Jacob Christensen

Another note to self:

The original paper – Geber, Alan, Malhotra, Neil: Can political science literatures be believed? A study of publication bias in the APSR and the AJPS

Some comments from the blogosphere:

• Crooked Timber – Kieran Hiely: Attractive Models
• Statistical Modelling – Something fishy in political science p-values, or, it’s tacky to set up quantitative research in terms of “hypotheses” (A typical case of academics being funny)
• Jeremy Freese – …and you thought astrosociology was dead

In case anybody gets the headline wrong: We are talking about the interplay of methodology and publication habits here, not political bias among political scientists.

I often tell my students that they should remember that an essay or a thesis with a negative result is as valid academically as one with a positive result or one presenting “new” knowledge.

What’s important to me is that the methodology is sound. But I’m not an editor, anyway.

If you look at regional variations in wins and losses in the Swedish election, a fascinating picture emerges:

• Left Party (biggest losses) – Stockholm City (-3,4), Västernorrland County (-3,2), Gävleborg County (-3,2)
• Green Party (biggest wins) – Stockholm City (+2,6), Gothenburg City (+2), Malmö City (+1,8); on the other hand the party also lost votes in a number of rural districts such as Norrbotten (-1,2), Dalarna (-0,4) and Gävleborg (-0,3)
• Social Democrats (biggest losses) – Stockholm City (-8), Skåne South (-7,5), Stockholm County (-7,1); the Social Democrats won 3,9 points in Norrbotten County due to the decline of the regional Norrbotten Party
• Centre Party (biggest wins) – Stockholm City (+4), Gotland (+4), Uppsala County (+3,1)
• Liberals (biggest losses) – Stockholm City (-9,4), Stockholm County (-9,1), Gothenburg City (-7,8)
• Conservatives (biggest wins) – Stockholm County (+15,9), Stockholm City (+13,9), Skåne South (+12,9)
• Christian Democrats (biggest losses) – Western Götaland South (-4,3), Skåne North and East (-4,3), Jönköping (-3,7)

My colleague Flemming Juul Christiansen did the maths on the two blocs and came to this result:

• Left bloc (losses) – Skåne South (-8,8), Stockholm City (-8,8), Stockholm County (-8,7)
• Centre-right bloc (wins) – Stockholm County (+7,4), Malmö City (+7,1), Stockholm City (+7,1); Kalmar, Värmland and Blekinge saw only small overall gains for the “Alliance”
• Combined changes – biggest in Stockholm County (16,1), Stockholm City (15,9), Skåne South (15,2), smallest in Norrbotten (3,6), Värmland (5,0), Kalmar (5,5)

Feel free to say “well, duh!” but the real battlegrounds of this election were the Stockholm region and Skåne. It would be interesting to correlate this to regional economic data but for now, my argument will be that the Social Democrats had their biggest problems in those regions with the fastest growth in service industries and population.

These calculations have been made from the Electoral Agency’s (Valmyndigheten) data for the final result of the 2002 elections and the preliminary results for the 2006 election.

SvT’s exit poll contains an interesting analysis of the issues Swedish voters found the most important for their choice of party.

If we focus on the top 3 issues for each party’s voters, the list is as follows:

• Left Party: a) gender equality (56%), b) employment (52%), c) health care (51%)
• Green Party: a) enviroment (81%), b) energy and nuclear power (51%), c) gender equality (49%)
• Social Democrats: a) health care (59%), b) education (54%), c) employment (54%)
• Centre Party: a) employment (59%), b) education (43%), c) conditions for businesses (44%)
• Liberals: a) education (61%), b) employment (59%), c) economy in general (52%)
• Conservatives: a) economy in general (64%), b) employment (63%), c) conditions for businesses (59%)
• Christian Democrats: a) education (59%), b) care for the elderly (59%), c) employment (59%)

For all voters the top 3 issues were: a) employment (56%), b) education (54%), c) health care (51%). It is worth noting that the Left Party’s and the Green Party’s differed from the mainstream in their emphasis on issues like gender equality (36% of all voters) and the environment (30% of all voters).

The extreme right Sweden Democrats attracted voters who emphasised immigration (80% of SD voters mentioned this as an important issue, followed by 70% emphasising “law and order” and 60% “own economy”); among all voters, however, only 25% listed immigration as an important issue in their choice of party.

Note to the editorial writer at Politiken: As I recall it, it was the Social Democrats and especially Poul Nyrup Rasmussen who – to my endless irritation – insisted on referring to the change of government in 1993 as “the regime change”.

The media and later the Danish People’s Party just followed the Social Democrats’ lead in abusing the Danish language.

But I’ll grant you points for the editorial’s very wicked headline. If the “New” Moderates – another label that I truly and utterly hate – turn out to be the “New Social Democrats” is something we will see during the next four years.

