Hi all,
As some of you know, we started a line of research back in 2016 to understand Wikipedia readers better. We published the first taxonomy of Wikipedia readers and we studied and characterized the reader types in English Wikipedia [1]. During the past 1+ year, we focused on learning about the potential differences of Wikipedia readers across languages based on the taxonomy built in [1]. We've learned a lot, and today we're sharing what we learnt with you.
Some pointers: * Publication: https://arxiv.org/abs/1812.00474 * Data: https://figshare.com/articles/Why_the_World_Reads_Wikipedia/7579937/1 * (under continuous improvement) Research page on meta: https://meta.wikimedia.org/wiki/Research:Characterizing_Wikipedia_Reader_Beh... * Research showcase presentation: https://www.mediawiki.org/wiki/Wikimedia_Research/Showcase#December_2018 * A series of presentations to WMF teams and community: Look for tasks under https://phabricator.wikimedia.org/T201699 with title "Present the results of WtWRW" for link to slides and more info when available. * We will send out a blog post about it hopefully soon. A blog post about the intermediate results is at https://wikimediafoundation.org/2018/03/15/why-the-world-reads-wikipedia/
In a nutshell: * We ran the taxonomy of Wikipedia readers in 14 languages and measured the prevalence of Wikipedia use-cases and characterized Wikipedia readers in these languages. * While we observe similarities in terms of the prevalence of the use cases as well as the way we can characterize readers, we can see that Wikipedia languages lend themselves to different distributions of readership and characteristics. In many cases, the one-size-fit-all solutions may simply not work for readers. * Intrinsic learning remains as the number one motivation for people to come to Wikipedia in the majority of the languages, followed by media. * In-depth reading and the reading of scientific oriented topics is highly and negatively correlated with the socio-economic status and Human Development Index of countries the readers in these languages are coming from. Long articles that may seem just too long for the bulk of our audience in US, Japan, and the Netherlands is in high demand in India, Bolivia, Argentina, Panamá, México, … * ...
This research was not possible without the extensive contributions by our formal collaborators: Florian Lemmerich (RWTH Aachen University) and Bob West (EPFL). On the WMF end, I was fortunate to work with Diego Saez on this project as well as more recently, Isaac Johnson. And all those in the Reading Web and Legal team who supported us throughout the process. I also want to underline the amazing work that the volunteers in the languages in the study did to support us heavily to learn more about their languages, not only through help with communications within their communities but also with the translation task which was not an easy one as they were asked to offer their time not only to translate but also do in-person meetings with us for us to make sure the intent of the question is translated the same way across the languages. Usernames Strainu, Tgr, Amire80, Awossink, Antanana, Lyzzy, Shangkuanlc, Whym, Kaganer, عباد_ديرانية, Satdeep_Gill, Racso, Hasive: Thank you!
Next we are going to extend this study to include demographics information. More information about it coming out in the next few weeks. (And I will send out a separate email to wikimedia-l about this topic and future research over the weekend. I need some time to finalize the message to make the message most useful for that audience.:)
Best, Leila
[1] https://arxiv.org/abs/1702.05379
-- Leila Zia Senior Research Scientist, Lead Wikimedia Foundation
Hi Leila,
I'm curious about the in-depth reading differences according to the socio-economic status. Why do you think such differences exist?
Regards, Micru
On Sat, Jan 19, 2019 at 1:15 AM Leila Zia leila@wikimedia.org wrote:
Hi all,
As some of you know, we started a line of research back in 2016 to understand Wikipedia readers better. We published the first taxonomy of Wikipedia readers and we studied and characterized the reader types in English Wikipedia [1]. During the past 1+ year, we focused on learning about the potential differences of Wikipedia readers across languages based on the taxonomy built in [1]. We've learned a lot, and today we're sharing what we learnt with you.
Some pointers:
- Publication: https://arxiv.org/abs/1812.00474
- Data:
https://figshare.com/articles/Why_the_World_Reads_Wikipedia/7579937/1
- (under continuous improvement) Research page on meta:
https://meta.wikimedia.org/wiki/Research:Characterizing_Wikipedia_Reader_Beh...
