Measuring Culture and Climate in Illinois Prisons

UPDATE: An interactive dashboard of the MQPL survey results described in this summary report is now available here.

The Center for Criminal Justice (CCJ) at Loyola University Chicago is working with the John Howard Association (JHA) of Illinois on a project designed to monitor and assess the quality of life in Illinois prisons, using data from the “Measuring the Quality of Prison Life” (MQPL) survey instrument.

The MQPL is a detailed survey that gathers prison residents’ views on the quality of life in the facilities where they are held. Originally developed by criminologists at Cambridge University in the early 2000s, the MQPL is now routinely used by the British government to monitor climate and social conditions in prisons throughout England and Wales, and has been translated and adapted for use in Germany, Switzerland, and elsewhere in Europe. It has been found to yield valid and reliable measures of multiple aspects of a prison’s social environment, including intangible but important qualities like humanity, respect, order, fairness, and staff professionalism.

Over the last few years, JHA has been working with prison monitoring groups in New York and Pennsylvania to bring the MQPL into American prisons. Under an agreement with the Illinois Department of Corrections (IDOC), JHA is now making an adapted version of the MQPL available in paper form to the roughly 28,000 people held in the IDOC’s 27 secure facilities. The surveys, which are voluntary and anonymous, are distributed in batches, prison by prison, and returned to JHA by mail. Since the effort began in 2022, more than 6,000 surveys from 22 prisons have been returned and analyzed.

Loyola CCJ staff and student research fellows, along with interns and volunteers, have been helping JHA staff code responses, analyze the data, and visualize findings in ways that will enable IDOC administrators and the general public to understand and act on them. It’s part of a new collaboration intended to benefit both organizations. A new undergraduate class on prison reform, taught jointly by JHA and CCJ staff, is another product of the partnership, as are JHA internships and fellowships for selected Loyola students that began last summer. The idea is to provide JHA with extra analytic capacity and research support, while giving Loyola students valuable hands-on experience and exposure to the real world of criminal justice research and reform.

What the MQPL Measures: Four Dimensions of Quality of Life in Prisons

The adapted version of the MQPL being used in Illinois contains 66 items. Some seek demographic information, some call for open-ended responses, and one asks for feedback on the survey itself. But most of the items are simple statements, positive or negative, regarding life in prison. Each statement concerns a specific aspect of prison life, often a very simple one, and respondents are asked to register a response for each, on a five-point scale from “strongly agree” to “strongly disagree.” Some of the treatment I receive is degrading…To get things done, you have to ask and ask and ask…This prison is controlled by incarcerated people rather than staff…You have to be in a group or gang in order to get by…Responses to the statements are given a numeric value, making it possible to “score” individual institutions across four fundamental dimensions:

• Safety. Safety is a central concern of everyone involved in the operation of prisons, and Illinois is not lacking in solid measures of prison safety: detailed tables in quarterly IDOC reports provide a regular accounting of the number of fights, assaults, uses of weapons, cell extractions, suicide attempts, requests for protective custody, and so on for each of Illinois’ prisons. What the MQPL can add is the perspectives of those who have the most at stake in a prison’s basic order and safety. The survey has some items that relate directly to what goes into creating a safe environment—the quality of supervision, the way alarms and incidents are handled, the level of staff control over living units—as well as others that get at feelings of safety. It also explores the kinds of compromises, trading, and allegiances that may be necessary to ensure personal safety.
• Wellbeing and Development. Another group of MQPL items is designed to reflect a prison’s responsiveness to the human need for growth. Here again, IDOC tracks and reports participation in the kinds of educational and vocational programming that can help prison residents develop their potential and make progress towards goals. It reports the number of people who received life skills training, the number who got various kinds of treatment, and so on. Those are important indicators, as far as they go. But the MQPL can tell us something about what the imprisonment experience actually feels like, to those going through it. Is it a chance to make positive changes, as one item puts it? Or just doing time rather than using my time, in the words of another?
• Professionalism. This dimension could also be called any number of other things—legitimacy, transparency, fairness, consistency—all of which relate to the way a prison’s staff and administration use their necessary authority over people in their custody. Are the rules clear? Is discipline predictable and proportionate? Are decisions accompanied by reasons? Prisons must have order, and rules, and the rules and their enforcement should be fair. But it’s also important that they be perceived as fair.
• Harmony. This is the hardest dimension of prison life to put a name to: the MQPL’s designers at Cambridge called it “harmony,” but humanity or decency might do just as well. Whatever it’s called, it implies a culture of basic respect and compassion towards people in custody. Staff-resident relationships that are courteous and supportive. It’s hard to define, but people tend to know it when they see it—and especially when they don’t.

MQPL Results in Illinois

For those who care about the welfare of people in prisons, results from the first-ever administration of the MQPL in Illinois are not especially encouraging. In fact, overall MQPL ratings of life in Illinois prisons are pretty dismal. Most people who filled out the survey did not rate their prisons highly in terms of safety, well-being and development, professionalism or harmony, and most were not very positive about the conditions of their lives. Maybe that’s no surprise.

People who took the survey were scattered across 22 IDOC prisons. Out of the whole group, only about 23% believed their institutions were safe and secure—a composite measure made up of reactions to statements about such things as supervision and control of the prison environment, protection from bullying and threats, and the presence of drugs and gangs inside the walls. Only 10% felt their prisons supported resident well-being and development, and only 6% thought the prison had a culture of respect, humanity and decency. Finally, only 3% found prison organization and staff to be professional.

But there is variation, and that matters. From one prison to another, residents’ ratings sometimes vary substantially. On the set of measures that make up the harmony dimension, for example, positive ratings ranged from a high of 86% at one institution to a low of 1% at another. Similarly, in response to questions about the extent to which they promote the well-being and development of their residents, the top three facilities had positive ratings ranging from 42% to 86%, while the bottom three were all under 5%.

