Got Color? Uh...No

Perhaps the ire over last year's Fall Color Contest becomes more understandable when we realize that cool does not equate with color in the minds of AT participants, despite large amounts of color therapy.

• About 50% of Smallest, Coolest entries show all-neutral color schemes.
• Another 35% show a neutral plus one or more colors.
• That leaves just 15% of all entries for all other color schemes, with cool+warm accounting for almost half of those.
  

Neutrals so thoroughly dominate Smallest, Coolest entries that it's impossible to find statistical support for regional differences or for any influence on voting patterns. Both beloved entries and derided entries rely on neutral-based color schemes, while colorful apartments are almost too few to count.

White, white, white is the color of our entire apartment!
Almost 75% of the all-neutral color schemes incorporate white as one of the neutrals.

• 22% of the all-neutral entries show all-white apartments.
• 27% show white plus wood as the two dominant "colors."
  

I see a red door and I want it painted white
If we look at colors individually, preferences still revolve around neutrality that Switzerland would envy:

• The single most popular dominant color is white (38% of all entries), trailed by gray (10%) and green (8%). Don't get your hopes up about the predominance of lime: about half of those greens are sage, almost a neutral itself.
• The single most popular secondary color is... wait for it... white (15%), followed by wood (13%).

If there's an AT aesthetic, it apparently involves neutral walls, neutral colors on the large furnishings, and pops of color limited to the accessories.

Don't Hate Me - I Do MCM

MCM is by far the dominant style in 2007 Smallest, Coolest entries, accounting for fully one-half of the entries. We can thank the East Coast and the Los Angeles basin for that dominance: the Midwest and the Pacific Northwest show substantial loyalty to a more Ethnic/Eclectic look, as well as to Scandinavian/Shaker style. A more "traditional" vibe -- often with Hollywood Regency flair -- appears consistently in about 15% of entries, regardless of region. The real surprise? The ultra-trendy Neo-Baroque promoted by Brocade Home and Anthropologie is the home-wide style of just one single entry.

While the pie charts show vivid regional differences, some styles were represented in numbers too small for statistical analysis. However, there is a statistically significant pattern that NY and LA entries favor MCM, while Chicago and SF entries favor other styles (p<.05).

So is it "MCM to win"?
Not exactly. Going MCM doesn't correlate at a statistically significant level with higher numbers of insta-finalist votes. What MCM does is protect the entrant from not my thang votes. MCM entries were less likely than non-MCM entries to receive large numbers of no votes (p<.005). <span style="font-style: italic;"><span style="font-weight: bold;">MCM can be seen as the "safe" conventional style for Smallest, Coolest entries.

Ethnic/Eclectic: Unsafe at Any Speed
Conversely, if you're playing to lose, go Ethnic/Eclectic: these entries get fewer insta-finalist votes (p<.001).

There is no similar pattern -- either direction -- for the other relatively popular style, Hollywood Regency/Traditional.

A follow-up question for later: is it the style that makes the voting pattern, or is it the typical level of clutter that goes with the style? Intuitively, MCM homes ought to be less cluttered. But are they?

Love is a Lucite Table

Despite many a cry of "If it has Eames, it has my vote!" (and contrasting cries of "not another Eames chair, please!"), no single Eames design won the heart of Smallest, Coolest entrants in 2007 with the fervor of lucite tables. Thirteen entries -- almost 12% of the total -- showed lucite tables in their initial three photos.

The nearest runners-up were the Eames "potato chip" chair in its various permutations and variants on the Eames fiberglass dining chair so familiar from elementary school cafeterias, circa 1974. Totaled together, iconic Eames designs appeared 28 times in 112 entries.

As the graph shows, Eames is running well ahead of the slatted coffee table that seems to be everywhere, and that West Elm lattice-back chair that was hot in 2006 is apparently so last year. Meanwhile, love of lucite doesn't extend from tables to chairs: the ghost chair shows up only once.

Does lucite lead to a win? Alas, no. There's no significant pattern in voting for lucite-loving and lucite-free entries.

A Flat-Screen TV Wants You!

At first glance, it would seem like there's a community consensus to kill your television. More than half of Smallest, Coolest entries show no television at all in their initial three photos. As far as TV types, old-school televisions and modern flat screens are about equally popular.

But when we look at the relationship between the number of insta-finalist votes* and the type of television, a different pattern emerges: entries that show flat-screen TVs get more insta-finalist votes than do entries that show no TV or an old-school TV. (p<.05)

*Insta-finalist votes are used here as a measure of enthusiasm, even though these votes alone don't necessarily control who goes to semi-finals.

