Two researchers from the University of Glamorgan found that when it comes to web design, what appeals to men doesn't float a woman's boat — uh, can I say that in public? The study focused on personal websites designed by college students at Glamorgan and then moved on to sites designed by French and Polish university students — I assume to measure if the conclusions were regional, which they weren't. The study's findings further concluded that women prefer websites designed by women.
"We started off by looking at the personal websites created by 60 university students, 30 male and 30 female, to discover whether there were any major design differences. We looked at factors such as language, visuals, and navigation - the differences were immediately apparent," explained Gloria Moss. "The statistics are complicated, but there is no doubt about the strength of men and women's preference for sites produced by people of their own sex," said statistician and co-researcher Dr Rod Gunn.
As a side note, in response to that line, "The statistics are complicated...", I'd like to know exactly when statistics are not complicated. I only know one man who had the ability to walk a person through a statistical data and the mathematics used to obtain the results without completely confusing the person and making them cry like a second runner in a Ms. America pagent. If you ever find yourself having to take Applied Statistics (AS300) at the University of Alaska Anchorage, fight your way into Hienz Noonan's class. That man is a well dressed genius, yet plain spoken and writes with nothing less than a Mont Blanc.
On a side, side note, I am happy to see the researchers included navigation as a design element in this study. I'm still amazed at how many people just shrug when it comes to the importance of navigation when it comes to the behavior of the web user. While I have yet to come across one navigation scheme that appeals to one sex over another it's still an important science that should not always be left up to borrowing conventions used in other websites.
Moving on, the report found that men are drawn to design with "straight lines straight lines (as opposed to rounded forms), few colours in the typeface and background, and formal typography." Uh hello?, can you say Subtraction? I need not say more.
As I could not find the actual report, I'm not sure what the study found to be appealing to women, but based off the comments left in the press release I left to assume that women like design that doesn't contain a lot of right angles, more colors than fewer, and typography that can be worn to a business meeting but dressed down for a cocktail party. In other words: a loose interpretation of a grid system, a color palette with more than four values, and it's time call Helvetica and tell her the love affair is over.
Ah Helvetica, It's like wearing a smoking jacket to breakfast.
Again I wish I had the full report to know what they interpret as having a female bias. I don't refute the findings but it would be interesting to read some of the remarks made by the women in the study.
I blame the male ego.
In the last ten years I've worked in a handful of positions that often put me in the middle of the decision making process. These decisions ranged from creating a color scheme to devising a strategy for raising eleven million dollars and some change. No matter the objective, the number of hours spent "planning", or the days spent devising "strategy" one key criteria was always missing: user/customer input.
I have to blame the male ego because whenever I would bring up the need for user input (male and/or female) you would have thought I had asked if everyone was sure they had a penis. Over the years I've heard every excuse for not needing to consider a customers preferences, from years of experience to the degree and length of a person's graduate education. Of course this wasn't always the case, there were times when user input was agreed to and sought after, although occasionally the results were questioned and the group reverted back to their *cough* dangling intuition.
It was never evident to my peers and superiors that being oblivious to user preferences was a big strategic error. And it happened time-and-time again. Had these meetings been a Major League Baseball game, every one of them would have earned at nice fat 'E' next to their name.
No matter how hard we will try, it's not possible for men to understand women, but it is possible for us to design to their needs and liking. This report's conclusion suggests that men can not possibly channel a woman's design preference and I'm not sure if I buy that. I mean if Guccio Gucci can do it so can we all, but it requires knowing what's important to the intended audience and that most definitely requires research: asking questions, listening to the responses, and formulating the appropriate art direction.
Join the fray by reading through and commenting at the end.
Good grief. I would like to have seen some examples. Say, like "Airbag" is from Mars, "Dooce" is from Venus?
You don't think Helvetica can be worn to a business meeting and dressed down for a cocktail party? I think I disagree. The beauty of Helvetica is in it's versatility. :)
In any case, interesting read. Would be nice to see the full report. This is all actually astonishingly relevant to a project I'm working on with a small visual design team consisting of three men and two women. At the moment, we're having a hard time agreeing on certain things, amongst them the female's preference for a much larger color palette than us men were striving for. Interesting.
Helvetica is like Christopher Walken, brilliant when used sparingly and in the right role.
Greg-
I agree (and nice analogy), but I think that stems primarily from it's overuse in past decades, more than from the relative beauty of the typeface itself. Helvetica is historically great in many roles, but we're tired of seeing her, and this makes her less effective. Thus, use her sparingly and in unique roles. :)
Yeah, I've found that girls who visited my site before my redesign didn't like it as much as my current version (though I didn't either, hence the redesign, :) ). Anyway, my old site was basically greyscale and this one is in color, some of it vibrant.
I'm with Beerzie -- couldn't they have given us some examples? I'd love to see the sites they had participants look at. This would be a tough study, I would think -- what if the more "feminine" sites had design flaws that had nothing to do with the "genderness" of the site, thus skewing the statistics?
I guess you could spend some time at csszengarden and see if you can spot a gender difference in the entries there. All the content is the same, so that shouldn't bias things.
