Criminological Highlights Vol. 21, No. 1 - September 2023

This issue of Criminological Highlights addresses the following questions:

1. Is morality in our society really declining?
2. What businesses increase firearms homicides?
3. Do mothers who were incarcerated neglect their children’s education?
4. Why are Black defendants less likely to get pretrial release?
5. What if police strength in a community changes?
6. Does skin darkness make a difference for people other than Blacks charged with offences?
7. Does climate change contribute to crime?
8. Do body worn cameras improve the reputation of police?

Article 1

People believe that the level of moral behaviour in their communities is declining even though systematically collected data show convincingly that people do not see a change in the behaviour of those around them with whom they interact. This belief is widespread (across countries) and across time. For politicians the belief in the decline provides simple justification for the need for policies such as “tough on crime” measures.

Why do people believe that their fellow citizens are less moral than they once were? One possibility is that there is an actual decline, though one wonders why we haven’t hit rock bottom since concern about the decline has been with us for more than 2000 years. Alternatively, it may be a misperception based on an illusion that people around the world are susceptible to.

This study first examined data from 177 survey items that people in the U.S. (n=200,772) had been asked over a 70-year period. These included questions such as “Do you think that over the last few decades our society has become less honest and ethical…” (p. 782). On 84% of these items, the majority of Americans in the various surveys thought that there had been a decline. The perception of decline was not related to the year in which the survey was carried out (1949-2019). People were, however, more likely to report a decline if asked about a long period of time. 

In another study, survey results from people in 59 countries who had been asked similar questions (between 1996 and 2007) showed the same result: Most people saw a decline. People all over the world believe that morality has declined. Older people perceived more moral decline during their lives simply because they were thinking about a longer period of time. Interestingly, however, when data were examined to see if people perceived there to be a moral decline in the years before they were born – they saw none. In other words, they believed that people were as honest, nice and good in the year that the respondent was born as they were 20 years before the respondent’s birth. “Participants believed that moral decline began at about roughly the same time they appeared on Earth” (p. 787), whatever year that happened to be. People attributed moral decline to decline within individuals and to the belief that younger people were more likely to embrace immorality.

Generally, it seems that people are especially likely to attend to negative information about people in general (e.g., whether they once committed a criminal offence). Furthermore, the mass media often focus on negative events (e.g., violent crimes in public places) and seldom put these events in context (e.g., whether they are unusual). Decreases in crime may not get the same amount of attention as crime increases.   When thinking about the past, however, positive events are more likely than negative events to be recalled. The past looks moral compared to the present.

But when looking at people’s own personal experiences (how they themselves reported they had been treated), the results were quite different. The researchers found 107 questions about people’s own experiences that had been asked at least twice in surveys administered at least 10 years apart between 1965 and 2020. People’s reports of current morality were, in fact, stable over time – both in the US and in other countries. In other words, when asked about the morality of people in their personal worlds – “the people with whom you currently interact, in person or otherwise, in your everyday life… friends, family members, coworkers, classmates, neighbours, etc.” (p. 786) – people saw no decline in the previous 15 years even though they reported general decline.

Conclusion: The data show that the misperception that there has been a moral decline in society is widespread. This misperception is not caused by people’s personal experience with the behaviour of others. With large numbers of people in western countries believing that addressing the moral decline of society should be a high priority, we run the risk of political leaders capitalizing on the public’s misperception that dramatic negative change has taken place and moving toward policies (e.g., increased imprisonment) that are not going to improve society.

Reference: Mastroianni, Adam M. & Daniel T. Gilbert (2023). The Illusion of Moral Decline. Nature, 618, 782-797.

Article 3

Children of mothers who have experienced incarceration experience more school-based involvement by their primary caregiver (the mother or, often, a grandmother). It would appear that these primary caregivers are trying to compensate for the fact that the child’s mother has been taken away.

There are many studies showing that the incarceration of fathers has negative consequences for their children (see Criminological Highlights Special Issue on The Effects of Imprisonment: Specific Deterrence and Collateral Effects). In contrast, this paper examines the effects of the imprisonment of a child’s mother on a specific aspect of the child’s upbringing: the involvement of the primary caregiver (PCG) in their child’s school.

