Words: 3571
Journal of Technology Research The Hypothesis Testing of Decision Making Styles in the Decision Making Process Enable Content Shenandoah University Abstract: The objective of this study is to test the effectiveness of various decision making styles in the decision-making process. Four broad categories of decision making styles are utilized in this simulation study. The methodology is illustrated with a complex, constructed problem often used to train and evaluate management personnel. In order to test the efficacy of these styles, two prototype systems will be constructed.
The Decision Support Systems architecture serves as a control and the Just-in-time intelligent Decision Support Systems as the experimental architecture. The experiment will test whether the use of either of the two systems offers a significant improvement in the process of and outcome from the four decision- making styles. The paper closes with a conclusion on the results of the experiment and their implications on Information Systems Research in relation to the decision making process.
Keywords: Decision Making, Decision Making styles, Decision-making Process, Simulation, simulation modeling The Hypothesis Testing, Page 1 Introduction The decision making process is directly linked with the need for problem solving and or decision making. The right choices we make in solving problems and making decisions depends on how correctly we follow the steps through in the decision making process. This paper addresses the effectiveness of the process and outcomes of decommissioning styles, in the decision-making process, puts forward a methodology for determining their effectiveness.
The merits of Non-subject Designs The non-subject design approach assumes that human subjects will not be used in adhering data for the simulation study and in evaluating information system architectures. Using human subjects can present some serious scientific, technical, and economic problems. It will be time consuming and potentially costly to get human subjects because of the selection, training, and motivational issues involved in the acquisition of subjects (Power, 2002); (Hoover and Perry, 1990); (Oaken and Spectacle, 1999).
In addition, there may be political considerations (obtaining consent from subjects and going through bureaucratic hurdles from the Institutional Research Board – RIB) involved in selecting and utilizing human subjects in experiments. Also, humans in an experimental setting may not behave the same way sample of human subjects. Even if the sample is representative of the defined subject group, it may not be representative of the population of potential information system users.
As a result, it may be difficult to generalize the results from the subject-based experiment. Human subjects, however, may be unnecessary to conduct simulation studies of information systems, especially when information is available about user behavior. One such instance involves studies that involve decision making support systems (DMS). There are various studies that define decision making behavior for the general population. Different decision-making styles will be generated based on a stochastic process (based on random variables).
This is regulated by means of the hypothesized decision-making styles as found in the literature and specifically in: Turban and Ransom’s Decision Support Systems and Intelligent Systems (1998) up. 62-3. Decision Style is the way and manner in which decision makers think and respond to or address problems. Decision style is also about their cognitive response to decision situations and their individual and additional differences in beliefs and values. Decision making is not linear. That is to say the emphasis, time allotment and priorities differ from individual to individual and as well as from situation to situation.
Gordon et al [1975] identified 40 processes in looking at 9 types of decision and (Miniature, 1973) identified 7 basic styles with a lot of variations. (Turban and Ransom, 1998). Once a simulated manager or decision maker is assigned to a certain decommissioning style category, the corresponding choice logic (parameters) will be applied to his or her decision behaviors within the simulation model. Based on these parameters, values for decision variables and uncontrollable variables will be generated and entered.
The simulation results, such as values for profit after tax, investment, marketing and Research and Development will be obtained for the simulated quarter or quarters. Profit status The Hypothesis Testing, Page 2 reports and sensitivity analyses will be utilized in arriving at a recommended policy (set of decision variables) for the simulated business organization. Since each simulated user will be utilizing the alternative information system architectures, the inferences need not be precise. Moreover, sensitivity analysis can be used to test outcome and process sensitivity for inferred values.
The suggested approach can be used to generate a very large number of simulation runs for the full variety of potential users within a very short time. More importantly, the approach is timely and cost effective. Since all potential, rather than a sample of users is considered, the approach is likely to be more generalize than human subject based experiments (Content and Forgiven, AAA). Figure 1 Simulation: Experimental Design Start Specify Decision Types Decision Types Generate Random Values Run Simulation and Compute Results Analyze and Report Results 3. Prototype ADDS and SIDES The non-subject simulation experimental approach can be illustrated with The Hypothesis Testing, Page 3 prototype ADDS and SIDES The methodology is illustrated with a complex, semi-structured problem often used to train and evaluate management personnel. The problem being simulated here involves a market in which an organization competes for a product’s four-quarter total market potential on the basis of price and marketing. The demand for the organization’s software products will be influenced by, (1) its actions, (2) a major impetigo’s behavior, and (3) the economic environment.
