Who cre­at­ed all these web­sites? What were their job titles? Their ages, gen­ders, eth­nic back­grounds, com­pen­sa­tion? Did their sex affect their salary? Their title? The speed with which they advanced? If half the peo­ple we knew in the busi­ness were women, why were so few of them speak­ing at tech con­fer­ences, and how could we change that?

Nobody knows the pixels weve seen

In 2007, we set out to answer these ques­tions. And the first thing we learned was that there was no data, and nobody had even asked the ques­tions. Shock­ing­ly, in the 17th year of the webs exis­tence there was not a byte of data to be found about its makers.

The respect­ed research firm we hired to answer our ini­tial ques­tions could not. There was, they told us, no data what­so­ev­er about web design­ers and devel­op­ers. They asked us if IT pro­fes­sion­als meant the same thing. Of course it does­nt, but they did­nt even know that. The inter­net was prac­ti­cal­ly the only healthy part of the econ­o­my at the time, but nobody knew jack about it or the peo­ple who made it.

We would have to get our own data. So began the A List Apart sur­veys, in coop­er­a­tion with An Event Apart­which had an inter­est in get­ting more great women onstage.

How it worked

We spun up the first sur­vey in 2007. Although our inter­est in diver­si­ty in web design had led us to cre­ate the sur­veys, the sur­veys pre­sent­ed neu­tral­ly, as a series of busi­ness ques­tions with no overt agen­da. About 35,000 design­ers respond­ed each year. The sur­veys were dynami­chow you answered a ques­tion deter­mined which ques­tion you would be asked next, so no two respon­dents took exact­ly the same journey.

Slic­ing and dic­ing each years results in com­plex ways enabled us to begin see­ing and shar­ing pat­terns of pow­er and prej­u­dice in our field.

Painstakingand some­times elab­o­rate­com­pu­ta­tions were required to make the slic­ing and dic­ing work. Then we had to write up and find ways to pic­to­ri­al­ize the results. The first year we favored dif­fi­cult, abstract ways of doing thissuch as con­vert­ing XML data to CSS, care of CSS wiz­ard Eric Mey­er of An Event Apart. We includ­ed anonymized raw data in a vari­ety of for­mats so read­ers could crunch their own num­bers and slice and dice to sat­is­fy their own agendas.

Draw your own conclusions

Heres a PDF of the 2007 sur­vey find­ings in our ini­tial visu­al style, and here’s the land­ing page from that year, designed by A List Apart cre­ative direc­tor Jason San­ta Maria. The pre­sen­ta­tion style changed after that, and we switched from PDF to web pages­be­cause, duh, were a web design firm, ALA is a web design mag­a­zine on the web, and An Event Apart is a web design conference.

Enjoy:

• Find­ings from the A LIST APART Sur­vey, 2008
• Find­ings from the A LIST APART Sur­vey, 2009
• Find­ings from the A LIST APART Sur­vey, 2010
• Find­ings from the A LIST APART Sur­vey, 2011

In 2011, to stop invent­ing the wheel each year, we switched from hand­spun back­end tech to Poll­Dad­dy, an Automat­tic prod­uct. Using Poll­Dad­dy cut out many hours of tedious work prepar­ing hand­spun ques­tion­naires, but forced us to sim­pli­fy our ques­tion inter­de­pen­den­cies somewhat.

By 2012, per­haps because so few data points seemed to change from year to year, the com­mu­ni­ty that had par­tic­i­pat­ed so vig­or­ous­ly dur­ing the first five years of sur­veys began to fade. Faced with mount­ing resource chal­lenges and the indif­fer­ence of the design, tech, and main­stream press, we have put the sur­veys on hold for now.