Taking a long, slow look at this graph shows us what's going on:

Interpreting the graph

The bottom axis represents the number of insta-finalist votes, lumped into categories according to their distance from the average. (Average = 108, standard deviation = 97, and the 61-156 category is 1/2 a standard deviation to either side of the average.)

The side axis shows what percentage of entries with that type of television fell into each category. This lets us compare the behavior of entries with no television and entries with a flat-screen TV without worrying about the fact that there were about half as many entries with flat-screen TVs as with no TV.

If you hated the two paragraphs above, just look at the shape of the colored lines. Entries with no television or an old-school television had most of their voting action below our average of 108 insta-finalist votes per entry. Entries with flat-screen TVs had most of their voting action at or above that average.

Quick, Where's the Nearest Best Buy?

Before you whip out your credit card for your flat-screen TV, do be aware that correlation is not causation. Entries with flat-screen TVs elicit more rampant enthusiasm than entries without them... but that doesn't mean the flat-screen TV is pixelating voters' brains. Most likely, flat-screen TVs are typically part of a larger array of design choices that express the community's wisdom on "good design."

Have Zebra Rugs Jumped the Shark?

It's an interesting image, especially from the shark's point of view (zebra belly! yum!). And it does seem that lately, zebra rugs have been everywhere on AT.

However, a quick count reveals that zebra rugs, cowhide rugs, and even insouciantly tossed-about furs and sheepskins are far outnumbered by shag area rugs in SC entries.

"But I'm Seeing Zebra Everywhere!"

If we look at appearances of these furry friends as a percentage of total entries in each region, we quickly see why there are so many remarks on the New York site about zebra rugs being overdone. Zebra rugs appear in 10% of the New York entries. True, that's only four entries -- but if a visually dramatic item appears several times within a few days (as these entries did), it will make a disproportionate impact on viewers' memories. Also, it's likely that AT participants don't make a clear distinction between what they see in SC entries and what they see in other postings on the site. (We'll see later that a trend like arranging books by color can be perceived as "overdone in SC" when it's barely done at all.)

While zebra is very much with us in New York, it hasn't even found the shark in Chicago or San Francisco, much less taken a running leap at it.

Shagadelic, baby!

If there's a rug trend in danger of becoming overdone, it's shag--which has attracted much less comment. Perhaps the secret is that the shag rugs usually appear in neutral colors that don't grab the eye with the force of a zebra pattern. (Or perhaps it's that no shags were killed to produce these rugs, so the comments section of these entries remain quiet on issues such as whether it's okay to wear shag if you're a vegetarian or whether faux shag upholds the same culture of cruelty as real shag.)

Shag rugs will return in a later posting on regional styles. At 25% of Los Angeles entries (which came almost entirely from the L.A. basin and San Diego), shag rugs arguably define a Southern California Style for 2007.

Implications

Bold design moves work best if you're the first person to make that move. By the third iteration, the idea may be deemed "tired" or "cliched." In the design blog world, you have to run like Alice's Red Queen just to stay even with fashion.

Is Design on the March?

In one of today's postings, our fearless leader Maxwell Gillingham-Ryan declared that:

The post Baby Boom generation is looking for new voices, separating themselves from the styles (and names) of the past: Eames, Starck, Noguchi, Rashid, Newson, van der Rohe... We want to find the new, new thing.

As far as Smallest, Coolest entries go... don't count on it. While I'll take up the obsession with all things Eames and Mid-Century Modern in a later post, we can get a sense of the current importance of the "small design shops" featured in the AT posting by looking at what SC entrants list as their favorite resources/inspirations.

There's No Furniture Like Old Furniture

Hands-down, the most popular approach to home design is bringing home bargains from thrift stores, junk stores, flea markets, the sidewalk, and generous relatives. Add eBay, Craig's List, and antique stores, and you've accounted for almost half of all resources suggested. With local stores -- not just local design shops but all local stores other than hardware and art supply -- we have half of the shopping inspirations for SC entrants. Designers and design magazines account for a much smaller share of inspiration.

Resources by Category, All Regions

IKEA vs. DWR Smack-Down

If we look at named resources more closely in the rather frightening rainbow-hued graph below, some surprises emerge.

Resources by Name, All Regions

What do we find?

• IKEA inspires more entrants than does DWR.
• The impact from design magazines is almost entirely limited to Dwell.
• The three designers who are named -- Jonathan Adler, Le Corbusier, and Armani Casa -- are hardly the young turks of design, and none receives more than two mentions.