The market simulation process is centered on the formulation of a software development policy that would generate as much total profit as possible over a forequarter planning period. Policy making requires: (1) setting the levels of four decision variables (the product price, marketing budget, research and development expenditures, and plant expansion investment) and (2) forecasting the levels of four key uncontrollable variables, that comprises the competitor’s price, marketing budget, a seasonal product sales index, and an index of general economic conditions.
These eight variables will Jointly influence the profitability of the simulated business organization. In both systems, twelve additional variables, including plant capacity, raw materials inventory, and finished goods inventory, will remain fixed from trial to trial and thereby become the scenario for decision-making. As in any competitive business environment, this problem is dynamic in nature, I. E. , a decision made in one quarter affects decisions and outcomes in the current and subsequent quarters. In this dynamic environment, it also is difficult to recover from initially poor decision trainees within the simulated time frame.
In this situation, the major focus of the users will be on the key uncontrollable events – competitors’ marketing and price, the seasonal index, and the economic index – and the major controllable actions – price, marketing, research and development, and production. Ranges for these values are available for scenarios specified in the training exercises. A DMS in general, can test alternatives and events specified by the user, offer information about the relationships between the variables, guide users toward a desirable action, or more. Different architectures can be specified for simulation experiment.
For each architecture, the user would specify the major controllable actions and key events. Table 2 below discusses cognitive style decision approaches and attempts to characterize the different decision styles based on their various problem-solving dimensions. Specifications for the variables as described above in the forgoing paragraph would be inferred from Table g’s characteristics. For example, an analytic user might select values of price close to the competitor’s values. On the other hand, an autocratic user might gamble with a higher than nominative price.
A simulation model of the problem will be necessary to test the specified controllable actions and key events for any of the alternative DMS architectures The Hypothesis Testing, Page 4 Decision Situation Two systems are constructed: one for the base ADDS, which is the control system and another one for the Intelligent Just-in-time Decision Support Systems (SIDES) which represents the experimental system. Guidance in the experimental system will be elicited by selecting pushbuttons that will aid in the setting of some inputs for the problem analysis, evaluation and profit reports.
Each of the two systems includes a basic strategic-management-specific ADDS, which has (1) internal organizational data, (2) external competitive data, (3) environmental data, (4) a model base of mathematical expressions and, (5) profit status reports and sensitivity analyses Users in the experimental group (SIDES) will elicit guidance by accepting advice from the KIDS. Simulation Models The objective of this simulation study is to test whether Intelligent Just-in-time Decision Support Systems (SIDES) offers a significant improvement in the process of and outcome from decision-making in comparison to other existing Decision Support
Systems. Table 1, which is adapted from Ransom and Turban’s Decision Support Systems and Intelligent Systems (1998) up. 62-3, summarizes the variety of potential behaviors. The Hypothesis Testing, Page 5 Table 1 Cognitive Style Decision Approaches Problem-solving Dimension Approach to Search Analysis Scope of Analysis Basis for Inferences Heuristic Analytic Learns more by acting than by analyzing the situation and places more emphasis on feedback Uses a planned sequential to problem-solving; learns more by analyzing the situation than by acting and places less emphasis on feedback.
Uses formal rational analysis Uses trial and error and Spontaneous action Uses common sense, intuition, and Develops explicit, often feelings quantitative, models of the situation Views the totality of the situation Reduces the problem as an organic whole situation to a set of Rather than as structure underlying causal functions. Constructed from specific parts Looks for highly visible Locates the similarities or situational differences that vary commonalities by comparing with time In effect, Table 1 defines classes of users for DMS and their potential behaviors.
In a simulation study, ranges of values for the controllable actions and key events can be obtained. Using the characteristics from Table 1, we can draw reasonable inferences about the values of actions and events within the ranges for the various classes of users. Decision Making Matrix A specific problem scenario was used to test the efficacy of each DMS. The scenario is described in Table 2.
