Diversity Filter Coverage – Women’s Names

booleanstringsBoolean, Diversity 3 Comments

Diversity Sourcing is not easy. There is no clear way to search for diversity categories on social sites or Google. Our less-than-perfect but necessary approach consists of “shortcuts” – ways to search that are likely to bring up groups of potential diversity candidates. As an example, Jonathan Kidder has a collection of diversity Boolean strings. Glen Cathey’s blog offers quite a few approaches. (In addition to using the shortcuts, we sift through “everyone” in professional search results and notice other profiles of interest.)

It makes sense to combine all possible “shortcuts” – such as the search for female names, associations, schools, etc., for the most inclusive approach. It also makes sense to get an idea of how large a population every approach covers.

Here is some research on finding professional women by the common first names in the US. (For other countries, this exact approach may or may not be possible – I am sharing it as an example).

First, I got a list of 1,000 most common female names in the US by Googling. I also found lists from the Social Security Baby Names page. In the end, the results for the names from both sources were similar. (But the popularity of the names has varied over the years.)

I did the research on LinkedIn, so it is affected by the data LinkedIn has – but it is LinkedIn that we use to search, so these conclusions are not affected by people (many!) who are not members.

I created a 1,000-long OR name string and tried it in LinkedIn Recruiter. While it was “too much” for it to search, it did show the numbers.

In the US, LinkedIn has 49% female members (compared to 43% worldwide) and a total of 170M+ members, i.e., it has 83M+ women. Our OR name search has found 73M+ results. So, a search for 1,000 common names (which you would have to do in portions to get results) amounts to 87% of women in the US. (If you were wondering about 1,000 common men’s names, the percentage is even higher.)

The results are affected by LinkedIn’s first names interpretation, which cannot be turned off – we cannot search “verbatim” (either in Recruiter or business account). It affects our results in good ways since we will see also nicknames and variations. However, some names (like “Andrea”) can be both men’s and women’s, and LinkedIn’s variations will include “Jerry” and “Gerry” if you search for “Geraldine”. That introduces false positives. But, examining results by adding extra filters, I can tell that the percentage of those is small. (We look at each result when sourcing anyway.)

There are name variations across locations. Here in the San Francisco Bay Area, the population is diverse ethnically, which results in fewer “American” names found – more like 50%, but it is still a high percentage. (To increase it, we can add lists of ethnic names). If you narrow to industry – for example, Software – the numbers will go down – but it only reflects the uneven numbers of men and women in the industry.

If you compare the high percentage we got with numbers for other filters (such as “she” or “her”, schools, and  memberships), it becomes clear that name search is powerful. The conclusion is – for the US, female name search is an excellent filter. Just make sure you include 1,000 and search in other ways as well.

Is your team sourcing for Diversity? Join us for the Certified Diversity Sourcing Professional (CDSP) Program August 2021! (June is now sold out.)





Comments 3

  1. Pingback: Diversity Filter Coverage – Minority Times

  2. Pingback: Three Ways to X-Ray LinkedIn for Diversity - Diversity Employer

  3. Pingback: Scraping Values to Feed the Boolean Builder | Boolean Strings

Leave a Reply

Your email address will not be published. Required fields are marked *