Monday, January 28, 2013

Gender and the Internet

Society is tougher on women than it is on men, and the Internet doesn't offer a much-improved environment. For a long time, the internet had the stereotype of being full of men. A large portion of notorious online communities still maintain a male-heavy user base, and internet-savvy women will receive constant reminders about this.  For a very long time, it was a prevalent joke among online circles that there were "no girls on the internet". If a girl was to make her gender known in many of these same online circles, then other users would accuse her of lying. That, or post pictures of herself. Overall, this was not the most encouraging environment.

The internet has changed drastically since the days that these attitudes first came to exist. However, with the scope of the current internet and its many communities, it's difficult to pinpoint a general attitude towards the question of gender. The modern internet user is constantly bombarded with information, and it's easy to run into conflicting (and sometimes contradictory) indicators of the modern demographics and attitudes.

Let's try to sort through this issue. First, where do men and women hang out on the internet?

To answer this question, we'll be using Alexa, a company whose main service is collecting information on web traffic. Websites are ranked based on the volume of traffic they received, and are assessed for legitimacy of visits (bot or not), and the demographics of visits. Although Alexa's accuracy is questionable among lower ranking sites (and, like any other web statistics site, questionable in general), we will allay this concern by only assessing more popular and more established websites. This is also to our advantage when talking about what people mostly do on the internet, since the demographics of the 1000th-ranked website will be more representative of worldwide activity than the demographics of the 100,000th-ranked website. These won't be exact measurements, but they'll give us a good ball-park idea of where the gender lines are drawn along the web.

I'll also point out that I pulled Alexa data in mid-January, then sat on it for a little while. Some figures might have changed between then and the time of this blog's post. However, I am assuming that the statistics of high-traffic, well-established websites probably didn't experience an extreme demographic shift between then and now.

First, let's look at the demographics breakdown of the most popular websites on the internet. These are all websites ranked in the top 15 most-visited websites.

Social networks eke out a slight female lead, but overall, it's pretty even.

With all of these websites, the genders are very close to the center, suggesting an approximate 50/50 split.* This makes sense - Google and Wikipedia are websites that serve very general purposes. YouTube and the social networking sites also offer services that don't skew any preference to one gender or the other. Everyone likes to socialize, everyone likes to be entertained, and everyone can use a reference manual now and then. These websites' statistics are more telling of our humanity than they are of our gender roles.

Now, the astute reader will point out that I only picked five of the top fifteen websites on Alexa. What did I exclude? Well, Amazon, for one. The online shopping website does feature a more pronounced over-representation of the female population. However, the reason why I did not include it is because Amazon is not an internet community. As in, people don't go on Amazon to hang out. Google was included for its #1 spot in the ranks, and Wikipedia was included because of its open-ness to user contribution. However, Amazon is a market - a place for transaction. It's not really a place where people "hang out" for its own sake (like they would on YouTube or a social networking site), so I left it out. It's worth pointing out that Alexa is free to use, so if you feel like I've neglected to point a website out, then by all means, find it and point it out.

So, what about the other online communities? The social hubs where people can go on the internet to talk to one another? When would I start seeing some clear gender division? I trekked through Alexa's top 500 list to see for myself.

It didn't take very long to find what I was looking for.

Well, that was unexpected.

Livejournal is the lowest-ranked on that list, sitting at rank #139 in Alexa's top 500 list. If you exclude Livejournal and Instagram, then all of these were found in the top 50 most-visited websites on the internet.

Not only do these websites represent a significant amount of the Internet's traffic, they also boast a very large proportion of women online. In fact, Tumblr and Instagram is listed as over-representing women online, and Pinterest is listed as greatly over-representing women online. Women make up more of the blog-o-sphere than men as well, with the female demographic having a clear distance to the right of center.**

To re-emphasize, I'm omitting websites that don't quite fit my criteria of 'social hub'. I'm not including adult websites (because a: the results wouldn't surprise you, and b: nobody goes on those to talk to people), and I'm not including services like Netflix or Pandora (which also favors women more than men). We're talking about places that people can go to socialize and expose themselves to dynamic interaction with other internet users.

After poking around the rest of the global top 500 websites, I assembled other recognizable websites that function as social hubs:

Yes, game forums (like those found in IGN) count. Though, clearly not for much.

And then I started manually searching for other websites that weren't in the top 500, but were sites that I thought covered the breadth of other places that people interact. They all conveniently fell within the top 1000 websites in traffic.

