Comments and Observations on Election Surveys

 Comments and Observations on Election Surveys

Past election surveys show that the most accurate were done just about a week or 2 before elections. Current pre election surveys are not reliable gauge because of still many unknown factors at this early stage and doubtful survey methods used by many survey groups. Worst, some are conducted for mind conditioning of the public or just part of  election strategy. Among the more reliable are the well - established, Pulse Asia and Social Weather Station, which follow standard survey methods. The following are comments and observations on some current election surveys:

1. Kalye Survey - Several street surveys already made with the latest showing BBM as leading in Mindanao (Headline of The Manila Times 1 Nov 2021). It claims the use of YouTube vlogger in real time real man survey on the street. Since no factual evidence are presented, the method remains contradictory and partial by its very nature. Assuming that it was really conducted, a top Filipino statistician who previously was head or member of major statistical bodies both local and international, Jose Ramon Albert, said that Kalye survey is highly unreliable  in measuring trend of voting for national positions.

2. On - Line Survey - Non Face to Face survey done on-line particularly Facebook is beset with problems because of the proliferation of troll farms and Buying Facebook  Likes Services. This was shown by the recent Facebook survey conducted by Rappler and Manila Bulletin which were participated in  by fake accounts. An online tool Sparktoro likewise discovered fake followers of some presidential candidates, an indication of existence of their own trolls. In the case of  Rappler, its survey registered that Leni was leading from 7-12 October but on the night of 12 to the morning of 13 October or about 10 hours the survey site was swamped by BBM likes even coming as far away as Pakistan and Arab account holders, which led to the system take down. Relatedly, a tech editor of Manila Bulletin, Art Samaniego, revealed that 100 Facebook Reactions cost 3$US while 20000 Facebook Reactions cost 29 $US.

3. Publicus Asia - Performs better than the Kalye survey and On-line/ Facebook survey. However, its 3QTR survey where BBM registers huge lead is vague without explaining even in general terms how it was arrived at. Its claim of using 1, 500 respondents from a total of 200, 000 panel currently maintained by an unnamed Singaporean company for market research did not help as it raises more questions. The academic credentials of Publicus owner, Maria Lourdes Tiquia and Chief Data scientist, David Yap look impressive enough. However, their works at Publicus need to show that they are not influenced by the fact that Ms Tiquia is a mainstay of The Manila Times while Mr Yap was a senior adviser of Sen Imee Marcos in the 2019 election.

********