17 Mar

Relationship between desire to migrate and trust in government

Recently Armenian political figure Vartan Oskanian joined Twitter. He tweeted a speech that he gave where  he claimed that lack of trust in government is a cause of migration in Armenia.

Thanks to him for giving me a hypothesis to test!

Using the preliminary 2011 Caucasus Barometer, I looked at the relationship between desire to temporarily or permanently migrate and trust in government. And guess what? Oskanian is right!

For those that want to short-term migrate, they are statistically significantly less likely to trust these institutions than those that do not want to short-term migrate:

education system, army, courts, parliament, president, media, and the church.

Desire to short-term migrate had no effect on trust in these institutions:

healthcare, banks, NGOs, Prime Minster, police, local government, ombudsman, EU, and UN.

And for those that are interested in permanently migrating, it gets even more interesting.

Those with a desire to permanently migrate were statistically significantly less trusting in these institutions:

healthcare system, banks, education system, army, court system, NGOs, parliament, Prime Minister, president, police, media, local government, church, and ombudsman.

There was no impact on trust in the EU or UN.

So, yeah… this is interesting!

For more on who wants to migrate, check this out from 2010. Also I have a paper on trust in the government in Armenia here.

14 Mar

Networked Authoritarianism and Social Media in Azerbaijan

Networked Authoritarianism and Social Media in Azerbaijan is an article by myself and Sarah Kendzior about Azerbaijan and Internet control.

Here is the abstract. (Translation via Google translate.) And here is the link to the study. (Contact me if you need a copy because of lack of library access.)


The diffusion of digital media does not always have democratic consequences. This mixed-methods study examines how the government of Azerbaijan dissuaded Internet users from political activism. We examine how digital media were used for networked authoritarianism, a form of Internet control common in former Soviet states where manipulation over digitally mediated social networks is used more than outright censorship. Through a content analysis of 3 years of Azerbaijani media, a 2-year structural equation model of the relationship between Internet use and attitudes toward protest, and interviews with Azerbaijani online activists, we find that the government has successfully dissuaded frequent Internet users from supporting protest and average Internet users from using social media for political purposes.


Digital media diffuziya hər zaman demokratik nəticələr yoxdur. Bu qarışıq-metodların öyrənilməsi Azərbaycan hökuməti siyasi aktivliyi İnternet istifadəçiləri dissuaded inceliyor. Biz digital media ağ avtoritarizmi, digital aracılıklı sosial şəbəkələr üzərində manipulyasiya açıq senzura daha çox istifadə olunur, orada keçmiş Sovet dövlətləri ümumi Internet nəzarət formasında istifadə necə yoxlamaq. Azərbaycan media, internet istifadəsi və etiraz münasibət arasında bir 2-il struktur tənlik model və Azərbaycan online fəalları ilə müsahibə 3 il tərkibi analiz etməklə, biz hökumətin uğurla etiraz dəstək olan tez-tez İnternet istifadəçiləri dissuaded ki tapmaq və siyasi məqsədlər üçün sosial media istifadə orta İnternet istifadəçiləri.


Распространение цифровых носителей не всегда демократическими последствиями. Это смешанные методы исследования рассматриваются как правительство Азербайджана отговорили пользователей Интернета от политической активности. Мы рассмотрим, как цифровые медиа были использованы для сетевой авторитаризм, формы контроля интернет распространен в бывших советских республиках, где манипуляции над цифровым опосредованной социальной сети используется более чем откровенная цензура. С помощью контент-анализа 3-х лет азербайджанские СМИ, 2-летняя модель структурного уравнения связи между использованием интернета и отношения к протеста, а также интервью с азербайджанскими активистами в Интернете, мы обнаружили, что правительство успешно отговорил частых интернет-пользователей от поддержки протеста и обычных пользователей Интернета от использования социальных медиа в политических целях.

14 Mar

Armenian ICT adoption – updated

Images from my #ictd2012 poster, plus some updates on April 1.

Some fun bits:

– 37% of Armenians have used the Internet — that’s some major growth, as you can see on slide 2.

– And Armenians are using the Internet more often. 22% of Armenians are online daily and 8% are online weekly. Big increases from the past few years.

– Those that are NOT online are more likely to be rural, poorer, less educated, and older — but when they were asked, most said that they didn’t get online because they don’t have a computer, followed by a lack of interest or need.

– From the original post  I have the PC, mobile, and home Internet connection data from the Caucasus Barometer, the ITU, and Gallup. Based on methodology, I give the most weight to the Caucasus Barometer, but I figured that it’d be interesting to see how the three measure up.

