Sourcing for Diversity: Expanding Boolean Strings

Welcome to the second episode of sourcing for diversity, expanding Boolean search strings. In our first episode, we explored how to get started when searching for diversity and we covered topics such as defining diversity, avoiding tokenism, and creating inclusive environments. So today we want to dig into things a little bit more and talk a bit about the technical aspects of sourcing and focus really on Boolean strings.

So Robyn, let’s start things off with the basics. For those that may be unfamiliar, what are Boolean strings? So Boolean strings are a concept we learn about in math, combining different words and phrases together in order to create more robust searches. So with Boolean strings you use “and” “or” and “not” as ways to search different candidates.

So using my X-Men and Avengers example here, if you were looking for somebody who was an X-Men and an Avenger, you would find Rogue, Beast, and Storm because they both have been on both schemes. Right? If you were looking for X-Men who hadn’t been Avengers, you’d be looking for professor X, Gambit, Nightcrawler.

So and then X men, just X men, not Avengers you would get Gambit, Professor X and Nightcrawler. Whereas if you did X-Men or Avengers, you would get everybody who’s in this current circle. So by using Boolean strings, you can refine your search in order to find the candidates you’re looking for depending on your needs.

So it’s important to note that when you’re working with Boolean strings, that a lot of Boolean strings are based off of old programming language. So I actually recently learned this at source con. I’d always heard of stop words, so it’s just words that when you’re typing into Google you say, “Where is it

the closest steak and shake?” Google’s going to skip over the “where” “it” “is” and “the” and it’s just going to look for steak and shake, right? So some examples that are ignored by various search engines, are “she” “her” “they” “them” and then “go”, which I included because that’s the language for Golang.

A lot of examples of diversity sourcing for gender have talked about including pronouns, which are important. And a lot of people like myself here include their pronouns in their profiles now, but you can’t actually search using them depending on what you’re doing. So what I always recommend is when you’re building out these search strings, play around with them, add different words, try out different stop words to see if it’s actually registering or not, and don’t commit to one Boolean string too quickly. Moving on, as recruiters and sourcers we know it’s important to do research on roles that we might be unfamiliar with before we start our search. So we do this really to become more familiar with the language that’s used to describe particular skillsets, which is known as natural language. Can you maybe describe natural language searches for people that aren’t super familiar with that term? Sure. So a lot of people, they’ll scan the JD and they’ll just take key terms out of the JD and throw them into a search string, which is great, but you’re going to get

people who maybe only talk about their work in one way. So as you’re delving into a search, it’s important to just take the terms that you might not know as well. You might know a high-level version of, and just like read articles that are written by- my example’s tax because I’m an ex sourcer.

Right? So read articles that are written by developers about the stack, like intro documents, just so that you can get an understanding of different words that might be in the same framework, right? So Ruby, ROR, and Ruby on Rails, they all indicate slightly different things, but it’s the same language.

So you would want to include those in parentheses rather than just using the word Ruby. And you would want to make that as a big part of your Boolean string, including other parts of your Boolean string. You touched on it a little bit, but for those that are new to sourcing for diversity, like what type of research or topic should folks cover to familiarize themselves more with natural language, searching for diversity?

So when it comes to diversity, you kind of want to take the same approach that we talked about last time with avoiding tokenism but with a more specific tint towards finding out where people that you might be interested in welcoming into your organization exist in the world. So for example, last time

I suggested reading articles written by community members to get an understanding of where the community is standing, what problems they face, what issues they have. The next step with that would be to find out what organizations exist in the world that work for the communities that you want to include.

So for example, if you are trying to increase gender diversity within your organization and you’re looking for primarily non-binary people or women or people of any sort of identification, you would want to look for organizations like women in tech. That’s the only one I can think of off the top of my head, but also there’s so many out there.

So I would say like reading articles about the organization, women in tech, chick tech, those were two examples of just female focused groups and you would want to read about the organization, see what events they put on, see what events make sense for you to attend.

You want to make sure that as a recruiter, you’re not stepping into spaces that you’re unwanted in. For a lot of reasons, first of all, you’re unwanted in that space so that’s uncomfortable for everybody involved. But also it’s not going to endear you to people who you might want to recruit. So make sure that like, if you’re planning on attending any events in a manner that is not intrusive. Along with the tokenism, I said last time you want to listen to members of the community to try and learn their perspective and experience

but one step further than that is when you’re learning about different experiences and organizations, you can then use the organizations in your Boolean strings to help expand your search. So I can give an example of how that might look. I’m focusing on universities here in three different types of universities.

There are of course not women- these aren’t like the intent- I just wanted like a snippet of things to fit on a slide. A lot of people talk about HBCU, which are great at historically black colleges and universities. There’s also tribal colleges and universities that focus on Native American populations, if you are sourcing in the United States and you want to

make sure that you’re including everyone. And then also there are universities that focus on deaf populations- which not to brag, but my mom went to an NITD and then my uncle went to the Gallaudet. So I wrote here just a couple examples of these different versions of colleges. When you are sourcing using universities, keep in mind that these are universities that serve a predominant population-

they don’t only serve that population. So if you are trying to find somebody who’s deaf and you include Gallaudet in your string and you find somebody who’s hearing, don’t be like, “Hey, I actually wanted a deaf person, so nevermind” because that’s 1. Rude. It’s also counterproductive because even if somebody went to Gallaudet and they themselves are not deaf, they do know deaf people and they have deaf friends and they would probably refer deaf people into the organization.

It’s also important to make sure that when you’re sourcing for diversity, that you’re looking for all sorts of diversity and you’re not limiting to just one type of diversity. These are how you would combine University search strings to different natural language search strings that you might create. The Ruby on Rails example

I already demonstrated. Then the second examples, how you might expand a sales search string to include different indicators of good performance. Excellent. These have been super helpful tips to create more inclusive search strings. What are maybe some more creative ideas to help create more inclusive Boolean search strings?

So there’s a few more things, right? It starts off pretty easy, you know, just expanding your Boolean strings. And then it gets into just going off the rails and finding stuff on Twitter and LinkedIn. So I will start with kind of the more easy step, which is trying to find organizations that serve the populations you’re targeting.

So search for local groups that focus on diversity, organizations by and for POC is important. The by part is important. You want to make sure that you’re supporting organizations that are run by the population that you want to support. Alumni organizations, if you’re very limited in your university search student unions and student organizations. And then you kind of- when you start sourcing for these organizations, what you’re going to start finding are conferences, right?

There’s a few sources out there in the world and if you’re more interested on how to do this, so I would recommend going to source-con and hearing from them yourselves on how to find conference like lists and how to take those lists and extrapolate them and contact everybody on the list.

But if you have kind of just like a couple hours and you want to make sure that you’re sourcing for something, for example, chick tech, actually black and tech has a conference that they do every year. So what you would do is you would go to the conference page and you would look for hashtags that people are using for the conference on Facebook, LinkedIn, and Twitter.

And then you would go on those three services and see what you could find. People who are responding to panelists, people who are interacting with other comments, and this way you can find people who are really active in the space and really excited about technology. Also, you know, go to business events and spend a lot of time learning. Today we’ve covered a ton, lots of tips and tricks on hands-on sourcing for diversity and expanding Boolean search strings, and really finding more creative ways to make more inclusive searches. So stay tuned for the next episode, which is sourcing outside of the box.