Gender Differences and Sentencing: A Critical Literature Review


This review focuses on various pieces of literature that surrounds the perceived differences in sentencing gender. Also, literature examining the reasons why these differences are taking place between genders, and theories that could be applied when explaining these differences, will be scrutinised in order to give an indication as to whether a reason for gender differences in sentencing has been identified. The two theories that will be focused are the Chivalry theory (including Selective Chivalry) and the Double Deviance/ Evil Woman theory. Some other factors effects on sentencing, and the literature surrounding them, were also looked at as it would be negligent to say that only one factor could cause the perceived disparity between male and female sentencing. This review mainly focused on bodies of work based in the United States of America. This is because a large amount of research has been done in this area in the United States. Therefore, any questions answered will mostly be only applicable to that country due to cultural and legal differences in other parts of the world. Throughout the review a lot of bodies of research can be seen to be relatively supportive of the ideas that Double Deviance and Selective Chivalry has on the sentencing process, less so for regular chivalry. This is because various other factors seem to have some sort of effect as well as gender. Therefore, it is perhaps inaccurate to point to gender being the factor that decisively affects the sentencing outcomes. More research should be done in this area to fully grasp the relationship between gender and sentencing outcomes, while taking into account a larger number of relevant factors (legal and extra-legal) in order to not over attribute the outcomes to gender.

Author: Kieran Malon, April 2020

BA (Hons) Criminology with Psychology

1.   Introduction

A question that has long been discussed in various forms of academic literature is why there seems to be a difference in how genders are treated during the sentencing phase of trials. Within the United States the male population in prisons massively outnumber the female population. This may suggest a difference in how genders are treated at some stage during the criminal justice process in the United States. The stage that will be focused on within this review will be the sentencing phase of the system. A focus will also be made on two theories that have been looked at in various pieces of academic literature as well as other factors that may be contributing to this disparity in treatment. The literature will then be reviewed, and its validity will be looked at in relevancy and ability to test and explain the differences between genders.

The sentencing phase of the criminal justice process was chosen because there is evidence to suggest that this area is where the most difference in how the differing genders are treated in relation to what they are sentenced to do. This is suggested by academic literature like authors such as Steffensmeier et, al (1998), who used the gender effect as one of the factors that affect how an individual is sentenced, and Gelsthorpe (2013) who also looked at this factor as well as whether that number is justified in the crimes the crimes they commit. The justifiability may come in the form of whether the different genders get the same treatment when it comes to being sentenced to a crime that is similar in nature. If after looking at that justifiability differences do appear, we will then begin the process of looking at why that may be. Some theories that have been hypothesised will be looked at in relation to any differences in treatment found and the literature surrounding these theories will be reviewed and scrutinised in order to find if they have any relevant effects on modern sentencing outcomes.

The two theories that will be focused on when it comes to this topic are the Double Deviance theory and the Chivalry theory. These two are theories that have been discussed frequently in literature when it comes to this area of the criminal justice system. The literature surrounding these theories will be further discussed later in the text and their relevance, or lack of, will also be discussed further on. Beyond these two theories however, I will also be briefly looking at further factors that have been hypothesised to affect this stage as well as the gender of a defendant. This may include popular factors in research in this area such as race and age of a defendant. Also, as these factors may interrelate with each other it is important to discuss how these may be advantageous to some groups of people and disadvantageous to other groups. This may lead to different sentencing being given to different groups of people depending on characteristics that may be out of their control.

Review Structure

Within this review, three studies have been chosen as the focus for each of the theories (Chivalry and Double Deviance/ Evil Woman). These studies will be analysed as well as various other forms literature on these theories and relevant studies will be mentioned in reference to whether they support the studies in focus. An attempt at looking at a variety of different crimes has been made to investigate whether the theories apply or not across a spectrum of different sentencing events. In order to take into account other factors that may contribute to any differences, other than gender, in sentencing, two legal and two extra-legal factors have been chosen for further discussion. However, it should be noted that there are a range of other factors that will not be discussed in as much detail due to the large amount. But it is noted that these other factors exist that may influence sentencing decisions. This will be finished with a discussion on future studies that could be conducted as well as the limitations of this review and a conclusion on what can be found from the reviewed literature.

2. Chivalry, Selective Chivalry, and Paternalism Theory

This is an idea that was put forward by Otto Pollak (1950) to suggest that women within the criminal justice system are treated much more leniently than men due to the idea of chivalry towards women. Later Paternalism would be identified as something that frequently follows the chivalry aspect. It is suggested under this theory that law officials/judges see women as child-like and defenceless in their behaviour (Herzog and Oreg, 2008) and therefore are in need of protection, this leads to said leniency I favour of women. It can be said that Pollak’s research could be seen as outdated, we look at whether elements of the theory can be seen in today’s criminal justice system when looking at sentencing. Paternalism has led to modifications to the Chivalry theory however, this has been called Selective Chivalry. It is suggested by Farnworth and Teske (1995) that the leniency that comes with this chivalry idea is only open to white women and those who have wealthy backgrounds (Jeffries and Bond, 2013). There have been studies conducted on a range of different crimes which then have looked at how chivalry could possibly influence how sentencing decisions turn out the way in which they do.

Study One

Holland and Prohaska (2018) conducted a study in which they looked at whether females were more likely than males to receive shorter sentences while also controlling for relevant factors that could possibly affect the sentences also. One factor that they did want to account for and investigate further, in addition to gender, was race. Racial effects will be discussed in more depth further in the review, however they did want to see, as well as if there are differences between men and women, whether there are differences in sentences between women of different racial groups. This would allow a view into whether just gender could possibly effect sentencing between the males and females, or if other factors also need to be present to effect sentencing. Therefore, a second hypothesis was that white women would receive more lenient sentences than women of colour, which would support the work of Farnworth and Teske (2008) who suggested chivalry would only apply to white women. They also took into account geography when making various assumptions about what the results may show in their hypotheses. They hypothesised that women in the south of the country will be sentenced differently from women in other regions. The data collected looks at all federal cases that spanned the year between the 1st of October 2014 and 30th of September 2015. Also, due to the database containing information on a range of controls for legal and extra-legal factors, it means that an in depth analysis can be done to measure the various factors influence on the sentencing process and seeing how they could interact with the gender factor to lead to a sentencing outcome.