It will serve as a drawing board for the reader to have an idea of the scenario data and the values for the various variables in the previous quarter prior to the quarter for which the actual market simulations are being carried out. Furthermore, an analytic type decision maker was assumed, for example, as hardhearted in the preceding sections would want to consult with such data. The Hypothesis Testing, Page 6 Table 2 Firm’s Scenario Data: Internal and Environmental Data Controllable Actions Key Events price = $ 100 Plant Investment = $ 500,000 Economic Index = 1. 00 Seasonal Index = 1. 0 Marketing= $ 550,000 = $ 600,000 Competitor Price = $ 100 Competitor Marketing= $ 500,000 From table 2 above, the following user types were proposed to be utilized for the purpose of assigning values for input into the interface for the decision variables and uncontrollable variables: Analytic-Autocratic, Heuristic-Autocratic, Multidimensionality and Heuristic- Consultative types. Decision making style Matrix The figure above describes the decision making styles on the basis of two dimensions: 1) the amount of information used and (Analytic/Heuristic) 2) the number of alternatives considered (Consultative/Autocratic).
Using the matrix in figure 2 above, four classes of decision styles were proposed based on the combined The Hypothesis Testing, Page 7 The amount of information used Heuristic managers, for example, generally base their decisions on minimal information. They do not normally stay long searching for information. They believe: 1) that additional information will only aid in confirming what they already know, (2) that the data will be too slow in arriving, (3) that additional information will be confounding and distracting.
Analytic managers, on the contrary spend more time exhaustively analyzing information before reaching conclusions about decisions to be made. Number of alternatives considered Autocratic managers or users, for example, are called “satisfiers”. This is because, they stop searching after finding one satisfactory solution to their problems and then adopt that solution. Consultative managers continue to explore multiple or additional elution and consider a range of feasible alternatives to the problem at hand before making a decision. They are called “maximizes. Autocratic and heuristic managers focus on speed, efficiency and consistency of their decisions. They are highly action-oriented. The plans of these managers tend to hinge on short-range objectives and they prefer to work in an organizational structure with clearly defined rules and specific plans. Decision Making Styles The following sections will treat how values for the various variables are allocated and will also attempt to Justify why these values were assigned. The major considerations are the decision-makers’ inclinations to risk-taking which is reflected by high value ranges.
On the contrary, the conservative ranges represent the non-risk taking or non-gambler types. A combination of these two extremes will mean the assignment of moderate or midrange. Other assumptions are their propensity to compare with the range of prices of their competitors in the market, ranges in the previous quarters and above all, consulting with colleagues and co-workers. Analytic-Autocratic: These are users who are highly analytic in terms of information processing. They are also autocratic with regards to their choice of alternatives and their propensity to consult with others.
They utilize a large amount of information before making a decision. On the other hand, they consider very few alternatives, and consult little with other staff members. In other words, they are unilateral in their decision making. This makes them rather autocratic. Based on this combination of attributes, and given their analytic inclinations, it is assumed that they are moderate risk takers. They would analyze information with regards to the firm’s scenario data and specifically the product prices of their organizational competitors.
These reasons account for the choice of value ranges for the analytic-autocratic decisions style as espoused in table 4 below. For example, the The Hypothesis Testing, Page 8 analytic-autocratic decision maker, being a moderate risk taker, would enter values that are not high. This risk taking tendency is furthered tempered by the fact that in combination they are analytic. Therefore looking at the ranges in the price column, “P” for instance, this type of decision style has the next-to-lowest range ($150 – $175), after the analytic-consultative style users, who are most conservative in their choice of values.
It is assumed the user will compare and analyze prices of the competitors in the market, the market conditions, and the firm’s product prices in the previous and current quarters. The competitor’s price in this case is $110 and the product price in the previous quarter was $100. It is also assumed the autocratic gambler type would input $200 but the analytic attribute would influence this user to input mid-way values of the range $1 50$175. Such considerations and analyses will temper the user’s liberal price ranges and bring them down mid-way.
This explains my choice of values for this decision style user in the table 5 below. Furthermore, looking at the marketing column “M”, this decision style user also has the next-to-lowest range ($650000 – $750000) after the multidimensionality style whose range is $500000 -$600000. Heuristic-Autocratic A manager with a Heuristic-Autocratic style uses minimum amount of data to arrive at a reasonably satisfactory decision. Heuristic-Autocratic managers when making their decisions prioritize certain factors like on speed, efficiency and consistency.
They are highly active and results-oriented. Their plans are centered on short-range objectives and they prefer to work in a well-structured organization with well-defined ales and detailed plan descriptions. They spend very little time consulting with their work colleagues and other professionals. They tend to like reports that are brief and precise and discard or ignore long and detailed reports. The table below shows values that a Heuristic-Autocratic user would enter for variables based on his or her predefined preferences above.
