This document discusses the relationships between tests and scales, populations, reliability, and validity in quantitative research methods. It provides examples of using surveys and experiments to collect numerical data from populations about their opinions or the possible outcomes of situations. Specifically, it describes an experiment conducted using blog debates that surveyed a population of voters on their preferences for online debate formats using a Likert scale. The scale was found to be a valid measurement tool for that population and could potentially be used for other populations when collecting Internet-based responses.