back to Home
Treat surveys with caution
01-Apr-2014, The Sun Daily Online
MERDEKA Center's recent survey is insightful and shouldn't be dismissed by those who disagree with its conclusions. Nevertheless, as American politicians have discovered, all opinion polls should be treated with circumspection – as just an indicator of voters' current views.
Furthermore, because questions asked by pollsters often require either a "yes" or "no" answer, this could gloss over responses that are more nuanced. Including the "yes, but …" into either category could distort the overall finding.
According to Merdeka Center, its latest survey showed 56% of respondents were dissatisfied with the government's handling of the economy while the satisfied segment comprised only 33%.
Additionally, four issues that require the federal government's urgent attention are corruption, inflation, safety and security issues as well as affordable housing.
Confined to Peninsular Malaysia, the 1,005 respondents polled by Merdeka Center comprised 60% Malays, 31% Chinese and 9% Indians – a demographic similar to that of the country.
Among those surveyed, the three largest age segments were those aged 31 to 40 years who accounted for 30% of respondents, those between 41 and 50 years (24%) and those between 21 and 30 years (20%). This means individuals below 40 made up half of the respondents.
That 65% of respondents have internet access also suggests this demographic group lives in urban areas.
A heavy weightage of urbanites and the young may explain one curious disconnect. Those who are less well-off cited corruption as a major concern. This is puzzling as they are less likely to have the means to pay bribes.
Equally mystifying, although the cost of living is one of four major concerns of respondents, those who earned a monthly income of less than RM1,500 were happier with the government's economic performance compared with their higher-paid counterparts.
"Perception towards the government improved the most among rural voters and those living in households earning under RM1,500 per month," Merdeka Center suggested. "In our opinion, the improvements could be attributed to the BRIM cash transfer payments to 5.4 million recipients nationwide which began at the end of February and through March 2014."
While the BRIM payments may explain the government's higher favourable rating among rural folk, it would be unwise for political leaders to assume cash disbursements can be relied on continually to secure political support. More worrying is another statistic.
Although 49% of those working for the government and for government-linked corporations (GLCs) were satisfied with the government's handling of the economy, this is marginally higher than the 46% that were dissatisfied.
Additionally, the gap between the two groups is just 3% – close to the survey's estimated margin of error of 3.09%.
In contrast, detailed analysis of respondents' perception towards Barisan Nasional show only 40% of civil servants and GLC employees are happy with the federal government – much lower than the 45% who are dissatisfied or angry.
Regardless of the mixed message these two sets of data present, this statistic is important. If BN cannot rely on civil servants and GLC employees to provide the base of electoral support, is there another growing bloc of voters it can count on to secure victory in the next general election?
Ironically, if the economy continues to expand, BN's electoral support base, as the Merdeka Center's poll suggests, – those less well-off, living in rural areas and who don't have internet access – will continue to shrink.
Because the survey was carried out from March 7 to 20, Merdeka Center believes issues such as Opposition Leader Datuk Seri Anwar Ibrahim's conviction for sodomy and declining racial relations could have influenced respondents.
Like all pollsters, a major challenge for the Merdeka Center is this: can it identify turning points in voter perception?
One event that underscores the limitations of opinion polls is Barack Obama's re-election as US president in November 2012.
In the months leading up to the November 2012 election, newspapers harped on the sluggish American economy, in particular continuing high unemployment, and Obama's failure to effect change – the platform that won him an unprecedented victory as the first African-American president four years earlier.
On the eve of polling in November 2012, among likely voters, Gallup found Obama's approval rating in handling the US economy was just 42%. In response to a question whether Obama deserve to be re-elected, Gallup found 51% of respondents said he did not.
Based on its findings, Gallup predicted a close fight while other pollsters indicated the Republican challenger Mitt Romney would win.
Strenuous efforts to lift the turnout rate among key Democratic constituencies – women and minorities – plus a targeted campaign to win critical electoral states like Ohio – coupled with Romney's complacency earned Obama a decisive victory in the electoral college and popular vote.
As the name suggests, opinion polls are opinions, not cast-iron predictions.
Opinions expressed in this article are the personal views of the writer and should not be attributed to any other organisation she is connected with. She can be contacted at firstname.lastname@example.org