From the results, they found that their first hypothesis was proven correct. Women in general did receive shorter sentence lengths in comparison to males. This is also with legal factors considered and supports the chivalry theory. This supports the various bodies of work that have shown similar results through various experiments they have conducted (Doerner and Demuth, 2010; Rodriguez et al., 2006). This supports the idea that women are not seen by law officials as being as culpable for the crimes they committed as men are. However, the results contradicted the selective chivalry claims that would suggest that white women would mostly benefit from the leniency hypothesised in the chivalry theory. Hispanic and black women got shorter sentences when sentenced for federal drug crimes. This is surprising as the reverse has been shown in respect to selective chivalry as white women have benefitted the least from the suggested leniency shown in the results of this study. However, that may be because they have been adjudged to have been more out of line with traditional gender roles leading them to punished as doubly deviant rather then being viewed as needing protection.

There are some limitations on this study even if it does cover a population across a large area (Across multiple states). This is a problem in itself as guidelines differ across different states. This means that some judges will have more discretion than others, allowing them more freedom in decision making on sentences. Therefore, that has to be taken into account when looking at the results. Therefore, this may need to be built upon by more studies looking on a state by state basis accounting for those guidelines and taking them into account. There are also multiple other variables that could be explored. These may explain any contradiction with other forms of research that show support for selective chivalry as it does not have any detail on if these other factors could have had a effect on the sentencing process.

Study Two

Embry and Lyons (2012) focused their study on the discrepancies in the way that male and female sex offenders are sentenced and how chivalry theory could possibly have an influence in these sentencing decisions. They give an initial idea of what they expect to find, which is based upon previous literature (Jeffries, Fletcher and Newbold, 2003; Curry, Lee and Rodriguez, 2004), that females receive more lenience than males do when it comes to sentencing. For their study, they collected data from the National Corrections Reporting program in order to do secondary data analysis. The sample they used was spread over ten years but they originally had more than this. They used the most recent ten years in order to get the most relevant results. This gave them a more modernised picture of what is affecting the current sentencing process as values and views can change over time. Also, using a large time frame allowed them to offset another problem which is the low number of female offenders who have been sentenced on being a sex offender. Therefore, looking at a data set over a larger period allows them to have a large sample of female offenders in the data to look at and analyse.

From the study, the evidence showed that although there is no difference in sentencing rates between men and women who commit sexual offences, men do tend to get harsher sentences. This shows that although judges can see that women should be charged (because they have committed a crime), they may not believe that they are as perhaps dangerous as male sex offenders are. Embry and Lyons (2012) earlier talk about this stereotypical image of a sex offender, which is usually a male offender with victim coming to mind as a young female according to them. This could be proven to have some accuracy if you look at the perceived leniency that could be inferred from the results of this study. Due to women not fitting the stereotypical image of a sex offender, even if they have in fact committed the crime, they may still be deemed as less dangerous as a male sex offender. This shows that although the decision that all genders need to be punished for criminal offences is equal, the severity of the punishment across genders is not equal.

There are areas that could be developed in this study in order to perhaps improve its scope and relevancy to a broader population as well as limitations that can be identified in this study. Although the number of women that were included in these studies was a fairly even split, this may not always be a study that can be compared to real life. This is because although it was a fairly even numbers, compared to a lot of studies, the offending rates of women are way below the offending rates of males. Therefore, even if they did get a better idea how the factors and theories may affect a trial. Therefore, if you did want to investigate these discrepancies women will based on even figures, women will be sampled far more than they generally offend. These results may simply seem to point to one answer when it is simply just a question of numbers in terms of offending rates. There is also an issue with the fact that the study they conducted was built from basic figures taken from the data base they were sourced. Although we can assume from the studies that support the chivalry theory that sentence lengths were affected by the gender of the individual being sentenced, we can be certain due to the lack of specificity within the statistics. Therefore, we cannot rule out the fact that there was a higher percentage of males who had committed a more severe form of the crime that they had committed in comparison to the females who were being sentenced. A final limitation would be that although they would class this as a cautious generalization, they could only possibly say that it is a cautious generalization of the population in the United States, where it was based. The sample population that had been sentenced that they were looking at was entirely from the United States, if they broadened their sample to include statistics from various other countries a much larger cautious observation could be made.

Therefore, if they were going to look at doing a further study with this as the basis, a few steps should be taken to expand on this body of research. A more detailed data set would be needed in order to see more information about crimes committed or perhaps so we can find out more about the defendant being sentenced. We would hopefully be able to see whether Chivalry was in fact taking affect in the sentencing phase, if there were legitimate reasons for sentencing for one group being to harsher degree or if other theories and factors may be able to be more relevant in affecting the process. Finally, a sample of offenders that are from a range of different countries in order to give it the best chance of it being more generalizable to the rest of the world. Different countries have different views, values and offending rates. It would therefore be interesting to see if these theories can be applied across more than just one country.

Study Three

A final study that looks at the Chivalry theory is a study conducted by Spivak et al. (2014), who looked at an area that is unique from the other pieces of literature that were focussed upon. They looked at status offences committed by juvenile offenders. Status offences committed by juveniles include but not limited to, truancy, consumption of alcohol or tobacco or running away. This is an interesting area as previous literature done on this suggests that status offences are the only area in the juvenile system in which female offenders outnumber male offenders (Tracey et al., 2009). It is generally assumed that males do commit more crime than females (Messerschmidt, 2007) and so to find a category in which males do not outnumber females, and in fact females outnumber males, makes it an area for further study. The study was conducted in Oklahoma and the data was collected by a local agency that collects basic data on juvenile cases. They then cut down the cases to only look at the relevant cases in relation to the type of crime they were looking at (Status offences). In relation to this review, two of the hypotheses included looking at the cases as to whether the chivalry theory could apply to these cases. More specifically, they wanted to see if girls’ cases were filed for review, in comparison to boys. They suggest that if proven this may show a want to make sure girls’ cases are scrutinised to make sure they get a correct judgement.