"strong tendency for each sex to prefer the output of its own sex" if that's true, then the female users prefer 7% of the web? No wonder all the bloggers have that macho attitude!
So far all the comments have been male.
Men and women typically differ in preferences when it comes to a lot of things -- clothing, cars, professions, etc -- so it isn't all that surprising if the same principle applies to web site design. However, a generalization such as this -- should we call it a genderalization? -- only goes so far; I'm quite sure there are men who prefer "feminine" designs and women who prefer "masculine" designs as well, just as there are young people who prefer traditional designs and older people who prefer trendy designs. The key is to identify the drivers behind your target audience's actions and tastes, and tailor your product with those drivers in mind -- be they gender, social status or whatever.
Glamorgan is spelt Glamorgan, not Glamoragn!
Its weird to see your Uni mentioned when you open up Airbag, of all places..
The key is to identify the drivers behind your target audience's actions and tastes...
Which is exactly why I'd like to read the actual findings.
Glamorgan is spelt Glamorgan, not Glamoragn!
Mybad. So Ben, can you go in as an alumn and get a copy of the report?! Tell them you'll donate a new library or something.
I dont go back to Uni for a couple of weeks, so I did the next best thing - I emailed them and asked!
I'll get back to you if I hear anything.
Ben
I'm curious to know if the study conducted any surveys or acually talked to any actual users.
Or if they just took 120 websites designed by college students, and based their study off of that.
Being a college student myself, I know that many of the personal websites of female students (at my university anyway), are not typical of your typical website. They do have large type in some font resembling comic sans, they are filled to the brim with animated .gif images, they have pictures of stars, hearts, and their puppies. Then thrown into the mix is a little bit of whatever they feel like.
I'm not trying to go against the study, or say that it is not reputable. I am just curious if they based the study off of anything else, or just studied personal websites designed by college students.
typical of your typical website.
wow, its early in the morning and I can't find an edit button... bear with me =/
Intersting article, but like others have said, it would have been nice to see the actual report - maybe Ben will be able to get his hands on it for us to share. A better approach would have been for the study to offer a good control group of websites (such as CSSZenGarden as mentioned above and popular ecommerce sites are a perfect example of good design with the same content), assuming that they did not do this. Then they should have spent some time finding men/women of different age groups with different social/ecomnomic backgrounds for their user set. If they had done this, I would think the study would have had shown less of a gap between the genders. You must also consider what the intent/purpose of the site was (personal to showcase work and family pictures versus ecommerce to sell something) as this would also change the results. In addition, men/women in different countries would show slightly different results as we all have biases that are learned from our environment.
It wish someone would do a study of this magnitude... I am sure there must be a graduate student in statistics or computer science who could find grant money for this type of study...
Thanks for the interesting commentary on the article - it made not having the actual report bearable.
This report raises some interesting questions. I mean, how do I know if my site is effeminate or not? Are there wedgies and locker-stuffings in my future, because I operate the virtual equivalent to a transvestite? Or perhaps my site is only appropriate for fraternity parties and ultimate fighting championships? I should have created male/female oriented style-sheets...
Hey... that's actually not too bad an idea...
"No matter how hard we will try, it's not possible for men to understand women, but it is possible for us to design to their needs and liking."
Isn't knowing anyone's needs and liking a substantial part of understanding them?
A lot of this report jives with what I've experienced working as the lead designer for a women's college website. Our team has one senior woman on the team, who has helped our designs evolve by serving as a kind of "reality check" before and after user testing. My personal experience is that there are often dramatic differences in gender preferences, and have had to learn over the years what these are, given the predominant female audience we design for. This being said, I'd modestly point to our site as an example that male designers can, with reflection, discussion and feedback, "channel" female design preferences successfully.
LintHuman, yes, I think that is a crucial element to tapping into the targeted audience. How can you pull off a successful campaign without understanding the needs and tendencies of your audience?
Ken, I like the tidbit about Gucci and how it is possible for men to design something that appeals to women. However, in regards to Gucci specifically, he is respected as a designer for trend-setting clothes. I don't think it has to do with design as much as it does with the appeal and allure of following current fashion trends. After all, that is what appeals to women.
I would like to see a copy of the full report as well for the reasons that have already been presented.
jish, just to clarify, animated gifs and puppy dogs don't appeal to all female college students. I think that is a generalization that should be avoided. However, college students as a whole fall short of mass reach online and aren't exactly qualified for such a study.
So far all the comments have been male.
Miko, we airbag readers are not all male.
We're not even all graphic designers!
Personally, I prefer colorful, clean, and fast loading pages. Much like this one, actually.
This study has helped me understand that I am actually a man.
I think that it was a great fault of the study that they only researched college-designed websites. I recall the student designed websites in my college, and if it didn't make me vomit, then it often appealed to me greatly.
From what I know of the study (primarily the press release) they did some very bad science.
