Evaluating Racial and Ethnic Invariance Among the Correlates of Guilty Pleas: A Focus on the Effect of Court Legitimacy,Attorney Type,Satisfaction,and Plea-Offer Evaluation

Abstract

Objectives

Research has identified racial and ethnic disparities in rates of guilty pleas relative to trial where minorities are more likely to proceed to trial, though little research has explored the source of this disparity.

Methods

vUsing an adult nationwide sample and a vignette methodology, this research uses Ordinary Least Squares regression to explore differences in White, Black, and Hispanic defendants’ willingness to accept a guilty plea (WTAP).

Results

Though there are not significant direct effects of race or ethnicity on initial WTAP, there are racial/ethnic differences in theoretical antecedents of WTAP such as perceived probability of conviction, court legitimacy, and attorney type. However, there are not differences in the effects of theoretical antecedents on WTAP across race or ethnicity. Significant differences by race/ethnicity also emerge following a defense attorney's evaluation of an offer and are conditional on guilt.

Conclusions

Racial/ethnic differences in rates of plea acceptance are likely due to cumulative racial/ethnic differences in antecedents of WTAP, differences in the effect of attorney evaluation by race/ethnicity, and/or factors that were not directly examined in this study such as variation in plea offers. Implications for future research on the nexus between plea bargaining and race/ethnicity are discussed.

Keywords

plea bargaining race court legitimacy defense attorney satisfaction

In the early stages of the American criminal justice system, jury trials were a relatively common method of resolving criminal cases (Friedman, 1979; Johnson, 2019). Much has changed, however, as 90–95% of criminal convictions are currently resolved via guilty pleas, which are frequently the result of negotiations between a defendant, their attorney, the prosecutor, and other courtroom actors (Devers, 2011; Reaves, 2013). This modern reliance on pleas has drawn increased attention to factors influencing defendants’ willingness to accept a plea (WTAP), with much extant research focusing on case characteristics such as guilt, probability of conviction, and potential sentence severity (e.g., Bordens, 1984; Dervan & Edkins, 2013; Redlich & Shteynberg, 2016). Recent work has also explored the role of defense attorneys and framing (Garnier-Dykstra & Wilson, 2019; Henderson & Levett, 2018; Lee, Jaynes, & Ropp, 2021). Though defendant race and ethnicity have not yet been a focal point of this body of work, sentencing research underscores differences in final outcomes by race and ethnicity; Black and Latino defendants are often more likely to go to trial than White defendants, suggesting a potential underlying difference in WTAP (e.g., Lee & Richardson, 2020; Testa & Johnson, 2020). Although individuals always have the right to reject a plea offer, and this decision ultimately rests with the defendant (Henderson & Levett, 2019), racial and ethnic disparities in plea acceptances are particularly noteworthy because of the well-documented “trial penalty,” which demonstrates that those who elect to reject a plea offer and are found guilty at trial, on average, receive harsher sentences (Bushway & Redlich, 2012; LaFree, 1985).

There is, however, little research aimed at understanding the causal mechanism behind the racial and ethnic disparity in plea rates. Though the disparity may be the result of differences in case characteristics across racial and ethnic groups, an alternate explanation is that some defendants simply are less willing to accept a plea offer from a prosecutor because of their views of the court system and court actors. Plea bargaining is a less than transparent process (Bibas, 2016) that involves placing trust in the decision-making of legal authorities (Testa & Johnson, 2020). One could hypothesize, then, that a lack of trust in legal authorities may influence one's WTAP. Specifically, lower levels of court legitimacy among racial and ethnic minorities may lead to plea avoidance because it would require placing their trust in the decision-making of legal authorities. Relatedly, those who do not perceive their defense counsel, another key player in the courtroom workgroup (Eisenstein & Jacob, 1977), as competent and zealous may be less willing to rely on them (Clair, 2021). In fact, recent research suggests those who perceive the court system as more legitimate and their defense attorney as more satisfactory are more likely to accept a plea (Lee et al., 2021). This is especially plausible given that minorities have lower perceptions of the criminal justice system's legitimacy (Higgins, Wolfe, Mahoney, & Walters, 2009; Madon, Murphy, & Sargeant, 2017; Sun & Wu, 2006; Wu & Cao, 2018) and are more likely to be represented by some type of public counsel – public defenders or assigned attorneys (Cohen, 2014; Harlow, 2000).

