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 effects of political design and organizational dynamics on structural disaggregation and integration in Norway 1947–2019

In countries with long-standing agency traditions, the creation of new agencies rarely comes as a large-scale reform but rather as one structural choice of many possible, most notably a ministerial division. In order to make sense of these choices, the article discusses the role of political design—focusing on the role of political motivations, such as ideological turnover, replacement risks and ideological stands toward administrative efficiency—and organizational dynamics—focusing on the role of administrative legacies and existing organizational palettes. The article utilizes data on organizational creations in the Norwegian central state between 1947 and 2019, in order to explore how political design and organizational dynamics help us understand the creation of agencies relative to ministry divisions over time. We find that political motives matter a great deal for the structural choices made by consecutive Norwegian governments, but that structural path dependencies may also be at play.