ISSUE: Data obtained by interview or questionnaire are routinely reported with summary statistics, frequency, mean and range.  Can we obtain more informative results by interpreting the data using risk analysis techniques? OVERVIEW: Probabilistic methods can be used when data is sparse and address areas of uncertainty.  Quantitative analysis of interview or questionnaire data typically involves presenting frequencies of response and summaries of aggregate data with descriptive statistics. These are used often to address research questions including assessing market access opportunities, pricing and reimbursement scenarios, filling gaps in health economic data such as resource use. But do we always get the most from the data we have? How can we make better informed decisions? Let us consider a hypothetical question where respondents are asked to rate a series of attributes (A, B, C) using a scale of 1 to 10. We end up with a distribution of answers that are summarised as averages for each attribute. Often we are then faced with interpreting a series of average scores that do not differ markedly between attributes.  Uncertainty in the ratings provided by respondents can be used to improve our interpretation. One method would be to use bootstrap techniques, to sample with replacement the raw data, and running a simulation to obtain the bootstrap uncertainty distribution for the mean. From this we can determine the probability that attribute A is better or worse than B or C. With this information at our disposal we are better positioned to make an informed decision.  CONCLUSION: Decision analytic methods add value, improve communication about risk, support decision making, and identify research opportunities for reducing uncertainty when interpreting interview and questionnaire data.