- Research showcase presentation:
https://www.mediawiki.org/wiki/Wikimedia_Research/Showcase#December_2018
- A series of presentations to WMF teams and community: Look for tasks
under https://phabricator.wikimedia.org/T201699 with title "Present the results of WtWRW" for link to slides and more info when available.
- We will send out a blog post about it hopefully soon. A blog post
about the intermediate results is at https://wikimediafoundation.org/2018/03/15/why-the-world-reads-wikipedia/
In a nutshell:
- We ran the taxonomy of Wikipedia readers in 14 languages and
measured the prevalence of Wikipedia use-cases and characterized Wikipedia readers in these languages.
- While we observe similarities in terms of the prevalence of the use
cases as well as the way we can characterize readers, we can see that Wikipedia languages lend themselves to different distributions of readership and characteristics. In many cases, the one-size-fit-all solutions may simply not work for readers.
- Intrinsic learning remains as the number one motivation for people
to come to Wikipedia in the majority of the languages, followed by media.
- In-depth reading and the reading of scientific oriented topics is
highly and negatively correlated with the socio-economic status and Human Development Index of countries the readers in these languages are coming from. Long articles that may seem just too long for the bulk of our audience in US, Japan, and the Netherlands is in high demand in India, Bolivia, Argentina, Panamá, México, …
- ...
This research was not possible without the extensive contributions by our formal collaborators: Florian Lemmerich (RWTH Aachen University) and Bob West (EPFL). On the WMF end, I was fortunate to work with Diego Saez on this project as well as more recently, Isaac Johnson. And all those in the Reading Web and Legal team who supported us throughout the process. I also want to underline the amazing work that the volunteers in the languages in the study did to support us heavily to learn more about their languages, not only through help with communications within their communities but also with the translation task which was not an easy one as they were asked to offer their time not only to translate but also do in-person meetings with us for us to make sure the intent of the question is translated the same way across the languages. Usernames Strainu, Tgr, Amire80, Awossink, Antanana, Lyzzy, Shangkuanlc, Whym, Kaganer, عباد_ديرانية, Satdeep_Gill, Racso, Hasive: Thank you!
Next we are going to extend this study to include demographics information. More information about it coming out in the next few weeks. (And I will send out a separate email to wikimedia-l about this topic and future research over the weekend. I need some time to finalize the message to make the message most useful for that audience.:)
Best, Leila
[1] https://arxiv.org/abs/1702.05379
-- Leila Zia Senior Research Scientist, Lead Wikimedia Foundation
Wiki-research-l mailing list Wiki-research-l@lists.wikimedia.org https://lists.wikimedia.org/mailman/listinfo/wiki-research-l
Hi Micru,
One hypothesis that we have is that in countries with lower socio-economic status, the reader may not have access to a variety of different sources for their reading needs (the availability of printed material, books, ... can be more limited in these countries.). At the moment, we don't have enough data from a diverse enough subset of countries to be able to look into this. I'm hoping that in the future iterations we can sample by country and collect enough data to be able to validate or reject this hypothesis.
Do you have other hypotheses as why this may be happening?
Best, Leila
On Wed, Jan 23, 2019 at 3:02 PM David Cuenca Tudela dacuetu@gmail.com wrote:
Hi Leila,
I'm curious about the in-depth reading differences according to the socio-economic status. Why do you think such differences exist?
Regards, Micru
On Sat, Jan 19, 2019 at 1:15 AM Leila Zia leila@wikimedia.org wrote:
Hi all,
As some of you know, we started a line of research back in 2016 to understand Wikipedia readers better. We published the first taxonomy of Wikipedia readers and we studied and characterized the reader types in English Wikipedia [1]. During the past 1+ year, we focused on learning about the potential differences of Wikipedia readers across languages based on the taxonomy built in [1]. We've learned a lot, and today we're sharing what we learnt with you.
