EDA for HLM: Visualization when Probabilistic Inference Fails

e-mail: jwbowers{at}umich.edu (corresponding author) e-mail: kwdrake{at}umich.edu Nearly all hierarchical linear models presented to politicalscience audiences are estimated using maximum likelihood undera repeated sampling interpretation of the results of hypothesistests. Maximum likelihood estimators have excellent asymptoticproperties but less than ideal small sample properties. Multilevelmodels common in political science have relatively large samplesof units like individuals nested within relatively small samplesof units like countries. Often these level-2 samples will beso small as to make inference about level-2 effects uninterpretablein the likelihood framework from which they were estimated.When analysts do not have enough data to make a compelling argumentfor repeated sampling based probabilistic inference, we showhow visualization can be a useful way of allowing scientificprogress to continue despite lack of fit between research designand asymptotic properties of maximum likelihood estimators.     

Risky Facilities: Crime Radiators or Crime Absorbers? A Comparison of Internal and External Levels of Theft

Objectives

To undertake the first exploration of the nature of the relationship between internal crime (those that happen within facilities) and external crime (those occurring outside but in the nearby locale of facilities). The following questions are addressed. Do those localities that suffer high volumes of crime internally within their facilities also suffer high levels of crime in their immediate external environment? How is this influenced by the distribution of internal theft across facilities? What are the likely mechanisms for any relationship found?

Methods

Spatial regression is used to explore these relationships using data for 30,144 incidents of theft from a Metropolitan area of the UK arranged into small 50 × 50 m grid squares. Variables used in the analysis include counts of external and internal theft, counts of victimized and ‘risky’ facilities, indicators of land-use and a proxy for the on-street population.

Results

There is found to be a strong positive relationship between internal and external theft that appears to be strengthened by the existence of facilities suffering particularly high crime volumes. Results suggest that internal theft problems precede external ones and that the physical concentration of chronically risky facilities is a particularly strong predictor of external theft problems.

Conclusions

An argument is made that risky facilities act as crime ‘radiators’, causing crime in the immediate environment as well as internally. This has implications for crime prevention policy in terms of facility placement and management.     

Challenger Quality and the Incumbency Advantage

Most estimates of the incumbency advantage and the electoral benefits of previous officeholding experience do not account for strategic entry by high‐quality challengers. We address this issue by using term limits as an instrument for challenger quality. Studying US state legislatures, we find strong evidence of strategic behavior by experienced challengers. However, we also find that such behavior does not appear to significantly bias the estimated effect of challenger experience or the estimated incumbency advantage. More tentatively, using our estimates, we find that 30–40% of the incumbency advantage in state legislative races is the result of “scaring off” experienced challengers.