Comparing health survey data from Internet- and paper-based convenience samples of lesbian women in Germany.

Background: Members of populations that are measured online should be compared with paper-based samples in order to determine whether data from these different sources can be merged or must be analysed separately due to substantial differences. Methods: A sample of lesbian women recruited via the Internet were compared with a paper-based sample. Both groups used a questionnaire consisting of demographic variables, questions regarding dealing with homosexuality, psychological burden, access to care, and discrimination experience within the German healthcare system. Results: Only small differences emerged in dealing with homosexuality between both groups. Moderately more women in the Internet-based sample suffered from nervous tension and the impression of being out of balance. Moderately more women in the paper-based sample stated that their main contact person regarding health matters was informed about their sexual orientation. Most differences had small effect sizes. No substantial differences occurred between the two groups in the area of discrimination experience. No large differences in socioeconomic data, psychological burden, healthcare utilisation, or dealing with or experiences with homosexuality were found. Conclusions: Health survey data derived from Internet- and paper-based samples of lesbian women in Germany are not too dissimilar. Further research in additional contexts is needed to decide whether they can be merged for further analyses. [ABSTRACT FROM AUTHOR]