Some pointers:
- Publication: https://arxiv.org/abs/1812.00474
- Data:
https://figshare.com/articles/Why_the_World_Reads_Wikipedia/7579937/1
- (under continuous improvement) Research page on meta:
https://meta.wikimedia.org/wiki/Research:Characterizing_Wikipedia_Reader_Beh...
- Research showcase presentation:
https://www.mediawiki.org/wiki/Wikimedia_Research/Showcase#December_2018
- A series of presentations to WMF teams and community: Look for tasks
under https://phabricator.wikimedia.org/T201699 with title "Present the results of WtWRW" for link to slides and more info when available.
- We will send out a blog post about it hopefully soon. A blog post
about the intermediate results is at https://wikimediafoundation.org/2018/03/15/why-the-world-reads-wikipedia/
In a nutshell:
- We ran the taxonomy of Wikipedia readers in 14 languages and
measured the prevalence of Wikipedia use-cases and characterized Wikipedia readers in these languages.
- While we observe similarities in terms of the prevalence of the use
cases as well as the way we can characterize readers, we can see that Wikipedia languages lend themselves to different distributions of readership and characteristics. In many cases, the one-size-fit-all solutions may simply not work for readers.
- Intrinsic learning remains as the number one motivation for people
to come to Wikipedia in the majority of the languages, followed by media.
- In-depth reading and the reading of scientific oriented topics is
highly and negatively correlated with the socio-economic status and Human Development Index of countries the readers in these languages are coming from. Long articles that may seem just too long for the bulk of our audience in US, Japan, and the Netherlands is in high demand in India, Bolivia, Argentina, Panamá, México, …
- ...
This research was not possible without the extensive contributions by our formal collaborators: Florian Lemmerich (RWTH Aachen University) and Bob West (EPFL). On the WMF end, I was fortunate to work with Diego Saez on this project as well as more recently, Isaac Johnson. And all those in the Reading Web and Legal team who supported us throughout the process. I also want to underline the amazing work that the volunteers in the languages in the study did to support us heavily to learn more about their languages, not only through help with communications within their communities but also with the translation task which was not an easy one as they were asked to offer their time not only to translate but also do in-person meetings with us for us to make sure the intent of the question is translated the same way across the languages. Usernames Strainu, Tgr, Amire80, Awossink, Antanana, Lyzzy, Shangkuanlc, Whym, Kaganer, عباد_ديرانية, Satdeep_Gill, Racso, Hasive: Thank you!
Next we are going to extend this study to include demographics information. More information about it coming out in the next few weeks. (And I will send out a separate email to wikimedia-l about this topic and future research over the weekend. I need some time to finalize the message to make the message most useful for that audience.:)
Best, Leila
[1] https://arxiv.org/abs/1702.05379
-- Leila Zia Senior Research Scientist, Lead Wikimedia Foundation
Wiki-research-l mailing list Wiki-research-l@lists.wikimedia.org https://lists.wikimedia.org/mailman/listinfo/wiki-research-l
-- Etiamsi omnes, ego non _______________________________________________ Wiki-research-l mailing list Wiki-research-l@lists.wikimedia.org https://lists.wikimedia.org/mailman/listinfo/wiki-research-l
I agree lack of alternative sources are likely to be a factor.
Another factor might be the time available. The high s-e countries may have a lot more competing for their eyeballs, eg Netflicks etc.
Also, on my experience of foreign students attending Australian universities, there may also be some cultural difference in how people think about education and learning. Some countries do appear to put a lot of emphasis on memorising and regurgitating slabs of textbooks and other rote learning rather than developing skills in applying that knowledge to solve a problem. Whereas I find myself increasingly learning by trying to absorb the main concepts and principles of a new topic but figuring I can lookup the detail if/when I actually need to apply that knowledge (particularly so in the WWW era, how to open a milk carton?, just look at the YouTube video!). So I think there is a shift to just-in-time learning happening but this relies on the meta-skill of information searching. I suspect poorly paid and poorly educated teachers in low s-e countries are more likely to teach using the same methods that they were taught by. This may lead to the mindset of “if I read this Wikipedia article many times through, I will have mastered the topic”.
