Search

AIMS: To examine changes in drinking behavior among United States (US) adults between March 10 and July 21, 2020, a critical period during the COVID-19 pandemic. DESIGN: Longitudinal, internet-based panel survey. SETTING: The Understanding America Study (UAS), a nationally representative panel of US adults age 18 or older. PARTICIPANTS: A total of 4298 US adults who reported alcohol use. MEASUREMENTS: Changes in number of reported drinking days from March 11, 2020 through July 21, 2020 in the overall sample and stratified by sex, age, race/ethnicity, household structure, poverty status, and census region. FINDINGS: Compared with March 11, the number of drinking days per week was significantly higher on April 1 by an average of 0.36 days (95% CI = 0.30, 0.43), on May 1 by an average of 0.55 days (95% CI = 0.47, 0.63), on June 1 by an average of 0.41 days (95% CI = 0.33, 0.49), and on July 1 by an average of 0.39 days (95% CI = 0.31, 0.48). Males, White participants, and older adults reported sustained increases in drinking days, whereas female participants and individuals living under the federal poverty line had attenuated drinking days in the latter part of the study period. CONCLUSIONS: Between March and mid-July 2020, adults in the United States reported increases in the number of drinking days, with sustained increases observed among males, White participants, and older adults.

Survey questionnaires often contain skip patterns, which let respondents skip over entire sections or a set of follow‐up questions that do not apply to them, and thus allow them to proceed through the interview faster. Survey designers might ask how filter and follow‐up questions can be presented to reduce the risk of motivated underreporting. We designed a series of experiments in which we varied the salience of the repetitive nature of filter and follow‐up questions. The results show that changes in topic removed the salience of the filtering patterns. Using slightly varied follow‐up questions and reducing the repetitiveness of the task increased endorsements to filter questions and thus successfully mitigated the effect of motivated underreporting. On the other hand, a visualization of the filtering by graying out items that no longer need to be answered reduced endorsements. Implications for questionnaire design involving filter questions are discussed.

Online-Befragungen sind zu einer ublichen und haufig praferierten Datenerhebungsmethode geworden. Die hohe Flexibilitat und Anpassbarkeit ist ein wesentlicher Vorteil dieser Technologie. Die technischen Moglichkeiten werden oft von Umfrageentwicklern verwendet, beispielsweise um die Befragten mithilfe automatischer Filter durch den Fragebogen zu fuhren. Andere Features wie beispielsweise Mausbewegungen konnen eingesetzt werden, um einzelne Befragte zu identifizieren, die besonderer Aufmerksamkeit bedurfen. Forscher aus verschiedenen Disziplinen haben insbesondere die zuruckgelegte Distanz, den Pfad der Maus und andere Bewegungsmuster analysiert, um damit Interesse, Unsicherheit und aufgetretene Schwierigkeiten beim Befragten zu messen. Die aktuelle Studie strebt die Entwicklung von Indikatoren und automatischen Prozeduren an, mit deren Hilfe Schwierigkeiten des Befragten diagnostiziert und quantifiziert werden sollen. Zu diesem Zweck wird auf vielversprechende Indikatoren aus der vorherigen Forschung und auf jungste methodologische Fortschritte aus der Psychologie zuruckgegriffen. Die psychologische Literatur schlagt vor, auf Basis von Mausbewegungen den kognitiven Zwiespalt zwischen einzelnen Antwortalternativen bzw. die Unsicherheit bei der Auswahl zu beurteilen.