From the results of the study conducted, both hypotheses relating to chivalry seem to be supported. The results show in this that girls were more likely then boys to have their cases further reviewed. An idea of why this could be explained by the chivalry theory is they want to try and protect girls from being guilty through further looking at their cases and the circumstances behind them. This perhaps leading to mitigating circumstances being shown on their behalf which could result in them receiving lesser sentences. Therefore, if chivalrous and paternalistic attitudes can be found even when it comes to looking at cases involving juveniles, it suggests that the want to protect females may start from juvenile court and be seen through most age groups once moved to be judged and sentenced as an adult. However, in this case it must be noted that although it was shown that girls did tend to get lesser sentences than boys, the relationship between gender and lesser sentences was very weak. Meaning that more studies must be done in this area as it is inconclusive when it comes to whether sentencing may differentiate between female and male juveniles, even if there is a slight relationship in favour leniency towards girls.

As is a regular problem when it comes to a lot of research in this area, a lot focus on one state for their research. This means that it cannot be generalisable as there are many differences in population and justice processes across the world and even in the United States. Therefore, more research in this area would be needed specially to help the more inconclusive aspects of this study. The database used is also quite dated for this study as they admit. This means that changes might be viewed if data was collected for juvenile cases now. If policies have been brought more recently, this may lead to a difference in results and lead to different hypotheses being drawn.

3.   Double Deviance / Evil Woman

Double Deviance is theory based on the point of view that certain women are punished under the view of doing two things wrong. They are viewed as having broken societal norms and expectations of how a woman behaves, as well as breaking the law. They are then judged upon the basis that they have done doubly wrong.  Murphy and Brown (2000) suggest that under this theory it creates a situation where women can either be demonised or can be shown more leniency depending on if they broke these societal norms on what is expected of women. Double Deviance theory which is sometimes referred to as the Evil Woman theory. Although this idea may seem like selective chivalry in the way in which some women may be treated more leniently. It differs greatly in the notion that women who break these societal ideas of gender norms are punished even greater than men do when they commit the same crime. Women who fit into this theory and are seen as doubly deviant are seen as more blame worthy in this case which leads to harsher sentences, even if the crime they have committed is the same (Herzog and Oreg, 2008; Tillyer, Hartley and Ward, 2015).

Study One

Tillyer et, al (2015) based their study on looking at the perceived unfairness that exists within the court systems. Although they noted various factors may contribute to these differing sentencing outcomes between various groups, they chose to focus this study on gender and the different theories surrounding the gender factor. The crime they chose to investigate was narcotic cases taken from a federal data base. Narcotic cases have been chosen because it is a crime that will be viewed as breaking traditional gender roles. Therefore, if the theory is to be accurate the results will show that women get a harsher punishment then men who commit the same crime. Some factors may effect this as well however such as criminal history as well as evidence showing that race may have an effect as well (Spohn and Holleran, 2006) therefore that also has to be taken into account when looking at the gender and sentencing differences. The dates taken from the database in order to be analysed has been specifically chosen due to it have the highest amount of women to have committed crime within this time frame, with females being most populous being sentenced for this crime than any other in the dates analysed. This should allow for a good comparison to be made between male and female sentencing cases as it gives more cases that will be analysed then other crimes and dates. According to the authors they look to answer two hypotheses. They want to test whether women with limited criminal history will get more lenient sentences and those with extensive criminal history get a harsher sentence then their male counterparts.

The results in this case showed support for both the hypotheses in their assumptions. The first hypothesis which suggested more lenient sentences for women who had limited criminal histories was correct. This can be assumed under the Double Deviance theory that it has led to those who are deemed to have acted in accordance with the image of how an idealised woman should act, which has led to them getting a more lenient sentence overall. The results gained from the database also suggested that the second hypothesis was also correct in it assumption that women who have been deemed to have broken the law and the norms of societal conduct for a women have been punished more severely than men who committed a similar offense. This assumption supports the hypothesis of Bontrager et, al (2013) and Herzog and Oreg (2008) that women are punished further for factors and occurrences that allow them to be viewed as doubly deviant. However, this could also show why there is a gap between the genders within the prison numbers. Although some women are overly punished for breaking a conceived notion of how a woman should act, some are also being given lesser sentences then men for committing similar crimes, which may explain some huge gaps in numbers between the genders in prisons. This study also shows the importance of how criminal history interacts with gender to influence the sentencing phase. That is a unique aspect about this study and has allowed it identify a key factor that has effected women who are being sentenced and may allow law officials to become more aware of factors that may be unconsciously effecting their decision making process. This sort of identification, if it increased awareness, could lead to more equal sentences in regards to these factors.

The limitations that have been noted does increasingly affect the ability to make this study generalizable. A large limitation that seems plague this study is the effect that new policy changes have on the sentencing phase. These policy changes may give judges more or less discretion when it comes to sentencing. For example, if a policy came in after this study was released that limited a judge’s discretion on sentencing for crimes like drug crimes, it may lead to a more equal distribution regardless of gender. This would be because a judge would then have less freedom when they are passing a sentence and may have to stick to more rigid guidelines when sentencing. If a policy such as this did come into law, then another study would have to be held in order to investigate how these changes may have affected sentencing and may lead to theories like the double deviance theory losing its validity. A further limitation is absence of a few different factors that may provide more information on the complexities when it comes to examining gender and sentencing. For example, they only limitedly consider factors such as family status and whether they are depended upon by others, which may include children. This is identified as a factor that could influence a sentencing decision and would be helpful if a more in depth study was done to look at how factors such as this could impact the findings and if they would give them a different look to lead to different findings.

