News coverage of elections is all about position and momentum – who’s ahead? Who’s gaining? Whose support is dropping?
Opinion polls are the single best way of determining how people are going to vote in an elections and the only meaningful way of answering the questions above.
There are more polls than ever these days, partly because the advent of online polling makes it cheaper and quicker to run them. We have loads of data on what voters plan to do.
Unfortunately, all over the world, reporting of opinion polls is poor. This might be because traditional political journalists have an instinctive, personal, human approach to politics rather than an affinity for spreadsheets and graphs. It might also be because the polls themselves are pretty dry and reporting them honestly doesn’t make good headlines.
One such headline was in Haaretz on Tuesday.
Haaretz poll: Likud-Beiteinu weakening as Bennett’s rightist party rises
Voters are punishing Prime Minister Netanyahu for attacking rival who said that if he were in the army, he’d refuse an order to evacuate settlers.
Some context. Last Thursday the 20th of December, HaBayit HaYehudi (Jewish Home) leader Naftali Bennett was reported as supporting the idea of IDF soldiers refusing orders to evacuate settlements in the West Bank.
He was attacked by politicians from other parties, especially the Prime Minister himself. Bennett later distanced himself from his comments, claiming he had never sanctioned soldiers refusing orders.
Is Haaretz right to claim that the row has actually helped Bennett and his party?
The graph below shows HaBayit HaYehudi’s performance in voting-intention polls since early October. For methodological reasons, we can’t show movement between polls of different companies, so this graph shows a line for each polling company.
The graph is pretty clear – since October, HaBayit HaYehudi’s predicted seat haul has risen substantially, from 6-10 seats to 10-16 seats in late December.
Two events may be triggers in this rise. Most obviously, the “Biberman” pact between Likud and Yisrael Beiteinu boosted HaBayit HaYehudi, probably due to national-religious switchers from Likud. Since Naftali Bennett won the party’s leadership, it has steadily improved its polling position.
Short-term, though, the picture is less clear, because we have less data.
There have only been four polls since Bennett’s remarks last Thursday – by Panels, Geocarteography, Teleseker and Dialog. Comparing these with the previous poll from each company shows
- One major increase: Teleseker (+4 seats)
- One minor increase: Dialogue (+1 seat)
- One no-change: Panels (-)
- One major decrease: Geocarteograhy (-3.5 seats)
This is a mixed picture, especially as the Dialog change is so small that it’s as likely to be sampling error as anything else.
Back to Haaretz’s claim that Bennett had gained from the refusal row. That story was based solely on the Dialog poll, which shows nothing at all. Haaretz also artificially boosted the size of the change by comparing Tuesday’s poll numbers with an older poll – something they’ve done before.
It’s too early to say what the impact of this row will be. Habayit Hayehudi could well end up profiting from it in the long-term, because Likud/Habayit Hayehudi waverers are likely to be a constituency that would have sympathy with Bennett’s position on refusing to evacuate settlements. But, contrary to the Haaretz report, the data doesn’t show that – yet.