Marketing Education Review
ments is partial formalization (Hunt 1991). To do so, each of the theoretical statements in the model should be summa- rized and ordered as follows:
Premises or assumptions that are not defined within the model and
Derived statements that can be logically deduced from the premises (Hage 1972; Hunt 1991).
The set of ordered statements should form a chain syllo- gism (Blumberg 1976), such as the following:
Premise: (1) If A implies B,
Premise: (2) and B implies C,
Derived: (3) then A implies C.
In the chain syllogism just given, statements 1 and 2 are premises and statement 3 is a derived statement. The basic theoretical statement "As extrinsic satisfaction decreases, job seeking is likely to increase," can be derived by using statements in the explanation as premises:
Decreasing extrinsic job satisfaction (A) -^ perception of a problem (B) (lack of adequate pay): A > B.
Perception of a problem (B) > motivation to search for a solution (C): B ^ C.
Motivation search for a solution (C) —> search for a more rewarding job (D).
Decreasing extrinsic job satisfaction (A) —> search for a more rewarding job (D): A » D.
Ensuring that the statements composing the theoretical argument are logically interrelated is a critical step. In a well-known critique of consumer research, Kassarjian (1971) states that attempts to relate general personality char- acteristics, such as sociability, to brand choice failed because the investigators did not develop a specific explana- tion of why sociability should be related to selecting a brand of toothpaste. Kassarjian (1971) cited a study by Jacoby (1971) that overcame the problem. Jacoby predicts that low dogmatics will be more innovative than high dogmatics on the basis of the explanation that high dogmatics will avoid the perceived risk of switching to a new brand.
The power of a theoretical statement to serve as a foun- dation for deriving research hypotheses is rooted in the very broad scope of the statement relative to research hypotheses. The theoretical statement, with definitions of its concepts and an understanding of the proposed explanation linking the concepts, can support a large number of empirical inquiries.
The statement "As extrinsic satisfaction decreases, the worker is likely to perceive that a problem exists" could lead to numerous studies. As an example, one could predict that as extrinsic satisfaction decreases, the worker is likely to search for means of receiving more rewards from the current job. The possibility of increased rewards may depend on the structure of the sales position. Salespeople working on com- mission may allocate more time to calling on prospective new accounts and spend less time providing services to low- volume current accounts. A prediction based on a theoreti- cal statement will move the research to the second major
step in the theory-setting-test sequence: the setting and proposition(s).
The S Step: Theoretical Statement to Setting and Proposition
The theoretical statement is abstract and broad and does not refer to specific phenomena that could become the topic of empirical research. The researcher must move from the abstract world of thought to a real-world setting. The theo- retical statement to setting and proposition step accomplish- es that end. A major tool of hypothesis development is appli- cable to this step: a ladder of abstraction.
To discuss the ladder of abstraction, we use the theoreti- cal statement used previously as an example: "As worker extrinsic job satisfaction decreases, job seeking increases." By replacing the abstract concepts of the theoretical state- ment with concepts that are specific to a setting, a proposi- tion could be stated, "As the extrinsic job satisfaction of industrial salespeople decreases, the salesperson is more likely to take steps to leam of other sales jobs that offer more material rewards." The same negative operational linkage between concepts is found in both the theoretical statement and the proposition.
The ladder of abstraction is the idea that the domain of a concept includes referents ranging from the broad and abstract to the narrow and concrete (Osigweh 1989; Zaltman, Pinson, and Angelman 1973). A ladder of abstrac- tion for extrinsic job satisfaction is as follows:
Worker extrinsic job satisfaction,
Sales worker extrinsic job satisfaction,
Industrial sales worker extrinsic job satisfaction,
Computer manufacturer salespeople extrinsic job satisfaction.
The ladder of abstraction can be of service to the researcher in two ways. First, a proposition should be logi- cally tied to the theoretical statement. Specifically, each cor- responding proposition level concept should be a subset of the domain of the theoretical concept, and the ladder pro- vides a means of checking for congruency. As an example, many consumer behavior attitude models have been drawn from theoretical statements in psychology that apply to humans in general. Thus, such models should apply to con- sumers in any setting (Ladder: Humans, Consumers, Consumers of Product X ...). However, the setting specific subjects in tests of the models should be consumers of the focal product, not just consumers. That line of argument has resulted in what appears to be a general rule in marketing that student subjects are acceptable if students are con- sumers of the focal product.
Second, the ladder of abstraction can open the door to many potential research topics, because it is possible to come down the ladder of abstraction to develop many dif- ferent propositions. A new proposition would be that for retail salespeople, as extrinsic job satisfaction decreases, job search increases. Dubin (1978) has proposed a general rule for coming down the ladder of abstraction and that is to test only new propositions that will be of scientific interest. Once sufficient empirical evidence has been obtained testing a theoretical statement, retesting it with new propositions is of little interest. However, if a theoretical reason exists to