Robert Thomson | Hokkaido University (original) (raw)

Papers by Robert Thomson

Proceedings of the International Conference on Internet Studies (NETs2013), Sep 7, 2013

Facebook’s global reach suggests good potential for recruiting research participants and collecti... more Facebook’s global reach suggests good potential for recruiting research participants and collecting objective behavioral data for cross-cultural research. Previous literature suggests the usefulness of Facebook advertisements to recruit participants in single-country studies. However, Facebook advert use in multi-country studies has not yet been reported. Nor are there any reports about soliciting Facebook user data (via Facebook applications) using Facebook advertisements. This paper contributes to this gap in Internet research literature, reporting the effectiveness of Facebook advertisements to recruit participants, and solicit anonymized Facebook user data, in a 20-country study about privacy concern on Facebook. In 7 days, 399 Facebook users from 18 countries responded to country-targeted advertisements in 13 languages. Response rates (ratio of advert clicks to valid responses) per country varied from 0% up to 14%. Overall, two-thirds of countries’ response rates were below 5%. We conclude for multi-national studies, Facebook advertisements may have potential for simple participant recruitment for surveys, but have limitations for soliciting Facebook user data. For user data collection, methods such as Amazon Mechanical Turk and snowball sampling may be more effective, but can be limited in their international reach.

This paper focuses on the micro-blogging service Twitter, looking at source credibility for infor... more This paper focuses on the micro-blogging service Twitter, looking at source credibility for information shared in relation to the Fukushima Daiichi nuclear power plant disaster in Japan. We look at the sources, credibility, and between-language differences in information shared in the month following the disaster. Messages were categorized by user, location, language, type, and credibility of information source. Tweets with reference to third-party information made up the bulk of messages sent, and it was also found that a majority of those sources were highly credible, including established institutions, traditional media outlets, and highly credible individuals. In general, profile anonymity proved to be correlated with a higher propensity to share information from low credibility sources. However, Japanese-language tweeters, while more likely to have anonymous profiles, referenced low-credibility sources less often than non-Japanese tweeters, suggesting proximity to the disaster mediating the degree of credibility of shared content.

Previous studies have shown clear cultural differences in how Japanese and American social networ... more Previous studies have shown clear cultural differences in how Japanese and American social network site (SNS) users interract with differing SNS platforms (see Barker and Ota, 2011; Fogg and Iizawa, 2008; Takahashi, 2010). In this study of 131 Japanese SNS users who use both Facebook and Mixi, self-disclosure, numbers of contacts, in-group numbers, and levels of perceived commitment on Mixi and Facebook were measured. The study found that such users showed a lower level of self disclosure, connected with fewer people and had fewer categories of contacts, and felt a higher degree of commitment on Mixi than on Facebook. It is suggested that these differences stem from differing responses to percieved relational mobility on each platform; Mixi being a socioecological environment which reflects low-relationally mobile Japanese society, and Facebook being a socioecological environment which reflects high relationally mobile North American social environments.

Posters by Robert Thomson

In recent years there has been an exponential increase in the amount of research making use of on... more In recent years there has been an exponential increase in the amount of research making use of online convenience samples from crowd sourcing websites, particularly Amazon's Mechanical Turk. While several studies have explored the features of MTurk's demographics there have been few comparable studies for other crowdsourcing platforms outside of the US.
To help address this we conducted the first large scale study of Lancers.co.jp, Japan's largest crowdsourcing website, and assessed its value as a participant pool for social science research, by addressing 3 Core Questions: 1. What are the common demographics of Lancers' users? 2. Do the responses of Lancers users meet acceptable standards of quality? 3. What do users consider to be fair rates of compensation? To collect our sample we posted a task to Lancers that required users to complete a quick survey (avg. 20 mins) for a small amount of compensation (avg. 500 JPY). To insure a broad representation of users we released the tasks in 10 staggered batches over 14 days (including weekends) in morning, afternoon, and evening slots.

Proceedings of the International Conference on Internet Studies (NETs2013), Sep 7, 2013

Facebook’s global reach suggests good potential for recruiting research participants and collecti... more Facebook’s global reach suggests good potential for recruiting research participants and collecting objective behavioral data for cross-cultural research. Previous literature suggests the usefulness of Facebook advertisements to recruit participants in single-country studies. However, Facebook advert use in multi-country studies has not yet been reported. Nor are there any reports about soliciting Facebook user data (via Facebook applications) using Facebook advertisements. This paper contributes to this gap in Internet research literature, reporting the effectiveness of Facebook advertisements to recruit participants, and solicit anonymized Facebook user data, in a 20-country study about privacy concern on Facebook. In 7 days, 399 Facebook users from 18 countries responded to country-targeted advertisements in 13 languages. Response rates (ratio of advert clicks to valid responses) per country varied from 0% up to 14%. Overall, two-thirds of countries’ response rates were below 5%. We conclude for multi-national studies, Facebook advertisements may have potential for simple participant recruitment for surveys, but have limitations for soliciting Facebook user data. For user data collection, methods such as Amazon Mechanical Turk and snowball sampling may be more effective, but can be limited in their international reach.

This paper focuses on the micro-blogging service Twitter, looking at source credibility for infor... more This paper focuses on the micro-blogging service Twitter, looking at source credibility for information shared in relation to the Fukushima Daiichi nuclear power plant disaster in Japan. We look at the sources, credibility, and between-language differences in information shared in the month following the disaster. Messages were categorized by user, location, language, type, and credibility of information source. Tweets with reference to third-party information made up the bulk of messages sent, and it was also found that a majority of those sources were highly credible, including established institutions, traditional media outlets, and highly credible individuals. In general, profile anonymity proved to be correlated with a higher propensity to share information from low credibility sources. However, Japanese-language tweeters, while more likely to have anonymous profiles, referenced low-credibility sources less often than non-Japanese tweeters, suggesting proximity to the disaster mediating the degree of credibility of shared content.

Previous studies have shown clear cultural differences in how Japanese and American social networ... more Previous studies have shown clear cultural differences in how Japanese and American social network site (SNS) users interract with differing SNS platforms (see Barker and Ota, 2011; Fogg and Iizawa, 2008; Takahashi, 2010). In this study of 131 Japanese SNS users who use both Facebook and Mixi, self-disclosure, numbers of contacts, in-group numbers, and levels of perceived commitment on Mixi and Facebook were measured. The study found that such users showed a lower level of self disclosure, connected with fewer people and had fewer categories of contacts, and felt a higher degree of commitment on Mixi than on Facebook. It is suggested that these differences stem from differing responses to percieved relational mobility on each platform; Mixi being a socioecological environment which reflects low-relationally mobile Japanese society, and Facebook being a socioecological environment which reflects high relationally mobile North American social environments.

In recent years there has been an exponential increase in the amount of research making use of on... more In recent years there has been an exponential increase in the amount of research making use of online convenience samples from crowd sourcing websites, particularly Amazon's Mechanical Turk. While several studies have explored the features of MTurk's demographics there have been few comparable studies for other crowdsourcing platforms outside of the US.
To help address this we conducted the first large scale study of Lancers.co.jp, Japan's largest crowdsourcing website, and assessed its value as a participant pool for social science research, by addressing 3 Core Questions: 1. What are the common demographics of Lancers' users? 2. Do the responses of Lancers users meet acceptable standards of quality? 3. What do users consider to be fair rates of compensation? To collect our sample we posted a task to Lancers that required users to complete a quick survey (avg. 20 mins) for a small amount of compensation (avg. 500 JPY). To insure a broad representation of users we released the tasks in 10 staggered batches over 14 days (including weekends) in morning, afternoon, and evening slots.