Looking at the values in table 5, the decision style user is not only autocratic but heuristic with the highest ranges in price “P’ ($180 – $200); marketing “M” ($800000-$900000); Research and Development “R” ($750000 – $850000) etc. Like in the previous decision style discussed above, the competitor’s price is $110 and product price in the previous quarter was $100. It is assumed that this decision type user neither analyzes information adequately nor consults to obtain information about market conditions.
The user would therefore be liberal in assigning input values. That is why, this user inputs a high value, in fact the highest value for the product price in the range of $180 – $200. The same is true for this user’s input for virtually all the remaining variables. This means, the user neither analyzes a reasonably large amount of information, nor consults and collaborates with others. In addition, the user does not consider a wide array of alternative choices; does not compare with prices of the competitor or the external environmental data.
This is also, because the user is rather hasty, and has to make quick decisions. This makes the user both autocratic and non-collaborative. This also means that the user is a very high risk-taker. The Hypothesis Testing, Page 9 Analytic-consultative Managers with an Analytic-consultative decision style like to process extensive information as they consider several feasible options or alternatives. They are creative and seek variety. They tend to be contemplative and rather cautious in taking action.
They also utilize or adopt a long-range perspective with regards to organizational planning, and their plans are adaptive because of the many alternative choices that they consider. Analytic-consultative managers prefer to work in organizations that are wildcatters with high degree of flexibility and delegation of duties. These managers like to receive long, detailed and analytic reports. Because of these qualities, they are able to use consultative and participative decision making effectively, thereby maximizing the availability of information for decision making and problem solving.
Managers who use an Analytic-consultative decision style tend to be wildflower and open to new information. They possess a broad vision of the organization and its distinctive mission, and they tend to generate creative solutions to organizational problems. Co-workers of Analytic-consultative decision makers are likely to describe them as empathetic and cooperative. The values in the table show that the multidimensionality user as the name implies considers a wide range of alternative choices, consults with colleagues and analyzes a lot of information before inputting values.
That is why this decision style user has the lowest ranges in all the columns which are closely similar to those in the scenario table 5. The values that this user inputs compare closely to those of the competitor’s product price which is $100 and also the product values in the previous quarter as mentioned at the value of $110. The assumption is that such a user is abreast of market information, does a lot of consultations with co-workers and colleagues and above all analyzes a lot of information regarding the product prices of competitors.
These considerations will therefore lead the user to utilize informed Judgment to be able to input values close to those of the competitor and the previous quarter. This user is therefore not the gambler type who is liberal with the values that she or he enters. For instance, this decision style user has the lowest price “P” ranges ($100 – $125); marketing “M” ($500000 – $600000); research and development ($500000 – $600000); the economic index “E'” (1. 25-1. 35); and the seasonal index “SSL” (1. 1 – 1. 40).
These are all aloes in the region of the competitors’ preferences in terms of parameters. This goes to show that this type of decision maker inputs values close to how the competitors price their products. And this decision maker also consults with co- workers and does exhaustive or ample research on market trends. Heuristic-consultative Managers who use a Heuristic-consultative decision style also rely on minimum information. They are willing, however, to consider several alternatives and reinterpret the information to arrive at possibly different conclusions.
These managers are also speedy and active in discharging their duties. They exercise some degree of flexibility and adaptability in their decision making. Heuristic-consultative managers prefer to work in an organization with a structure that allows them the directions depending on the conditions. Heuristic-consultative managers also prefer to receive from subordinates precise communications that contain a variety of specific steps or solutions to consider during decision making or problem solving.
They also prefer to work within settings that permit collaborative interactions between people and other staff member or Junior co-workers. They use their popularity and charm to influence and win others, and they also induce co-workers by providing them with incentives. Given the preferences of this decision style user, the assumptions are that this user is consultative but does not consider large number of alternative choices nor analyzes a lot of information.
This makes this user a semi-satisfier, in that the user’s tendency to accept the first satisfactory solution to the problem that comes up is tempered by the user’s consultative attribute. In other words, if the price of the competitor is $110, the heuristic attribute may lead the user to enter a 100% increase putting the price at $220. The consultative tendency will however bring it down to $1 55-$180. The below table (5) shows values (controllable actions and key events) that managers or users belonging to this user type would enter based on the preferences as defined above.