Four Point Elicitation Process

Experts have to settle on in the four point elicitation process a lower limit, upper limit, best supposition and self specified rating of the confidence with which the interval surrounds the truth. The expert has to answer the following questions.
1. What might be the logically lowest value?
2. What might be the reasonably highest value?
3. What might be the realistically most likely estimate?
4. How convinced s/he is that the interval produced from the lowest to highest will confine the true value?
The above mentioned value will be a number between 0% and 100%. Various studies propose that the four point procedure is significantly improved than the three point elicitation procedure. Although the four point elicitation process does not surmount

It exists for a number of reasons including natural variation, measurement error and bias, insufficent data and subjective judgement. Essentially epistemic uncertainty means there is a 'truth'; we just may not know it. The first four papers of this course taught you how to interpret, evaluate, capture and analyse data to take a scientific approach to finding out what the 'truth' may be.
Variability refers to natural biological variation or heterogeneity in a population- for example in food consumption, worm burdens, air temperature. The liveweight of chickens is a simple example of variability as liveweight varies within and between populations. One of the differences between incertitude and variability is that variability cannot be reduced through the collection of additional data.
The variability is going to be a source of uncertainity in your model. Is it worth collecting data to refine the estimate? Explain your

observation and selection bias). You learned that these factors can impact on the results of an epidemiological study in such a way as to make us question the internal and external validity of a study. In risk analysis, the effects of bias in the parameter estimates we use could result in an incorrect estimate of risk.
Imagine a test with poor sensitivity is used in a survey to estimate the prevalence of paratuberculosis in a cattle population. If we were to use the information from the survey as a measure of the likelihood of infection in imported cattle we could underestimate the risk.
Subjective judgement
Subjective judgements are those that are influenced by people's beliefs, feelings and values. Often we have insufficient data and have to rely on an expert’s judgement. These judgements are by their very nature subjective because they are based on the person's opinion, even if it is a well educated guess. In the second module, we will explore a range of methods that can be used to reduce uncertainty when we are eliciting an expert's