Investigating the Portability of an Automated Coding System of the Two-Step Method of Gender

摘要 这项拟议的研究汇集了一个研究小组，并利用了已经收集的数据， 美国大学健康协会国家大学健康评估（ACHA-NCHA），以检查 使用两步法和扩大的性别分组自动分类方法的可移植性 性别认同的选择。虽然两步法已被推荐为金标准， 评估出生时的性别分配和性别认同，如何分类和识别缺乏明确性， 性别扩张个体（即，个体的身份不仅仅是男性或女性）， 法这一点很重要，因为性别扩张的个人不认同强加的二元框架 根据出生时的性别分配和性别认同之间的不一致来分配性别认同。 此外，研究表明，健康状况（如物质使用）和 确定为变性人的个人与确定为 性别扩张我们的团队已经创建了一个自动化的编码方法， 少数民族（SGM）的人，考虑到多个性别选择，写在性别的反应，和性别 出生时分配的。在这项研究中，我们将改进和扩展编码方法，用于样本，包括 SGM和非SGM人：ACHA-NCHA。我们期望我们的改进的自动化编码方法可以 可以扩展到非SGM数据集，并可以改善使用写在响应的人的分类， ACHA-NCHA推荐的方法。我们还预计，我们的方法将大大减少数量 与ACHA-NCHA推荐的方法相比，我们还将测试 我们的方法能够更好地预测ACHA-NCHA推荐的物质使用参与 法这项研究将为严格和可复制的SGM健康研究奠定基础， 身份识别方法

ABSTRACT The proposed study brings together a team of researchers and leverages already collected data within the American College Health Association National College Health Assessment (ACHA-NCHA) to examine the portability of an automated method of classifying gender groupings using the two-step method and expanded options for gender identity. While the two-step method has been recommended as the gold standard for assessing sex assigned at birth and gender identity, there is a lack of clarity as to how to classify and identify gender expansive individuals (i.e., individuals whose identities are not masculine or feminine alone) via that method. This is important as gender expansive individuals do not identify with the binary framework imposed when assigning gender identity based on discordance between sex assigned at birth and gender identity. Additionally, research indicates that there are differences in the health status (such as substance use) and social determinants of health of individuals who identify as transgender versus individuals who identify as gender expansive. Our team has created an automated coding method developed with sexual and gender minority (SGM) people that takes into account multiple gender selections, write-in gender responses, and sex assigned at birth. In this study we will refine and expand that coding method to use on a sample that includes both SGM and non-SGM people: the ACHA-NCHA. We expect that our refined automated coding method can be expanded to a non-SGM data set and can improve classification of people who use write-in responses over the ACHA-NCHA recommended method. We also expect that our method will significantly reduce the number of people classified as missing when compared to the ACHA-NCHA recommended method. We will also test the capability of our method to better predict substance use involvement over the ACHA-NCHA recommended method. This study will lay the groundwork for rigorous and replicable SGM health research with robust gender identity methods.