It is broadly accepted that there are values to children of parental involvement in their education.  Few fathers who are incarcerated considered themselves to be the PCG for their children, whereas most (70%) of incarcerated women with children identified as their children’s PCG. This paper, therefore, focuses on incarcerated women and their children.

The study uses two sets of data: A US survey that oversampled unmarried mothers, interviewing them shortly after their child’s birth and following them through the time that their child was in primary school, and detailed interviews with 42 formerly incarcerated mothers.  In the survey, involvement in the school was assessed with 6 questions related to such things as attending events being held at the school for parents, and attending meetings related to their children.

Not surprisingly, children of ever- vs. never-incarcerated mothers experienced many additional forms of disadvantage. Their household income was lower. Their mothers were more likely to report substance abuse and/or depression, and were more likely to have had contact with Child Protective Services (CPS). Children of incarcerated mothers were more likely to be living with a PCG who was not their biological mother.  Given these facts, it was important to compare PCGs who were comparable on these and other variables.

Controlling for paternal incarceration, demographic differences, and maternal challenges, it was found that the children of mothers who had been incarcerated had increased likelihood that their PCG would be involved in their schooling (whether or not the primary caregiver was the child’s biological mother). This involvement of the PCG in school activities was not due to disciplinary or behavioural problems of the child.  The interview data suggested strongly that the involvement of formerly incarcerated mothers in the schooling of their children was linked to mothers’ desire to compensate for their absence in prison and to mothers’ hopes for social mobility of their children. Interventions by CPS were an ever-present concern and one reason that the PCG was often the mother’s mother. In the interviews it became clear that “Proving oneself as a mother in the eyes of one’s child and family was a repeated theme…” (p. 504).

Conclusion: Being incarcerated redefined what it meant to be a mother. Being involved in the child’s education was one way in which these disadvantaged mothers could attempt to compensate for the deficits experienced by their children. “Importantly, the primary audience of school involvement was not the school itself but children first and foremost, along with the alternative caregivers with whom relationships had been strained by maternal incarceration…. Mothers also understood their children as a source of pride and self-worth and viewed their children’s educational attainment as a barometer of their parenting…” (p. 508).

Reference: Branigan, Amelia R., Rachel Ellis, Wade C. Jacobsen, Anna R. Haskins (2023). System Management and Compensatory Parenting: Educational Involvement After Maternal Incarceration. Criminology, 61, 482-517.

Article 4

Black defendants in US federal courts are more likely to receive detention recommendations than are White defendants. The racial disparity in probation officers’ recommendations operates largely as a result of one factor: the criminal history of the accused. 

The use of pretrial detention is a concern for at least two simple reasons: it can be harmful to an accused who has not been convicted of an offence and it is, inherently, a punishment that is imposed by a court on a legally innocent person.

Although ultimately pretrial detention decisions are made by judges, other people typically give advice or suggestions to the judge. In US federal courts probation officers are required to provide information and recommendations to the judge. They are instructed to consider both the offence and the defendant’s criminal record in their recommendations as well as more social factors, such as community ties. In addition to descriptions of the accused on various specific measures, the probation officers complete a pretrial risk assessment (PTRA) measure which is given to the judge.

This paper compares the recommendations made by 3123 probation officers in 84 federal districts on the likelihood of a recommendation for pretrial detention for Black and White (non-Hispanic) defendants. Overall, Blacks were 34% more likely to receive a recommendation for detention. 

Racial disparities that disadvantage Black defendants were the smallest in districts that tended to detain lots of people (of either race) and where the PTRA suggested that the defendant was high risk. These obviously are cases in which there would appear to be little discretion for the probation officer: a recommendation for detention was “natural” if people were generally detained in that district and the accused person was rated as being of high risk.

However, in both the districts with high and low detention rates, Blacks with low criminal history scores were more likely to be disadvantaged by the recommendations of probation officers. For those with serious criminal records, Blacks and Whites tended to be treated in a more similar fashion. “Racial disparities generally are greater when the situation provides greater room for discretion. These ‘gray area’ cases require officers to exercise personal judgement to make a detention recommendation and, therefore, may provide greater room for personally mediated disparities” (p. 251). However, about 60% of the racial disparities could be explained by criminal history alone. 