Study Two

The study conducted by Koons-Witt et, al (2012) had a smaller focus when it comes to population they focused on. They conducted their study on the state of South Carolina, whereas the previous study looked at sentences across the country. However, within this study they looked at various crimes to see how the effect of gender changes across the various crimes that are committed. As well as looking at how gender has an effect when it comes to sentencing, they consider various other factors and how they interact with interact with gender sentencing processes. These include race and gender which will be discussed later in the review. Within the study they expect to find that women are treated more leniently when it comes to sentencing, which is in line with chivalry theory. However, as mentioned within this study the environment they a basing this study in must be taken into account. According to Koons-Witt et, al (2012) the state of South Carolina historically has rather conservative views of women which means that there may be stronger view, than in most places, in the traditional gender role of women. Traditional crimes for women are often non-violent crimes such as fraud according to Rodriguez et, al (2006). Therefore, when women commit crimes that are not viewed as traditionally crimes women commit (violent crimes), the evil woman hypothesis may have a stronger effect then it may do in states or countries with less conservative views.  The data for this study was collected from a now disbanded commission within the state. They focused on the latest data set that was made by the commission which was from 2001. This was due to various reforms that made the data collected before problematic to use. Therefore, they had 12 months of data to study and analyse.

The results for this study again showed support for the double deviance hypothesis. Similarly Tillyer et, al (2015), shows that only women with very limited criminal history are shown leniency in this case. Koons-Witt et, al (2012) mentions how the effect of women having an extensive criminal history background makes the leniency that is shown to women completely vanishes which then leads to them being sentenced on the same level as men. A woman with an extensive criminal history seems to then lose the protection that is often shown when it comes to sentencing. This could be as the law officials view that as the criminal history builds up, they are viewed as increasingly culpable for their actions and lose leniency that they are perceived to have. As a result, they receive greater sentences than females with lesser criminal histories might have. The support for the evil woman theory is further shown by this study because it again shows that women are shown leniency up until a certain point within the criminal justice system. Although this study focuses just on one state, it still supports a pattern that has been shown in many different pieces of literature on this topic (Belknap, 2007; Franklin & Fearn, 2008). This can also explain why there is a clear difference in how women and men are treated in this study. They note that in the sample they looked at the average for women with no criminal history was lower than the average for males with no criminal history. Therefore, this would then lead to a lack of cases where the double deviance hypothesis would take place which would, if women did have a higher average criminal history, theoretically lead to more women being sentenced on a similar or more harsh level then the males.

There some limitations to this study and suggestions on how to further this line of research in the future, if it has not already been conducted. The focus of this study was on, as the author says, a rather conservative state in relation to how women are supposed to behave. Due to this being on juts the single state, it would be interesting to see how this would compare to other states that are viewed as conservative in a similar way to South Carolina. This would give a better idea of how the ideological in the environment may change how different genders are sentenced. If similar results are seen, then it can be ruled out as an outlier and helps with generalizability. It would also be interesting to see how a more liberal state would sentence with a focus on differences between genders. This would give information on whether the conservativeness of the environment may have an effect of sentencing between genders and if it does, how much of an effect does it have. Due to it being secondary data they are working from as well, they cannot control what was recorded and what was not recorded. This has led to various factors being excluded from the recorded data that may have had an influence on the sentencing process. Although this may not have necessarily affected what was learnt from the results, some information that may have allowed a further, more detailed analysis from taking place. This would have looked at how these factors interacted with gender and sentencing outcomes, due to some of these factors often being stereotypically thought of as female roles, such as whether there are dependent children or the role of the individual with a family. These may have an effect due to women often being thought of as the primary care giver when children are involved compared to males. Daly (1987) suggested that women are more likely to receive leniency due to them holding important familial roles. This therefore that accounts for some disparity in sentencing outcomes between genders. However, without data recorded on this factor we cannot tell if this influenced sentencing outcomes in this case.

Study 3

Tasca et, al (2018) based their study upon analysing how parental status and sentence length interact with each other. The study was based in the state of Arizona with data collected through self-report and official data. The questions for their study were based around whether the parenthood factor, whether they have children that they care for, get affected by gender when sentencing. They also wanted to look at how sentencing lengths vary between offenders without children as well as parents who are involved with their children and those who are not involved. Various studies have already been done in this area on how familial status may affect sentencing outcomes (Daly, 1987; Freiburger, 2011). However, the unique aspect of this study is that it also considers how involved the offender may be with their children. This will give a more complex look into how the parenting dynamic affects sentencing outcome, while considering how gender plays into the dynamic. Within this study, they also look at various smaller factors within the parenting dynamic, such as how women incarcerated often tend to be single mothers with reliant children. Therefore, with factors like these in mind it may be easier to interpret the results as well as make initial predictions. Although within this study they are not necessarily looking into how the evil woman theory plays into this study, through the results it gives you an idea of how the theory’s ideas may have had a role in affecting the sentencing outcomes.

The results, if split between men and women parents, without breaking them down into the further categories could already suggest that women might be viewed as doubly deviant in the crimes they have committed. This is because the results show that female parents received longer sentences than fathers. Under the evil woman hypothesis, you could suggest that this may be because they have broken the traditional gender role of committing a crime when they are supposedly the primary caregiver in comparison males who are traditionally the economic support for the family. These initial results contradict previous research and hypothesis (Daly, 1987; Daly, 1989), however there may be other factors to be revealed that explain this further in the results section. Once the results were then further broken down, although it did show that women who lived with their children prior to arrest received more lenient sentences compared to mothers uninvolved with their children. However, it still shows that males still received less harsh sentences then the women. The fact that the results did show less lenient sentences on women who were uninvolved with their children could suggest that double deviance has an effect. It could be argued that the women who are uninvolved with their children are being more severely punished for not performing their motherly duties and are therefore being punished for more than just the crime they committed. It is also mentioned that the sample was drawn from offenders who are not new to being involved with the criminal justice system. Therefore, the criminal history factor also has to be taken into account as it has been earlier suggested that it can negate any leniency they may have been granted under the chivalry theory (Tillyer et al, 2015; Koons-Witt et al, 2012). Therefore, it could be argued that because this study was mainly populated by a sample of women who have a criminal history, most would lose the leniency effect that would be applied to those without a criminal history. Meaning that if the evil woman theory was applied here it could explain why the women had on average a higher length of sentence then the males did. As under the theory they would be treated more harshly for breaking traditional gender roles and committing the crime.

