Prospect theory tells us that individuals attitude towards risk vary. Some individuals are risk seekers while some individuals are risk averse. These attitudes also vary depending on whether it's a gain or a loss.
Before I talked about more about the prospect theory I would like you to introduce you to a problem that was later termed as the Allias paradox. Just have a look at the sample of questions below :
Participants are usually asked to choose one of these options.
Gamble A: A 100% chance of receiving £1 million.
Gamble B: A 10% chance of receiving £5 million, an 89% chance of receiving £1 million, and a 1% chance of receiving nothing.
After their decisions have been made they are then presented with two more options to chose from
Gamble C: An 11% chance of receiving £1 million, and an 89% chance of receiving nothing.
Gamble D: A 10% chance of receiving £5 million, and a 90% chance of receiving nothing.
Most people usually prefer gamble A to B and gamble Dto C. Expected Utility theory infer that if a person prefer gamble A over B then the person must prefer gamble C over D. That is because they are maximising the expected utility or wealth. If a person prefer A and D together then there is an inconsistent in the expected utility theory. Expected utility theory convey that rational decision makers should weight the utility of the outcome by their probability of occurrence. That is equal outcomes to added to each of the choices should cancel out.
A series of experiment has been conducted by Kahneman and Tversky using this paradox and in each case there has been a violation in the axiom. Thus people do not seem to prefer B and D and A and C. What they found was that people overweight outcomes that are certain relative to outcomes that are merely probable. They termed this the certainty effect. It also indicated how people are risk averse when it comes to sure gains and risk seeking when it involves sure looses. Most people prefer a sure gain over a chance of gain but when the outcomes are looses then most people prefer the risky option to the sure loss. Thus people prefer to gamble when they have something to loose.
Thursday, 8 December 2011
Thursday, 10 November 2011
Fast and frugal thinking among professionals
Usually it is expected of professionals to use all relevant information and weight and combine them appropriately when making decisions. Making a decision by this process is classified as the regression model. This is a very statistical way of thinking, it follows a mathematical sequence. High weight attached to a cue is compensated for a low weight when making a decision using the regression model
However there are other non-statistical and more cognitive simpler way of thinking alternative to the regression model. These are the Franklin's rule and the Matching heuristic. Franklin's rule is similar to the regression model but dissimilar in a way that it doesn't take into account the interactions between the cues.
The matching heuristic on the other hand is non compensatory. Decisions are made by searching through a subset of cues and basing the decision on one cue.
A research report of judges and jurors decision making on whether to bail or jail found that the matching heuristic was a better predictor of judges decision than the Franklin's rule (Dhami, M.K.,2003). This was a real life cases that went on for 4-month period. They divided the court into two , A and B and had 1 observer for each court. The matching heuristic and the Franklin's rule were used to capture the policies of each court separately because each court differed in the cases presented and the decisions made.
Whereas the Franklin's rule searched through all the 25 cues that were presented the maximum cue that was was searched by the matching heuristic was on average 3.
There is evidence that people switch to non-compensatory strategies that uses fewer cues as the number of cues increases and the time pressure increases.. It has also been found that group decision involving shared responsibility tend to use fewer cues (Weldon & Gargano, 1985).
The matching heuristic is therefore known as a simple or fast and frugal heuristic because the decision is based on one cue alone and not altered by the value of other cues. It is also flexible beacuse different cues can be used to make decisions on different cases.
However there are other non-statistical and more cognitive simpler way of thinking alternative to the regression model. These are the Franklin's rule and the Matching heuristic. Franklin's rule is similar to the regression model but dissimilar in a way that it doesn't take into account the interactions between the cues.
The matching heuristic on the other hand is non compensatory. Decisions are made by searching through a subset of cues and basing the decision on one cue.
A research report of judges and jurors decision making on whether to bail or jail found that the matching heuristic was a better predictor of judges decision than the Franklin's rule (Dhami, M.K.,2003). This was a real life cases that went on for 4-month period. They divided the court into two , A and B and had 1 observer for each court. The matching heuristic and the Franklin's rule were used to capture the policies of each court separately because each court differed in the cases presented and the decisions made.
Whereas the Franklin's rule searched through all the 25 cues that were presented the maximum cue that was was searched by the matching heuristic was on average 3.
There is evidence that people switch to non-compensatory strategies that uses fewer cues as the number of cues increases and the time pressure increases.. It has also been found that group decision involving shared responsibility tend to use fewer cues (Weldon & Gargano, 1985).
The matching heuristic is therefore known as a simple or fast and frugal heuristic because the decision is based on one cue alone and not altered by the value of other cues. It is also flexible beacuse different cues can be used to make decisions on different cases.
Reference
Dhami, M.K. (2003).Psychological models of professional decision making. Psychological Science, 14 (2), 175-180.
Weldon, E., Gargano, G.M. (1985). Cognitive effort in additive task groups: The effects of shared responsibility on the quality of multiattribute judgements. Organizational Behaviour and Human Decision Processes, 36, 348-361.
Thursday, 3 November 2011
Probabilities turned to frequencies
Previously I mentioned the fact that when making decisions we use representative heuristic which is also known as similarities which in turn gives room for error to what is know as the base rate fallacy or neglect. Base rate fallacy is when we do not take into account the proportion of people who are like the description given but are actually something else. Gigerenzer, 1994, proposed that when probability questions are turned to frequencies such errors are less likely to occur.
