Bernard: Well the party had an opinion poll done and it seemed all the voters are in favour of bringing back National Service.
Sir Humphrey:Well have another opinion poll done showing the voters are against bringing back National Service.
We all use polling but recently two issues stood out where we ought to be more careful in their use. The first is a false interpretation of what the data shows, the second is the flaws of using polling data on a complex issue to support your argument.
The first is simpe enough, take David Sirota’s argument at openleft, seems pretty straightforward, the US public opinion on trade between the two value statements of “opportunity for growth” and “threat to the economy”. As Mark pointed out; public opinion favours the latter statement when the economic growth was slowing down or in recession and it favoured the the former statement when the economy is growing.
David Sirota however, somehow drew this conclusion out of his magic hat:
Last week, Gallup reaffirmed what polls have been telling us for a few years now: Americans are sick and tired of rigged trade policies that they know are selling them out:
That isn’t what the polls show, they don’t discuss trade reform or the type of trade, they simply show that American favourability to trade is correlated with the growth of the economy,little else could be concluded without further questions.
However, in a stroke of irony, Sirota has the gall to bemoan Gallup’s own analysis which used the terms “anti-trade” and “pro-trade”
As always, those who want to reform trade are billed as “anti-trade” – not, for instance, “pro-trade-reform.”
In fact on that basis of that graph neither can be said to be entirely correct, because trade isn’t a simple for or against, much like Government isn’t for or against, people may want trade reform, they may want more trade with Japan for their cars but less with Mexico, Detroit workers might want protectionism but still want Chinese imports, but we don’t know that from this poll.
Even if the data was attached to a more specific value statement we have to be careful, take these two article, one is from Pejman Yousefzadeh at Redstate.com and the other is from Nate Silver at fivethirtyeight.
Pejman, is rather pleased a Rasmussen report, shows that before and after the stimulus, 59% of Americans support Ronald Reagan’s statement that: “government is not the solution to our problem; government is the problem.”
He concludes:
I have to think that all of this could pay dividends for Republicans if the GOP can win the public relations war over the size of the Obama stimulus package, the budget, the deficits the Administration is running up, the taxes they want to raise, and the way in which the Administration is working to bring back the era of Big Government. Ecstasies over a New New Deal notwithstanding, the public is not in the mood for bigger government and can be persuaded to give a big thumbs-down to the Administration’s efforts to expand government if only White House spin is not allowed to trump the facts of the Administration’s domestic program.
It may be Barack Obama’s White House. But it is still Ronald Reagan’s America. If the latter sees clearly that the Administration’s policy priorities are antithetical to its own, we will look back on talk of the GOP’s decline and laugh.
Pejman’s conclusions aren’t unreasonable but they could easily be wrong, as a Gallup poll also shows when it asked this more specific statement:
As you may know, Congress is considering a new economic stimulus packagae of at least $800 billion. do you favour or oppose congress passing this legislation?
Ah, and there we have it, contradiction and by 59% no less!
I’m not saying polls aren’t useful and we use polls here ourselves, but perhaps in our more reflective moments we ought to be more cautious on some issues, “Government” and “Trade” are complex terms that mean different things to different people and so polling without further exploration of what the respondent meant by their answer, can be misleading.
I’ll leave you with Sir Humphrey, a master of the dark arts of Government:

