Working with Misspecified Regression Models

Objectives

Conventional statistical modeling in criminology assumes proper model specification. Very strong and unrebutted criticisms have existed for decades. Some respond that although the criticisms are correct, there is for observational data no alternative. In this paper, we provide an alternative.

Methods

We draw on work in econometrics and statistics from several decades ago, updated with the most recent thinking to provide a way to properly work with misspecified models.

Results

We show how asymptotically, unbiased regression estimates can be obtained along with valid standard errors. Conventional statistical inference can follow.

Conclusions

If one is prepared to work with explicit approximations of a “true” model, defensible analyses can be obtained. The alternative is working with models about which all of the usual criticisms hold.

     

The Approved Narrative of the Srebrenica Massacre

The Srebrenica massacre has taken on both symbolic and mythical properties, all in the service of political agendas. The symbol is of Serb evil, manifested in an alleged cold-blooded and unprovoked massacre of innocents reminiscent of Nazi behavior during World War II. The myth is that 8000 Bosnian Muslim “men and boys” were executed in the vicinity of Srebrenica in July 1995, and that any executions there were unprovoked and carried out in accord with a genocidal plan. Both facts and language have had to be aggressively massaged to support these claims. The political agendas served by the massacre claims were those of the Bosnian Muslims, striving to get NATO to enter the Bosnian struggle more actively, the Croats, who needed a demonization of Serbs to carry out their own ethnic cleansing plans (most notably, removing 250,000 Serbs from Croatian Krajina in the very next month, August 1995), and the Clinton administration, under attack for a failure to intervene more actively on behalf of the Croats and Muslims and searching for an excuse to do so.This paper is partly drawn from and cites chapters in a forthcoming book on the Srebrenica massacre, Srebrenica: The Politics of War Crimes, written by George Bogdanich, Tim Fenton, Philip Hammond, Edward S. Herman, Michael Mandel, Jonathan Rooper, and George Szamuely. This book is referred to in the notes below as Politics of War Crimes. The author and his colleagues are indebted to Diana Johnstone, David Peterson, Vera Vratusa-Zunjic, Milan Bulajic, Milivoje Ivanisevic, Konstantin Kilibarda, and George Pumphrey for advice. Johnstone’s Fools Crusade is a fine basic statement of an alternative perspective on the Balkan Wars; George Pumphrey’s “Srebrenica: Three Years Later, And Still Searching,” is a classic critique of the establishment Srebrenica massacre narrative and repeatedly hit the target with facts and analyses still not rebutted.     

What Small Spatial Scales Are Relevant as Electoral Contexts for Individual Voters? The Importance of the Household on Turnout at the 2001 General Election

For many years, scholars of voting behavior have been thwarted in their attempts to identify micro spatial variations in turnout by data limitations. This has meant that most analyses have been ecological, which has implications for valid inference. Here, for the first time, a hierarchical approach is used to show the relative importance of several micro spatial scales, including the household, on voter participation. The findings highlight the importance of the household context. While those who live together often turn out together, the relative level of clustering within households as opposed to between geographical areas is found to be more important for two-person households compared to other households. Even after taking account of whether individuals are likely to self-select others from similar social backgrounds or with similar political attitudes, there is strong evidence of large and significant household effects on voter participation.     

The Civil Servant as Legislator: Law Making in British Administration

How are government policy commitments converted into legislation and what happens in the conversion? The role of civil servants in preparing legislation is far more important than is generally assumed. By looking at the work of four recent bill  teams in Britain – teams of civil servants given the task of developing Acts of Parliament – their crucial roles in initiating policies, placing them on the political agenda (even helping secure their place in a party manifesto), developing them, making sure they pass through parliament and enacting them once they have reached the statute books are assessed. The article explores the composition and working methods of bill teams. These teams work with considerable autonomy in developing legislation, but it cannot be assumed that they operate outside ministerial control. Teams see themselves as reflecting the priorities of the government in general and their ministers in particular. Yet ministers typically know relatively little about the law they are bringing in until they receive the submissions and briefings from their officials. Perhaps the biggest danger for democracy is not a civil service putting forward proposals which a minister feels forced to accept, but rather that ministers do not notice or fully appreciate what is being proposed in their name despite having the political authority to change it and a civil service which bends over backwards to consult and accommodate them.     

Cooperation and competition in public welfare policies

Present public welfare policies and various proposals for reforming them do not often exploit the advantages of cooperation between otherwise competing altruists. The article examines this principle from a number of perspectives. It shows first how the advantages of the existing cooperative payments structure could be lost in some widely-publicized welfare reform proposals, and it makes a suggestion for retaining these advantages. The article next shows how the same principle could be used to improve existing incentives for private giving. Finally, it uses the same logic to rationalize newly evolving notions of sharing responsibility between governments and the recipients themselves.