There are couple of limitations and suggestions for future studies that could be made based from this study. One limitation of this is the need for a larger sample. This is because varying court systems throughout the United States, as well as throughout the world, may show different results. It may have been more generalizable if they at least looked at various court judgements in different states within the country. Due to the different court systems between states giving judges varying degrees of discretion, that may prove to be a key factor in how sentencing patterns show in the results. Therefore, a larger more varied sample should be looked at in further research to see if the results and hypotheses drawn from the results are applicable to more than just Arizona. As the authors mentioned also, the majority of the sample had criminal backgrounds and that may have had a potential effect on the results. Therefore, a more varied sample in terms of whether they have a vast criminal history or not should be explored. This will allow a more balanced view on how much effect certain factors have on the sentencing process as it hard to measure the effect of factors such as criminal history when many of them have criminal histories.

4. Contradicting Literature

Although a lot of literature that has been analysed show a pattern that supports the hypotheses. There is going to be some literature that contradicts the pattern that is seen. This can be for various reasons and these reasons can be discussed later when looking at the literature. One piece of literature that was reviewed did seem to contradict the pattern of support and although there may be more, only one will be the focus here.

A study that contradicts the findings in support of the theory is one that looks at chivalry in relation to nonviolent offending. Specifically, the study undertaken by Koeppel (2012), looks at nonviolent property crime in rural areas in the state of Iowa, in the United States. This study is interesting as Rodriguez et, al (2006) mentioned that there are varying degrees of leniency in the sentencing process depending on the crime, but males do often receive harsher sentences. Rodriguez et, al (2006) and their body of work suggested that males were more likely to receive harsher sentence outcomes when it comes to property crime. This, if backed up by continuous research as well as the study of focus here, would look like it would be following a pattern on which females receive more lenient sentences across multiple types of crime. The author of this study (Koeppel, 2012) suggests that we may see differences because of the setting also, as the they are focussing on a rural setting. Steffensmeier et al. (1998) suggests that because of the volume of cases that an urban judge must get through, they succumb to using stereotypes and generalizations in order to make their decisions. Therefore, if this theory is correct, we should see a decline in the difference in sentences between male and female who commit the same crime. This could then show evidence against chivalry and suggest other factors may yet have a larger influence on sentencing.

The data was collected from five small rural counties that have similar demographics; this could then decrease the effect that different environments may have had on the process. The sample size in this case is a 507 split between 188 females and 319 males. Worth noting also was that 95% of the sample were white which will be a point brought up later when analysing this study. The presumed result for this study was based from previous research and that a clear pattern of leniency in favour of females would be present. However, the results of the study contradicted previous research and found that gender did not have a significant effect on the sentencing outcomes. This is surprising, if the Chivalry theory is to be believed, considering the nature of the crime itself. The crimes looked at were nonviolent crimes that would, in theory, be seen as crimes a woman would get more leniency under the Chivalry theory. Herzog and Oreg (2008) discuss how the judges would think of women as defenceless and weak under chivalry and paternalism. Therefore, because it is a white, female, nonviolent offender they fill that criteria in which chivalry/paternalism is suppose have an effect and lead to leniency in sentencing them. This may also give support to the idea that sentencing may vary between urban and rural with the idea of chivalry also having a different degree of effect. It also goes against Rodriguez et, al (2006) and their idea that chivalry can be seen across the entire spectrum of crime sentences and even if it does have an effect, that effect is negligible enough to not affect the results as much as it would compared to other crimes perhaps because of the non-violent nature of it. Also, due to there being no specific gender effect notable, the double deviance theory cannot be applied either as measures of traditional sex roles did not have any effect on sentencing length. Therefore, in this case the results suggest that no matter if they have broken traditional gender norms, they sentenced on equal basis with women who might not be viewed to have broken traditional gender norms.

There are some developments that could be made to this study that would allow researchers to develop a clearer and more real to life picture of this area of study. One such development would be to look at basing the study in more diverse areas or look at similar data from rural areas in different countries if they want to focus on the rural aspect of the study. This is because it has been suggested (Herzog and Oreg, 2008; Jeffries and Bond, 2013) that chivalry only really effectively applies to certain women who fit the ideal image of “womanhood” which according to Steffensmeier and Demuth (2006) is white and middle class. Therefore, because this study was done in an environment where there is a miniscule amount of individuals, compared to the sample population size, where from an ethnically diverse background, it makes it hard to be generalizable except from maybe to small counties with little ethnic diversity like the ones that have been researched in this study. Also, although it could show some partial evidence against general chivalry’s effects on sentencing for this crime, it would struggle with some other theories. It would struggle to show evidence against selective chivalry as it has very little diversity to show that certain women would get better treatment then others due to its lack of a diverse sample population. As previously referred, it would only be able to show evidence to support counties with similar populations to itself and it would struggle to provide evidence in other states and counties as the diversity varies around the country. Support for the generalization of the conclusions of these studies is very low as it would only apply to a very small group. They state themselves that the results may differ from rural counties in different states. A further limitation was the issue that for some cases they could not access their criminal background. Therefore, they were unable to take the factor of whether some may have a criminal history which would then affect the outcome. The resulting sentence could then be different because of this and would then have to be considered when looking at the results of this study. For example, the reason why in some of the cases women look like their sentence is just as long or harsh as their male counterparts is because they have a criminal history. If this was then it may be that women in this case may have only gotten similar judgements due to having a criminal history. Although, it may be unlikely that enough women had criminal histories that affected their sentence length to then affect the results and conclusion of this study, we cannot rule it out due there being no proof against this idea.

