The reliability of self-reported drinking in adolescence

Area of Science:

• Public Health
• Adolescent Health
• Survey Methodology

Background:

• Accurate measurement of adolescent alcohol consumption is crucial for public health research and interventions.
• Previous studies have raised concerns about the reliability and validity of self-reported alcohol intake in adolescents.
• Understanding perceived drunkenness is also important for assessing alcohol-related risks.

Purpose of the Study:

• To assess the reliability and validity of adolescents' self-reported alcohol consumption amounts.
• To compare the effectiveness of open-ended versus closed-ended survey questions in capturing drinking data.
• To examine the relationship between self-reported alcohol intake and perceived drunkenness.

Main Methods:

• Utilized data from two cross-sectional, school-based surveys (European School Survey Project on Alcohol and Other Drugs - ESPAD) in Finland (1995, 1999).
• Included representative cluster samples of 15-year-old respondents with high response rates (94% and 90%).
• Analyzed data using both open-ended and closed-category questions on recent alcohol consumption and perceived drunkenness.

Main Results:

• Closed-ended questions generally yielded comparable or slightly higher estimates of alcohol consumption compared to open-ended questions.
• Individual-level correlations between different question formats were consistently moderate to high (0.65-0.69).
• Self-reported drunkenness showed a logical association with the reported amounts of alcohol consumed.

Conclusions:

• Adolescent self-reports of alcohol consumption amounts appear reasonably reliable and valid at both population and individual levels.
• A set of closed-ended questions may be more effective than a single open-ended question for quantifying adolescent drinking.
• Findings support the use of well-designed survey instruments for studying adolescent alcohol use.