相关性(法律)
认识论
工作(物理)
社会学
实证经济学
贝叶斯网络
极限(数学)
消费者行为
贝叶斯概率
规则网络
营销
计算机科学
构造(python库)
经济
点(几何)
心理学
作者
John Lynch,Stijn M. J. van Osselaer,Patricia Torres
摘要
Abstract Consumer researchers often prize relevance but overlook how narrow assumptions about good theory limit it. Much of the work focuses on construct-to-construct theorizing, which involves introducing new constructs or new links among them. Far less valued is phenomenon-to-construct theorizing, which begins with real-world patterns and seeks to explain them by identifying the underlying active ingredient constructs. Examples include why GMO labels reduce demand, or why drip pricing leads people to choose higher-cost options. Our survey of authors in four leading journals shows that most believe only construct-to-construct work counts as theory. We argue this view is too narrow. Drawing on a Bayesian framework for updating beliefs within a theoretical network, we show that phenomenon-to-construct theorizing follows the same logic of scientific inference. Both approaches rely on established links in the nomological network to draw stronger conclusions about the focal link of interest. Using well-supported construct-to-construct mechanisms to explain real-world phenomena is therefore a strength, not a weakness. We clarify how phenomenon-to-construct theorizing differs from both “mere application” and “empirics-first” research. Embracing this form of theorizing can broaden the reach of consumer research by connecting abstract ideas to meaningful, actionable phenomena that matter to scholars, practitioners, and policymakers.
科研通智能强力驱动
Strongly Powered by AbleSci AI