Survey-Shmurvey

Behavioralism.....
An aspect of Behavioralism in aiding in predicting political behavior is basically marginalizing the normative theories because it doesn't help in the empirical science of politics. In respect to both SWS and Pulse Asia surveys, their attempts or pursuit in an coming up with an empirical statement is a method or chracteristic thats of behavioralism. They do surveys to come up a tangible truth that although can still be tested as to whether indeed true or false, can come out as something everyone can hold on to through large survey samples and very small error rates. To put this all short, their methods attempt to make all judgements made "scientific"

-------------

A thousand and two hundred sample?

Statiscally, its prefferable to use a bigger sample size than that.. the ideal actually would be 60%.. but to answer whether or not 1,200 respondents are enough, maybe it is, judging all statistical methods and analysis be done smoothly and with extreme care. Compared to the entire population, 1.200 is really so small to come up with a generalized answer or to generalize the opinions of the Filipino people. But if continuous polls are done within the same people so that changes in time would be considered because people's minds are totally unpredictable and the slightest things make people's thoughts about something change. Also take into consideration what kind of people these respondents are, whether they belong to a specific economic level in the society, etc. Their exposure to media and the entirity of the topic being asked about should be considered too. All factors that may affect these people's votes should be studied carefully to come up with a mathematical equation that would justify the validity of the results of the surveys. But like i said, 1,200 is too small of a sample size, and, although possible, will be far more difficult than just choosing a bigger size. Many discrepancies may arise and errors in outcomes will surely show up.

----------------

Alot can be done by carefully monitoring each and everything that goes on in these surveys, a deeper study on the statistical analysis and a constant adjusting of methods based on the attitudes of respondents might help make the results of these surveys more exact.