Formal and Informal Surveys for President

by admin | Mar 26, 2016 | Election 2016, Local News / Bohol Balita, Tech Talk

By: Jerome Auza

The last few weeks has been very interesting in the campaigns of the presidential candidates for the May 9, 2016 elections. Particularly interesting is the fact that formal surveys and informal surveys have widely contrasting results.

In recent formal surveys, Poe is leading while in mock elections in different universities nationwide, as well as online surveys and even in “cup” surveys in a 24-hour convenience store show Duterte consistently in the lead. The cup survey is done using cups for drinks printed with the face and names of the different candidates. The Duterte cups always get out of stock.

The formal surveys are conducted in a scientific method and the selection of respondents are done randomly in order to come up with results that are reliable and statistically representing the whole population.

The mock elections and online surveys are non-scientific because for one, they can only cover a certain group of people. For example, university surveys would include only students. In online surveys, these include only those with Internet access. In other words, they may not really represent the sentiment of the whole population.

However, there is the fact that the informal methods are done with a much larger sample of the population. In statistics, if you take a large enough sample, the tendency of the sample data is to match the whole sample. The mock elections held in universities are done nationwide, with sample size of around 1000 each or more. If you take into account that this is nationwide and the university students are obviously the youth, then it may be safe to say that the youth vote will go to Duterte. Results are varying in percentage but Duterte almost always leads the mock election results.

Now looking at the online surveys, I would like to use the Rappler online survey for February 2016 because the number of respondents is around 100,000. Also, another reason for using Rappler is that they learned how to ensure that only valid votes are counted because in an earlier survey, they discovered that votes for Roxas surged on a specific period with the votes identified as coming from China and Russia. That obviously was an attempt to manipulate the results. So assuming Rappler’s data for the February survey is filtered of invalid votes and with the number of respondents at 100,000, then at 39%, Duterte’s lead is quite significant.

Again, because of the tendency of a large sample to represent the whole population, it may be safe to say that the Rappler survey represents the Filipinos with access to the Internet.

Two large segments of our voting population, the youth and voters with Internet access are voting for Duterte. Obviously, there is an overlap of these two segments but if you combine these two segments, they are a very large percentage of all the voters.

Curiously, the most recent formal survey now shows that Duterte is about to gain the lead. Will Duterte eventually show as the significant leader in voting preference in the formal surveys in the coming weeks?

I would review the survey data again in a few weeks time to see where the numbers are. I have two questions for now:

Will the volunteerism of the Duterte supporters prove to be a force to be reckoned with versus traditional political machineries? Will social media ads of Duterte voluntarily done by supporters beat the other candidates’ heavily funded TV ads?

It remains to be seen.