Improving the Marriage of Modeling and Theory for Accurate Forecasts of Outcomes

This volume in the series has big objectives: describe the bad science practices now in use in most studies in business-to-business marketing strategy and describe a true paradigm shift to good science practices by replacing the variable-based linear-symmetric null hypothesis testing (NHST) approach in theory construction and testing--with case-based asymmetric models with somewhat precise outcome testing (SPOT). Whether the question refers to success or failure, wise executives ask, how did we get here? What's in store for the next decade? Unfortunately, the majority of scholarly articles examining the causes of success and failure offers scant useful information that is accurate in forecasting success or failure strategy outcomes. The majority of studies on strategy performance outcomes focus on variable relationships and testing for the directionality (positive or negative relationships) and effect size of relationships--using multiple regression analysis and structural equation modeling (MRA/SEM) using null hypothesis statistical testing (NHST). Research on the value of NHST indicates that such studies are worse than useless: such research does not focus on case-based outcomes and achieving a statistically significant relationship greatly depends on the sample size of firms in the studies. Researchers using NHST are answering the wrong questions in examining the net effects of independent variables on dependent variable of interest (e.g., net earnings per revenue). Here are the right questions to ask. What configurations of antecedent conditions combine to generate positive outcomes for our firm and similar firms? What configurations of antecedent conditions combine to generate negative outcomes for firms in our industry? Sound reasoning and empirical evidence supports the wisdom of business executives ignoring the scholarly empirical literature on forecasting successful and unsuccessful management strategies using the NHST of the size and directionality of relationships. Good science practice relies on the complexity theory tenets covered in the chapters in this volume. Good science practice includes matching case-focused theory with case-focused data analytic tools and using somewhat precise outcome tests (SPOT) of asymmetric models. Good science practice achieves requisite variety necessary for deep explanation, description, and accurate prediction. The fear of submission rejection is another reason for rejecting case-based asymmetric modeling and SPOT. Overcome such fear by learning to apply complexity theory tenets, constructing separate case-based, mid-range, models of successful versus unsuccessful outcomes, and testing for accuracy via SPOT. This volume provides tools necessary for you to accomplish this task.