PhD thesis published: Romania at voting age, Lambert Academic Publishing, 2011

https://www.morebooks.de/store/gb/book/romania-at-voting-age/isbn/978-3-8465-2445-9

Short presentation: The integrated analysis of electoral, socio-economic and geographic data brings new light in understanding deep social cleavages and their political translation in post-communist Romania. The impact of the social characteristics of the voters, the systemic factors of the political system and territory on the electoral behaviour of Romanians is observed and analysed during the first 18 years of democratic rule. One of the compelling conclusions is that democracy is a learning process. But in 18 years of democracy, the majority of voters have given up the democracy school. They do not turn out to vote as they are profoundly disengaged with politics. Voters, political system and even geography are changing. Although geography is stable throughout this book, joining the European Union is expected to have a significant further impact on electoral behaviour.

Quantile map of the PNL votes at the early elections of November 28th elections (Moldova)

Voting for the Liberal Party seems clustered in the districts with the highest shares of voters who declare Moldovan or Romanian ethnicity: Calarasi, Nisporeni, Straseni, Hinceni, Ialoveni, Chisinau.
Some additional comments based on geo-statistical analysis:
1. At this level of analysis, the highest spatial competition was between the PCRM and the PDM (Moran’s I = 0.27). They both have higher shares in the Northern districts: Briceni, Edinet, Riscani, Glodeni, Falesti, Ocnita, Donduseni, Drochia, Balti and Sangerei.
2. PLDM and PNL are in spatial competition; it is low but it exists (0.11). The districts of higher territorial competition are the ones in the centre of the country: Orhei, Calarasi, Ungheni, Nisporeni, Straseni, Chisinau, Ialoveni, Hincesti.
3. PCRM and PLDM have distinct territorial profiles. The only district they both get high shares of votes is Leova.
4. PDM and PNL have also distinct territorial profiles. However, they fight each other in three districts: Hincesti, Nisporeni and Calarasi.

Quantile map of the PDM votes at the early elections of November 28th elections (Moldova)

The Democratic Party has got higher shares in the districts of Edinet, Briceni, Singerei, Floresti, Nisporeni and Leova.

Quantile map of the PLDM votes at the early elections of November 28th elections (Moldova)

The spatial dependence is noticeable in the higher rates that the PLDM gets in the districts Orhei, Telenesti, Straseni, Ialoveni, Hincesti, Nisporeni and Stefan Voda.

Quantile map of the PCRM votes at the early elections of November 28th elections (Moldova)

Visually, the Party of Communists seems to be preferred in districts with higher shares of Russian, Ukrainian and Gagauze population: Ocnita, Balti, Gagauzia and Taraclia. Further analysis needs to be conducted in order to statistically establish a territorial dependence. Dubasari seems to be a special case.

Turnout at the early elections in Moldova (Nov. 28, 2010)

I put some results on the map drawn at district (raion) level. The source of the data is www.voteaza.md. Votes of the Moldovians from abroad were not included. As the final data do not match any of the official reports of CEC on turnout reporting, the formula used to compute electoral participation was number of participants divided by total number of voters (on permanent + supplementary lists). The number of participants is different (larger) from the number of valid votes as some ballots are invalid due to incorrect handling. The local electoral agency (CEC) makes however a difference between number of participants and number of persons having received a ballot. I was not able to detect where this comes from so far. Apparently, one may receive a ballot that s/he decides not to use for actually voting.

Cyberwar again, more seriously

The Air Force cyberwarfare manual published on the website of the LeMay Center for Doctrine Development and Education gives some hints about the shadowy cyberwar world. The entire document is available on their website: http://www.cadre.au.af.mil/main.htm.
Some excerpts:

• “In cyberspace, the time between execution and effect can be milliseconds. Nonetheless, the observe-orient-decide-act (OODA) loop remains a valid construct for examining the decision cycle in cyberspace. Ongoing operations can be considered those operations that span past the phases of warfare.” (page 30)
• “Even for ongoing operations, planning at the strategic level is imperative because cyberspace operations can create effects simultaneously at the strategic, operational, and tactical levels across multiple domains.” (p. 30)
• “Cyberspace operations can enable creation of many effects that formerly required physical attack to accomplish.” (p. 30)
• “Cyberspace operations normally are planned as part of major operations and campaigns, homeland operations, crisis response, and limited contingency operations.” (p. 31)
• In order to accomplish this assessment process within the interrelated and complex nature of many cyberspace operations, operational-level cyberspace planners and analysts should develop an intimate understanding of the linkage between cyberspace and the supported mission or operation. This requires direct feedback from those closest to observing the intended effects, such as the Airmen executing cyberenabled Air Force missions or the warfighters in theater, in order to assess the level of cyberspace performance and effectiveness. For example, the assessment of cyberspace operational effects in support of influence operations requires an in-depth understanding of the warfighter’s desired impact on behavior and the ability to measure any resulting behavioral changes. (p. 33)

See the whole AP story on KSL website.
James Lewis – a senior fellow at the Center for Strategic and International Studies – declared that the government still hasn’t decided whether offensive cyberwarfare is the province of the military or intelligence agencies.
Overall U.S. military cyberwarfare operations will be the job of the U.S. Cyber Command, which began limited operations in May. It will have components from the Army, Air Force, Navy and Marines.
Responsibility for civilian and government cybersecurity is less clear. Congress is debating between giving more power to the Homeland Security Department or the White House and the National Institute of Standards and Technology.

Homeland Security and the National Security Agency announced this month they would cooperate to strengthen the nation’s cybersecurity.

Cyberwar by US Army

That is a news worth reading: Military asserts right to return cyber attacks.
How to deal with attacks launched through proxies? You may destroy the proxies, cause damage in other countries and the attacker be far away and safe …

Results after recounting the canceled votes

The recounting of the canceled votes asked by the Constitutional Court reduced the error of the CSOP-IRES prediction of the election outcome up to 0.06%. Not bad!
The Central Electoral Bureau concludes that – following the recounting of the 137206 canceled votes – 2137 have been declared as being valid. Out of these, 1169 for Basescu and 968 for Geoana.
Doing the maths, it means 50.34% for Basescu and 0.06% margin of error of CSOP prediction of the electoral outcome on Dec. 6th on B1TV: 50.4% for Basescu and 49.6% for Geoana. The only correct estimation of the election outcome in terms of ranking and score of the candidates.

Core electoral areas

Third question, is there any regional pattern of the voting? How do they compare to the pattern of party voting that I noticed in the European Elections (June 2009)?

Geographical profiling of Basescu’s voters in the first round See PD-L party vote (Jun. 2009, European Elections)

Notes: Red= Homogeneous area of high vote for Basescu Blue= Homogeneous area of low vote for Basescu Pink= Disparity: Neighborhood with low vote for Basescu, unit with high vote for Basescu Light blue= Disparity: Neighborhood with high vote for Basescu, unit with low vote for Basescu

Geographical profiling of Basescu’s voters in the second round See PD-L party vote (Jun. 2009, European Elections)

Notes: Red= Homogeneous area of high vote for Basescu Blue= Homogeneous area of low vote for Basescu Pink= Disparity: Neighborhood with low vote for Basescu, unit with high vote for Basescu Light blue= Disparity: Neighborhood with high vote for Basescu, unit with low vote for Basescu

Geographical profiling of Geoana’s voters in the first round See PSD party vote (Jun. 2009, European Elections)

Notes: Red= Homogeneous area of high vote for Geoana Blue= Homogeneous area of low vote for Geoana Pink= Disparity: Neighborhood with low vote for Geoana, unit with high vote for Geoana Light blue= Disparity: Neighborhood with high vote for Geoana, unit with low vote for Geoana

Geographical profiling of Geoana’s voters in the second round See PSD party vote (Jun. 2009, European Elections)

Notes: Red= Homogeneous area of high vote for Geoana Blue= Homogeneous area of low vote for Geoana Pink= Disparity: Neighborhood with low vote for Geoana, unit with high vote for Geoana Light blue= Disparity: Neighborhood with high vote for Geoana, unit with low vote for Geoana

Geographical profiling of Antonescu’s voters in the first round See PNL party vote (Jun. 2009, European Elections)

Notes: Red= Homogeneous area of high vote for Antonescu Blue= Homogeneous area of low vote for Antonescu Pink= Disparity: Neighborhood with low vote for Antonescu, unit with high vote for Antonescu Light blue= Disparity: Neighborhood with high vote for Antonescu, unit with low vote for Antonescu