There is an additional related possibility that if racial and ethnic minorities perceive court actors to be less legitimate and competent, they may be less influenced by the advice of their counsel. Bolstering findings of Henderson and Levett (2018), Lee et al. (2021) recently highlighted how defendant decision-making is influenced by attorney evaluations of a plea offer – a finding which was moderated by both the defendants’ perceived satisfaction with their representation and perceptions of courtroom legitimacy. This suggests that while there may be initial differences in WTAP by race and ethnicity, there may also be differences in the influence of a counsel evaluation on WTAP based on the defendant's race or ethnicity which could affect final outcomes. However, Lee et al. (2021) relied on a relatively homogenous sample of White university students and did not test this assertion, leaving an important gap in the literature.

The current study offers insight into the potential factors contributing to minority disparities in sentencing. We focus on the perspective of defendants, who are often erroneously seen as “passive consumers of legal sanctions” (Clair, 2021, p. 2). To this end, we use a nationwide (United States, U.S.) sample of adults and an experimental vignette methodology to evaluate the factors that influence WTAP. Specifically, we explore racial and ethnic differences in WTAP and the effect of factors theoretically associated with WTAP, focusing on court legitimacy, defense attorney satisfaction, counsel type, and defense attorney evaluations of a plea offer. Below, we provide an overview of extant research surrounding racial and ethnic disparities in plea bargaining. We continue with a description of the survey and vignette instrument before describing our findings in detail.

In sum, our results indicate that contrary to expectations, race and ethnicity are not direct predictors of initial WTAP. Race and ethnicity also do not moderate the effect of other theoretical correlates of initial WTAP. Instead, established factors such as guilt as well as racialized perceptual correlates such as perceptions of punishment and court legitimacy predict initial WTAP. Significant 3-way interactions also emerged between attorney evaluation, race/ethnicity, and guilt – indicating that an attorney's influence on their client following an evaluation is based on characteristics of the case (e.g., guilt) as well as the defendant (e.g., race and ethnicity). We end with a discussion of our study's limitations, theoretical implications, and directions for future research.

Racial and Ethnic Disparities in Court Processes and Outcomes

A key and pervasive finding in courts and sentencing literature is the existence of racial disparities (Mitchell, 2005; Wu, 2016), which extends across multiple stages of the criminal court system, from pretrial detention to length of incarceration, and is often replicated amongst ethnic minority populations (see e.g., Kurlychek & Johnson, 2019; Kutateladze, Andiloro, Johnson, & Spohn, 2014; Omori & Petersen, 2020). Prior research suggests that racial and ethnic minorities, such as Black and Hispanic populations, are more likely to proceed to trial (rather than resolve their case with a guilty plea) than their White counterparts, though there is generally less research on plea bargaining and race/ethnicity, relative to studies of race/ethnicity and sanction severity (Johnson & Richardson, 2019). In a study of state sentencing data over forty large urban U.S. counties, Lee and Richardson (2020) found that Black defendants were more likely to go to trial than White defendants, though ethnicity effects were less consistent. In addition, a recent analysis of Maryland sentencing data found Black and Latino defendants were 40% and 50%, respectively, less likely to plead guilty and go to trial relative to White defendants (Testa & Johnson, 2020). This finding has also been observed in Florida state (Metcalfe & Chiricos, 2018) and federal sentencing data (Everett & Wojtkiewicz, 2002).

One critical limitation of this body of work, however, is that administrative datasets (typically collected for official use by agencies) rarely include information on prior offers – or plea bargaining in general – (Johnson & Richardson, 2019), leaving an open question of whether individuals are receiving disparate offers for similar cases. In addition, there is some evidence that racial and ethnic minorities may be receiving harsher offers from prosecutors to begin with, even after controlling for relevant characteristics such as offense severity and criminal history (Kutateladze et al., 2016). This paper seeks to improve on this limitation by employing vignettes which hold many case characteristics, including charged offense and plea offer, constant.