A probability question is like: this taken from Gigerenzer,1994 article.
"1. A device has been invented for screening a population for a disease known as psylicrapitis. 2. The device is a very good one, but not perfect. 3. If someone is a sufferer, there is a 90% chance that he will be recorded positively. 4. If he is not a sufferer, there is still a 1% chance that he will be recorded positively. 5. Roughly 1% of the population has the disease. 6. Mr. Smith has been tested, and the result is positive. The chance that he is in fact a sufferer is:______.
What was found here was that most participants gave a response of 85% which is close to the 90% hit rate given in the example despite the 1% chance. Making such judgemental error is what is known as the base-rate fallacy. However when the question was switched to frequency readers could digest the question more easily. The frequency question is as follows:
"Out of every 100 persons tested, we expect 1 to have the disease; and the device is nearly certain to say that he has. Also, out of that 100, we expect the machine to say that 1 healthy person has the disease. Thus, in the long run, out of every 100 persons tested, we expect 2 positive results, one of which will be correct and the other incorrect. Therefore the odds on any positive result being valid are roughly even".
Response to this frequency format was 50-50.
It is not to say that such errors do disappear when probabilities are turned to frequencies but peoples are answer are usually better in frequency questions than probability questions. The likelihood of making such judgements are reduced are less likely.
References
A probability question is like: this taken from Gigerenzer,1994 article.
"1. A device has been invented for screening a population for a disease known as psylicrapitis. 2. The device is a very good one, but not perfect. 3. If someone is a sufferer, there is a 90% chance that he will be recorded positively. 4. If he is not a sufferer, there is still a 1% chance that he will be recorded positively. 5. Roughly 1% of the population has the disease. 6. Mr. Smith has been tested, and the result is positive. The chance that he is in fact a sufferer is:______.
What was found here was that most participants gave a response of 85% which is close to the 90% hit rate given in the example despite the 1% chance. Making such judgemental error is what is known as the base-rate fallacy. However when the question was switched to frequency readers could digest the question more easily. The frequency question is as follows:
"Out of every 100 persons tested, we expect 1 to have the disease; and the device is nearly certain to say that he has. Also, out of that 100, we expect the machine to say that 1 healthy person has the disease. Thus, in the long run, out of every 100 persons tested, we expect 2 positive results, one of which will be correct and the other incorrect. Therefore the odds on any positive result being valid are roughly even".
Response to this frequency format was 50-50.
It is not to say that such errors do disappear when probabilities are turned to frequencies but peoples are answer are usually better in frequency questions than probability questions. The likelihood of making such judgements are reduced are less likely.
References
Gigerenzer, G. (1994). Why the distinction between single-event probabilities and frequencies is important for psychology (and vice versa). In G. Wright and P. Ayton (Eds.), Subjective probability. Chichester, UK: Wiley.
Thursday, 20 October 2011
Probability & Frequencies. Decisions under uncertainty
When making judgement or decision we often use heuristics, also known short cut. For example when I think of a model, I think of tall, slim, beautiful person with an eye for fashion. Not to say am wrong or anything but then I don't take into account that a person could attribute all those things and not be a model. When making such judgement we use what is known as representative heuristics, where we look for similarities, most common and most frequent. We attach a high level of importance to certain information which lead us to these judgments. In my case I have attached the importance to tall, slim, beautiful and eye for fashion but in this modern world where people are very weight conscious and always following trend these kind of attribute could fit anyone besides a model. When I don't take into account the proportion of people who are like that as opposed to something else I make what is known as the base rate neglect.
Talking about the attachment of importance to certain things bring me my next point where people often assume that having success on a win will show having success on the next. For example in the casino when people first win, most people tend to think that if they've won then the would probably win in the second round. The russian roulette is based on randomness and having a success in first round does not predict having a success in a second round.
Talking about the attachment of importance to certain things bring me my next point where people often assume that having success on a win will show having success on the next. For example in the casino when people first win, most people tend to think that if they've won then the would probably win in the second round. The russian roulette is based on randomness and having a success in first round does not predict having a success in a second round.
Thursday, 13 October 2011
Decision Making, Good or Bad???
Decision we make are usually informed by our judgements we have about the issue at hand. We make decisions in our everyday life, "should I take the train today or walk"?, "Should I buy the red or blue top?". The action we decide to take is the decision.
When making decisions we usually weigh the outcome, the consequences. Sometimes we make decisions which we might think is good but others might disagree.
But what makes a good decision? Is it a decision everyone will agree on or a decision we believe is right? Sometimes we might make decisions that we believe is good but opposite to others. What clues do you draw on when making decisions?. Sometimes we want to wait till we are absolutely certain before we make a decision and most times we end up not making any decision at all.
Nevertheless we should always know that the amount of time that goes into a decision is what makes a decision good or bad. Taking time to think through a problem will make one come to a reasonable decision.
When making decisions we usually weigh the outcome, the consequences. Sometimes we make decisions which we might think is good but others might disagree.
But what makes a good decision? Is it a decision everyone will agree on or a decision we believe is right? Sometimes we might make decisions that we believe is good but opposite to others. What clues do you draw on when making decisions?. Sometimes we want to wait till we are absolutely certain before we make a decision and most times we end up not making any decision at all.
Nevertheless we should always know that the amount of time that goes into a decision is what makes a decision good or bad. Taking time to think through a problem will make one come to a reasonable decision.
Subscribe to:
Posts (Atom)