Tag: surveys

The News Media, The President

(Updated) When the reporting of surveys becomes propaganda (Dear Tangé; Tangé is my term of endearment for friends)

             A tv news program  reported  an SWS survey by saying that Noli de Castro was still the top choice for President and gave the figures as follows: Noli : 29%; Manny Villar: 28%; Loren: 26% .

Photo by David.R.Stoecklein.Tail-End. Right-clicked from www.allposters.com, used here for non- commercial purposes, under the terms of , free service by blog-use of image provided by said site.
Photo by David.R.Stoecklein.Tail-End. Right-clicked from http://www.allposters.com, used here for non- commercial purposes, under the terms of , free service by blog-use of image provided by said site.

   

 

 

     Not that i like any of the presidentiables, i don’t; i’d vote for Chief Justice Puno even if he isn’t/ wouldn’t be a candidate: but one of the most  important feature to mention in reporting surveys is the margin of error. I checked the SWS site and the margin of error for this particular survey is  plus or minus 2 .5% for national percentages (bigger for regions: plus or minus 6% for Metro Manila, Mindanao, Visayas, and plus or minus 4% for “Balance Luzon”).

        That means that figures within plus or minus 2.5%, or a range of 5%, are statistical ties.

       That means the three are statistical ties (not that i like any of them). For that matter, figures from 24% to 29% are statistical ties (24% plus  a 2.5% margin of error or  a maximum of 26.5%;  and 29% minus 2.5% margin of error or a minimum of 26.5% are statistical ties).

     (the tv news program changed the headline to Noli, Manny, Loren top picks; as i was writing this. Anyway.) Update: the evening news (ANC) reverted to the news headline and “story”: “Noli de Castro leads the survey, etc.” without mentioning any margin of error at all, while the free tv version (TV Patrol Channel 2) also ran the same story: “Nangunguna si Kabayan Noli sa survey, etc.”  also without mentioning any margin of error at all.  All these are forms of propaganda and bandwagonning in favor of Noli de Castro in the guise of news stories. The required reading for corruption in the media is “News for Sale” volumes 1 and 2 by Chay Florentino-Hofileña for PCIJ, which investigated and documented how media organizations sold news airtime and space to politicians to promote them.

      In the 2004 elections, the government-owned or government- controlled tabloids then  always bannered that GMA was leading the pack according to surveys when, on those rare occasions, she was ahead by only 1%; they also never reported  when FPJ led the surveys. Those banner headlines and news  stories are grossly misleading (as stated, figures within a 5% range are usually statistical ties) and intentionally misleading. They’re classic examples of surveys as propaganda, or the reporting of surveys as propaganda (i know,  i know, i promised last year to summarize the studies on the reporting of surveys, they’re in the back issues of the Philippine Journalism Review, which are cited in my syllabus, etc. etc., i never got around to it, if you’re a reporter or if you’re reporting on surveys you have a duty to  look them up yourself).

      The other important features that should be mentioned in reporting surveys, of course, are: who sponsored/ financed the survey, the method, the time frame, sample size, etc., etc., they’re all in those studies i mentioned in reporting surveys.  

       I have to run along now i will edit this and put a picture later; as it is, it already  took 45 minutes from me!

       Here’s the SWS media release on this particular survey, from www.sws.org.ph

     Quote “7 November 2008. Third Quarter 2008 Social Weather Survey:

      Quote “Noli De Castro, Manny Villar, and Loren Legarda

are people’s top recommendees for 2010

     Quote “Social Weather Stations

      Quote “Vice-President Noli De Castro, Senate President Manny Villar, and Senator Loren Legarda are the top recommended successors to President Gloria Macapagal Arroyo in 2010, according to the Third Quarter 2008 Social Weather Survey, fielded over September 24-27, 2008.

        Quote “To the survey question, “Sa ilalim ng kasalukuyang Konstitusyon, ang termino ni Pang. Arroyo ay hanggang sa taong 2010 lamang at magkakaroon ng halalan para sa pagka-pangulo sa Mayo 10, 2010. Sinu-sino sa palagay ninyo ang mga magagaling na lider na dapat pumalit kay Pang. Arroyo bilang Presidente? Maaari po kayong magbanggit ng hanggang tatlong sagot.” [Under the present Constitution, the term of Pres. Arroyo is up to 2010 only, and there will be an election for a new President in May 2010. Who do you think are good leaders who should succeed Pres. Arroyo as President? You may give up to three names], 29% named Vice-President De Castro, 28% named Senate President Villar, and 26% named Sen. Legarda.

      Quote “No list of names was provided to prompt the respondents.

      Quote “They were followed by Sen. Panfilo Lacson (17%), Sen. Francis Escudero (16%), former President Joseph Estrada (13%), and Sen. Mar Roxas (13%).

      Quote “The survey found 1% mentioning Pres. Gloria Macapagal-Arroyo, MMDA Chairman Bayani Fernando, Sen. Antonio Trillanes IV, Sen. Juan Ponce Enrile, Sen. Francis Pangilinan, Sen. Mirriam Defesor-Santiago, Sen. Richard Gordon, Sen. Jinggoy Estrada, Cong. Jose De Venecia, Sen. Jamby Madrigal, and Sen. Pia Cayetano.

       Quote “Nine percent could not give an answer, and 9% had no one to recommend.

       Quote “Compared to the previous quarter, those with an increase in mentions were: Senate President Villar, by 3 points, Sen. Escudero, by 2 points, former President Estrada, by 2 points, and Sen. Lacson, by 1 point. The question wording has been exactly the same in the past five quarterly surveys shown in Table 1.

