication on the publisher's website