|
Mariel Gruppi , Universidade Federal de Minas Gerais
Collecting complete information has always been a concern in sample surveys, as well as refusal rates for items typically found in household surveys - i.e., income, education - have always been under the attention of experts. In parallel, few actions were done regarding the inclusion of questions about sexual orientation and gender identity in surveys. The implementation of these items has been slow, especially for developing countries such as Brazil, and little has yet been discussed in the literature about the quality of the data collected. Therefore, this work is dedicated to analyzing the non-response rate to the question of sexual orientation and gender identity in a sample survey carried out in one of the Brazilian regions. Hence, this work aims to characterize the pattern of non-response to questions about sexual orientation and gender identity based on sociodemographic variables, as well as to find the odds ratios for non-response considering covariates that may be important for predicting non-response. The results have shown the relevance of some of the covariates in the response refusal, such as race and marital status. Thus, it will be possible to contribute to the national and international literature regarding the implementation of questions that identify sexual and gender minorities, in addition to a better understanding for the implementation of measures that favor adherence and response to sexual orientation and gender identity.
Presented in Session 28. Flash session New and Critical Perspectives on Data Collection and Measurement