       Quote “The mentions of Vice-President De Castro’s name went down by 2 points, while the mentions of Sen. Legarda and Sen. Roxas were steady.

      Quote “Survey Background

      Quote “The Third Quarter of 2008 Social Weather Survey was conducted over September 24-27, 2008 using face-to-face interviews of 1,500 adults in Metro Manila, the Balance of Luzon, Visayas, and Mindanao (sampling error margins of ±2.5% for national percentages, ±6% for Metro Manila, Visayas and Mindanao, and ±4% for Balance Luzon). The area estimates were weighted by National Statistics Office medium-population projections for 2008 to obtain the national estimates.

       Quote “The survey item on the recommended leaders to succeed Pres. Arroyo is not commissioned, but is done on SWS’s own initiative and released as a public service, with first printing rights assigned to BusinessWorld. #” Closed-quote.

 

 

The News Media

“I beg your pardon”: the Survey

archi2.jpg

Last summer during the election campaign i promised to post some notes for discussion on surveys, but never did. Soweee; begging your pardon. Here are some notes on the survey on the “touted” presidential pardon of the accused in People vs. Estrada for plunder.

The following is a portion of the syllabus in a class, on the topic, media coverage of surveys. You might want to read these articles for a better, more comprehensive discussion:

quote “D. Covering Elections and Reporting on Opinion Polls

Articles: `Embedded’ Journalists: Philippine Style” by Evelyn Katigbak, Philippine Journalism Review, April-May June-July 2004

“Surveys in the News: Front Page Treatment, Mostly” by the editors, PJR, supra

“Reporting Election-Related Violence” by the editors, supra

“Testing for Reliability”, with sidebar:

“Reporting the Surveys” by Angela M. Banadios, Anna Patricia de Leon, and Venus Elumbre, supra” . closed-quote.

I usually pose these questions: Does it matter who sponsored the survey? The sponsors are not allowed to change the numbers, so does it matter who sponsored it, and is it necessary to provide context; and the side of other parties and their analysis?

In the opposition-sponsored SWS survey on the pardon of Erap, the angle of the Inquirer, gleaned from the choice of persons interviewed (Erap and his son), used quotes to produce a slant that an overwhelming majority was sympathetic to the ousted president.

Ironically, the supposed presidential pardon after a supposed verdict of guilt was a scenario foisted a couple of years ago reportedly by a group in Malacañang. When word of it spread, one of the volunteer lawyers for the prosecution, Atty. Nards de Vera, as I recall, exposed the supposed scenario in a media conference and said something like it negated all concepts of justice, etc.

In yesterday’s and today’s edition of the Inquirer, the banner story reported that the survey was opposition-sponsored. A coupleof years ago it was being “touted” reportedly by a group in Malacañang and today it’s being spread by the opposition. That might seem strange unless the reader sees the slant and is aware of what goes on in the release of information and the media coverage of surveys, and discerns how the story was “angled”.

There was an “excising” of context.

What was not fully reported in the news stories as background are the following:

1.As is generally known, there is a confidentiality agreement between the pollster and the sponsor such that the pollster is not allowed to disclose the results of any commissioned survey without permission from the sponsor. Pollster-organizations subsidize themselves and the staff generally through grants and commissioned studies.

2.If there were several surveys conducted, it is the sponsor who decides which of those surveys he/ she commissioned would be released or “leaked” to the public.

3.The sponsor (the person who paid for the study) decides which part of one survey, or what item of the survey, gets released to the public. For example, if there were ten items or ten questions, the sponsor decides or chooses what item or what questions and its results, are released or “leaked” to the public. For example, of the ten items, the sponsor can decide or choose that only Item Number 6, of the ten or so items, gets released publicly.

drumbeaters.jpg

Photo, detail of a bas-relief by Napoleon Abueva

4. So, hypothetically, a sponsor can decide or choose not to include the following items: 1)Do you think that the 4 billion pesos stashed in the Velarde account belong to Estrada? 2) Do you think that those funds were payoffs of businessmen to him? 3) Do you think that Estrada received jueteng protection money during his term? Etc.

If one looks at the slant of the article, one might realize that the sponsors care less about courting the ire of the Sandiganbayan for contempt than conditioning the mind of the public (albeit, maybe incompetently). The coverage of results of the question on the presidential pardon, slanted with quotes that people are sympathetic to Estrada is no less contemptuous of the administration of justice; they influence the administration of justice no less than the answers to the above-stated questions.

But what would be the result to the above-stated questions?

Isn’t the premise of a presidential pardon, as Estrada himself stated in rejecting it, and as Mayor Jejomar Binay himself stated in an interview: a conviction, or a finding of guilt beyond all reasonable doubt? That’s the only time a pardon can be exercised.

Since there was a question on possible presidential pardon and those were the results, don’t you wonder what the previous questions were? Don’t you wonder what the premises of the respondents were? As Mayor Jejomar Binay himself stated or asked, weren’t the respondents presuming guilt (or did they understand it). Or, to put it another way, did the respondents answer or wouldn’t the respondents have answered, in overwhelming majority, the question: is there guilt beyond doubt, with a Yes?

This is either a botched-up job on the part of Erap propagandists, because it implied guilt, or a wonderful job, because it implied sympathy.

Who the sponsor is (and maybe how competent a propagandist he/ she is) , matters; because the sponsor gets to control what information is released or “leaked” to the public. When reporters and editors are either (a) lazy; (b) negligent; (c) harassed (d) incompetent; (e) compromised or received something; the context (as discussed above) is not provided; no other side or analysis is presented. The sponsor being the only source of information therefore controls how the news story is developed, written, produced.

The sponsor becomes the reporter and the editor.

We do not know better because we don’t know the context.