Accuracy of Self-Reported Drinking in Young Adults
Accuracy of Self-Reported Drinking in Young Adults
Aims: As a formative step towards determining the accuracy of self-reported drinking levels commonly used for estimating population alcohol use, the validity of a 'last occasion' self-reporting approach is tested with corresponding field observations of participants' drinking quantity. This study is the first known attempt to validate the accuracy of self-reported alcohol consumption using data from a natural setting.
Methods: A total of 81 young adults (aged 18–25 years) were purposively selected in Perth, Western Australia. Participants were asked to report the number of alcoholic drinks consumed at nightlife venues 1–2 days after being observed by peer-based researchers on 239 occasions. Complete observation data and self-report estimates were available for 129 sessions, which were fitted with multi-level models assessing the relationship between observed and reported consumption.
Results: Participants accurately estimated their consumption when engaging in light to moderate drinking (eight or fewer drinks in a single session), with no significant difference between the mean reported consumption and the mean observed consumption. In contrast, participants underestimated their own consumption by increasing amounts when engaging in heavy drinking of more than eight drinks.
Conclusion: It is suggested that recent recall methods in self-report surveys are potentially reasonably accurate measures of actual drinking levels for light to moderate drinkers, but that underestimating of alcohol consumption increases with heavy consumption. Some of the possible reasons for underestimation of heavy drinking are discussed, with both cognitive and socio-cultural factors considered.
National surveys of alcohol consumption assume that participants are able to make an accurate estimate of their consumption levels, either as exact quantities on particular occasions or as accurate means (or modes) for an extended period of time. However, there has been little research undertaken to determine the validity of self-reporting measures through comparing participant estimates to independently observed data. This oversight is striking, given that much time, money and effort have been spent on national surveys in which the validity of measures employed are not sufficiently established. Indeed, population surveys of drinking are acknowledged to vastly underestimate the actual amount of alcohol consumed in the target population, with coverage rates typically between 50 and 75% (e.g. Stockwell et al., 2004). Studies that seek to compare survey reported drinking levels with diary-based reporting or daily telephone reporting (Hilton, 1989; Perrine et al., 1995; Leigh, 2000) are helpful, but they are only comparing methods of self-reporting—in other words, they employ self-reports to test self-reporting, and so are only of value to the extent that the more immediate reporting approach is accurate.
Studies of the accuracy of self-report measures using independent verification techniques are virtually non-existent in the alcohol research field. There has only been one study to our knowledge that has sought to verify self-reported estimates with independently recorded consumption levels taken at the time of drinking. Using data from a Finnish study undertaken in the 1950s involving a simulated drinking environment (Bruun, 1959), Poikolainen (1985) compared the estimates of 58 males (25–54 years) to the number of drinks observed by researchers behind a one-way mirror. Estimates were made 1–2 days after the session. Participants estimated, on average, 10.5 drinks, in contrast to the average of 11.4 drinks observed by researchers (ρ = 0.65, P < 0.001). They found that heavier drinkers (12 or more drinks) underestimated by 12%, in contrast to light drinkers (less than 12 drinks) who underestimated by 4% (t-test, P = 0.007).
The findings of the Finnish study are important in terms of assessing the value of recent recall approaches in population surveys. However, the artificial setting used in the Finnish study raises questions about its applicability to real-life settings. As part of a broader observational study of a purposive sample of young adults (aged 18–25 years) in Perth, Western Australia (plus some additional locations in New South Wales, Victoria and the UK), the opportunity arose to cross-validate observed drink levels with self-reported estimates taken 1–2 days after the recorded sessions. The estimates were based on drinking undertaken in natural settings (hotels, clubs, houses and outdoor events) rather than in a controlled setting, and therefore represents the first reported 'live' validation study of self-reporting measures.
Abstract and Introduction
Abstract
Aims: As a formative step towards determining the accuracy of self-reported drinking levels commonly used for estimating population alcohol use, the validity of a 'last occasion' self-reporting approach is tested with corresponding field observations of participants' drinking quantity. This study is the first known attempt to validate the accuracy of self-reported alcohol consumption using data from a natural setting.
Methods: A total of 81 young adults (aged 18–25 years) were purposively selected in Perth, Western Australia. Participants were asked to report the number of alcoholic drinks consumed at nightlife venues 1–2 days after being observed by peer-based researchers on 239 occasions. Complete observation data and self-report estimates were available for 129 sessions, which were fitted with multi-level models assessing the relationship between observed and reported consumption.
Results: Participants accurately estimated their consumption when engaging in light to moderate drinking (eight or fewer drinks in a single session), with no significant difference between the mean reported consumption and the mean observed consumption. In contrast, participants underestimated their own consumption by increasing amounts when engaging in heavy drinking of more than eight drinks.
Conclusion: It is suggested that recent recall methods in self-report surveys are potentially reasonably accurate measures of actual drinking levels for light to moderate drinkers, but that underestimating of alcohol consumption increases with heavy consumption. Some of the possible reasons for underestimation of heavy drinking are discussed, with both cognitive and socio-cultural factors considered.
Introduction
National surveys of alcohol consumption assume that participants are able to make an accurate estimate of their consumption levels, either as exact quantities on particular occasions or as accurate means (or modes) for an extended period of time. However, there has been little research undertaken to determine the validity of self-reporting measures through comparing participant estimates to independently observed data. This oversight is striking, given that much time, money and effort have been spent on national surveys in which the validity of measures employed are not sufficiently established. Indeed, population surveys of drinking are acknowledged to vastly underestimate the actual amount of alcohol consumed in the target population, with coverage rates typically between 50 and 75% (e.g. Stockwell et al., 2004). Studies that seek to compare survey reported drinking levels with diary-based reporting or daily telephone reporting (Hilton, 1989; Perrine et al., 1995; Leigh, 2000) are helpful, but they are only comparing methods of self-reporting—in other words, they employ self-reports to test self-reporting, and so are only of value to the extent that the more immediate reporting approach is accurate.
Studies of the accuracy of self-report measures using independent verification techniques are virtually non-existent in the alcohol research field. There has only been one study to our knowledge that has sought to verify self-reported estimates with independently recorded consumption levels taken at the time of drinking. Using data from a Finnish study undertaken in the 1950s involving a simulated drinking environment (Bruun, 1959), Poikolainen (1985) compared the estimates of 58 males (25–54 years) to the number of drinks observed by researchers behind a one-way mirror. Estimates were made 1–2 days after the session. Participants estimated, on average, 10.5 drinks, in contrast to the average of 11.4 drinks observed by researchers (ρ = 0.65, P < 0.001). They found that heavier drinkers (12 or more drinks) underestimated by 12%, in contrast to light drinkers (less than 12 drinks) who underestimated by 4% (t-test, P = 0.007).
The findings of the Finnish study are important in terms of assessing the value of recent recall approaches in population surveys. However, the artificial setting used in the Finnish study raises questions about its applicability to real-life settings. As part of a broader observational study of a purposive sample of young adults (aged 18–25 years) in Perth, Western Australia (plus some additional locations in New South Wales, Victoria and the UK), the opportunity arose to cross-validate observed drink levels with self-reported estimates taken 1–2 days after the recorded sessions. The estimates were based on drinking undertaken in natural settings (hotels, clubs, houses and outdoor events) rather than in a controlled setting, and therefore represents the first reported 'live' validation study of self-reporting measures.
Source...