« SL Flickr of the Week: Japanese Steampunk Adventures of a "Flightless Bird" | Main | "Be Creative -- Please!" Longtime Linden Lab Veteran Pleads With Company to Staunch Second Life's Revenue Loss With New Ideas »

Wednesday, April 12, 2017


Feed You can follow this conversation by subscribing to the comment feed for this post.

Nalates Urriah

Your testing is flawed. Thinking changing just skin color is the only significant perceptual change is omitting how humans see images.

It is hard to definitively prove anything in a simple surveys. But, with images of people we want to be able to 'see' the PERSON. It is much more difficult to 'see' a dark skinned person in images. Your sample is a good example. The dark skinned person is visually detail deficient. So, is it race people are biased about or quality of details visibility?

My thinking is progressives define a goal then look for any evidence to support it. No consideration is given to alternative possibilities. Asking others to find the same supporting evidence based on the same incomplete assumption isn't adding support. What other factors, in the same images, are people reacting to?

If you look at the number of views of images with close up views of avatars verses the number of views for avatars at a distance, I believe we can see a bias for closeups. How close is popular is debatable too.

Then there is the 'glamour' factor, sexual factor, and how well the photographer/artist can portray those with a more difficult dark skinned subject?

Carlos Loff

This survey is interesting because the difference is quite significant, but it shoul be taken in account the ethnicity of viewers that are interested or use virtual worlds - Inagine for example that 85% of users are caucasian, it would just mean there is a ballanced relation between preference for same race, witch is more cultural and natural than any supposed racial issue

Leonorah Beverly

Seconding Natales Urriah in every single point.

The picture with darker skin has a really terrible lighting on the avatar's face. Eyes are over-emphasized, body details are not really visible.
There's simply a lack of quality in the upper picture with the darker skin. (No offense meant!!!)
A really good picture of a dark-skinned avatar will get at least the same count of favs as a picture with a bright skinned avatar.
Just my opinion.

Tizzy Canucci

I disagree with you about 'unconscious bias' not being racism. Racism often articulates itself as the 'natural order of things', so commonsensical that it is obvious and need not be thought about. That's partly what makes it so pervasive and persistent. And discrimination is in part the manifestation of an accumulation of individual biases.

There is an implication in what you write that if you argue against your method, you are also arguing racism doesn’t exist. Racism exists, but your method is still fatally flawed.

Wagner J Au

"A really good picture of a dark-skinned avatar will get at least the same count of favs as a picture with a bright skinned avatar."

Quite literally why I wrote "Strive for the same level of visual quality in each."


Ok, this wasn't the most scientific test. It was two pictures on one persons Flickr account. If I was going to do it again I would have tried to take a better picture of the dark skinned avatar and then changed skin and reshot. Even then, the sample pool is too small. However, a few other interesting points I'd like to bring up. Some of the most popular images on my Flickr are super high key pictures without much contrast. However, they have less contrast in the opossite direction from the image of the dark skinned avi. They are very washed out, without a lot of detail.
Also, every year I do a calendar, so I take 12 pictures of various girls. This year, one was black. Her picture isn't the best, or the worst. It's in the middle somewhere. However, her picture has the fewest likes. Again, it's a small sample, but something to consider.
My avi is blonde with lighter skin, although it does change from time to time. People who look at my Flickr are probably more prone to like a picture of a blonde, lighter skinned girl.
All things to think about. It's an interesting discussion, but people will still take from it what they want.


Can't make assumptions about personal preferences and biases without knowing who the test subjects (the photo viewers) are.

Opal Lei

I usually don't take the numbers literally but relatively. If you divide the number of faves by the number of views, you get approximately the same percentage.

211 / 1708 = 12.35% of viewers voted for the fair-skinned avie
84 / 685 = 12.26% of viewers voted for the dark-skinned avie

The difference is statistically insignificant.

A better question is ... What is the number of views really? Is it the number of people who actually opened the page of that image (active viewing)? Or does it also include browsing the photostream as one view (semi-passive viewing)? How about slideshow viewing (more passive viewing)?

Another question would be the effect of the order that people see these two images. Maybe they think they've already seen the second one because it looks familiar (only one detail changed), so they ignore it. Therefore, it didn't get as many views, which translates to fewer faves.

Ciaran Laval

"many from people angrily denying this "proved" widespread racism exists against non-white avatars. "

This didn't really happen, Ysabelle Stewart actually suggested that racial bias probably does exist but most people were pointing out that to the naked eye, one picture simply looked better than the other. There are many reasons for this.

I covered avatar skins and racial bias based on the work of Jong-Eun Roselyn Lee a while ago, which was not perfect research due to low numbers, but very interesting, The Mary Sue covered it too, here's a link to their article on the subject :


Verify your Comment

Previewing your Comment

This is only a preview. Your comment has not yet been posted.

Your comment could not be posted. Error type:
Your comment has been posted. Post another comment

The letters and numbers you entered did not match the image. Please try again.

As a final step before posting your comment, enter the letters and numbers you see in the image below. This prevents automated programs from posting comments.

Having trouble reading this image? View an alternate.


Post a comment

Your Information

(Name is required. Email address will not be displayed with the comment.)

Wagner James Au
Samsung Edge computing reports NWN
Breakroom virtual meetings conferences-GIF
Dutchie ad SL couch
Ample Avi  SL avatars
Sinespace virtual world Unity free home
my site ... ... ...