Differences in Legitimacy Across Race and Ethnicity

In addition to the oft-observed differences in case outcomes based on race and ethnicity, another relevant area of research focuses on racial differences in perceptions of the criminal justice system. We posit that one potential explanation for the minority disparities in rates of guilty pleas is variation in perceptions of court legitimacy. Legitimacy in this context refers to the perceived obligation to obey and trust in institutional authority (Baker et al. 2015; Tyler & Huo, 2002; Tyler & Jackson, 2014), and criminological research has shown that high levels of legitimacy cultivate voluntary compliance and cooperation (Bolger & Walters, 2019; Mazerolle et al., 2013a, 2013b; Tyler, 2006). Institutions are thought to establish legitimacy by upholding social norms through impartial, ethical, and consistent processing; often referred to as procedural fairness (Sunshine & Tyler, 2003). Racial and ethnic minorities often perceive legal authorities to be less legitimate than Whites, a finding which is believed to be caused by differences in personal and vicarious experiences with the criminal justice system (Higgins et al., 2009; Madon et al., 2017; Sun & Wu, 2006; Woolard, Harvell, & Graham, 2008; Wu & Cao, 2018).

Racial and ethnic minorities indeed face harsher outcomes throughout the criminal justice system. For instance, research suggests Black individuals are significantly more likely to be stopped by the police (Baumgartner et al., 2018; Crutchfield et al., 2012), searched (Baumgartner et al., 2017; Rojek et al., 2012), and arrested (Baumgartner et al., 2017; Mitchell & Caudy, 2015). Within courts and sentencing, racial and ethnic minorities are also widely recognized as receiving harsher outcomes including a higher likelihood of pretrial detention (Omori & Petersen, 2020), less favorable plea offers (Kutateladze et al., 2016), and more severe final sentences (King & Light, 2019). This can further result in cumulative disadvantage (Kutateladze et al., 2014; Kurlychek & Johnson, 2019), an awareness of which would contribute to lower levels of legitimacy and higher levels of distrust amongst racial and ethnic minorities.

Although legitimacy research has primarily focused on policing, legitimacy may also be associated with compliance and cooperation in the court system (Hamm, 2020; Tyler & Huo, 2002; Tyler & Jackson, 2014). Because most criminal convictions are the result of a guilty plea, plea bargaining is arguably a form of compliance with court authorities (Lee et al., 2021). Defendants who are distrustful of legal authorities may be more cynical of plea offers from the state/prosecutor and may believe a trial by their peers will offer a more procedurally just process (Testa & Johnson, 2020). However, the “trial tax” (sometimes referred to as a “trial penalty” or a “plea discount”) is one of the most common empirical observations in sentencing literature (e.g., Johnson, 2019; Johnson et al., 2016; Ulmer et al., 2010; Yan & Bushway, 2018). Thus, one possibility is that lower levels of perceived legitimacy may lead defendants to opt for trials which can result in harsher sentencing if found guilty.¹ In support of these underlying assumptions, Lee and colleagues (2021) found higher perceptions of court legitimacy and defense attorney satisfaction were associated with an increased WTAP among a large sample of young adults from a university in the Northwest U.S. However, minorities were under-represented in the sample, raising questions as to the generalizability of these findings to a more diverse population. This is especially important given that racial minorities are well over-represented in the criminal justice system (Denbeaux, Kearns, Ricciardelli, 2016; Horowitz & Utada, 2018; Kovera, 2019; Seguino & Brooks, 2018) and there are differences in sanction severity perceptions across racial groups (Irizarry, May, Davis, & Wood, 2016; May & Wood, 2005).

Defendants and Their Attorneys

Though the defense attorney is a key player in the courtroom workgroup (Eisenstein & Jacob, 1977) and the main (often sole) source of information for a defendant (Henderson & Levett, 2019; Lee & Ropp, 2020), research exploring the influence of defense attorneys on their clients has only recently gained traction again. In one early study, Atkins and Boyle (1976) found that prisoners who received the shortest sentences reported the highest levels of satisfaction with their counsel. Similarly, recent work has found that clients who viewed their sentences favorably were more pleased with their attorneys (Boccaccini & Brodsky, 2002). There is also growing evidence that certain attorney skills (O’Brien et al., 1977) and traits, including the client's perception that they are permitted to be more involved in decision making (Bocaccini, Boothby, & Brodsky, 2004) influences trust and satisfaction.