Having travelled in lower s-e countries in Africa recently, many people do see education as the key to a better future and actively invest in their children’s education for that reason. In Kenya I often heard the term “lean family” which meant a small number of children getting the best education the family could afford rather than the traditional large family. I think this was something actively promoted by their government. Many parents proudly talked about making a personal sacrifice of some kind which was undertaken to have more money for children’s education.
I think many of us in more developed economies see Wikipedia as useful. I suspect there are others who see it as potentially life-changing.
Sent from my iPad
On 24 Jan 2019, at 7:33 am, Leila Zia leila@wikimedia.org wrote:
Hi Micru,
One hypothesis that we have is that in countries with lower socio-economic status, the reader may not have access to a variety of different sources for their reading needs (the availability of printed material, books, ... can be more limited in these countries.). At the moment, we don't have enough data from a diverse enough subset of countries to be able to look into this. I'm hoping that in the future iterations we can sample by country and collect enough data to be able to validate or reject this hypothesis.
Do you have other hypotheses as why this may be happening?
Best, Leila
On Wed, Jan 23, 2019 at 3:02 PM David Cuenca Tudela dacuetu@gmail.com wrote:
Hi Leila,
I'm curious about the in-depth reading differences according to the socio-economic status. Why do you think such differences exist?
Regards, Micru
On Sat, Jan 19, 2019 at 1:15 AM Leila Zia leila@wikimedia.org wrote:
Hi all,
As some of you know, we started a line of research back in 2016 to understand Wikipedia readers better. We published the first taxonomy of Wikipedia readers and we studied and characterized the reader types in English Wikipedia [1]. During the past 1+ year, we focused on learning about the potential differences of Wikipedia readers across languages based on the taxonomy built in [1]. We've learned a lot, and today we're sharing what we learnt with you.
Some pointers:
- Publication: https://arxiv.org/abs/1812.00474
- Data:
https://figshare.com/articles/Why_the_World_Reads_Wikipedia/7579937/1
- (under continuous improvement) Research page on meta:
https://meta.wikimedia.org/wiki/Research:Characterizing_Wikipedia_Reader_Beh...
- Research showcase presentation:
https://www.mediawiki.org/wiki/Wikimedia_Research/Showcase#December_2018
- A series of presentations to WMF teams and community: Look for tasks
under https://phabricator.wikimedia.org/T201699 with title "Present the results of WtWRW" for link to slides and more info when available.
- We will send out a blog post about it hopefully soon. A blog post
about the intermediate results is at https://wikimediafoundation.org/2018/03/15/why-the-world-reads-wikipedia/
In a nutshell:
- We ran the taxonomy of Wikipedia readers in 14 languages and
measured the prevalence of Wikipedia use-cases and characterized Wikipedia readers in these languages.
- While we observe similarities in terms of the prevalence of the use
cases as well as the way we can characterize readers, we can see that Wikipedia languages lend themselves to different distributions of readership and characteristics. In many cases, the one-size-fit-all solutions may simply not work for readers.
- Intrinsic learning remains as the number one motivation for people
to come to Wikipedia in the majority of the languages, followed by media.
- In-depth reading and the reading of scientific oriented topics is
highly and negatively correlated with the socio-economic status and Human Development Index of countries the readers in these languages are coming from. Long articles that may seem just too long for the bulk of our audience in US, Japan, and the Netherlands is in high demand in India, Bolivia, Argentina, Panamá, México, …
- ...
This research was not possible without the extensive contributions by our formal collaborators: Florian Lemmerich (RWTH Aachen University) and Bob West (EPFL). On the WMF end, I was fortunate to work with Diego Saez on this project as well as more recently, Isaac Johnson. And all those in the Reading Web and Legal team who supported us throughout the process. I also want to underline the amazing work that the volunteers in the languages in the study did to support us heavily to learn more about their languages, not only through help with communications within their communities but also with the translation task which was not an easy one as they were asked to offer their time not only to translate but also do in-person meetings with us for us to make sure the intent of the question is translated the same way across the languages. Usernames Strainu, Tgr, Amire80, Awossink, Antanana, Lyzzy, Shangkuanlc, Whym, Kaganer, عباد_ديرانية, Satdeep_Gill, Racso, Hasive: Thank you!