5. Other Factors

So far, the focus has been on how certain theories help explain the differences that may be seen during the sentencing phase of the criminal justice system. Various other factors have been mentioned when looking at the theories and how they may moderate and interact with gender and the theories surrounding differences in sentencing in relation to gender. However, this chapter will look more specifically at studies surrounding factors and how they may have an influence on the sentencing process. Four factors, legal and extra-legal, shall be looked at within this section, however there are many more factors that have been looked at in various research and are can be said to have some sort of effect. Each factor will look at how they interact with gender and sentencing, looking at whether they have a significant impact on how defendants are sentenced.

Racial Factor

Within this section the race of the defendant being sentenced will be looked at in reference to seeing if there are any differences to be seen between gender and race. There are substantial bodies of research that detail how race can affect how individuals may be sentenced, with white defendants often getting the more lenient sentence (Brennan & Spohn, 2008; Spohn, 2000). The focus for this however will be how race and gender can lead to differences in sentencing with a discussion on what those differences are and what they show about the sentencing system.

What is found through looking through various pieces of literature is that black women do not receive the same level of lenience from the court system as white women do (Koons-Witt et al., 2012; Brennan, 2006; Steffensmeier et al., 1993). Therefore, you can assume the leniency that is supposed to apply to women when it comes to sentencing only applies to certain women, which supports the theory of selective chivalry. This shows that there may be some king of racial stereotype in causing this and allows them to fall out of the archetype image of women that need this protection because they believe they are have a low risk to society. However, there are also studies that contend this theory that black females are treated less leniently then white females. Many of these studies still find differences in how people of different races are sentenced, however they find that the difference mostly occurs when it comes to black and white males as earlier mentioned (Steffensmeier & Demuth, 2006; Spohn & Spears, 1997). It can also be suggested that the area that is picked to base the study may affect the outcome and results of these studies. It is suggested by Zatz (2000) that studies based in the United States seem to see a pattern of where race appears to be a larger factor in the sentencing phase and these states where you see it are in the south of the country. Therefore, you may assume that the study may vary from area to area depending on how strong traditional gender roles and views on women are valued. For example, the Koons-Witt et, al (2012) study is based in South Carolina which is in the south of the country. They mention how conservative views on women are held within the state and the results showed that black women and men appear to be on the receiving end of less lenient sentences. A lot of studies also seem to only consider black and white sentencing disparities. There are studies however, that show Hispanic defendants can often face even harsher punishment then both black and white defendants (Brennan & Spohn, 2008). Therefore, although race does seem to have a definite effect in relation to men on a general basis, it is less certain in its effect when women are the defendant. A future study could look at the effects of gender and race and compare them between various states or countries in order to determine whether there is any connection between the factors.

Age Factor

The factor of age has not yet been discussed in this review though it is often included as control within most literature. Most of these studies often focus on the age of male defendants rather than across the gender divide. This maybe perhaps due to it having more of an influence over male sentencing then female leading to a lack of need to test how the factor may influence sentencing of women.

Steffensmeier et al. (1998) hypothesised this when looking at how a range of different factors interact when sentencing. They found that younger males are often sentenced more harshly than older individuals. They also found that as the defendant got older, the race and gender differences diminished. This may be because they are deemed too old to be deemed as much of a threat to the general public. This was then further built upon by Spohn and Holleran (2000) who narrowed down the age in which an individual was most likely to be punished harshly. It also must be borne in mind that they looked at which ethnic groups were mostly likely to receive harsher punishment as well as the age range in which those groups would likely receive this punishment. They identified that black and Hispanic male defendants between the ages of 21 and 29 while being unemployed were most likely to receive the harshest punishments out of the different groups. Therefore, it must be noted that age can show various patterns, when interacting with other factors, that can lead to various other avenues of research and groups that can be focused on that lead to much groups being identified for further examination.

Crime Severity Factor

How serious an offence is may seem like a more obvious indicator into how a defendant may be sentenced with the idea that the more serious the crime, the harsher the sentence. However, this may be a key factor to explain the why there is a lot more males serving out more severe sentences. According to literature women are a lot less likely to commit serious crimes and violent crimes (Belkanp, 2007; Rodriguez et al., 2006). Non-violent crimes, as previously discussed, like fraud and theft are seen as crimes that traditionally women commit. However, this may result in women who commit violent crimes being affected by the views that are brought up in the theories above.

This is possibly the most key feature that can lessen the disparity between sentencing leniency. It is shown in research that a lot of the disparity disappears when it comes to violent crimes (Rodriguez el al., 2006; Boritch, 1992). This is because under the selective chivalry theory, this would fall out of its purview as it is crime that outside of the gender role expectations (Koons-Witt et al., 2012). They, therefore, lose the leniency that they would get under that paradigm as well as being judged as doubly deviant because of that breaking of the gender norms. Warren et al. (2011) suggest that as the seriousness of the offence increases the less discretion the judge then has when giving a sentence. This would then lead to this factor outweighing other key extra-legal factors, such as gender, because the seriousness of the crime has more influence over the sentence than any possible influence these extra-legal factors could possibly have on the outcome of the sentencing. This does mean that when the crime is less severe, the extra-legal factors can seem to have some measure of effect. Rodriguez et al. (2006) can support this view as they saw differences when it came to property crime in the way in which leniency was shown in favour of women when it came to sentence them.

Criminal History Factor

Criminal history is factor that was discussed in one of the studies that was focused on earlier in this review. The results for that study suggested, in the case of narcotics cases at least, that women do tend to get more lenient sentences if they have a lack of criminal history in comparison to males with a lack of criminal history (Tillyer et al., 2015). This can be further backed by Spohn (2000) and Koons-Witt et al. (2012) in relation to chance of being incarcerated.

From the various pieces of literature that have reviewed (Tillyer et al., 2015; Spohn, 2000; Koons-Witt et al., 2012; Daly & Tonry, 1997), there is a large amount that suggest that criminal history is very useful factor to use when predicting how a defendant may be sentenced. Daly and Tonry (1997) suggest that when judges are deciding on how to sentence an individual, when they have some discretion on how sentence a defendant, they look at criminal history as means of predicting future behaviour. They suggest that they use it as a means of testing how much they respect the law. A large criminal history would suggest a lack of respect and give the impression that they have no reservations about breaking the law. If the judges have doubt that that the defendant will not commit another crime, then they will be a lot less likely to give sentence that will be lenient. They may have a lack of belief that the offender would learn from committing their crime and that it would expose the wider public to harm. All these different issues stemming from a larger criminal record would leave a law official less inclined to be lenient to what in their view may be a veteran offender.