Moreover, there are indications that levels of trust and/or satisfaction may moderate the defense attorney's influence on their clients. Recently, Henderson & Shteynberg (2020) used an online MTurk sample to assess how attorney characteristics in a DUI case affected the attorney's influence on their clients. Attorneys’ trustworthiness and expertise were randomized based on attorney behaviors and characteristics such as maintaining contact, following through, honesty, knowledge of the law, and experience defending DUI cases. Level of trust in their attorney significantly affected how likely respondents were to listen to the attorney's advice and the attorney's expertise also affected how confident respondents felt in their final decision. Along those lines, Clair's (2021) recent qualitative study found that many clients of public defenders did not trust their court-appointed counsel and instead turned to other sources of information in making decisions. Clair (2021) concluded that defendants are more active in the process than perhaps previously assumed, noting that the disadvantaged defendants in the study often questioned the authority of courtroom actors, including their own counsel. Thus, it is critical to explore the role that attorneys serve for their clients and whether their impact is affected by clients’ perceptions of them and the courts overall. Moreover, given the association between race, ethnicity, and disadvantage in this country (Parker, Stults, & Rice, 2005; Rodriguez, 2013; Wilson, 1987), it is imperative to assess these relationships across race and ethnicity.

There are also differences in attorney representation across racial and ethnic groups. In general, minority defendants are more likely to have a public defender or court appointed defense attorney (Cohen, 2014; Harlow, 2000). This is important as some evidence suggests that defendants represented by a public defender plead guilty more often than those with private counsel (Atkins & Boyle, 1976; Sudnow, 1965) and qualitative research supports this idea, based on the pressures of high caseloads on public defenders (e.g., Hussmann & Siegel, 2021). Research has also found differences in the quality of the plea “bargain” that an attorney believes they can obtain for their clients based on the defendant's race, which raises equality concerns (Edkins, 2011).

Several recent vignette-based studies have demonstrated that attorney recommendations (Henderson & Levett, 2018) or evaluations (Lee et al., 2021) post-offer can significantly influence a defendant's decision-making and that this can vary based factors such as guilt, court legitimacy perceptions, and satisfaction with one's attorney. In addition, Shook and colleagues (2021) found that juveniles who felt that their defense attorneys treated them more favorably were also more likely to possess more positive views on the legitimacy of police and courts. This study provides further support for the idea that legitimacy perceptions are affected by relationships and experiences with courtroom and other CJS actors.

An additional critical aspect to this discussion is whether an attorney's influence on defendants’ decisions varies by defendant race and ethnicity. Thus, a key extension of the extant literature is an exploration of whether White defendants differ from racial and ethnic minority defendants in how much credence they provide their attorney and thus how much of an effect their counsel can have on their plea choices.

The Current Study

The current study seeks to advance knowledge on plea bargaining within the criminal court system by evaluating factors which influence defendant decision-making. Given recognized racial and ethnic disparities in the criminal justice system (Mitchell, 2005; King and Light, 2019), we focus on potential racial and ethnic differences in WTAP. We first seek to explore whether there are initial differences in WTAP by race and ethnicity, while holding all other case characteristics constant (with the exception of guilt). While there may be explanations for potential racial and ethnic differences in going to trial vs. accepting a plea offer (Lee & Richardson, 2020; Testa & Johnson, 2020), we focus on whether legitimacy is a significant predictor of guilty plea decisions capable of explaining observed racial/ethnic differences in WTAP. Based on a significant body of research indicating that racial and ethnic minorities are treated more harshly (Mitchell, 2005; Kutateladze et al., 2014, 2016; King & Light, 2019) and that racial and ethnic minorities typically have lower perceptions of legitimacy relative to Whites (Higgins et al., 2009; Sun & Wu, 2006), we also assess whether there are moderating effects of race and ethnicity on the influence of legitimacy. Second, we focus on defense attorneys. As satisfaction with one's defense attorney may vary by attorney type (Clair, 2021), which is associated with race and ethnicity (Cohen, 2014; Harlow, 2000), we explore similar questions with regards to potential confounding and moderation regarding defense attorney type (public or private) and perceived satisfaction with one's defense counsel. We complete our analyses by exploring whether the influence of an attorney evaluation of a plea offer varies by race and ethnicity, as the arguments regarding legitimacy and trust in court authorities may also extend to whether a person trusts the perspective of their defense counsel. Specifically, we seek to answer the following key research questions:

Are there differences in initial WTAP by race and ethnicity?

