Research questions primarily address three fundamental problems. Descriptive research addresses questions of what, who, when, and where. Causal research deals with the questions of why and how. Predictive research involves questions of could, should, or would an event occur in the future.
An argument was presented for how the research question is connected to design.
Research design is (1) a logical plan for selecting and arranging the evidence that answers the research question(s); (2) a framework for specifying the relationships among a study’s variables; (3) a blueprint outlining each procedure from the hypotheses to the analysis of data; (4) a way of configuring the study to its paradigm (e.g., achieving findings that are numeric, systematic, generalizable, and replicable or obtaining non-numerical observations that reveal and detail the experience and constructed meanings of individuals, groups, situations, or contexts); and (5) a means for determining how to allocate limited resources of time and budget.
The nonexperimental quantitative design was defined as a “systematic empirical inquiry in which the scientist does not have direct control of independent variables because their manifestations have already occurred or because they are inherently not capable of being manipulated. Inferences about relations among variables are made without direct intervention, from concomitant variations of independent and dependent variables.”46
A case was made for business problems posing perplexing academic questions and managerial dilemmas where nonexperimental designs appear as good alternatives especially when conditions are not amenable to experimentation.
The logical structure of the study dominates logistics in organizing evidence to answer the research question. Design should not be confused with a research method or a technique for gathering data.
There has been little agreement in business, social or behavioral science, or education as to an acceptable design typology, but one is needed because of numerous theoretical and pedagogical advantages for educators, researchers, managers, and students.
The author reviewed 30 current research methods textbooks in business and other fields and found that confusion has continued to exist for almost 3 decades, resulting from the practice of treating designs, methods, and techniques for data collection alike rather than isolating the design as the logical structure of the study. Only the descriptive design met the theoretical tests for typological usefulness.
A 3 × 3 classification of nonexperimental quantitative designs was proposed that resulted from answering fundamental questions about the objectives of the study (producing descriptive, explanatory, and predictive designs) and the time dimension for collecting data (longitudinal,
cross-sectional, and retrospective). Definitions and studies to illustrate the resulting nine cells of the matrix followed. The studies were linked to functional areas of business along with how the typology met the criteria for successful identification of design.