IN BRIEF 2015/6 ELECTION OBSERVATION IN SOLOMON ISLANDS NICOLE HALEY AND KERRY ZUBRINICH State, Society & Governance in Melanesia ips.cap.anu.edu.au/ssgm In November 2014, Solomon Islands conducted its ninth general election since independence. With accreditation from the Solomon Islands Electoral Commission and financial support from the Australian Government through the Department of Foreign Affairs and Trade, the State, Society and Governance in Melanesia (SSGM) program and the Centre for Democratic Institutions deployed 12 teams to undertake systematic observation, monitoring and assessment of the elections. Overall, the group comprised academics and professional staff from both programs, affiliated academics, and 60 locally engaged domestic observers. The approach employed built on observations of the 2007 and 2012 Papua New Guinea elections (Haley and Anere 2009; Haley and Zubrinich 2013). Observations were undertaken in six provinces (Western, Isabel, Central, Malaita, Guadalcanal and Makira) and 12 separate constituencies, spread across the country to ensure a national perspective. As in PNG, constituencies were purposefully select- ed to ensure a representative mix. Included were rural and urban constituencies; some with large enrolments and some with small enrolments; some with high incumbent turnover and some without; different results patterns and different histories of electioneering and electoral competition. Through our work in PNG over the past decade, SSGM scholars have developed an election observa- tion tool that generates quantitative and qualitative data in a systematic way to inform: improved elec- tion management, future donor engagement, politi- cal economy analysis, and academic research con- cerning electoral politics and the changing nature of political culture in Melanesia. In the Solomon Islands case, individual observers were engaged to undertake 20 days’ observations over a five-week period and were given the task of making observa- tions during the campaign, pre-poll, polling, and counting periods. In total, observations were carried out at 148 polling stations. Observers also conducted citizen surveys during the pre-poll and post-polling periods. In total, 2564 citizens were surveyed. Election Monitoring Tools A comprehensive observer journal remains the major observation tool. It is intended not merely to collect quantitative data but to assess the conduct of the election and citizen experiences of it through the campaign, on polling day and in the immediate aftermath of the election. The journal is organised in chronological order starting with voter registration and concluding with the declarations. Local observers were trained to use the journal in late October 2014 and were deployed to their respective constituencies soon thereafter. They undertook detailed observations from 10–21 November. For the first time ever, tablets were used to record citizen responses to two surveys carried out pre- and post-polling periods. This reduced data-entry errors and the time needed for data analysis. Given financial and administrative constraints, however, a rigorous debrief workshop was not possible. The journal is divided into 12 sections, with particular attention given to: the new Biometric Voter Registration (BVR) System introduced by the Solomon Islands Government for this election, the performance of key electoral personnel, the nature of campaigning, the influence of political parties, the influence of ‘money politics’, vote buying and political gifting, the conduct of polling, voting irregularities, and the impact of mobile phones on the election. Importantly, observations and reflections are triangulated against citizen experiences, with individual topics explored in a number of ways. Key Findings A key success was the new electoral roll compiled with the use of BVR. It was well received, enjoys popular support, raised confidence in the integrity of the electoral process and makes it difficult for vot- ers to double vote. Procedural inconsistencies and some instances of electoral fraud were identified on polling day, although these did not appear to be widespread nor significant enough to undermine the overall integrity of the process.