The weeds are deep in this one.
All the findings: @3ieNews has mapped out existing evidence on citizen-state relations, put together a linked matrix organized according to the interventions and outcomes measured, plus confidence levels. It includes “18 completed systematic reviews and two systematic review protocols, 305 completed impact evaluations reported in 280 papers, 60 ongoing impact evaluations reported in 59 papers.” And everything is linked. And it’s not ugly. Swoon. h/t
Meanwhile, @bbcmediaaction blogs on their new data portal, which collects survey data on “rarely polled” people in 13 developing countries, providing insights on media use, governance and freedom of expression perspectives. Continue reading “research links w 14-17”
Politically marginalized groups have less access to the internet, worldwide. This shocker based on network measurements over 8 years and identification of politically relevant groups as defined by the Ethnic Power Relations (EPR).
The relationship between online and offline activism is messy, according to a survey of 1023 adolescents from five Balkan countries, while a year-long study in Uganda and Kenya documents ways that citizen-generated data can be used to improve service delivery and policy, but finds that relationships matter, and that measurement is hard.
Why governments implement e-participation: Governments are most willing to implement e-participation schemes when they enjoy strong ICT infrastructure and human capital, according to a review of archival data from 153 countries (pulled from UN E-Gov surveys and the World Bank’s Development and Governance Indicator sets from 2010-2012). Most interestingly, quality of governance did not positively correlate with willingness to implement e-participation, and the authors suggest that advocates should accordingly push for better ICT infrastructure and human resources, “to move up the ladder of e-government maturity.” Also worth noting, willingness to conduct e-consultations was the only form of willingness negatively associated with e-government maturity. On this last point, the authors speculate is because governments are afraid that consultative processes will slow down e-government processes.
Continue reading “research links w 12- 17”
Twitter advocacy can bypass mainstream media that excludes non-elite voices, according to a study of how #IfTheyGunnedMeDown was used following 2014 police shootings in Ferguson, Missouri. That’s good news for digital advocacy innovators, but important to remember that people don’t feel safe online and don’t understand how their personal information gets used, but most aren’t willing to get trained. This according to survey of “Mozilla community members” (n=30,000+) by the Mozilla Foundation.
All the FOIA researchin the US:
Freedom of information in the US has been suffering in recent years from reduced access to information and increased FOIA denials, and a survey of “336 freedom of information experts” suggests that it is only going to get worse. On the other hand, “data pulled from over 30,000 FOIA requests” by @MuckRock suggests that the US govt is much better at responding to FOIA requests than we tend to think. Digging into the details, US FOIA gets used by business actors more than any other group, single actors dominate requests from most groups (@MuckRock among news groups, @JudicialWatch among NGOs) and Democratic committees dramatically out-FOIA their Republican counterparts (81/7). These and other fascinating insights in ‘s analysis of 229,000 requests to 85 government agencies.
Continue reading “research links w 11-17”
Papers & Findings
Using the internet leads to civic engagement. Sometimes. Kind of. This according to structural equation analysis of US college survey data (n=2000), which finds “both positive and negative effects” of internet use on engagement patterns (students who share political opinions online tend to have less political conversations offline) but also identifies “feedback loops” between online and offline engagement activities (the effects of offline activity on online activity being slightly stronger than the other way around), and online info-gathering as having particular predictive power for engagement, both online and off.
Crowdsourcers, microtaskers and distributed team wranglers, choose your platforms wisely. Online groups do not display “collective intelligence.” This based on a replication of the 2010 study that demonstrated collective intelligence among groups, but using online groups. The likely explanation is that collective intelligence relies on social sensitivity of group members rather than individual intelligence, and that social sensitivity in turn relies on social cues absent in the type of online group-work tested in this study.
Open data enthusiasts and civic techies are good at exploiting norms around #open for mobilization, but “open data intermediaries lack a shared culture and political understandings necessary for broader and more impactful action,” with “exceedingly fragmented” perspectives on what data can and should do. This from work with focus groups in a mid-sized US city. Continue reading “research links w 6-17”
Papers and Findings
A cross-disciplinary team of researchers has developed an NLP method that can predict judgements in the European Court of Human Rights with 79% accuracy, based on an analysis of case documents. AI to replace judges? Perhaps. More comforting is their conclusion that this finding supports the theory of legal realism, “suggesting that judicial decision-making is significantly affected by the stimulus of the facts.”
100 Stories: The Impact of Open Access. That’s the bombastic title of a forthcoming paper focused on public access to scientific research, aiming to change “how we talk about the impact of open access.” They do indeed present 100 stories, but in short format and not of “impact” per se. The bar for inclusion is unsurprisingly low. An example (in full) from the category of advancing innovation:
“New patents matched against University’s patent portfolio Iowa State University patents have been downloaded over 16,000 times by 275 institutions. 35% of the patent downloads have been from high-profile corporations such as IBM (33), Unilever (11), Dow-Corning (7), Hewlett-Packard (6), and Deere & Co (5).”
Also, it’s rife with depressing word clip art charts.
Continue reading “research links w 43-44”