In the next few days I’ll post online activities and some demographic breakdowns. As always, if you have questions, I’m happy to do some analysis for you. And, please link back to this post and/or give the proper attribution if you use these stats. This is the result of many hours of analysis and I appreciate your respect for my intellectual labor and property.

Creative Commons License
Armenia ICT Adoption by Katy Pearce is licensed under a Creative Commons Attribution-NonCommercial-NoDerivs 3.0 Unported License.
Based on a work at www.katypearce.net.

13 Mar

Where do Armenians get online?

I’m at the #ictd2012 conference right now and I’ve been asked about Armenian Internet use. Business Internet use is actually quite low, especially compared to household use. This surprises a lot of people.

So I did a quick analysis to find out where Armenians get online most often. (This is rough 2011 Caucasus Barometer — will revise if different in the final version.) And this is of ALL Armenians.

19% of all Armenians get online at home.

3% on a friend’s computer.

2% at work.

2% via the mobile.

1% at an Internet cafe.

07 Mar

Georgian Women in 2011

Happy International Women’s Day.

This month I’m going to share initial early analysis of the 2011 Caucasus Barometer that focuses on women. The first set is Georgia.

This is early data, so there is a possibility of error. I apologize in advance if there is and will update this blog post if any errors are found.

Also, I’ve had some issues with my analysis, graphics, and slides being re-appropriated. Please respect the hours of work that I’ve put into this and attribute properly by linking to this blog post, including my @katypearce name on Twitter, or any other suitable form of attribution.


– Georgian women’s life satisfaction is fairly normally distributed.
– And more Georgian women are happy than not happy.
– Religion is very important to many Georgian women.

– In terms of values, the majority of Georgian women do not believe that it is justified to engage in many forms of “cheating.” Two-thirds of Georgian women believe that it is never justified to have an abortion (despite the abortion rate being quite high). 71% of Georgian women do not believe that it is ever justified to be a homosexual.

– Many Georgian women want large families.

– Georgian women’s PERSONAL monthly income is on average about US$100. However, this is only one part of a total family income, so do not extrapolate household income from this.

– Over a third of Georgian women are interested in temporary migrant, while much less are interested in permanent migration. Half of Georgian women have a family member living abroad.

– 17% of Georgian women played a video game in the last 6 months!
– 10% of Georgian women volunteered in the last 6 months.
– 19% of Georgian women use the Internet daily – mostly for information seeking and social networking sites.
– But nearly two-thirds of Georgian women have NO computer skills.

– More Georgian women think that politics are going in the right direction than the wrong direction, although there are mixed opinions about how the government treats people.
– 86% of Georgian women would vote in a presidential election and over two-thirds would vote for a woman!

As always, if anyone is interested, I’m happy to do breakdowns by age, region, education level, or whatever on any of these.

I hope to post the Armenian and Azerbaijani women’s results in the coming weeks.

Happy International Woman’s Day!

Creative Commons License
Georgian Women 2011: Early analysis from the 2011 Caucasus Barometer by Katy Pearce is licensed under a Creative Commons Attribution-NonCommercial-NoDerivs 3.0 Unported License.
Based on a work at www.katypearce.net.
Permissions beyond the scope of this license may be available at http://www.katypearce.net/cv/georgia.

10 Jan

What do Armenians do on social networking sites?

More from the ArMedia dataset…

With so many Armenians on social networking sites (85% of mobile Internet users, 83% of Armenians that have both mobile and PC-based Internet, and 54% of PC-based Internet users in Armenia, as of a year ago), what are they doing on there?

Mostly communicating with friends! While only a little bit more than a third (37%) of Armenians that access social networking sites via a computer say that they communicate with friends, three-quarters of mobile Internet SNS users and 83% of those with both mobile and PC Internet and use SNSs say they’re communicating with pals.

Some use messaging to communicate with friends (27% of PC SNS users, 23% of mobile SNS users, and 19% that have both mobile and PC for SNS access).

Those PC SNS users are busy with something else — they’re sharing information (well, 37% of them are), but compared to only 13% of those with both and 17% of those with mobile for SNS access, this is a lot!

Those PC users are givers, not takers — they are busy sharing information, but they’re not really seeking information. Only 8% of PC SNS users are seeking information on social networking sites, while a quarter of mobile users and those with both PC and mobile for SNS access do.

What about new people? In my qualitative work I’ve heard a lot about Armenians trying to meet dating partners on social networking sites – and maybe this is the case in this data as well. A quarter to a third of Armenian SNS users are trying to meet new people on social networking sites.