If so, do factors such as perceived defense counsel type, defense attorney satisfaction, and court legitimacy explain these differences?

Are the effects of theoretical antecedents of WTAP moderated by race and ethnicity?

Are there differences in the effect of an attorney evaluation of a plea offer on WTAP by race/ethnicity, holding all else constant?

This study serves as a replication and extension of prior research (Lee et al., 2021) which was limited by the homogeneity of its sample to study the guilty-plea process across minority groups.

Data

This study draws on data from a large nationwide sample of U.S. adults who were surveyed about their court perceptions, prior experiences with the criminal justice system, and their WTAP in specific circumstances. Data were collected using Qualtrics’ online surveying platform.² Qualtrics has a diverse pool of users recruited to participate in research through various methods such as website intercept recruitment, permission-based networks, social media, and targeted email lists (Qualtrics, nd). This platform is increasingly utilized in social science research (Bushman et al. 2012; Fox et al., 2021; Moule et al., 2019; Wright & Skagerberg, 2012) and is advantageous because it offers a cost-effective means to access samples which approximate the national population. Indeed, research suggests that Qualtrics’ respondents closely resemble U.S. Census demographic characteristics (Heen et al., 2014). Within this study, participants were selected from Qualtrics’ pool using a stratified quota sampling procedure to approximate the 2010 U.S. Census in terms of sex, household income, and age. Because of the nature of our research questions, and the overrepresentation of minorities within the criminal justice system, we elected to oversample Black and Hispanic individuals.

Table 1 provides sample demographic information. In total, 659 individuals completed the survey. We excluded those who identified as Asian (8%) or “other” (2%), resulting in a final analytic sample of 593 respondents, including White (46%), Black (29%), or Hispanic (24%) individuals.³ About half the sample is male (51%), with an average age of about 45 years-old (range 18–83). The average household income is between $50,000 to $69,999 a year, and about half (52%) of respondents believed they could afford to retain a private attorney if charged with a crime. About 13% had been arrested or convicted of a crime, while 35% had a close friend or family member who was. Approximately 8% of the sample had a prior experience with plea bargaining.

Table 1.

Descriptive Statistics for the Aggregate Sample (N = 593).

Variable	Mean	Standard Deviation	Minimum	Maximum
WTAP_i	34.34	34.61	0	100
WTAP_e	34.62	33.69	0	100
Guilt	0.47	–	0	1
Attorney Recommendation
Good	0.34	–	0	1
Average	0.34	–	0	1
Bad	0.32	–	0	1
Probability of Conviction	53.00	29.22	0	100
Severity Discount	20.17	23.71	0	90
Defense Attorney Satisfaction (DAS)	0.02	0.97	−3.00	1.46
Court Legitimacy (CL)	0.02	0.97	−2.62	1.96
Private Attorney	0.52	–	0	1
White	0.46	–
Black	0.29	–	0	1
Hispanic	0.24	–	0	1
Age	44.55	16.98	18	83
Male	0.51	–	0	1
Household Income	6.64	4.68	1	16
Risk Aversion	4.45	0.58	1	5
Prior Arrest	0.13	–	0	1
Family Prior Arrest	0.35	–	0	1
Previous Plea	0.08	–	0	1

Methodological Design

The study employed an experimental survey in which factual guilt (guilt vs. innocence) and attorney evaluation of the plea offer (“good” vs. “average” vs. “bad”) were manipulated in a 2 × 3 fully-crossed factorial design. After demographics, respondents were asked to indicate their level of court legitimacy, if they could afford to hire a private attorney, expected satisfaction with their attorney, and their prior criminal justice system-related experiences. Individual perceptions of court legitimacy and their defense attorney were evaluated prior to the vignette to ensure these variables were not conflated with vignette-related factors such as the terms of the plea offer. Although the scenario we presented was hypothetical, it is assumed that most people have a strong perception about the quality/type of attorney they would be able to retain.

