No abstract

This paper reports the processes by which an interdisciplinary research project overcame the usual problems of antagonism among disciplines and succeeded in introducing a number of social innovations into a community. In terms of the experience five "principles" of successful social innovation are suggested: (a) the principle of maximum investment, (b) the principle of co-optation, (c) the principle of egalitarian responsibility, (d) the principle of research as creative play, and (e) the principle of ideological research leadership. A high degree of staff involvement is made possible by a relaxation of superego prohibitions against novelty and originality, by allowing maximum feasible participation of all project personnel, and by providing realistically wide but unambiguous boundaries and limits. Under these conditions the research group may take on certain of the emotionally charged, proselyting characteristics of a small social movement and may have an impact on the community out of all proportion to its size.

Small, irregular terraces on hillslopes, or terracettes, are common landscape features throughout west central China. Despite their prevalence, there is limited understanding of the nature of these topographic features, the processes that form them, and the role humans played in their formation. We used an interdisciplinary approach to investigate the geology, ecology, and cultural history of terracette development within Jiuzhaigou National Park, Sichuan Province, China. Terracettes occur on south facing, 20° slopes at 2500 m elevation, which appears to coincide with places people historically preferred to build villages. Ethnographic interviews suggest that traditional swidden agricultural cycles removed tree roots, causing the loess sediments to lose cohesion, slump, and the terrace risers to retreat uphill over time. This evidence is supported by landslide debris at terracette faces. Archaeological analysis of terracette sites reveal remains of rammed spread soil structures, bones, stone tools, and ceramics dating from at least 2200 years before present within a distinct paleosol layer. Radiocarbon and optically stimulated luminescence dating of terracette sediments ranged in age from between 1500 and 2000 14C yr BP and between 16 and 0.30"ka, respectively. These multiple lines of evidence indicate a long history of human habitation within Jiuzhaigou National Park and taken together, suggest strong links between terracette formation and human-landuse interactions.

Google Insights for Search provides a new and rich data source for political scientists, which may be particularly useful for state politics scholars. We outline the prior uses of Google Insights for Search in social and health sciences, explain the data-generating process, and test for the first time the validity of this data for state politics research. Our empirical test of validity shows that Google searches for ballot measures' names and topics in state one week before the 2008 Presidential election correlate with actual participation on those ballot measures. This demonstrates that the more Internet searches there were for a ballot measure, the less likely voters were to rolloff (not answering the question), and establishes the construct validity for this data for one important topic in state politics research. We also outline the limitations to this data source.

We evaluate the effect on test scores of a UK education reform which has increased funding of schools and encouraged their specialisation in particular subject areas, enhancing pupil choice and competition between schools. Using several data sets, we apply cross-sectional and difference-indifferences matching models, to confront issues of the choice of an appropriate control group and different forms of selection bias. We demonstrate a statistically significant causal effect of the specialist schools policy on test score outcomes. The duration of specialisation matters, and we consistently find that the longer a school has been specialist the larger is the impact on test scores. We finally disentangle the funding effect from a specialisation effect, and the latter occurs yielding relatively large improvements in test scores in particular subjects.