I really don’t know why almost no one else complains

I wrote about these election polls years ago, but nothing has changed. (Once, a pollster argued my post but only on one point—how telling.)

Part of the problem seems to be the news agencies that don’t record the margin of error anymore and who commissioned the poll or when.

The biggest problem is the sample size. Is this greed of the contractor or laziness of the pollsters? Below 600, the margin of error is plus or minus 4%, while the election threshold of 4 seats is lower than that. Every time, numerous parties in their prediction fall just above this minimum. We see Labor and Meretz per poll getting respectively 0 and 4 or 4 and 0 seats. So, results for the other parties getting 5, 6, or 7 seats are meaningless too.

All kinds of hypotheticals are asked all the time. After me harping about it for months, finally, a poll asked who people would vote for if Meretz and Labor would run jointly, predicting 8 seats for them. I already guessed it would be 7-9 seats, which is far more than all the previous predictions of 4, with every time, one or the other falling below the cut-off norm, which underestimates the number of leftwing voters. Netanyahu was only able to form his last government because Meretz and Labor ran on separate lists. It is unthinkable they would repeat their disastrous mistake.

Still, the polls are taken seriously by journalists (who generally understand math below the level of an average 6-year-old) and politicians alike. Political leaders get fired and governments go to elections, only based on these frivolous statistics. Netanyahu called for new elections four times within two years to get more seats in the Knesset based on polling, and thrice, he got fewer. And not caused by last-minute scandals or so.

Every time polls turn out significantly different from the election results, but the pollsters just blame the voters. Imagine a life insurance company blaming physicians, patients, or the Angel of Death for wrong statistics.

There seems also no standard, norm, or controlling body to examine any scientific rigor behind these polls. Anyone can just ‘poll’ 500 people and publish results. I was called thrice to give my opinion, but twice, I was cut off because they already had enough people of my age or religious group to feign representativeness of the sample. How fake do you want it?

There a small lies, big lies, and election polls. Only fools believe them.