The vignette section was modeled after prior research (Bordens, 1984; Lee et al., 2021; Tor et al., 2010) and presents a vehicular manslaughter scenario in three sections. An advantage of this scenario is that even largely law-abiding individuals could envision themselves facing this circumstance. First, the following scenario (a) was presented:

Suppose you were driving around town one night. You had not had any alcohol or taken any illegal substances that day. After driving around a particularly sharp corner, you hit a person who died the next day from their injuries. Because of this incident, you now face criminal charges of vehicular manslaughter. You know that you [did/did not] exceed the speed limit while going around the corner, but there were no cameras near the incident.

The outcome of the trial, if you choose to go to trial, depends on whether the prosecutor can prove that you were speeding at the time of the incident. If they succeed in proving it and you are found guilty, you are eligible for the maximum penalty; if they fail, you go free. In this state, vehicular manslaughter carries a maximum of a 10-year sentence in prison.

Within vignette (a), guilt and innocence (i.e., speeding or not speeding) were randomly assigned. Respondents were then asked to rate their perceived probability of conviction and perceived severity of the maximum penalty (10 years in prison). Within the next section of the vignette (b), respondents were told:

Now, just before trial, the prosecutor offers you a plea bargain: a sentence of 4 years in prison if you plead guilty to vehicular manslaughter.

All respondents viewed the same offer. Respondents were then asked to indicate the likelihood they would accept the plea offer (4 years) on a scale ranging from 0% to 100%. Respondents were also asked to rate the severity of the plea offer using the same severity scale as the initial rating of the 10-year maximum sentence.

Finally, the vignette concluded with the following section (c):

Now, imagine that you asked your defense attorney what they thought of the plea offer. Your defense attorney responded that based on their experience in criminal defense, they believe this offer is a [good, average, or bad] one.

The defense attorney's evaluation of the plea offer (good, average, or bad) was randomly assigned, and respondents subsequently indicated their willingness to accept the 4-year plea offer (0–100%) for a second time. Following the vignette, additional questions were asked to gauge between-individual differences in factors like risk aversion that could potentially influence the constructs of interest (all variables are explained further below).

To ensure response quality, Qualtrics excluded all responses that were below one-half of the median soft launch response time. We also included 6 attention checks throughout the survey and excluded respondents who failed to accurately answer any of them. One of these attention checks followed the vignette, in which participants were asked to indicate whether they exceeded the speed limit in the scenario and the correct answer was dependent on their vignette condition. Those in the guilty scenario were directly told they “did exceed the speed limit” (emphasis added) whereas those in the innocent condition were directly told they “did not exceed the speed limit.” The correct response to this attention check not only ensures participants read the vignette, but also that they understood their guilt/innocence, as the vignette explained to participants that if they were speeding, they were guilty.

Though there have been questions as to whether vignettes eliciting future intentions (e.g., willingness to offend, intentions to exercise, or willingness to plea as implemented in this study) are strong proxies for actual behavior (see Exum & Bouffard, 2010), there is evidence of a moderately strong association in support of the methodology (Beck & Ajzen, 1991; Sheppard et al., 1988; Steiner et al., 2016). Construct validity for the specific vignette here is challenging, given that plea offers, and plea acceptance rates are not well documented within administrative data (Johnson et al., 2016; Johnson & Richardson, 2019). However, we believe our vignette includes a great deal of contextualization (e.g., guilt, facts and circumstances of the incident, relevant laws, and victim behavior) which is frequently assumed to increase construct validity (Steiner et al., 2016). Furthermore, a large majority of our sample (over 85%) reported driving a vehicle in the last year, suggesting an increased likelihood that respondents could imagine themselves in this scenario. Our findings with regards to the initial WTAP distribution and its observed association with known correlates such as guilt, probability of conviction, and perceived offer severity are also consistent with prior research using similar scenarios amongst different samples (Bordens, 1984; Lee et al., 2021; Tor et al., 2010), providing further evidence in support of our study's validity. In addition, given that our survey responses were drawn from a large nationwide anonymous online sample, we believe it is unlikely that our findings are strongly impacted by social desirability bias. Yet is important to acknowledge that our vignette methodology relies on a hypothetical scenario with stated intentions rather than actual behavior, and thus our results should be interpreted with some caution. Our study's methodology is particularly valuable, however, as our design allows us to hold constant case and plea-offer characteristics, which a great deal of prior research has not been able to fully account for. This is important, as in practice there may be baseline differences in these factors across racial/ethnic groups which could confound findings.

