Predicting behavior based on attitudes Attitudes can be very complex, a product or service can be composed of many attributes or qualities, some of which may be more important than others for some people. Additionally, a person's decision to act on their attitude is influenced by other factors, such as approval from family and friends. For this reason, multiattribute models have been built that are extremely popular among marketing research. Many models have been built but the most influential model is Fishbein's 1973 model. Fishbein's multiattribute model argues that attitudes can predict behavior. It measures three components of attitude. Salient beliefs that people have about an attitude object, the object-attribute links, and the evaluation of each. The example of Sandra's college choice (Solomon, Bamossy, and Askegaard) illustrates this. Sandra considers which attributes are important in forming an attitude towards each college and then assigns a rating regarding each college's performance on that attribute, for example reputation, cost, party atmosphere etc. Then determine how important each attribute is to her. The overall attitude score (A) is obtained by multiplying your rating of each attribute by the importance rating of that attribute. This model has many strategic applications, for example, capitalizing on relative advantage, if Sandra rates a college highly in terms of party atmosphere but doesn't think this attribute is important, the college can emphasize why it would be. Other strategies might be to strengthen links between attributes, add a new attribute, or influence competitor ratings. These strategies are all ... middle of paper ... feel like citizens of their own country. This study also focuses on how changes in the attitudes of different age groups can occur. This program is useful for identifying change agents and future scenarios that may more accurately predict actual behavior than TRA and TPB. (Greenwald 1987). Although the theory of reasoned action has successfully predicted a wide range of behaviors, including dental hygiene and family planning, and has provided a framework for consumer research, attitudes and behaviors are only weakly related: people they don't always do what they mean to do. There are many limitations, the model lacks heuristic value and ignores differences in cultural values ​​and attitude trends over time. Turpin and Slade (1998) conclude that current models are still quite poor at predicting actual behavior.