Yeah, seriously. Digg's still in the top 1000. I was surprised too.

As someone who basically grew up with an internet connection, I came to be familiar with sites like Newgrounds, reddit, 4chan, and so on. Those were websites that other people online would talk about. I actually deliberately neglected to post rankings about other once-popular websites like ebaumsworld, homestarrunner, weebl's stuff, and YTMND, because it's hard to argue for their relevancy to the current internet audience. I also thought about including fark, gamefaqs, and Something Awful. I ultimately decided against it because IGN and gamefaqs have such similar demographics, SA only falls into the top 10,000 websites nowadays, and nobody seems to talk about Fark anymore.

However, their sites' traffic pales in comparison to the mostly-female internet websites listed earlier. Every single female-dominant website I listed in the second picture is higher ranked than reddit, which sits at 142 on global rankings. It's can't be said for sure, given the qualitative nature of this data, but it's possible that a website like pinterest or wordpress may have more female members than reddit has male members.

This, however, leads us to a very important point.

The jokes around "no girls on the internet"  are clearly unrepresentative of the internet. But it seems that the way that genders hang out online is pretty segregated. Someone on 4chan or reddit could legitimately be fooled into thinking that most girls don't spend much time online.

There are whole swathes of the internet that are primarily male-dominated, and whole swathes of the internet that are primarily female-dominated. The most integration that you see are on social networking sites and on YouTube. But a site like YouTube is so complex, it might be important that we break down its demographics even further. What if the genders hang out in different corners of YouTube?***

In any case, it is clear that men and women tend to isolate themselves from each other on internet communities.

Now, we must ponder why this is the case.

Is this a question of site format? Unlikely. Pinterest, tumblr, and instagram all have a heavy emphasis on pictures. However, there are male-dominated websites that emphasize pictures just as much - see 4chan, technically an imageboard. Is this a question of site content? Maybe, but it seems misleading to factor this in. After all, content on these websites is user-generated. A website's content reflects on its user base as much as it would influence it.

If I were to hazard a guess, the big reason that comes to mind is mainly a transitional one. There was once a point in time when these male-centric communities (4chan, newgrounds, etc) were the few social hubs that the Internet had to offer. Since the internet during this time period had not yet gone mainstream for its social services, the type of male that would normally be found on these forums probably had non-mainstream interests. The internet provided a common linker for people with these non-mainstream interests to talk to one another, which allowed these people to develop their interests and cultivate their outlooks on life, people, etc.

Because these interests and interactions were so removed from mainstream social scrutiny, the directions that those outlooks could take were relatively unrestricted. If you liked hobby X, and if you found an obscure online community where you found people interested in hobby X, and if most or all of those hobby X enthusiasts had other demographic points in common (male, white, etc), then maybe that would begin to influence your world-view. The creation of this sort of feedback loop by hanging out in isolated social circles has a term - cyberbalkanization. It can create some severe bias within these social circles, including bias against gender.

The development of popular internet hubs (like Facebook and YouTube for example) brought more traffic to the web. This meant that there were more people present online, from all walks of life. If the ventured deep enough online, they would find social circles previously shaped by a mostly male presence, with their own interests, inside jokes, and biases.

For lots of women, not only were they walking into a mostly male-populated subculture, they were also walking into a subculture that had developed around the idea of being a mostly male-populated subculture.

People like being a part of communities and groups, but if the existing communities are no good for you, then what do you do?

That's easy - it's the internet. You can simply start a different community.

Let's take 4chan and Reddit as examples. 4chan was launched in 2003, while Reddit was launched in 2005. Reddit gained its momentum after 4chan had already achieved its popularity. Reddit's growing audience emulated the portions of 4chan that it found funny (like its meme fixation), but without its seedier qualities. In fact, some people on Reddit like to refer to their website as some variant of "4chan lite", implying that the two websites are comparable, but one isn't as deep and dark as the other. Notice that, between Reddit and 4chan, Reddit has the more balanced gender ratio.

Newer internet projects also tend to have smarter marketing than the ones that precede them. 4chan was around at a time when the internet was still relatively young. Websites like YTMND and homestarrunner were at their peak during a time before Facebook. Newer websites, like Reddit, rose in popularity at a time when people were beginning to really latch on to social networking. This made it easier for new websites to reach out to the internet population, while old communities got buried.