But in addition, “a simple combination of defendants’ criminal history (a policy-centric factor) and socio-economic status (a policy-peripheral factor) explains nearly as much of the racial disparity in officers’ detention recommendations as the best summary of policy-centric factors” (p. 254). The criminal history of the accused person appeared to be a central factor in the probation officer’s recommendation.

Conclusion: In sum, there is evidence of racial disparity in recommendations regarding pretrial detention, especially in the ‘gray area’ cases. This disparity appears to operate “through policy-centric factors rather than personally mediated bias” (p. 254). The focus of the decision appears largely to be the criminal history of Black accused people. If, ultimately, the goal is to reduce the racial disparity in recommendations concerning pretrial detention, this might be done by restricting the weight given to criminal history to ensure that its effect does not go beyond what might predict serious offending and failure to appear. 

Reference: Skeem, Jennifer, Lina Montoya and Christopher Lowenkamp (2023). Understanding Racial Disparities in Pretrial Detention Recommendations to Shape Policy Reform. Criminology & Public Policy, 22, 233-262.

Article 5



Misdemeanor arrests in the U.S. dropped dramatically between 1997 and 2018. The decline is related most closely to lower numbers of police officers per capita and lower police expenditures.

Misdemeanor arrests increased in American cities between 1980 and 1997 and then decreased by 45% between 1997 and 2018. Although some of this decline could have been a result of changing values and laws related to marijuana use, the decline in “quality of life” offences (e.g., loitering, disorderly conduct, public drunkenness) was also dramatic, peaking in the early 1980s and declining 77% between then and 2018. 

A number of possible explanations exist for this decline including the implementation of various forms of community policing models. Police have a great deal of discretion on whether to arrest people for misdemeanors, especially since many contacts between police and citizens related to misdemeanors are not the result of reports to the police. This paper examines the relationship of the drop in misdemeanor arrests to changes in police strength, measured by the number of officers and police expenditures per 1000 residents. Data were obtained for 940 U.S. cities between 1990 and 2016. Hence the analyses look at the impact of variation in staffing levels and expenditures on policing across time (within cities) and across cities.

The paper looks at three measures of low-level (misdemeanor) arrests: the rate, the percent of all arrests that involved misdemeanors, and the rate of “quality of life” arrests. The results were very similar for the three measures. Various controls were also included: the violent crime rate (as an indicator of the amount of serious crime in each city), race, economic disadvantage, the percent of young men and foreign born in the city.

The number of police officers and the amount of police expenditures per 1000 residents predicted all three outcome measures. As these measures of police strength decreased, misdemeanor arrests (and quality of life arrests) also decreased. 

For about 40% of the cities, a measure of the implementation of community policing was available. It consisted of an index created from measures of 5 separate dimensions of what typically is considered community policing (e.g., consistent patrol beats for officers, encouragement of officers to assess and solve problems). The presence of community policing had no impact on changes in the outcome measures (misdemeanor or quality of life arrests).

“When cities shrank their police staff or their police spending, they saw associated declines in low level arrest rates” (p. 342).  Police strength did not change dramatically during this period, suggesting that even small changes can affect what happens to those susceptible to arrest for minor matters. As a New York City police commissioner once suggested, “When you hire more officers, they make more arrests” (p. 342). Police, once hired, understandably want to show that they are productive; discretionary arrests are a mechanism to do this. And there is more discretion for minor matters (misdemeanors) than there is for more serious offences.

There was little evidence to suggest that ordinary citizens were at risk from lower rates of misdemeanor arrests: “Cities that saw large [overall] crime reductions experienced larger declines in both misdemeanor and quality of life arrest rates” (p. 343).

Conclusion: The rate of low-level arrests for minor matters is under policy control.  Increased police strength and police expenditures would appear to be likely to result in an increase in the rate of arrests for minor matters. Even though arrests (and convictions) for minor matters do not necessarily lead to severe criminal justice outcomes, the effects on the employment, finances, school performance, immigration status, etc., of the person being arrested can be dramatic. Furthermore, as other research has shown, “For young people, especially Black adolescents, a single arrest increases the likelihood of a subsequent re-offence” (p. 332).

