There is an increasing need for all radiologists, particularly pediatric radiologists, to have a good working knowledge of the relative radiation dose of the examinations we perform or supervise. We must be capable of providing such information to our referring physicians, patients and their families, for internal quality-assurance purposes and in the future, perhaps to outside regulatory or advisory bodies. In an ideal world, there would be a universally accepted common dose parameter applicable to all medical examinations involving ionizing radiation. That parameter would be derived with a user-friendly standard methodology that accounted for age and developmental variability and involved minimal extrapolations and assumptions. We do not have such a parameter. Currently, the most practicable common dose assessor is effective dose (ED). Defined by the ICRP and measured in miliSieverts (mSv), ED is the sum of the absorbed doses in all tissues and organs in the radiation field, each multiplied by a tissue-specific weighting factor. It provides an assessment of overall radiation detriment from a non-uniform dose distribution in terms of a uniform or whole-body exposure. The merits of ED are that it allows us to compare the relative dose impact of common radiographic, fluoroscopic, CT and nuclear medicine examinations, enables comparison to annual background radiation, and can be used together with risk models to provide a broad estimate of future excess malignancy risk. However, there are reservations associated with its use in medical imaging, certainly when applied to an individual patient. Calculation involves multiple levels of intrinsic uncertainties, extrapolations and error bars, generating an ED value that is a broader estimate than often appreciated. Expert assessment suggests that we should expect an uncertainty of +/−40%, whether using organ dose and weighting-factor methods; Monte Carlo simulations; or age-, regionand modality-specific conversion factor methods. Systematic and random components to error exist at each stage in calculation, and for all modalities. A few of the many potential contributors include dosimeter thresholds, accuracy, and placement; inter-observer variability in field of view specification with implication on partially included organs; whole-population based (not pediatric) weighting factors that are periodically updated with new epidemiological evidence; console-displayed CT dose parameters based on previously calculated standardized cylindrical phantom doses, the use of Monte Carlo or conversion factor data derived from older CT scanners, and uncertainties in models of radionuclide uptake, distribution and retention. A decade ago, there was a paucity of available ED data for the practicing pediatric radiologist. Isolated publications addressed individual examinations, mostly radiographic and fluoroscopic studies, and there was some dose data from national surveys but rarely providing ED values. Although conversion coefficients were available, they were not widely appreciated beyond physics circles. There was almost no dose data for pediatric CT. Some ED values, when investigated, did not relate to children but to the more extensive, although still not comprehensive or collated, adult-based data. In the years since, publications involving organ dose measurement and ED estimation in pediatric imaging have escalated. The information available to all of us has significantly increased, although there is ongoing debate as to the merits of various approaches and a need to improve the methodology and inconsistencies that especially plague pediatric CT dosimetry. Disclaimer Dr. Thomas has no financial interests, investigational or off-label uses to disclose.