Comparison of experienced and novice cost estimator behaviors in information pull and push methods

Semiha Kiziltas, Burcu Akinci, Cleotilde Gonzalez

Research output: Contribution to journalArticlepeer-review


Construction cost estimators mainly use analogy-based decision-making processes while estimating activity production rates in new bids. They need to search information items that will help them in understanding the differences and similarities between the current and previously completed projects to select appropriate production rates and make necessary adjustments to the selected production rates. The objective of the study is to understand the effect of information pull-push methods on the behaviors of cost estimators when they refer to information items from historical sources in structuring their decisions. Eleven experienced estimators and 11 novice civil engineering candidates participated in an experiment involving card games, which simulated the information pull method, and in another experiment involving a prototype system, which simulated the information push method. Results showed that novices can behave like experienced estimators when information relevant to a decision is pushed to them. This result has implications for the design of information systems. Results show that it is possible to improve the estimates of novice estimators by providing them with decision support tools designed with the right information delivery method for their level of experience.

Original languageEnglish (US)
Pages (from-to)290-301
Number of pages12
JournalCanadian Journal of Civil Engineering
Issue number2
StatePublished - Feb 2010


  • Construction cost estimating
  • Expert-novice comparison
  • Historical data
  • Information push and pull
  • Menu-driven information delivery
  • Project histories

ASJC Scopus subject areas

  • Civil and Structural Engineering
  • General Environmental Science


Dive into the research topics of 'Comparison of experienced and novice cost estimator behaviors in information pull and push methods'. Together they form a unique fingerprint.

Cite this