First, they started with personal sites created by students. They make no mention of normalizing that sample for such things as the designer's past experience with computers or website design. Neophytes often go crazy with colors, graphics, and personal stuff. As they gain more experience with the Web, both as a user and a designer, they move to a "less is more" style, upon realizing that their gaudy pages are hindering what they're actually trying to communicate. For cultural reasons, young men often have significantly more computer experience than young women, and that experience is more functionally-oriented.
Then, they showed them to a test group whose makeup they do not specify but, seeing as it's a university study, I'd guess that it consisted of undergraduate students. (the universal guinea pigs) Once again, they apparently made no attempt to normalize for computer and net experience. Sites designed by newbies tend to appeal more to newbies. (is there a "newbish aesthetic"?)
Also, it appears that the site content was retained, from the mention of formal language, etc. Not replacing it with the usual "lorem ipsit" (with apologies to Cicero) gibberish means that the site content influenced the opinions, and in ways that relate only to personal/vanity sites. That brought a social context into it. The exact same female site-raters might have evaluated online bookstores, for instance, totally differently. They expectia social experience from a personal site, and hence lean towards the most social looking ones, sites that appeared to be made by people they would interact with socially, whereas if they were buying books, they'd want a site where they could order their books as efficiently as possible and get on with their day.
Apparently that's where their usage of actual human website evaluators ended. They created a list of characteristics for "appeals to males" and "appeals to females" based solely on which personal websites appeal to college students of similar background to the website creators. Then they treated that list as an authoritative evaluation for other sites which were not personal sites created by college students, with absolutely no justification for why this would be valid. They rated those sites not on their actual appeal to actual men and women, but on whether certain elements of their design were the same as those found in vanity websites created by college students.
Finally, they considered only one factor in the differences: sex. They apparently did not take any other factors into account, such as amount of online experience, which (especially for such a young group) might split along sex lines. They extrapolated from personal/social websites to commercial websites with no apparent justification. It seems like they had their conclusion in hand at the start, and went seeking evidence to support it.
I think the key text is one line at the bottom: The University is now offering a consultancy service to businesses seeking advice on tailoring their websites to their business goals.
Women do a lot of things better, but I prefer my websites to be designed by men. For now.
And how would you know who designed them?
Does "not floating their boat" mean that they don't have a "Glamorgasm" on this web site?
Wanderer-I like the your assessment of the study's press release. I agree; the study means nothing unless they show that they took all measures to make the study as airtight as possible. The press release makes me uneasy; all it tells me is that college girls like college girls' personal websites more than they like college guys' personal websites. The transition from personal websites to commercial ones bothers me the most, perhaps. Though I believe Shaun when he says that there are distinct differences, the press release isn't giving me much to work with.
This being said, I think it's funny that my current website just happens to be using another person's template (technically less savvy) and was chosen because it's "colorful". Note also, the softened edges and lack of right angles...
Nod- my mistake.
Great post, and thanks (I think?) for the nod to Subtraction as an archetypal example of male design tendencies. It's ironic because I may design in a very manly way, but everyone says I throw like a girl.
Khoi, your inability to throw is nothing that can't be cured at Man Camp.
you know, i noticed that when i moved the sexy pictures of me and the other writers on our blog up from the bottom of the sidebar to the top the stats went through the roof...go figure!
Granted that this, indeed, is an interesting post, but I've just got to know, are you updating your feed purposely so many times just to get people to read it or what?
Every day since this post was written (Friday), we have seen at least 3 "updates" a day. This could all be due to my using Bloglines and nothing to do with your website. I still had to ask though.
Could the repoeats be due to the number of comments being included in the text in your feed?
I've received 25 copies so far with 31 comments to your post.
*blush*
Yes you are right Michel. I didn't even realize the feed had the number of comments listed in it. It's not like I haven't been reading Airbag for a few years now or anything ;)
Greg, did you recently start including the number of comments in your feed or I am just oblivious the fact that this has been going on for quite some time?
ugh *repeats*, of course!
Greg, did you recently start including the number of comments in your feed or I am just oblivious the fact that this has been going on for quite some time?
A few days ago I altered the RSS feed to include the entire post as well a link to the comments. It does display the number of comments made but I had no idea this would cause problems with repeating RSS entries. I'll repair that now, let me know if that fixes the problem.
Personally, I loved this article.
Problem number one is that people get defensive when their own preferences are called into question.
It's human nature to want to be liked, to want to be seen as smart. The notion that someone else's way might be better scares a lot of people. And since most sites are designed by men, the result is obvious. "Nuh-uh!" People look for flaws in the research rather than considering the validity.
I'm certain there are flaws in the research. We are human beings after all. Human beings make mistakes.
But I'm also sure there is a lot of validity to the research.
Problem number two is individuality. People are unique. Of course there are going to be women who's tastes are more typical of men's tastes. And there will be men who's tastes and preferences are more typical of a woman. Designing for individuality is impossible unless you're designing for yourself. That's why we use commonalities. Who are the people in your target audience? Where's the happy medium?
I work with radio stations that target women, and it amazes me how male they sound. It amazes me how many of their websites are based on a black background with a really abrasive look. What can I say? Most radio stations are run by men.
I do beleve it's "losing," not "loosing." Good read.
Whoops! Wrong post :D