Our research looks to assess if and to what extent New Jerseyans support four areas of sentencing reform in the state. To examine this research question, we employ a particular type of methodology, an experimental online survey, to assess the attitudes of a sample of the public from New Jersey. This means that the participants in our survey are people who live in the state of New Jersey, and they take this survey on the internet for our research. Here, we briefly explain the details of this methodology and its utility in understanding attitudes towards practices and policies in the criminal-legal system.
In our study, we present an online survey, built on the survey platform Qualtrics, to our participants. However, what makes our study an experiment is that we randomly assign participants to different conditions of the survey, but we ask them to answer the same set of questions that ask participants to rate their feelings and views on a Likert scale (e.g., from not at all (1) to completely (7)).
For example, to study one of the areas of sentencing reform, participants are presented with several descriptions of a potential rehabilitative release program that would allow incarcerated people who reach a certain age and length of incarceration to apply to a judge for resentencing by proving that they have been sufficiently rehabilitated. They are then asked to rate how much they support that program. These descriptions come from real sentencing recommendations by the New Jersey Criminal Sentencing and Disposition Commission. But in our survey, the descriptions of the program vary the defendant’s age at which he is eligible for the program (e.g., age 60 or 62), the type of crime he was incarcerated for (e.g., murder or another serious felony), and how much time he has served so far (e.g., 20 or 30 years in prison). The goal of varying these factors is to be able to isolate their effects and test whether the inclusion or omission of these factors significantly affects how participants rate their support for the program they read about. Indeed, if other parts of the program description are the same and held constant, and only specific factors are varied, we can be more confident that they are the reasons that people are more or less likely to support that program.
Further, our survey is administered to participants on the computer over the internet by a survey firm (Qualtrics Panel), as opposed to being handed out physically in person. There are several benefits to this method, as well as a few shortcomings. The first to consider is that we want to collect responses from 1,500 participants across New Jersey. Doing this online, as opposed to in-person, saves us weeks to months of time. Online survey responses can be collected quickly, such as within hours or days; therefore, the large sample procurement is a huge benefit.
Logistically speaking, online surveys are also very cost-effective. There is no need to develop physical materials, travel, and physically store data. When it comes to data validity, they may need to be cleaned, but they do not need to be transferred from paper to computer, preventing many mistakes and general human error. Further, due to the length of the survey, we are also able to ask additional questions of participants that can be included in our analyses to account and control for many other variables, such as sociopolitical views, race, moral positioning on punishment, and other background characteristics; this allows us to draw stronger conclusions regarding the strength and patterns of such findings and allows us to be more confident in possible causal claims.
However, a key limitation of this methodology is that, of course, there is also a large population of people who do not have access to the internet or, particularly, older individuals who may not actively use technology. Therefore, even if not captured by our study, other inquiries must help to fill those gaps through different methods that capture the attitudes and experiences of those out of reach.