People don't really try to fix internet communities. They just see what's wrong with other internet communities, and decide to make their own community instead. The new community promises to avoid the pitfalls of the community before it. The old community doesn't go anywhere, though; it persists along with its problems.

This brings us back to our female-weighted websites. Tumblr was founded in 2007. Instagram, in 2010. Pintrest, also in 2010 and through 2011.

Entrepreneurs are jumping into the website business on a very different internet than the one from ten years ago. The demographics are different, and the websites are offering a fresh start to online communities. They're addressing a need that older websites cannot supply, because the older websites built up their communities on demographics from a different time period.

Every new social hub on the internet is a new opportunity for you to be a part of something. And that's very appealing when you find it hard to be part of already-existing social hubs.

I have a feeling that this is a trend that will continue as well. As more new websites emerge that offer new chances at online communities, the older communities will find it harder to get favorable attention. It's hard to say if we'd be able to determine a "turnover rate" among social websites, but it's likely that newer sites will have more refined - and possibly more balanced - demographics. If the extremely female-skewed presence on newer sites is a reaction to the extremely male-skewed presence on older sites, then maybe newer websites will find a more moderate, integrated line. This could point towards newer, better-tempered attitudes toward diversity in communities.

On the other hand, maybe there will come a time when we hit a saturation point for social sites. There's far more competition between social websites than there was ten years ago, and it can only get harder for a new online venture to stand out. If there isn't a sizable new wave to bring more people online, then new websites will have to vie for the attention of already-settled internet users. Perhaps the current gender lines that are drawn online are going to stick around for a while. Perhaps that means that people will have to step up their efforts to fight the bad habits in developed internet communities.

Only time will tell.


*-Let's clarify what this information is telling us. The bars that you see measure the website's traffic against the general population of the internet. If a given gender on a given website has a bar that extends to the right, then that website over-represents that gender compared to the actual demographics of the internet. Likewise, if a given gender on a given website has a bar that extends to the left, then that website under-represents that gender. The closer a demographic's bar is to the middle, then the more similar the website's demographics are to the demographics of the general internet.

"But wait!" some readers might be saying at this point. "Since these measurements are relative to the total demographics of the internet, then shouldn't we know how many total men and women use the internet first?" Yes indeed, you're correct, and we'll define the general demographics of the internet for you as 50/50 men and women. This figure makes sense. However, we already acknowledge that there will be some skewing based on how Alexa collects demographic information. Let's keep this in mind as we very qualitatively assess online demographics. Okay, back to the article!

**-I should note that blogger and blogspot are listed as separate domains on Alexa, despite my Google searches of blogspot re-directing me to blogger. I ultimately went with listing blogger instead of blogspot because I couldn't find a standalone "" anywhere. This is important, because although blogger favors women to men, blogspot very slightly favors men to women, and is higher-ranked than blogger. Maybe blogger is an organization that came to replace blogspot, implying that, at one point in internet past, blogs were more male-dominated? Maybe blogspot exists by some trick of blogger's server, or perhaps the other way around? It's hard to say, but it's worth pointing out that this discrepancy exists. Okay, back to the article!

***-Luckily, Alexa can search statistics on individual YouTube channels, provided that you search the specific channel in the Alexa search function.  Let's look at the gender breakdown for the six most popular YouTube channels.

Ryan Higa and Rihanna are the most split down the middle. The rest? Eeeeeeh.

Despite YouTube's egalitarian total userbase, the visitors to its most popular channels are mostly male. If we are to assume that both collections of data offer an accurate picture, then what are all the women watching? Are the more female-centric youtube channels lower in rank, but greater in quantity (therefore spreading out views)? Is one gender simply more likely to visit the YouTuber's channel (as opposed to simply getting linked to a video) than the other? It's also worth noting that these individual channels all fall above the top 100,000 most viewed websites on the internet, putting them at a very far-removed range than our initial exploration of the top 1000.

Maybe this is worth exploring in greater detail in a future blog post. But, this line of inquiry is kind of a tangent to this blog post, so maybe we'll talk about it some other time! Okay, back to the article!

1 comment:

  1. Kurti tops can be worn with a salwar, churidar, leggings, jeans and even capris. you can get all these kind of quality shopping from this shop buy kurtis online. We provide great quality kurti tops and capris for as well as corporate kurtis and office wear kurtas. kurtas for women