Games are a bit popular too, especially with mobile users — a third of those with both PC and mobile and a quarter of mobile-only SNS users play games. Only 18% of PC SNS users do though.

Photos, videos, and music are posted by 17-27% of SNS users too.

** If anyone is interested, I can do a Facebook versus Odnoklassniki breakdown. **

New data is coming soon (hoorah!), but this gives a pretty good snapshot of Armenian social network site use in January-February 2011.

17 Dec

Kazakhstan Internet

With all of the activity in Kazakhstan right now, I wanted to pull some findings from a conference paper/ article in submission about Central Asian Internet. (I can’t post the entire paper while it is still in submission – sorry!) If you have questions or are interested in more detailed analysis, contact me and I’d be happy to do what I can.

Here’s Kazakh Internet penetration according to the ITU (which frequent readers know I’m not a fan of because it uses the telecom companies’ information to determine adoption):

Internet Adopters (% of population that have ever used) (ITU Development Index, 2011)















And here’s the data from the paper, national, from spring 2011:

Aware of Internet


Of total, Adopted Internet


Of total, Use Internet



Less than monthly








In all of my work I emphasize the importance of ‘frequent use’ as a more salient category of Internet use than ‘ever’ used.


In our paper, we model antecedents to Internet awareness, adoption (ever used), and frequent use.

In Kazakhstan, the strongest determinants (in order from strongest) of AWARENESS that the Internet exists were: age, urbanness, education, and economic wellbeing (although economic wellbeing wasn’t strong – statistically significant, but not as strong as the others). (This model explained 13% of the variance in awareness.)

In plain English this means that younger people, more urban people, those with more education, and those with more wealth are more likely to be in the 80% that know what the Internet is.

The strongest determinants (in order from strongest) of ADOPTION (ever having used) were age, economic wellbeing, urbanness, and education. (And this model was set up to control for the effect of these variables on awareness first…) (This model explained 33% of the variance in Internet adoption.)

In plain English this means that younger people, wealthier people, more urban people, and those with more education are more likely to be in the 50% that have ever used the Internet.

The strongest determinants (in order from strongest) of FREQUENT USE were age, urbannesses, and education. Economic wellbeing wasn’t a significant factor, but I’d imagine that the economic barrier exists more at the ‘ever used’ stage and once you’ve gotten over that, frequent use isn’t as much of a cost issue. (This model explained (a whopping!) 49% of the variance in use. (This is really high.))

In plain English this means that younger people, more urban people, and those with more education are more likely to be in the 29% that use the Internet frequently (at least weekly).

So what are Kazakhs doing online?

Of those that are online at least weekly,  85% do email, 77% read news (as a side note, this is quite high compared to what I’ve found in my work in the rest of Central Asia and the Caucasus), 73% are on social networking sites (this is normal compared to the Caucasus) (47% on Odnoklassniki, 43% on vKontake, 14% on Facebook, 7% on Twitter), ~40% interact with blogs (this is very high compared to the rest of Central Asia and the Caucasus), and 23% watch YouTube.

Access points vary.  2/3rd get online via the own PC, most at home (60% of frequent users). Public places are less common that home — 44% at work, 33% at school, 33% at a cafe on a public computer, 35% at a cafe with their own laptop. Mobile Internet is used by 55% of frequent users.

For what it is worth, the Kazakhstan Internet landscape is very different from the rest of Central Asia and the Caucasus — demographically and in the activities engaged in.

05 Dec

Regional differences in mean of accessing the Internet

Based on data from the Armenian Media landscape report, here are the regional breakdowns of means of accessing the Internet in Armenia (in early 2011).

You can see that for the country as a whole, most Internet users are personal computer based – although mobile is not insignificant. In the capital, PC dominates, but for rural users, mobile is catching up to PC. And those mobile users are often NEW users.

In an upcoming study we discuss how much the way people access the Internet influences what they actually do once they’re online as well as the demographic differences between those that use a PC and those that use a mobile phone as their primary Internet access device.

This is the entire country.

This is Yerevan.

This is regional cities.

This is rural areas.

(All images by Janine Slaker)




Regional city


N = 420

N = 215

N = 145

N = 60

Mobile Internet access





PC-based Internet access





Both mobile and
PC based Internet access








Regional city


N = 1420

N = 504

N = 443

N = 473

No Internet access 70%




Mobile Internet access





PC-based Internet access





Both mobile and
PC based Internet access