Next we are going to extend this study to include demographics information. More information about it coming out in the next few weeks. (And I will send out a separate email to wikimedia-l about this topic and future research over the weekend. I need some time to finalize the message to make the message most useful for that audience.:)
Best, Leila
[1] https://arxiv.org/abs/1702.05379
-- Leila Zia Senior Research Scientist, Lead Wikimedia Foundation
Wiki-research-l mailing list Wiki-research-l@lists.wikimedia.org https://lists.wikimedia.org/mailman/listinfo/wiki-research-l
-- Etiamsi omnes, ego non _______________________________________________ Wiki-research-l mailing list Wiki-research-l@lists.wikimedia.org https://lists.wikimedia.org/mailman/listinfo/wiki-research-l
Wiki-research-l mailing list Wiki-research-l@lists.wikimedia.org https://lists.wikimedia.org/mailman/listinfo/wiki-research-l
Hi Leila and Kerry,
Thanks a lot for your insights. Indeed, the lower availability of sources, and the relationship with knowledge might be the main factors. It seems that there are so many topics, that just by knowing that they exist is already an accomplishment.
It would be interesting if there would be a system to keep track of the topics one is familiar with.
Regards
On Thu, Jan 24, 2019 at 1:39 PM Kerry Raymond kerry.raymond@gmail.com wrote:
I agree lack of alternative sources are likely to be a factor.
Another factor might be the time available. The high s-e countries may have a lot more competing for their eyeballs, eg Netflicks etc.
Also, on my experience of foreign students attending Australian universities, there may also be some cultural difference in how people think about education and learning. Some countries do appear to put a lot of emphasis on memorising and regurgitating slabs of textbooks and other rote learning rather than developing skills in applying that knowledge to solve a problem. Whereas I find myself increasingly learning by trying to absorb the main concepts and principles of a new topic but figuring I can lookup the detail if/when I actually need to apply that knowledge (particularly so in the WWW era, how to open a milk carton?, just look at the YouTube video!). So I think there is a shift to just-in-time learning happening but this relies on the meta-skill of information searching. I suspect poorly paid and poorly educated teachers in low s-e countries are more likely to teach using the same methods that they were taught by. This may lead to the mindset of “if I read this Wikipedia article many times through, I will have mastered the topic”.
Having travelled in lower s-e countries in Africa recently, many people do see education as the key to a better future and actively invest in their children’s education for that reason. In Kenya I often heard the term “lean family” which meant a small number of children getting the best education the family could afford rather than the traditional large family. I think this was something actively promoted by their government. Many parents proudly talked about making a personal sacrifice of some kind which was undertaken to have more money for children’s education.
I think many of us in more developed economies see Wikipedia as useful. I suspect there are others who see it as potentially life-changing.
Sent from my iPad
On 24 Jan 2019, at 7:33 am, Leila Zia leila@wikimedia.org wrote:
Hi Micru,
One hypothesis that we have is that in countries with lower socio-economic status, the reader may not have access to a variety of different sources for their reading needs (the availability of printed material, books, ... can be more limited in these countries.). At the moment, we don't have enough data from a diverse enough subset of countries to be able to look into this. I'm hoping that in the future iterations we can sample by country and collect enough data to be able to validate or reject this hypothesis.
Do you have other hypotheses as why this may be happening?
Best, Leila
On Wed, Jan 23, 2019 at 3:02 PM David Cuenca Tudela dacuetu@gmail.com
wrote:
Hi Leila,
I'm curious about the in-depth reading differences according to the socio-economic status. Why do you think such differences exist?
