|Authors||M. Jørgensen and S. Grimstad|
|Title||Avoiding Irrelevant and Misleading Information When Estimating Development Effort|
|Afilliation||Software Engineering, Software Engineering|
|Publication Type||Journal Article|
|Year of Publication||2008|
Software development effort estimates are reported to be highly inaccurate and systematically over-optimistic. We provide empirical evidence that suggests that this problem is caused, to some extent, by the influence of irrelevant and misleading information, e.g., information about the client's budget, present in the estimation material. The only really effective means of eliminating this influence is to avoid exposure to this type of information. Other means, such as more use of formal effort estimation models, improved analysis of requirement specifications, and better selection of estimators, had a positive effect but did not eliminate the influence. We propose process elements that are designed to avoid irrelevant and misleading information and illustrate how this process may lead to more realism in effort estimation.