Preference Evaluation Example A. Direct-Objective Evaluation Direct-objective evaluation can be in the form of test result which connected the object with the objective of the test directly through the test result Examples : 1. Evaluate the student learning objectives by collecting information on student performance on tests.
If the objective is to make students able to master a particular skill, and that skill is tested in a few questions on an overall exam, the instructor can evaluate attainment of the skill by looking at just those relevant questions on the exam which will give direct measure between the test result and particular skill they have. 2. Evaluate aircraft engine performance after shop visit. The objective is to get optimum airflow through the engine. The object is the clearance between the blades and the shrouds. Method : * measure the length of all blades * measure the diameter of the shrouds * measure clearance between blades and shrouds check on the manual book about the clearance limitations * make adjustment to get optimum clearance refer to manual This method will directly effect on the engine performance by resulting an optimum airflow through the engine which will maintain the engine bypass ratio on its optimum level. B. Ordinal Evaluation Ordinal evaluation allows us to sort the rank of the object that we measure. The limits of the values are not clear, so that can be compared only if the value is higher, the same or lower than the others but we can’t say how much different interval between the values. Examples : 1. The T-shirt size : S - Small M - Medium
L - Large 2. The scale of frequency : 1- Always 2- Often 3- Rarely 4- Occasional 5- Never C. Pair Wise Comparison Pair wise comparison generally refers to any process of comparing entities in pairs to judge which of each entity is preferred, or has a greater amount of some quantitative property. The method of pair wise comparison is used in the scientific study of preferences, attitudes, voting systems, social choice, public choice, and multi agent AI systems. (Wikipedia) Example : Choosing to buy a motorcycle with criteria as follows : * Price * Economical * Design Weight for each criteria : * Price three times more important than design Price is two times more important than economical * Economical one-half times more important than design Pair wise comparison matrix : Criteria| Price| Economical| Design| Priority Vector| Price| 1| 2| 3| 0. 5455| Economical| 0. 5| 1| 1. 5| 0,2727| Design| 0. 333| 0. 667| 1| 0,1818| Total| 1. 833| 3,667| 5. 5| 1| Based on weight of each criteria, price is the most important, economical is the second important and design is the least important. D. Nominal Evaluation Nominal evaluation classified the criteria but with no relative meaning. Each criteria not represented which one is better or bigger, only categorized it ased on identification and fact of the object. Examples : 1. Number of DKI Jakarta citizen based on religion (2005) : Religion : Islam, Christian, Catholic, Hindu, Buddha Province| Islam| Christian| Catholic| Hindu| Buddha| Total| DKI Jakarta| 7,157,182| 501,168| 336,514| 28,508| 313,217| 8,336,589| 2. Evaluate the number of people based on marriage status : 1 - Married 2 - Single 3. Blood type classification : A, B , O or AB 4. Ethnic group classification : Javanese, Sundanese, or Betawi 5. Classification of part based on color : red, blue, yellow, green, etc