Making the Study
​How did we do it?
To quantify LGBTQ+ history, we recorded the entries of archival materials into a dataset. We input information such as name, ethnicity, gender, sexuality, and many more in order to make direct comparisons across each entry of data.
​
​
​
Unit of study
When designing our dataset, we wanted to question how history is represented. For example, a local LGBTQ+ activist may get only one entry at their local archive, but famous figures like Sir Ian McKellen may get multiple entries from several archives. Should they be considered to have the same weight in the dataset (as a person) or different weights (with different entries)?
​
​We decided that the audience sees LGBTQ history as entries, and therefore histroy is represented as entries. As such, the unit of study would be entries, and famous figures would be more weighted.
​
​
Is this representative?
It may seem that materials chosen by a small team of students are not considered a representative sample. We disagree.
​
​We believe that we are not only researchers but also the audience of the LGBTQ+ archives. The majority of the audience is very similar to us, in being interested and curious about LGBTQ+ history. They also have similar, if not less competent, tools to locate these materials. Therefore, the data that we have chosen and have accessed can be considered a relatively representative sample.
Validity
We did a lot of research to ensure validity of our data.
​
Language-oriented:
Labels such as “gay” and “lesbian” are used differently in the past. We record the language used in each resource. To complement this, we also attempt to record demographic information. Therefore, (at least) two pieces of information are included for each entry.
​
For instance, historically, people used the term “gay” to refer to men who were feminine as well as those who identified with homosexual sexuality. Therefore, some trans people may also be mistakenly recorded as "gay", so we need to research enough materials to identify their "true" identity, while simultaneously keeping the original language used when the term was stigmatized; hence they may not be recorded in the archives.
​
Outside research
​
Information on gender and sexuality may not be explicitly stated in archives. Therefore we have the “Not available” option to account for this absence.
​
​We also conducted research ourselves from outside sources, to determine certain demographic information. We documented the amount of research and made sure they are also within public reach and counted as "representation".
​
Limitations
​While quantitative data provides representative data that is directly comparable, it does not paint the full picture of LGBTQ+ history. The experience of LGBTQ+ people has been fluid, changing, personal: these features mean that quantitative methodology may not be the best strategy to record LGBTQ history.
​
​Gender identities are changing. Many individuals in our dataset have taken on multiple gender identities in their lifespan. Even though our dataset can accommodate the recording of multiple identities, the dataset cannot record the reason for these changes, the direction of these changes or the factors influencing them; this is why qualitative research is important. Qualitative research finds personal and contextual details; it also discovers new factors for potential quantitative research (for example, police brutality and Stonewall).
