Sunday, 25 May 2014

Vote-match sites (EU elections): is there a bias?

For today’s EU elections a couple of websites aim to help voters to decide which party they prefer. The usual structure is that thirty or so statements are presented (say, something like: ‘European integration should continue’) and you have to click whether you agree, are neutral, disagree, or want to skip the question.
   Of course, the way these statements are presented may steer users towards particular answers. For example, there may be a preference for users to choose ‘neutral’ (as things are often more complex than the simple statements may indicate). Now, if one always chooses ‘neutral’, which of the main parties is seen as the most appropriate one?
  • Euandi has a slight centre-left preference. For England the order is Labour (73%), Libdems (67%), Conservatives (63%), UKIP (58%), Greens (57%); for Germany: FDP (70%), SPD (67%), Left Party (63%), AfD (63%), CDU (62%), Greens (61%). Similar for France and Italy.
  • EUvox is ambiguous. For England the order is Labour (26.7%), Libdems (25.8%), Conservatives (22.5), UKIP (14.8%), Greens (12.5%), but for Germany it is CDU (23.3%), Greens (20%), SPD (19.2%), AfD (18%), Left Party (14.7%), FDP (14.3%).  
  • Electio prefers right-wing Eurosceptics. For the UK the order is UKIP (30%), Conservatives (20%), Greens and Labour (5%), Libdems (0%). Here, it is also possible to get a rank of the EU-wide party groups: European Conservatives and Reformists Group (20%), Nordic Green Left (10%), Europe of Freedom and Democracy Group, European People's Party, Greens  (5%), and then the others (0%).
  • Votematch tells you that when you click ‘neutral’ for all questions, it cannot give you a reliable results (which may be preferable indeed).
Overall, this shows that the design of the questions may well matter (in particular considering Electio). Of course, there are also more subtle  ways of ‘nudging’ users to particular answers – which I leave for others to explore.

Saturday, 17 May 2014

The Method and Culture of Comparative Law – and how this relates to business cards in Japan and David Hasselhoff in Germany!

I’m in lovely Gent for the book launch of The Method and Culture of Comparative Law - Essays in Honour of Mark Van Hoecke (Adams & Heirbaut eds): see here for the conference and here for the book.
  My own contribution is on "Overfitting Legal Transplants". The working paper version is available here and the abstract reads:
Sometimes cultural phenomena are more popular abroad than at home. For example, modern business cards arrived in Japan in the late 19th century but in today’s Japan they are considered more important than in Western countries. Similarly, and even more surprisingly, in Germany the US actor David Hasselhoff is considered as one of the greatest pop stars while his musical career has not been very successful in the US.  This chapter suggests that a similar phenomenon may also exist in law. This is meant to refer to the situation that legal transplants may work even ‘better’ in the transplant than in the origin country, here called ‘overfitting legal transplants’. This suggestion departs from the current debate about legal transplants which holds that, at best, a transplant may work almost as good as in the origin country (and, at worst, it may be irrelevant, or even harmful to the transplant country). Thus, it is the main aim of this paper to relate the idea of overfitting legal transplants to the current discussion, to outline possible categories and examples, and to show how law-makers may be able to make use of them.

Saturday, 10 May 2014

Taxonomies and Leximetrics

Forthcoming chapter in The Oxford Handbook of Corporate Law and Governance (Jeffrey Gordon and Wolf-Georg Ringe, eds.). Since the eventual publication may still take a bit, I have made the working paper version available here. Abstract:
Taxonomies are frequently used in the literature of corporate law and governance. This chapter outlines four types of typologies: limited typologies that may either focus on legal or non-legal questions, and general typologies that may either have a legal or a non-legal focus. Subsequently, it discusses the challenges to these taxonomies, in particular the severe criticism directed against the quantitative research on “legal origins”. This is not meant to reject the possibility of quantifying legal differences. Indeed, the subsequent section presents how such “leximetric” tools can contribute to the question of whether or not there are distinct legal families in corporate law. This is illustrated through a cluster analysis of investor protection data of the World Bank’s Doing Business Report. The findings confirm the skepticism of this chapter about broad taxonomies. 

Friday, 2 May 2014

The main source of faculty infighting: playing different ‘games’?

                 Source: my translation of cartoon by Otto Waalkes.
In any organisation (firm, government body etc) there may be disagreements; yet, I feel that in universities things may be special – and possibly more pronounced! To understand those conflicts, the following analogy may be useful: academics perceive universities and their jobs in different ways, namely as similar to one of the following:
  1. Athletics: there are clear performance indicators, say, publications in top journals that determine promotion etc.
  2. Figure skating: as previous one but the indicators are more subjective.
  3. Poker: you have to ‘beat’ your colleagues to make progress: say, only some academics get promoted each year (or, eg, who has to teach a particularly burdensome undergraduate course)
  4. Football: the main aim is to ‘beat’ other universities (or other departments within your university) with each ‘player’ having his or her strengths and weaknesses.
  5. Theatre assemble: you want to perform well (as the previous category) but the main aim is not to ‘beat’ another assemble but just to be as good as possible to the benefit of the ‘audience’ (students etc).
  6. Military unit: the main aim is to pursue a common aim (similar to previous category) but this is done within a strict hierarchical model.
  7. Extended family: in contrast to all of the previous categories, there is no clear aim but university life is just an extension of ‘normal life’, ie one should be friendly, helpful, considerate etc.
  8. Shared house: similar to previous category but the connection between the participants is more loose, ie in principle everyone is doing his or her ‘own thing’ (whatever that may be in terms of teaching and research).
Now, the problem is that, say, if academic A perceives her job as similar to athletics, B perceives it as similar to a theatre assemble, and C as similar to a military unit, conflicts are bound to arise. More elaborately, one could do a cross-table with the eight categories and identify why and how persons ‘clash’.
    The question is whether this can be avoided. A tempting response may be that it just needs to be clarified in advance which ‘game’ is played in a particular university or department. But would this really be feasible – or is it not in the very nature of university that these eight (or more) models overlap? I’ll leave this open.