By the way: The Swedish Conservatives – consciously or inadvertently – used nearly the same slogan (“Sverige behöver ett nytt arbetarparti”) during this campaign as the Danish Conservative Party did in the 1975 election campaign (“Borgerlige stemmer der arbejder”). The difference was that the Danish party polled their worst election result ever in 1975, the Swedes their best.

I think that we will have to admit that Lars Ohly has a wicked sense of humour. But how does giving a female opponent the Finger square with the Left Party’s Feminist agenda?

Aftonbladet: Lars Ohly Gives “The Finger” during Breakfast TV Show.

And why, oh why, do I watch the wrong TV channels?

(This is not Lars Ohly. But it’s a pretty cool picture).

I really don’t think that I have anything profound to add to the debate about why the Social Democrats lost the Swedish elections right now so I’ll just add a link to Nick Aylott’s round-up of events and comments which is posted at the Political Science Association’s homepage.

It catches the general line of argument among pundits and academics quite well.

One intriguing question remains: Why did the Social Democrats choose to overlook employment policy in their campaign? Did they know that they would lose the argument on this issue and hoped that if they ignored employment, they would be able to shift the focus of the campaign to other issues?

It’s official: If you want to earn money at a Danish university, you should become a manager, not a researcher or a teacher.

And bigger units mean more managers, not more money to research and education.

Oh, and if you want a say in what you work with, Helge Sander thinks that you really should consider going somewhere else.

I just wanted to call some attention to an article in Monday’s Berlingske Tidende which pointed out that Denmark’s two most important export markets are Germany and Sweden – followed by the U.K. as a distant third.

And so what?, you may ask.

The thing is that when we – or rather politicians and bureaucrats – talk about internationalisation it is always in terms of learning and speaking English and adjusting to U.S. standards in general. The Social Liberal Party recently took this stance to its logical conclusion and suggested that English should be a – in practice the – official language of Denmark.

In the eyes of politicians and bureaucrats, Denmark is too small a country to sustain its own language.

What I wanted to rant against, however, was the narrow focus on English in Danish education policies. In practice, the latest reform of higher secondary education has managed to kill off third and fourth languages.

I would argue that in the long run the decline in linguistic skills will be a liability to Denmark. Say what you will about the Germans, but they are not Americans and neither are the Swedes (even though the Swedes like to pretend that they are the true Americans). As younger generations of Danes lose the ability to speak, read and write German and Swedish, Danish firms also lose some important tools to navigate in the German and Swedish societies.

The irony is of cause that Berlingske Tidende not too long ago renamed its business section from “Erhverv” to “Business” – just to prove that the paper had an international (i.e. American) outlook – and that this rant is written in some kind of English.

The headline is a hidden quote from one of the 18th Century Danish – or was that: Norwegian? – writer and dramatist Ludvig Holberg’s comedies where one of the protagonists despairs at the phony French used by academics and lawyers to impress ordinary people.

The quote in Danish is from Jeppe på Bjerget and goes:

“Tal Dansk, din sorte Hund! saa skal vi nok svare for os”

I have never been yelled at by Germans. Perhaps because I actually speak the language.

In The New York Times, no less:

“There’s a sense that the government has become tired, that it’s been drifting, really,” Nicholas Aylott, a political scientist at Sodertorn University, said in an interview. “It hasn’t been clear where the Social Democrats want to lead Sweden.”

[....]

“The election’s big issue was the poor functioning of the labor market,” Mr. Aylott said. “The fact that so many people don’t have work has to be considered a major failing of the Social Democratic government, given that they’ve presided over such a long period of economic growth.”

And here’s a Reuters report on the Washington Post’s homepage. No Aylott here, though.

Finally, let’s throw in a comment in Die Zeit for good measure and call it a night:

Insgesamt war die Wahl, darüber sind sich die schwedischen Wahlbeobachter einig, ein persönliches Votum gegen Göran Persson.

Sören Holmberg would beg to differ on that conclusion.

Given that the four centre-right parties have won the majority in the new Swedish riksdag, it wasn’t surprising that Göran Persson would resign as leader of the Social Democratic Party during the coming electoral term.

What did surprise me was that he in his speech (at 23.00) not only made a clear admission of defeat but also immediately announced that he would resign as party leader in March 2007. The question now is how long it will take the Social Democrats to find a new party leader and how messy the procedure will be.

Oh, and as Sören Holmberg just commented: 2006 yielded the best election result for the Conservatives since 1928 and the worst result for the Social Democrats since the 1920s. The Danish Social Democrats would be ecstatic at the prospect of winning 35% of the vote – but then again the Danish left has consistently polled between 10-15% of the vote since the 1960s.