6. Conclusion and Discussion

In terms of volume of research in the area of sentencing disparities between genders, there is large amount, as shown above. However, there are a few interesting areas of future research that could be partaken in to further understanding with more specific factors. For example, an underlying factor in the United States may be what region the sentencing is taking place in (Zatz, 2000; Myers and Talarico, 1986). As mentioned earlier certain states seem to have conservative views regarding gender roles as well as Zatz (2000) mentioning that race effects can generally be seen to be stronger. A more recent investigation into the differences that can be seen across a variety of states would be an interesting area. As a detailed comparison between states that have historically been seen either conservative or liberal in their views would be able to give an idea how much of an effect the historical views of the state they are based in, is having. More studies in this area conducted outside of the United States would also be interesting. There is a very large body of work existing in this this area in the United States and not as many in other countries. Therefore, more studies done in other parts of the world may show if cultural differences still allow the ideas of the above theories to have a basis there. As well as how much the other extra-legal and legal factors have in relation to sentencing. This is a large limitation within this review as the most popular studies in this area are mostly conducted in the United States, making it harder to generalize it on continental or worldwide basis due to lack of a large body of research in other countries. Also due to there being so many different extra-legal factors that exist and could influence the sentencing process, more studies at looking into precisely how much each factor may influence a sentencing procedure. The current studies that have been conducted use databases that only give so much information on different factors. Therefore, a future study on the subject should try to address this issue which can be done by a range of measures. One such way could be creating a database that is more qualitive in its approach and includes more extra-legal factors with more information on the offender and their offences. This can allow the analysis of more than on factors, in order to establish their influence, they may have on decision, on the same sentencing cases. More research into race and gender sentencing outcomes may also be an avenue of further research as although there is research in this area a lot of them only investigate the black and white race comparison. This approach risks missing out on how other ethnic groups experience the sentencing process and if there can be witnessed differences in how they are treated in that phase of the justice system

Based from the evidence shown in the studies above and the results given. It can be suggested that Selective Chivalry and the Evil Woman/Double Deviance theories have some relevance in showing how sentencing decisions are made. Gender does appear to be a factor when determining how an offender is sentenced. It can be shown that a large body of work, in which I reviewed, shows a pattern that supports the idea that women are shown more leniency when it comes to sentencing. Yet are judged more harshly sometimes then men when other factors contribute that could break the traditional gender roles and traditional gender stereotypes such as committing violent crime, as opposed to non-violent crime. However, they also show that other factors may need to be taken into consideration in order for it to have a significant effect. Further research into how extra-legal may affect sentencing, due to the sheer number of factors that there is, and how they may be assisting the gender disparity when it comes to sentencing outcomes. Overall, the literature reviewed does generally point to the idea that gender does play a role in sentencing outcomes, but also that other factors are also considered in order to give sentence and not just gender. Chivalry, Selective Chivalry and Evil Woman/Double Deviance theories all cannot be discounted in having elements of the ideas within them being relevant to sentencing within the United States justice system.


Belknap, J. (2007). The invisible woman: Gender, crime, and justice (3rd ed.). Belmont, CA: Thomson Wadsworth Publishing Company

Bontrager, S., Barrick, K., & Stupi, E. (2013). Gender and sentencing: A meta-analysis of contemporary research. The Journal of Gender, Race, & Justice, [online] 16, 349-635. Available at: [Accessed 3 April 2020]

Boritch, H., (1992). Gender and Criminal Court Outcomes: An Historical Analysis. Criminology, [online] 30(3), pp.293-326. Available at: [Accessed 3 April 2020].

Brennan, P. and Spohn, C., (2008). Race/Ethnicity and Sentencing Outcomes Among Drug Offenders in North Carolina. Journal of Contemporary Criminal Justice, [online] 24(4), pp.371-398. Available at: [Accessed 3 April 2020].

Curry, T., Lee, G. and Rodriguez, S., (2004). Does Victim Gender Increase Sentence Severity? Further Explorations of Gender Dynamics and Sentencing Outcomes. Crime & Delinquency, [online] 50(3), pp.319-343. Available at: [Accessed 3 April 2020].

Daly, K., (1987). Discrimination in the Criminal Courts: Family, Gender, and the Problem of Equal Treatment. Social Forces, [online] 66(1), p.152. Available at: [Accessed 3 April 2020].

Daly, K., (1989). Rethinking Judicial Paternalism: Gender, Work-Family Relations, and Sentencing. Gender & Society, [online] 3(1), pp.9-36. Available at: [Accessed 3 April 2020].

Daly, K. and Tonry, M., (1997). Gender, Race, and Sentencing. Crime and Justice, [online] 22, pp.201-252. Available at: [Accessed 3 April 2020].

Doerner, J. and Demuth, S., (2010). The Independent and Joint Effects of Race/Ethnicity, Gender, and Age on Sentencing Outcomes in U.S. Federal Courts. Justice Quarterly, [online] 27(1), pp.1-27. Available at: [Accessed 3 April 2020].

Embry, R. and Lyons, P., (2012). Sex-Based Sentencing: Sentencing Discrepancies Between Male and Female Sex Offenders. Feminist Criminology, [online] 7(2), pp.146-162. Available at: [Accessed 3 April 2020].

Farnworth, M. and Teske, Jr., R., (1995). Gender Differences in Felony Court Processing: Women & Criminal Justice, [online] 6(2), pp.23-44. Available at: [Accessed 3 April 2020].

Fernando Rodriguez, S., Curry, T. and Lee, G., (2006). Gender Differences in Criminal Sentencing: Do Effects Vary Across Violent, Property, and Drug Offenses? Social Science Quarterly 87(2), pp.318-339. Available at: [Accessed 3 April 2020].