Reference: Beck, Brendan, Eaven Holder, Abigail Novak, & Jacob Kaplan (2023). The Material of Policing: Budgets, Personnel and the United States’ Misdemeanor Arrest Decline. British Journal of Criminology, 63, 330-347.

Article 6



Hispanics charged with criminal offences are more likely to be convicted and imprisoned if they have darker coloured skin.

Previous research has shown that Black accused people with darker skin colour are more likely than those with lighter skin tones to be treated harshly by the criminal justice system (see Criminological Highlights 16(3)#1, 18(3)#4). This paper extends these findings to Hispanics and examines whether the same finding holds for Hispanics in the U.S.

In this study, mugshots of male and female adults with Hispanic names who had been arrested in Miami-Dade county, Florida, were rated on a 7-point scale (very light to very dark). Raters, who did not have the names or the case characteristics of the defendants, were able to do this very reliably. Cases that went to trial were excluded from the study. Of the remaining cases, about 40% were convicted (by way of a guilty plea) and, overall, about 7% were incarcerated. 

Various demographic factors were used as control factors including sex, whether the person was an immigrant, age, poverty and criminal justice factors (criminal record, number and nature of the charges). The study used data (and controlled for) the police service involved in the case: 3 large police services plus a number of smaller ones that were pooled into one category. 

The arresting police had characterized each person, on the arrest form, as either Black or White. In order to be able to examine the data for males and females separately and for those described by the police as Black/White, samples of about 1630 individuals from each of these four groups (male and female who were either Hispanic or white) were examined.

Both the probability of conviction and the probability of incarceration (of those convicted) increased with the darkness of the defendant’s skin colour. For both those described by the police as Black and those described as being White, those darker in colour tended to receive harsher treatment.

The effect of skin colour on the criminal justice outcome appeared to be linear. The lightest skinned defendants (those rated 1 on a 7 point scale) had a probability of being convicted of 36%. This increased with every point of rated skin colour to 53% for those with a rated skin colour of 7. Looking only at those convicted, those with the lightest skin colour (rated as “1”) had a probability of incarceration of 5%. Those with the darkest skin colour rating (rated as “7”) had a probability of incarceration of 25%.

Conclusion: Once again, it appears that in the criminal justice system – at least in Florida – the actual colour of a person’s skin makes a difference in terms of how punishing the treatment will be from the criminal justice system. Specifically, those defendants rated by independent raters as having darker skin were more likely to be convicted and were more likely to be incarcerated than those with lighter coloured skin. This effect holds when various relevant factors (offence, criminal history, etc.) are controlled.

Reference: Lanuza, Yader R., Nick Petersen, and Marisa Omori (2023). Colorism in Punishment among Hispanics in the Criminal Justice System. Social Problems, 70, 275-296.

Article 7

Climate change may change the pattern of crime.

In the early part of the 19^th century, it was suggested that there was a strong association between warm weather and crimes against persons. This paper looks at the relationship between one of the most important aspects of climate change – temperature volatility – and crime. Hence it examines not long term changes in crime, but changes in crime associated with unusual temperatures on particular days. More specifically, the study examines unusual seasonal temperatures in 28 large US cities and relates their occurrence, on a given day, to changes in the rates of robbery, aggravated assault, and homicide during a 7-year period (2015-2021).

Generally speaking, violent crime (in the US) tends to be higher in the summer than in other seasons, though the findings are not entirely consistent. There are obvious theories as to why violent crime rates might be higher in the summer, yet there are few data that might be considered to be ‘tests’ of these theories. The routine activity hypothesis, for example, suggests that certain weather events – e.g., pleasant weather that puts people together outside or in public places – may provide opportunities for interactions that might not otherwise take place. And these interactions may provide an opportunity for crime to take place.

The study looks at the 30-day rolling average temperature for each day in each of the 28 cities and identifies those days that are two standard deviations above (and below) the rolling average temperature. In other words, it looks for those days that have very unusually high (or low) temperatures for that immediate period. 2.5% of the days had unusually warm temperatures and 4.3% of the days had unusually cool temperatures. Obviously, what a ‘high temperature’ (or low temperature) anomaly looked like – and what season they took place in -- varied with the city. Essentially, the idea was to look at what the crime rate was for these three violent crimes (robbery, aggravated assault, homicide) on days when the weather was unusually hot or cold.

