Subject Respondent Robotics: Simulating Responses to Likert-Scale Survey Items Message body The semantic theory of survey response (STSR) proposes that the prime source of statistical co-variance in survey data is the degree of semantic similarity (overlap of meaning) among the items of the survey. This can be computed a priori through natural language algorithms. The known semantic structure of a survey can be used to compute missing values with an unprecedented precision. This study demonstrates the predictive value of STSR in an experimental way replacing increasing numbers of real responses with semantically predicted ones. A sample of 153 randomly chosen respondents to the Multifactor Leadership Questionnaire (MLQ) were used as target. We developed an algorithm where data from digital text analysis of the survey items served as input. As we deleted increasing numbers of real responses, we compared how the "robotic" responses compared to the "real" responses on usual psychometric criteria such as alpha reliabilities, score levels, and factor structures. Depending of the criterion for success, the robotic responses could replace This was not the case for the same algorithm if the semantic information was replaced with random values in the same range. Our study opens for experimental research on the effect of semantics on survey responses. Attachment See attached pdf-file