Research Question 1 MISSING MID-LEVEL MODEL GRANULARITY: For encompassing broad phenomena like “culture”, must we either drown in catalogs of concepts or be stifled by 4 or 5 extremely general variables that though they may constitute “a theory” are practically useless due to their breadth and generality? Research Question 2 TOOLS FOR MID-LEVEL MODEL GRANULAITY: Are there intellectual tools that allow some mid level of granuality in models -- more than overally general 4 to 5 variables per model, less than catalogs of dozens of variables per model? Research Question 3 WAY TO CAUSALLY ORDER THE MANY CULTURE-RELATED VARIABLES: For encompassing broad phenomena like “culture”, must we publish little causal snippets two or three causal linkages in length or can we publish causal links much longer without confusion and imprecision? How do we do so? RESEARCH IDEAL -- We need models of vast phenomena like “culture” with tunable levels of model granularity, maintained in the same long causal chains--so we can tune for few variables in the chain or a great many more articulate/specific ones, as needed. TWO WEAKNESSES ADDRESSED -- This article addresses two weaknesses in current research publishings on culture: one, a confusing number of culture components and dimensions with no clear consensed on causal ordering among them; two, the vastness of culture handled by highly abstract sets of four or five factors that, however appealing to academics, utterly frustrate practitioners who “wonder how to be communal not individual now that I am here in this Japanese police station”. TEN COMPONENT MODEL -- It offers ten common components in research publishing on culture in a clearly defined (not necessarily correct -- that has to be gradually demonstrated) causal order, to handle the first weakness. It offers large (16 and 64 factor) well ordered categorical models, of each of the ten components in that causal model, to move factors towards concreteness enough, to be of use to practitioners, while, because of being well ordered, being easy enough for researchers to handle, with the same ease and accuracy, as they now handle models having only five to ten factors. NEW INTERFACES FOR RESEARCH -- This article presents two new interfaces for culture study -- all the variety of culture components in a clear causal model as one interface, and a unique regularized fractal concept model interface that allows 64 factors to be handled as easily and accurately as people now handle five to ten factors in models. Method 1 REPLACE HOFSTEDE-BOND-HAMDEN-TURNER GENERATION 8 TO 10 DIMENSIONS OF CULTURES: Models of culture having few dimensions make theorizing easy but are practically worthless, each dimension being far too general for guidance in actual case situations Method 2 USE GREENE’S PRIOR 10 COMPONENT MODEL OF HUNDREDS OF CULTURE-RELATED VARIABLES: Culture purports to be everywhere in everything human encompassing all in civilization so models of a dozen or two variables of it must be uselessly vague, general, and un-insightful. Method 3 RECOGNIZE IDENTICAL “ICEBERG EFFECT” IN CULTURE, HIGH PERFORMANCE, AND SELVES Method 4 FRACTAL ORDERING OF VARIABLES WITHIN VARIABLE TYPES: Order variables having the same type fractally on multiple levels with ordering copied from top to lower levels by analogy. COMPANION ARTICLE ON CULTURE POWERS -- This article is based on a definition of culture presented in a companion article (“Powers of Culture, Connecting Culture with High Performance”). KEY DISTINCTION -- That article makes crucial distinctions between culture theory (what anyone consciously recognizes or says culture is) and culture itself (always vast amounts of unconscious routines imbibed while growing up locally or joining some group). If culture is vast amounts of stuff inside of people that they unconsciously learned while growing up or being involved in some group, models that reduce that vastness to four or five dimensions or the like, while making intellectual work simpler, ruin practical work. Highly reduced abstract models reduce number of factors to remember but at a cost of increasing greatly the work of recognizing abstractions in cases and of grounding abstraction features in cases where the abstraction is being applied. So journals tend to favor “cogent” models, highly reduced and abstracted, while practitioners benefit most from mid-level models having well-ordered dozens of factors, requiring less work to recognize and ground. RESULT 1 FRACTAL CONCEPT MODELS Fractal ordering of great numbers and diversity of variables -- Fractal concept models allow dozens of variables to be ordered on multiple levels with orderings analogously repeated across levels, so granularity of model treatment can be adjusted per case situation to be handled by the model. RESULT 2 CAUSAL PATH OF SUCH FRACTAL CONCEPT MODELS: Allowing us to order dozens of culture-related variables while maintaining “tunable” levels of model granularity BENEFIT: MODELS OF BOTH MANY AND FEW VARIABLES WITH TUNABLE VARIABLE GRANULARITY -- This article applies one new interface to culture models and contents, fractal concept models, an interface derived from artificial intelligence research on handling extremely large databases and visual displays. With this new interface, a mid-level of theorizing and empirical work opens up, wherein enough detail is preserved in models to reduce recognition and grounding work for practitioners while still preserving cogency of factors for academic researchers. Also, this article applies another new interface, one that presents ten common components of culture research studies in a causal flow. This involves making several careful distinctions not common in current research literature on culture that empower research while connecting the essence of culture and the essence of power and the essence of high performance theoretically and practically. The article closes with seven research agenda implications of these two new interfaces for study of culture
