• Replies 5 replies
  • Subscribers 359 subscribers
  • Views 64 views
  • Users 0 members are here
Related

Interesting article on gender values in databases

Levi Sauerbrei
$organization

Some time ago I had made an inquiry with Tess support about the possibility of altering the potential values of the gender dropdown in the application. The answer at the time was that making such a change wasn't possible in the current application, but that some consideration might be given to it in the design of the NextGen product.

This morning I came across a blog post from one of the developers on the Diaspora project. (Diaspora is intended to be a Facebook-esque social network in which the backbone consists of "nodes" installed onto servers by the users of the system.)

The blog post, Why Gender is a text field on Diaspora, is a brief look at the developers thought process in making gender a very flexible piece of data.

Four years ago, at my first rails job, I worked at a company with a mostly-*** customer base. It turns out, in that context, knowing if someone is “male” or “female” gives you almost no useful information. The *** community has other widely-accepted categories of gender, but the company’s internal order tracking software — a well-known package from a national vendor — offered only male or female.

...

I made this change to Diaspora so that I won’t alienate anyone I love before they finish signing up.

The comments seem to be running about 50/50 between "So happy you've done this!" and "Meh".

I know there are a lot of design considerations being made in the planning for the NextGen project.  Just wanted to share this link as a bit of a thought provoker and conversation starter.  I suspect that some Tessitura licensees find themselves with unique constituent populations that might benefit from this type of flexibility.

-Levi

 

Parents
  • $organization
    Kudos for raising a sensitive but legitimate issue, and for asking a good question: why?

    I'll add a Marketing department perspective on how we might use gender: who is making decisions in a family is an oft-discussed topic.  Moms are often said to be the main decision makers regarding activity programming for the family.  In a recent nation-wide survey of museum patrons, entire segments of data analysis are being given to analyzing what "kinds" of moms and what "kinds" of dads use museums, how those groups differ, and how we can get under-engaged fathers more involved.  Movie theaters have "Movies for Moms" type activities; a museum (or other institution) might want to offer something similar.  Having a gender field allows you to identify these customer segments, so that you can address the "Movies for Moms" postcard to any female adults on the membership, for example.  Similarly, an institution might want to target some marketing or events at dads, and would need a similar field for that purpose.  It would also help in data mining; if I understand the theoretical new architecture correctly, we'd be able to see which parent (male vs female) actually made the reservation for a given visit, which speaks to my first point about who makes the plans for the family.

    I'll also say that personally, I think it's always wise to consider all your options, so I'd love to see us consider how to accommodate various gender identities; however, I don't think I'd try to find the solution by myself.  If it were up to me, I'd do some quick research on the top gender-identity organizations in the US (and other member countries) and request a quick consult on their organization's perspective.  Better to ask people what they want directly than to try to guess on their behalf.  Free-text may not be the only - or best - option.  Let's ask the experts.

    Beth

Reply
  • $organization
    Kudos for raising a sensitive but legitimate issue, and for asking a good question: why?

    I'll add a Marketing department perspective on how we might use gender: who is making decisions in a family is an oft-discussed topic.  Moms are often said to be the main decision makers regarding activity programming for the family.  In a recent nation-wide survey of museum patrons, entire segments of data analysis are being given to analyzing what "kinds" of moms and what "kinds" of dads use museums, how those groups differ, and how we can get under-engaged fathers more involved.  Movie theaters have "Movies for Moms" type activities; a museum (or other institution) might want to offer something similar.  Having a gender field allows you to identify these customer segments, so that you can address the "Movies for Moms" postcard to any female adults on the membership, for example.  Similarly, an institution might want to target some marketing or events at dads, and would need a similar field for that purpose.  It would also help in data mining; if I understand the theoretical new architecture correctly, we'd be able to see which parent (male vs female) actually made the reservation for a given visit, which speaks to my first point about who makes the plans for the family.

    I'll also say that personally, I think it's always wise to consider all your options, so I'd love to see us consider how to accommodate various gender identities; however, I don't think I'd try to find the solution by myself.  If it were up to me, I'd do some quick research on the top gender-identity organizations in the US (and other member countries) and request a quick consult on their organization's perspective.  Better to ask people what they want directly than to try to guess on their behalf.  Free-text may not be the only - or best - option.  Let's ask the experts.

    Beth

Children
  • $organization in reply to Beth Varro

    Beth,

    Thanks for a marketing take on things.  Being a DBA means I love data and its always good to see how it is being used by different disciplines.

    My hope would be that the gender field would become something that organizations could customize to their needs (either through system tables or something similar).There will obviously be organizations who never touch the default values, but for those who do see a need to customize it, it would be a great tool to have.

    There was some follow up to this conversation on the web through a few other blog posts.  The most interesting one I saw was from Metafilter.com.  One of the DBAs did some quick analysis on their database.  It is a realtively small sample size (< 10,000 records with a value entered for gender) but even with a free text field in place, they were still able to identify over 75% of the values as indicating a preference for being identified as either male or female.  The full comment has a detailed breakdown of the remaining values (along with his caveats), but it is interesting to see what kinds of things people will enter when left to their own devices.