Hi everyone. My name is Sebastian, and this is  joint work with my colleagues from the University  
Hi everyone. My name is Sebastian, and this is  joint work with my colleagues from the University  

of Basel, the Hamm-Lippstadt University of Applied  Sciences and the University of Washington.
of Basel, the Hamm-Lippstadt University of Applied  Sciences and the University of Washington.

CHI often provides results and ideas  that influence technology innovations  
CHI often provides results and ideas  that influence technology innovations  

designed and developed in technology hubs in  Western countries, by people who are arguably  
designed and developed in technology hubs in  Western countries, by people who are arguably  

Western, Educated, Industrialized, Rich,  and Democratic [pause] or short, WEIRD.
Western, Educated, Industrialized, Rich,  and Democratic [pause] or short, WEIRD.

Now, many studies have shown that  the use and perception of technology  
Now, many studies have shown that  the use and perception of technology  

varies across countries and cultures. This means that we, the HCI community,  
varies across countries and cultures. This means that we, the HCI community,  

should not only study people that are WEIRD,  but also people from around the world to develop  
should not only study people that are WEIRD,  but also people from around the world to develop  

technology that is globally useful and usable. The CHI community already tries to increase  
technology that is globally useful and usable. The CHI community already tries to increase  

international diversity by SIGCHI Executive  Committee efforts, activities like RepliCHI,  
international diversity by SIGCHI Executive  Committee efforts, activities like RepliCHI,  

HCI Across Borders, ICTD as well as researchers who studied diverse  
HCI Across Borders, ICTD as well as researchers who studied diverse  

groups in Western and Non-Western societies. Our goal is to support these efforts and to shed  
groups in Western and Non-Western societies. Our goal is to support these efforts and to shed  

light on the current international diversity  of CHI by answering the following questions:  
light on the current international diversity  of CHI by answering the following questions:  

To what extent are participant samples in  CHI papers recruited in Western, Educated,  
To what extent are participant samples in  CHI papers recruited in Western, Educated,  

Industrialized, Rich, or Democratic societies?  and Which countries are over- and understudied?
Industrialized, Rich, or Democratic societies?  and Which countries are over- and understudied?

To answer these questions, we conducted a  systematic quantitative content analysis  
To answer these questions, we conducted a  systematic quantitative content analysis  

of over 3000 articles of the  last five proceedings of CHI.
of over 3000 articles of the  last five proceedings of CHI.

Our analysis showed that participant  samples of CHI publications  
Our analysis showed that participant  samples of CHI publications  

are not representative of the world s population.
are not representative of the world s population.

More than half of all countries in the world  are not covered by CHI research studies.
More than half of all countries in the world  are not covered by CHI research studies.

73% of findings are derived from studies  with Western participant samples.
73% of findings are derived from studies  with Western participant samples.

97% of these papers are fully WEIRD. This means that almost three quarter  
97% of these papers are fully WEIRD. This means that almost three quarter  

of the knowledge we produce at CHI is based  on roughly 12% of the world s population.
of the knowledge we produce at CHI is based  on roughly 12% of the world s population.

We identified positive trends,  too. CHI participant samples  
We identified positive trends,  too. CHI participant samples  

become steadily more geographically diverse. There is a reduction of participant samples  
become steadily more geographically diverse. There is a reduction of participant samples  

from the US by 13% and almost a doubling  of non-Western samples to around 30%.
from the US by 13% and almost a doubling  of non-Western samples to around 30%.

This could be explained by: different conference  locations, general efforts to diversify CHI,  
This could be explained by: different conference  locations, general efforts to diversify CHI,  

the usage of online and big data studies,  and research projects in the area of ICTD.
the usage of online and big data studies,  and research projects in the area of ICTD.

Our second goal was to identify which  countries are over- or understudied. It  
Our second goal was to identify which  countries are over- or understudied. It  

turns out that many small European and  WEIRD-countries like Finland, Denmark,  
turns out that many small European and  WEIRD-countries like Finland, Denmark,  

Sweden, Luxembourg, Switzerland, and Great Britain  
Sweden, Luxembourg, Switzerland, and Great Britain  

are strongly over-represented at CHI, compared  to their country s population size.
are strongly over-represented at CHI, compared  to their country s population size.

So, it seems that our average CHI study  participant tends to be more western, educated,  
So, it seems that our average CHI study  participant tends to be more western, educated,  

industrialized, rich, and democratic  than most of the world s population.
industrialized, rich, and democratic  than most of the world s population.

Based on our results we developed  several ideas to make CHI less weird.  
Based on our results we developed  several ideas to make CHI less weird.  

Keep in mind that all authors of  this paper are WEIRD themselves. So,  
Keep in mind that all authors of  this paper are WEIRD themselves. So,  

this background has influenced our view  of the following potential solutions.
this background has influenced our view  of the following potential solutions.

We could see in our data that most authors  recruit local participants. Balancing the number  
We could see in our data that most authors  recruit local participants. Balancing the number  

of authors and co-authors from non-Weird  and WEIRD countries would likely lead to  
of authors and co-authors from non-Weird  and WEIRD countries would likely lead to  

a more balanced composition of study samples. Our study found that online studies usually have  
a more balanced composition of study samples. Our study found that online studies usually have  

more diverse participant samples. They are limited  to participants with access to the Internet.  
more diverse participant samples. They are limited  to participants with access to the Internet.  

But we would like to encourage the CHI  community to conduct further research  
But we would like to encourage the CHI  community to conduct further research  

on how to increase data quality and allow  quantitative as well as qualitative methods.
on how to increase data quality and allow  quantitative as well as qualitative methods.

The CHI community has a lot  of experience in conducting  
The CHI community has a lot  of experience in conducting  

research with different methodological  settings across countries and cultures.
research with different methodological  settings across countries and cultures.

We should start to share best and  worst practice examples gained when
We should start to share best and  worst practice examples gained when

designing, recruiting for,  and conducting these studies.
designing, recruiting for,  and conducting these studies.

In the case of cross-cultural  studies, replications can uncover  
In the case of cross-cultural  studies, replications can uncover  

interesting variations that may help us to better  understand whether the innovations we make are  
interesting variations that may help us to better  understand whether the innovations we make are  

equally suitable for people around the world. Such replications and extensions of studies should  
equally suitable for people around the world. Such replications and extensions of studies should  

continue and be promoted by the community. 56% percent of the analyzed papers did not  
continue and be promoted by the community. 56% percent of the analyzed papers did not  

present detailed information on their  participants country of origin.
present detailed information on their  participants country of origin.

By including this information, we would  facilitate replications, extensions,  
By including this information, we would  facilitate replications, extensions,  

meta-analyses, and could track the international  representativeness of participant samples.
meta-analyses, and could track the international  representativeness of participant samples.

Altogether, our takeaway message  for today is that CHI is WEIRD,  
Altogether, our takeaway message  for today is that CHI is WEIRD,  

but we re hopeful that this is changing! Thank you for listening and I m excited to  
but we re hopeful that this is changing! Thank you for listening and I m excited to  

discuss our results and ideas for making  CHI less WEIRD.
discuss our results and ideas for making  CHI less WEIRD.