Interpreting MQPL Results

Unpacking these differences begins with acknowledging that not all prisons serve the same functions, or house interchangeable populations. For example, some of the most negative MQPL scores pertain to aspects of life in large, maximum-security facilities. More positive responses—and these are very much outliers—come out of smaller, minimum-security prisons and “Life Skills Reentry Centers.” That makes sense.

We also explored whether variations in MQPL composite scores could be accounted for by the individual demographic and other characteristics of those taking the survey, rather than which facility they were held in. Through multivariate statistical analyses, we sought to measure the extent to which MQPL composite scores were influenced by any of the things we also knew about survey respondents, including their current age and age at first incarceration, their race, how close they were to release, whether they reported getting at least two hours of out-of-cell time per day, and whether they had a work or school assignment.

Our preliminary analyses found that there were some individual characteristics that did explain some of the variation in composite scores, as well as facility-to-facility differences in composite scores that remained even after we statistically accounted for the influence of the individual-level characteristics of respondents. But we still found a substantial amount of variation in scores that could not be explained by either kind of factor—known individual characteristics or holding facility.

In some ways this is disappointing, because it means, in plain English, that there’s a lot we don’t know about what may drive MQPL scores up or down. The encouraging thing, however, is that the factors that were found to have the most influence on quality-of-life ratings were things that are at least potentially under the operational control of correctional authorities.

For example, the factor having the strongest effect on Wellbeing/Development ratings was whether individuals reported getting at least two hours a day out of their cell or sleeping area per day. Overall, Wellbeing/Development scores averaged 2.2 on a scale from 1 to 5. But scores for those that said yes to the question about out-of-cell-time, after statistically controlling for the influence of other known characteristics, were .30 higher than those who said no. Similarly, those that reported having either a school or work assignment had an average rating that was .22 higher than those that had neither type of assignment.

The same two factors had similarly strong effects on composite Harmony ratings. Again, the overall Harmony score average was 2.2 on a scale from 1 to 5, but those who reported having out-of-cell-time had an average score that was .30 higher than those who did not. And those reporting either a school or work assignment had an average rating that was .19 higher than those with neither one.

Open-Ended Responses

In addition to standardized agree-or-disagree items, the MQPL has some open-ended questions, which give survey-takers space to write about positive and negative aspects of the institutions they’re in, offer suggestions for improvement, and share any comments or feedback they may have about the survey itself. These are “data points” too, though they may be harder to chart or reduce to numbers. That’s why CCJ student volunteers are helping to ensure that they don’t get lost, by transcribing and coding written responses and marginalia from the MQPL survey. Occasionally, they flag indications of dangerous situations in particular prisons, such as physical assaults or sexual abuse, that must be called to the attention of IDOC. But mostly they’re cataloguing hopes, frustrations, cynicism, anguish, boredom, irony—the whole range of individual human reactions to the experience of imprisonment—and thereby helping to fill out a more authentic picture of prison life.

Using MQPL Results

The MQPL is designed to elicit information that isn’t easy to get any other way, and to give voice to people who may not always feel free to speak. From the point of view of correctional leaders and prison administrators, residents’ insider views of prison functioning are potentially useful for performance management and improvement purposes—especially when prisons vary along the same measures, or change over time. Reliable information about how people in prison feel and perceive things can also illuminate incident data that IDOC already tracks, on assaults, uses of force, disciplinary segregation, and the like. And given the relatively high rates of post-prison recidivism in Illinois and elsewhere, it makes sense for leadership to pay attention to information about the ways prisons may be failing those who pass through them, especially when it comes from those who really know.

JHA is already using MQPL survey responses to supplement its usual sources of information about conditions in individual facilities, such as published administrative data, letters from people in custody and their families, and observations during monitoring visits. For example, a recent JHA report on the Kewanee Life Skills Reentry Center featured charts showing responses to a few key MQPL questions, based on completed surveys that had been received at the time the report was issued. It wasn’t a big sample (36 residents, 18% of the population), but the results suggested how radically an institution like Kewanee—a repurposed juvenile facility with newer buildings, a smaller population, more programming and opportunities, and a clear institutional focus on preparing residents for release and successful reentry—can differ from other prisons in the IDOC system, at least from the point of view of the people who live there.

MQPL results can be of immense value in educating the general public as well. CCJ is working with JHA to create an online dashboard that will provide visitors to the JHA website with MQPL results and analysis in a comprehensible visual format. The tool allows users to compare institutions across various MQPL dimensions, to zero in on particular issues like staff-resident relationships, and to filter results by demographic and other characteristics. When launched, the dashboard will help JHA fulfill a central part of its mission, which is to “shine a light” on institutions that the public cannot otherwise see.

The Future of the MQPL

European corrections professionals have been using the results of quality of life surveys for decades, to monitor prison climate and improve prison effectiveness. But the American effort is still very much at the beginning stages.

Two of JHA’s peer organizations (the Correctional Association of New York and the Pennsylvania Prison Society) are mounting similar efforts using the adapted MQPL in their states. So one day, multi-state comparisons of prisons and prison systems may be possible, along with the sharing of learning among states. State-to-state and prison-to-prison variations in MQPL results can be explored, promising outliers can be examined more closely, changes over time can be tracked, progress can be celebrated and spread. Where deliberate efforts are being made to improve prison climate—like the “Little Scandinavia” project in Chester, PA, which is attempting to replicate Norwegian corrections practices in an American prison unit—the MQPL can help measure and contextualize the results.

Loyola’s CCJ is committed to staying involved in this work, providing analytical, data visualization and other research support to the MQPL project as part of our ongoing partnership with JHA. We’re still learning, and we hope to share what we learn with our students and the broader field.