IMPROVING SELF-REPORTS OF DRUG USE: A COGNITIVE SET MOD

改进药物使用自我报告：认知模式

• 批准号: 2733580
• 负责人: GEORG E MATT
• 金额: $ 7.6万
• 依托单位: SAN DIEGO STATE UNIVERSITY
• 依托单位国家: 美国
• 财政年份: 1997
• 资助国家: 美国
• 起止时间: 1997-07-01 至 2000-06-30
• 项目状态: 已结题
• 来源: https://reporter.nih.gov/project-details/2733580
• 关键词: alcoholism /alcohol abuse; behavioral /social science research tag; bias; clinical research; data collection methodology /evaluation; data management; drug abuse; health surveys; human data; human subject; interview; marijuana abuse; mathematical model; model design /development; psychometrics; self concept; statistics /biometry; tobacco abuse

Retrospective self-reports play a central role in basic and applied research on drug use. The proposed study introduces and tests a fuzzy-set model to improve retrospective self-reports of the frequeny, onset, ad amount of drug use (i.e., tobacco, alcohol, marijuana). Different from traditional modelsof self-reported data, this fuzzy set model assumes that self-reprots of high-frequency behaviors cannot be properly represented by a single point estimate. Instead, self-reports ae better measured, interpreted, and analyzed as membership distributions of possible, partially valid estimates. Six-hundred aldult residents of the San Diego metropolitan area will be recruited to participate in three distinct experiments to probe the validity of this model. Experiment 1 is designed to quantify and distinguish fuzziness in self-reports from other sources of error and uncertainty. Experiment 2 compares the susceptibility of traditional point and fuzzy set estimates to socilly desirable response sets. Finally, Experiment 3 is designed to compare the criterion-related validity of traditional point estimates and fuzzy estimates of health and social behaviros in different gender and ethnic groups. In summary, this study investigates an innovative approach to representing self-reports to questions regarding the frequency, duration, or amount of drug use. The proposed study builds on our previous research and advances the state-of-the-art by combining cognitive models of retrospective self-report and psychometric theory. The proposed fuzzy-set model promises to yield measures of behaviors that more completely reflect what can be recalled about past drug use, it provides a framework for studying bias, precision, and uncertainty of self-reprots. Findings from this study could have significant implications for improving the collection, analysis, and interpretation of self-reports in large-scale national surveys (e.g., national helath Interview Survey, national Householf Survey of Drug Abuse), in epidemiologic and behaviroal research, and in everyday life.

回顾性自我报告在基础和应用方面发挥着核心作用 药物使用研究。 拟议的研究介绍并测试了 模糊集模型来改进回顾性自我报告 吸毒的频率、发作时间和数量（即烟草、酒精、 大麻）。 与传统的自我报告数据模型不同， 这个模糊集模型假设高频的自我报告 行为不能通过单点估计来正确表示。 相反，自我报告可以更好地衡量、解释和分析 作为可能的、部分有效的估计的成员分布。 圣地亚哥都会区的 600 名成年居民将 被招募参加三个不同的实验来探索 该模型的有效性。 实验 1 旨在量化和 将自我报告中的模糊性与其他错误来源区分开来 不确定。 实验2比较传统方法的敏感性 点和模糊集估计到社会期望的响应集。 最后，实验3旨在比较与标准相关的 传统点估计和健康模糊估计的有效性 以及不同性别和种族群体的社会行为。 在 总之，本研究探讨了一种创新方法 代表自我报告有关频率的问题， 持续时间或药物使用量。 拟议的研究建立在我们的基础上 以前的研究并结合先进的技术 回顾性自我报告和心理测量的认知模型 理论。 所提出的模糊集模型有望产生测量结果 更完整地反映可以回忆的行为 关于过去的吸毒情况，它提供了一个研究偏见的框架， 自我报告的准确性和不确定性。 这项研究的结果 可能对改善馆藏产生重大影响， 对大型全国性自我报告的分析和解释 调查（例如，全国健康访谈调查、全国家庭 药物滥用调查），在流行病学和行为研究中， 以及日常生活中。