Looking across all seven years (and ignoring the season) violent crime rates were significantly higher on anomalously hot days. For example, an unusually warm day was associated with a 9.4% increase in aggravated assaults (compared to the surrounding days). On anomalously cold days there was an approximately 10% lower rate than the rate on the surrounding days. Crime rates for these three crimes were also higher when the 30-day average temperature was higher (i.e., in the summer).

The effect of unusual temperature increases appeared to be most important during the winter months (defined as January-March) for robbery and homicide (but not for aggravated assault). Unusually cold days in the winter were associated with lower rates of robbery. 

Conclusion: The data suggest that “urban assaults, robberies, and homicides are meaningfully associated with the temperature volatility predicted to become more common as climate change accelerates” (p. 7). This effect is independent of general seasonal variation in crime rates in these cities. The mechanism that might account for these effects is at this point unknown. However, the data suggest not only that the ‘warming’ effect of climate change could have an effect on crime, but also the volatility of the day-to-day weather will have an independent effect on violent crime rates.

Reference: Thomas, Christopher & Kevin T. Wolff (2023). Weird Winter Weather in the Anthropocene: How Volatile Temperatures Shape Violent Crime. Journal of Criminal Justice, 87, 1-15.

Article 8

The use of body-worn cameras by the police does not improve residents’ views of the local police or residents’ assessments of their own interactions with the police.

A survey carried out in New York City in 2016 suggested that most residents favoured the use of body-worn cameras (BWCs) by the police and hoped that their use would improve police-community relations. Previous research, however, has found that citizens from various minority groups are more skeptical than whites that the use of BWCs will contribute to holding the police accountable for their behaviour. This study examines whether BWCs improve ordinary citizens’ perceptions of the police.

Forty New York City police precincts with high numbers of complaints against the police were identified. Pairs of precincts were matched on various characteristics (demographic measures of the citizen, crime rates, police activities) and then, within each of the 20 pairs, one precinct was randomly assigned to have BWCs deployed by the police officers assigned to that precinct. Police officers in the other precinct in each pair did not use BWCs. In the experimental precincts, uniformed police officers and those working anti-crime assignments used BWCs from 3 PM to midnight each day. Other research suggests that police wearing the BWCs were more likely to file reports about stops and that complaints against officers wearing BWCs were less frequent than they were in precincts in which the police did not wear BWCs.

The evaluation in this study was carried out by surveying a representative sample of adults in each of the 40 precincts both before and after the implementation of the BWCs. Adults 18-34 years old were oversampled to ensure that there were sufficient respondents who had experienced interactions with the police. About 3000 people were interviewed at each time period.

Respondents answered, on 4- or 5-point scales, questions concerning such matters as their perceptions of and trust in their neighbourhood police, whether they had been treated well during any police stop (car, pedestrian) in the previous 12 months, and whether they had generally been treated well by the police in any matter in which they were seeking police assistance.

Because the study had measures prior to the implementation of BWCs, any changes that might have taken place in the precincts in which BWCs were implemented could be compared to changes that might have taken place (for any reason) in the control precincts.

The results were very consistent: there were no changes between the pre- and post-intervention periods that were statistically significantly different from changes in the control areas. Said differently, residents’ perceptions of the police did not change as a result of the deployment of BWCs.

Conclusion: In this study, not everyone who was surveyed (before or after the deployment of BWCs in the experimental precincts) had experienced contact with the police. Nevertheless, the results suggest that “BWCs are unlikely to lead to short-term changes in public perceptions of the police” (p. 303). Whether the same results would be found in locations in which police have considerable contact with members of the public is yet to be determined.

Reference: Braga, Anthony A., John M. MacDonald & Lisa M. Barao (2023). Do Body-worn Cameras Improve Community Perceptions of the Police? Results from a Controlled Experimental Evaluation. Journal of Experimental Criminology, 19, 279-310.

This issue of Criminological Highlights was prepared by Anthony Doob, Rosemary Gartner, Sara Fruchtman, Maria Jung, Tyler King, Audrey Macklin, Roxy Shlapak, Jane Sprott, and Danielle Van Wagner.