A Critical Review of Opinion Polls relating to Iranian Voting Intentions: Problems of Research Methodology as applied to Complex Societies

After the 10th Iranian Presidential election on June 12, 2009, several public opinion polls taken in Iran attracted the attention of policy-makers and journalists around the world because of the political crisis that followed. In this paper I first review critically the polls conducted by the WPO (WorldPublicOpinion.org), PIPA (Program on International Policy Attitudes) at the University of Maryland. I also review an essay by Steven Kull, which is based on the aforementioned poll results and which in my opinion leads to false conclusions concerning Iran’s political prospects. I also discuss “An Analysis of Multiple Polls of the Iranian Public,” published by WPO-PIPA on February 3 2010. The present paper arrives at the overall conclusion that it is impossible to obtain an accurate image of political opinions in societies as complicated as that of Iran by concentrating on only one technique of research and analysis, especially when the political and social situation in the society concerned is in a state of constant flux.

Ramadan school holidays as a natural experiment : impacts of seasonality on school dropout in Bangladesh

In 2000, Ramadan school vacation coincided with the original annual exam period of December in Bangladesh. This forced schools to pre-pone their final exam schedules in November, which was the month before the harvest begins. 'Ramadan 2000' is a natural experiment that reduced the labor demand for children during the exam period. Using household level panel data of 2000 and 2003, and after controlling for various unobservable variations including individual fixed effects, aggregate year effects, and subdistrict-level year effects, this paper finds evidence of statistically significant impact of seasonal labor demand on school dropout in Bangladesh among the children from agricultural households.