I Left My Copy of Dwell in San Francisco

If small design shops are on the march, they still have miles to go before crossing the Rockies, and their power is diluted even in areas where they might be strong.

• In NY (East Coast), hardware stores are almost as popular as all local stores (which includes design shops but also fabric stores and local retailers such as Pearl River).
  
  
• In Chicago (Midwest), online shopping vies with local stores for popularity.
  
  
• In Los Angeles (Southwest), Craig's List is just as popular as local stores and eBay is close.
  
  
• In San Francisco (Pacific Northwest), local stores have few fans, while Dwell, Apartment Therapy, and resources like old movies are unusually popular.

Regional Preferences, Resources with at Least 5 Votes

Implications

While SC entrants may be interested in the new, new thing in design, spending big bucks on it is not on their to-do list. Demonstrating bargain-hunting savvy has a much larger role in design decisions. Hot young designers' work may not bubble down to the ordinary AT participant until the designer does a line for a big-box store such as Target.

Do Professional Designers Have an Edge?

Professional designers and architects entering their own homes represent 15% or fewer of entries across all sites. Another 10% of New York entrants definitely hired a professional designer.

If an entry did not use a designer, it is unlikely to receive a strong consensus of insta-finalist votes. Of the seven entries with 66% or more insta-finalist votes in proportion to total votes, five were submitted by designers or by homeowners who mentioned hiring a designer. If an entry received 33% or fewer insta-finalist votes in relation to total votes, it almost certainly did not involve a designer.

Percentage of insta-finalists

However, in the past, AT has not advanced entries to semi-finals based solely on insta-finalist votes. Rather, insta-finalist and has potential votes are added together and no way votes are subtracted from this total. If this I+P-N number is normalized to remove the influence of receiving an especially large or especially small number of votes, then only New York entries show a significant advantage in involving a professional designer (assuming AT continues to select semi-finalists by this method).

Proportion of designer-assisted entries in Top 4 vs. all entries

Implications

While the situation in New York may seem to make a case for banning professional designers from entering their homes in Smallest, Coolest, it’s important to note that two of the three professionally designed entries that qualify for the Top 4 under the normalized I+P-N method were entered by homeowners who had hired a professional. Removing the “professional advantage” would require policing how much help entrants had received, as well as entrants’ own professions.

If apartments that received professional design assistance are banned from the contest, be prepared for a wave of complaints that 2008 entries are not as strong as entries in prior years. As the bar charts above show, there is an apparent pattern that the best non-designer-assisted entries do not earn the universal raves of the best designer-assisted entries. This pattern is highly significant when all entries are considered as a whole (p<.01); separating entries by site shows a significant edge for designer-assisted entries only in Los Angeles (p=.05).

What's This All About?

What?
This site provides an ongoing analysis of entries and finalists in Apartment Therapy's Smallest, Coolest Contest, starting with regional entries in 2007.

Why?
As smaller-space living becomes recognized as a major trend and AT continues to grow, the AT SC contest provides a provocative snapshot of how real people who are proud of their small homes live. Analyzing these results is one way to get a sense of where the trend is going, as well as information for continuing to improve contest management.

Who?
Let's talk about what I'm not:

• I'm not an employee of AT.
  
• I'm not a professional designer.
• I'm not a graphic designer, either, so don't send angry emails if you hate the standard Blogger template.

Or what I am:

• I earn my money as an analyst and am starting an MBA in fall 2007.
  
• I've published some book chapters and presented some conference papers on trends in home life and leisure.
• I'm doing this project as a hobby, because apparently I don't see enough spreadsheets during the day.

When?
New postings will appear as I have time. I'm aiming for about once a week but will post more if I can.

A note for statisticians
Right, I know!

• The data is self-selected rather than a random sample (with an extra "mystery level" of selection when entries were pre-screened before posting, so that some entries don't make the cut). All survey data on the site is self-selected, but self-selected by different people than the contest entrants.
  
  
• There's no assurance that the voters are the same across entries (though there's some reason to believe in a fairly consistent voting pool on any given site, for reasons I'll discuss in a later post).
  
  
• This is essentially a pilot study because there's no second or third rater to assure inter-rater reliability on matters of perception like dominant color or level of clutter.
  
  
• Sample sizes within each regional site are often too small for extensive analysis.

So what you'll see here are numbers and graphs that raise questions as much as answer them. However, it's a start.