Scales are the primary assessment method for subjective data in psychological science. However, validating the scale's hypothesized factor structure using a confirmatory factor analysis (CFA) requires a large sample, which is not always feasible. Therefore, we propose an approach to validating existing scales on a large sample without new data collection. By systematically searching for publicly available data from studies that used the original unmodified scale, a CFA can be conducted on the combined datasets. This approach is beneficial in cases where the scale has not been validated on large samples and where new validation is needed using samples from different populations and contexts. We applied this approach to validate two widely used scales in human–robot interaction research and found only moderate support for their original factor structures. In addition, we identified limited generalization across stimulus types, highlighting the need for validation to ensure appropriate scale usage.