DOES PUBLIC ACCOUNTABILITY WORK? AN ASSESSMENT TOOL

In recent years, there has been a drive to strengthen existing public accountability arrangements and to design new ones. This prompts the question whether accountability arrangements actually work. In the existing literature, both accountability ‘deficits’ and ‘overloads’ are alleged to exist. However, owing to the lack of a cogent yardstick, the debate tends to be impressionistic and event‐driven. In this article we develop an instrument for systematically assessing public accountability arrangements, drawing on three different normative perspectives. In the democratic perspective, accountability arrangements should effectively link government actions to the ‘democratic chain of delegation’. In the constitutional perspective, it is essential that accountability arrangements prevent or uncover abuses of public authority. In the learning perspective, accountability is a tool to make governments effective in delivering on their promises. We demonstrate the use of our multicriteria assessment tool in an analysis of a new accountability arrangement: the boards of oversight of agencies.     

A COMPARISON OF DEFENSE STRATEGY AND RACE AS INFLUENCES IN DIFFERENTIAL SENTENCING

This study compares the influence of defense strategy and the race of the defendant on the variance in felony sentence length for a sample of 2,090 convicts in a state penal system Defense strategy includes the subsequent plea of the defendant and the type of defense counsel. Control variables are age, marital status, urban-rural childhood, education, prior criminal record, and type of crime. A comparison is made for both burglary and homicide. The findings for butglay show defense strategy explains 16.5% of the variance in sentence length (because of the type of plea), while race of the defendant explains only I% of the variance (due to the positive relationship between being black and sentence length). For homicide the unique explanation of the defense strategy model accounts for 3% of the variance (due more to type of attorney than plea), while the race of the defendant accounts for another 3% (because of the negative relationships between being Indian or Mexican-American and sentence length).