Regards, Micru
On Sat, Jan 19, 2019 at 1:15 AM Leila Zia leila@wikimedia.org wrote:
Hi all,
As some of you know, we started a line of research back in 2016 to understand Wikipedia readers better. We published the first taxonomy of Wikipedia readers and we studied and characterized the reader types in English Wikipedia [1]. During the past 1+ year, we focused on learning about the potential differences of Wikipedia readers across languages based on the taxonomy built in [1]. We've learned a lot, and today we're sharing what we learnt with you.
Some pointers:
- Publication: https://arxiv.org/abs/1812.00474
- Data:
https://figshare.com/articles/Why_the_World_Reads_Wikipedia/7579937/1
- (under continuous improvement) Research page on meta:
https://meta.wikimedia.org/wiki/Research:Characterizing_Wikipedia_Reader_Beh...
- Research showcase presentation:
https://www.mediawiki.org/wiki/Wikimedia_Research/Showcase#December_2018
- A series of presentations to WMF teams and community: Look for tasks
under https://phabricator.wikimedia.org/T201699 with title "Present the results of WtWRW" for link to slides and more info when available.
- We will send out a blog post about it hopefully soon. A blog post
about the intermediate results is at
https://wikimediafoundation.org/2018/03/15/why-the-world-reads-wikipedia/
In a nutshell:
- We ran the taxonomy of Wikipedia readers in 14 languages and
measured the prevalence of Wikipedia use-cases and characterized Wikipedia readers in these languages.
- While we observe similarities in terms of the prevalence of the use
cases as well as the way we can characterize readers, we can see that Wikipedia languages lend themselves to different distributions of readership and characteristics. In many cases, the one-size-fit-all solutions may simply not work for readers.
- Intrinsic learning remains as the number one motivation for people
to come to Wikipedia in the majority of the languages, followed by media.
- In-depth reading and the reading of scientific oriented topics is
highly and negatively correlated with the socio-economic status and Human Development Index of countries the readers in these languages are coming from. Long articles that may seem just too long for the bulk of our audience in US, Japan, and the Netherlands is in high demand in India, Bolivia, Argentina, Panamá, México, …
- ...
This research was not possible without the extensive contributions by our formal collaborators: Florian Lemmerich (RWTH Aachen University) and Bob West (EPFL). On the WMF end, I was fortunate to work with Diego Saez on this project as well as more recently, Isaac Johnson. And all those in the Reading Web and Legal team who supported us throughout the process. I also want to underline the amazing work that the volunteers in the languages in the study did to support us heavily to learn more about their languages, not only through help with communications within their communities but also with the translation task which was not an easy one as they were asked to offer their time not only to translate but also do in-person meetings with us for us to make sure the intent of the question is translated the same way across the languages. Usernames Strainu, Tgr, Amire80, Awossink, Antanana, Lyzzy, Shangkuanlc, Whym, Kaganer, عباد_ديرانية, Satdeep_Gill, Racso, Hasive: Thank you!
Next we are going to extend this study to include demographics information. More information about it coming out in the next few weeks. (And I will send out a separate email to wikimedia-l about this topic and future research over the weekend. I need some time to finalize the message to make the message most useful for that audience.:)
Best, Leila
[1] https://arxiv.org/abs/1702.05379
-- Leila Zia Senior Research Scientist, Lead Wikimedia Foundation
Wiki-research-l mailing list Wiki-research-l@lists.wikimedia.org https://lists.wikimedia.org/mailman/listinfo/wiki-research-l
-- Etiamsi omnes, ego non _______________________________________________ Wiki-research-l mailing list Wiki-research-l@lists.wikimedia.org https://lists.wikimedia.org/mailman/listinfo/wiki-research-l
Wiki-research-l mailing list Wiki-research-l@lists.wikimedia.org https://lists.wikimedia.org/mailman/listinfo/wiki-research-l
Wiki-research-l mailing list Wiki-research-l@lists.wikimedia.org https://lists.wikimedia.org/mailman/listinfo/wiki-research-l
wiki-research-l@lists.wikimedia.org