Franklin, C. and Fearn, N., (2008). Gender, race, and formal court decision-making outcomes: Chivalry/paternalism, conflict theory or gender conflict? Journal of Criminal Justice 36(3), pp.279-290. Available at: [Accessed 3 April 2020].

Freiburger, T., (2011). The Impact of Gender, Offense Type, and Familial Role on the Decision to Incarcerate. Social Justice Research, [online] 24(2), pp.143-167. Available at: [Accessed 3 April 2020].

Gelsthorpe, L. (2013). ‘Sentencing and gender’ in Sheehan, R., McIvor, G., Trotter, C. (ed.) What Works with Women Offenders. Hoboken: Taylor and Francis.

Herzog, S. and Oreg, S., (2008). Chivalry and the Moderating Effect of Ambivalent Sexism: Individual Differences in Crime Seriousness Judgments. Law & Society Review, [online] 42(1), pp.45-74. Available at: [Accessed 3 April 2020].

Holland, M. and Prohaska, A., (2018). Gender Effects Across Place. Race and Justice, [online] p.215336871876749. Available at: [Accessed 3 April 2020].

Jeffries, S. and Bond, C., (2013). Gender, Indigeneity, and the Criminal Courts: A Narrative Exploration of Women’s Sentencing in Western Australia. Women & Criminal Justice 23(1), pp.19-42. Available at: [Accessed 3 April 2020].

JEFFRIES, S., FLETCHER, G. and NEWBOLD, G., (2003). PATHWAYS TO SEX-BASED DIFFERENTIATION IN CRIMINAL COURT SENTENCING. Criminology, [online] 41(2), pp.329-354. Available at: [Accessed 3 April 2020].

Koons-Witt, B., Sevigny, E., Burrow, J. and Hester, R., (2012). Gender and Sentencing Outcomes in South Carolina. Criminal Justice Policy Review, [online] 25(3), pp.299-324. Available at: [Accessed 3 April 2020].

Messerschmidt, J., (2007). Masculinities, Crime and. The Blackwell Encyclopaedia of Sociology, Available at: [Accessed 3 April 2020].

Murphy, E. and Brown, J., (2000). Exploring gender role identity, value orientation of occupation and sex of respondent in influencing attitudes towards male and female offenders. Legal and Criminological Psychology, [online] 5(2), pp.285-290. Available at: [Accessed 3 April 2020].

Myers, M. and Talarico, S., (1986). The Social Contexts of Racial Discrimination in Sentencing. Social Problems 33(3), pp.236-251. Available at: [Accessed 3 April 2020].

Pollock, O. (1950). The criminality of women. Philadelphia: University of Pennsylvania Press

Spivak, A., Wagner, B., Whitmer, J. and Charish, C., (2014). Gender and Status Offending. Feminist Criminology, [online] 9(3), pp.224-248. Available at: [Accessed 3 April 2020].

Spohn, C. (2000). Thirty years of sentencing reform: The quest for a racially-neutral sentencing process. In J. Horney (Ed.), Policies, processes, and decisions of the criminal justice system: Vol. 3 (pp. 427-501). Washington, DC: National Institute of Justice.

SPOHN, C. and HOLLERAN, D., (2000). THE IMPRISONMENT PENALTY PAID BY YOUNG, UNEMPLOYED BLACK AND HISPANIC MALE OFFENDERS. Criminology, [online] 38(1), pp.281-306. Available at: [Accessed 3 April 2020].

Spohn, C. and Spears, J., (1997). Gender and Case Processing Decisions. Women & Criminal Justice 8(3), pp.29-59. Available at: [Accessed 3 April 2020].

Steffensmeier, D. and Demuth, S., (2006). Does Gender Modify the Effects of Race–ethnicity on Criminal Sanctioning? Sentences for Male and Female White, Black, and Hispanic Defendants. Journal of Quantitative Criminology, [online] 22(3), pp.241-261. Available at: [Accessed 3 April 2020].

STEFFENSMEIER, D., KRAMER, J. and STREIFEL, C., (1993). GENDER AND IMPRISONMENT DECISIONS. Criminology, [online] 31(3), pp.411-446. Available at: [Accessed 3 April 2020].

STEFFENSMEIER, D., ULMER, J. and KRAMER, J., (1998). THE INTERACTION OF RACE, GENDER, AND AGE IN CRIMINAL SENTENCING: THE PUNISHMENT COST OF BEING YOUNG, BLACK, AND MALE. Criminology, [online] 36(4), pp.763-798. Available at: [Accessed 3 April 2020].

Tasca, M., Cho, A., Spohn, C. and Rodriguez, N., (2018). The Role of Parental Status and Involvement in Sentence Length Decisions: A Comparison of Men and Women Sentenced to Prison. Crime & Delinquency, [online] 65(14), pp.1899-1924. Available at: [Accessed 3 April 2020].

Tillyer, R., Hartley, R. and Ward, J., (2015). Differential Treatment of Female Defendants. Criminal Justice and Behaviour, [online] 42(7), pp.703-721. Available at: [Accessed 3 April 2020].

Tracy, P., Kempf-Leonard, K. and Abramoske-James, S., (2009). Gender Differences in Delinquency and Juvenile Justice Processing. Crime & Delinquency, [online] 55(2), pp.171-215. Available at: [Accessed 3 April 2020].

Warren, P., Chiricos, T. and Bales, W., (2011). The Imprisonment Penalty for Young Black and Hispanic Males.Journal of Research in Crime and Delinquency, [online] 49(1), pp.56-80. Available at: [Accessed 3 April 2020].

Zatz, M. S. (2000). The convergence of race, ethnicity, gender, and class on court decision-making: Looking toward the 21st century. In J. Horney (Ed.), Policies, processes, and decisions of the criminal justice system: Vol. 3 (pp. 503-552). Washington, DC: National Institute of Justice.

Leave a Reply

Your email address will not be published. Required fields are marked *