Making Appropriate Choices
Everyday people make choices. Some of the decisions that they make are trivial and have little effect on their overall well-being. However, other choices will impact the rest of their life. There are many different factors that people care about when they make these decisions and they weigh each category differently, which can make the decision-making process labor-intensive and confusing. How does one make such important decisions? A recommended strategy for making informed choices is to leverage one of the multi-attribute choice methods. Applications include making personal decisions such as where to work and live. By using multi-attribute choice, they can simplify the process and not use a “gut feeling” to arrive at conclusions.
Compensatory vs. Non-Compensatory
Multi-attribute choice, also known as multicriteria analysis, is a tool to structure the decision-making process and assist in selecting an optimal choice. There are several different choice strategies that are either compensatory or non-compensatory, alternative or attribute, and exhaustive or non-exhaustive. An alternative strategy processes one alternative (for example, a job applicant) before analyzing the next one. Processing by attribute involves analyzing every possible choice for a single category before moving on to the next category. For example, each potential home will be analyzed based on their price before moving on to another category such as location of the house. An exhaustive strategy is where all attributes and alternatives are processed and analyzed before a choice is made.
Compensatory
A compensatory strategy trades low values on one dimension for high values on another. For example, a prospective homeowner may want to trade number of bedrooms for number of baths. One of the strategies that can be used in a hiring search is a labor-intensive, alternative and exhaustive process called the linear additive model. In this model, we give weights adding up to one-hundred percent to each attribute by their importance to us. Next, we consider each alternative one by one and assign values to each attribute. Lastly, we add the weighted values to see which is the “best” option. Another strategy is an exhaustive attribute system called the additive difference model. This model compares two alternatives side by side, sums the differences across all attributes to get one overall difference score that is carried on to the next alternative for comparison.
Non-Compensatory
A non-compensatory strategy involves eliminating alternatives if they fall outside of the predefined boundaries. In the home search, we may want to rule out any homes that have red flags in terms of age, maintenance, or seller’s inclination to sell. Dominance is an exhaustive, alternative strategy that finds alternatives that are better or worse than the other alternatives for every attribute. Throw the worst alternative out if it is lower in every attribute or choose the “dominant” one that is higher for each attribute. For example, if home B has a better score on each attribute than homes A, C, D and E, then home B is the clear choice. An alternative, but non-exhaustive strategy is satisficing. Satisficing, also known as the conjunctive model, uses cutoff points set for each attribute. Any alternative that does not meet the cutoff points for each attribute is eliminated from the set. Another alternative, non-exhaustive strategy is the disjunctive model. Finally, Elimination-by-Aspects (EBA) is a non-exhaustive, attribute strategy where an attribute is picked at random and all alternatives that do not meet the cutoff point for that attribute are eliminated. This continues for each attribute until there is only one alternative left. If the cutoff point for maintenance (on a 1-10 scale) is a 4, any candidate with a score more than 4 for maintenance is eliminated from contention.
Common Mistakes
There are three very common errors when it comes to decision making. First, a non-exhaustive, non-compensatory, alternative strategy, the recognition heuristic, is used when people are poorly informed on the decision that they are trying to make. Thus, they rely on name recognition and choose the first alternative that they recognize. With respect to the home search, this would be if the prospective buyer chose the home with a well-known or big brand name realtor with high commissions and low personal service. However, this strategy does not take into account preferences or alignment with other important considerations, and, as such, is not recommended. This leads to the second error in decision making: limited time. The home buying process is ideally a short process. Lastly, a common error when making decisions with multiple attributes that do not have a common measure is a lack of standardization. Standardization is important because it takes each attribute and, regardless of its measure (home age, commissions, etc.), equalizes them all on the same scale (z-scores) and thus, makes the candidates comparable. In the home search we can use z-scores by taking the individual attribute score, subtracting it from that attribute’s mean, dividing the sum by the attribute’s standard deviation, and lastly multiplying the product by the attribute’s weight. Adding all of the weights gives simple sums that can be ranked. These errors could be avoided by making use of the multi-attribute choice processes for decision making in the home buying process.
Multi-Attribute Choice in the Home Buying Process
Edit the document to insert your own attributes, weights, and scores here. Warning: most homeowners prefer to buy houses with lower commissions and less maintenance so the standardization formulas start with a negative sign. When editing the form make sure that all positive attributes do not have this and negative attributes do.
In this example, we can see that a buyer is considering five home candidates in a linear additive model and are taking into consideration the commissions, number of beds and baths, location of the home, size of the lot, and amount of maintenance. The most valued attribute is commissions, followed by maintenance, location, number of beds and baths and lastly the size of the lot as seen in the weight row with a checksum to confirm our weights add up to 100 percent. It is also clear that the six attributes are scored in different ways which makes standardization necessary to be able to compare them. After filling out scores for each candidate, excel will use the inputted formulas to get the average and standard deviation of the scores, find the standardized z-score for each cell, add up the z-scores for each house candidate, and then rank the homes based on the results. In this case, home candidate 1 is in the top two for all categories so with the current weights it is ranked as the best option. Leveraging a multi-attribute choice system, such as this linear additive model, streamlines the home buying process and makes it easier for the prospective owner to choose a home based on their priorities.
Multi-Attribute Choice Lays the Foundation for Successful Home Buying Decisions
There are several different multi-attribute choice strategies that are compensatory or non-compensatory, alternative or attribute, exhaustive or not exhaustive, and range from easy to a great deal of mental effort. Each strategy, however, is intended to structure the decision-making process and assist in finding an optimal decision when faced with important choices such as the home buying process. It is structured this way to avoid the damage of a poor purchase.