Dependent Variable

Willingness to Accept a Plea (WTAP)

After presenting the plea offer (4 years, vignette part b), respondents were asked to report the likelihood that they would accept this plea, on a scale from 0 (no chance) to 100 (100% certainty). This is our first dependent variable: one's initial willingness to accept a plea (WTAP_i). The average WTAP_i is 34% (see Table 1), suggesting that individuals were relatively unlikely to accept the plea offer. After the attorney evaluation of the plea offer (vignette part c), respondents were asked to again evaluate the likelihood that they would accept the offer, on the same scale. This is our second dependent variable, referred to as WTAP_e as it indicates WTAP following an attorney evaluation of the plea offer. The average of this variable was approximately 35% (see Table 1).

Independent Variables

Defense Attorney Satisfaction (DAS)

Before the vignette, respondents were asked to indicate how satisfied (on a scale from “very dissatisfied = 1” to “very satisfied = 5”) they expected they would be with several aspects of representation from the hypothetical defense attorney they believed they would retain. These items included satisfaction with their attorney overall and the amount of time their attorney would be able to spend focusing on their case (a list of the scale items is available in Appendix A). A 5-item defense attorney satisfaction (DAS) scale was then created through factor analysis, with items demonstrating strong internal reliability (Cronbach's alpha = .93). Higher values indicate increased perceptions that one would be satisfied with their defense attorney.

Court Legitimacy (CL)

To establish court legitimacy, respondents were asked questions regarding their perception of the court which were adapted by Lee et al. (2021) from measures developed within the policing literature (Reisig, Bratton, & Getz, 2007; Sunshine & Tyler, 2003; Tankebe, 2013). Respondents were asked the degree to which they agreed or disagreed with statements such as, “the courts use rules and procedures that are fair to everyone,” on a scale from “strongly disagree” ( = 1) to “strongly agree” ( = 5). They were also asked to assess how effective they believed the court system functions on a scale from “very poorly” ( = 1) to “very well” ( = 5) regarding 7 court roles such as convicting guilty individuals. A 14-item scale was created through factor analysis with items demonstrating high internal reliability (Cronbach's alpha = .88).⁴ Higher values indicate the individual viewed the courts as more legitimate.

Private Attorney

To provide additional context regarding an individual's expectations regarding their defense counsel, respondents were asked to indicate if they were charged with a crime, what type of lawyer, a private attorney ( = 1) or a court appointed attorney ( = 0) they believed they or their family would be able to hire to represent them. About half of this sample (52%) believed they had the resources to hire a private attorney.

Additional Measures

Probability of Conviction

Perceived probability of conviction was established by asking respondents to evaluate, on a scale from “no chance” ( = 0) to “100% certain” ( = 100), the likelihood that they believed they would be convicted if they went to trial. Overall, respondents felt there was about a 53% chance of being convicted if they went to trial.

Severity Discount

Respondents also rated how severe they felt the maximum penalty (10-year prison sentence) was on a scale from 1–10 (1 = least severe, 10 = most severe). This represents the perceived severity of the maximum penalty (severity_max). Post plea offer, respondents were asked once again to rate the severity of the plea offer (4-year prison sentence), on the same scale. This represents the perceived severity of the plea offer (severity_plea). To establish plea severity discount, the percentage change in perceived severity was calculated ((severity_plea-severity_max)/severity_max). Positive values were top coded at 0 for ease of interpretation and the absolute value was taken to create the perceived discount in severity associated with the plea offer. Higher values indicate that respondents felt the plea provided a larger discount in severity (a less severe punishment).

Demographics and Additional Control Variables

Because risk aversion has been identified as an individual difference that influences plea-bargaining (Bibas, 2016; Garnier-Dykstra & Wilson, 2019; Lee et al., 2021; Tor et al., 2010), we include a composite measure based on two metrics: portions of a previously validated risk instrument by Weber, Blais, and Betz (2002) and four questions from Gardner and Steinberg (2005).⁵ Factor analysis indicated that all responses loaded onto a single factor and items demonstrated high internal validity (Cronbach's alpha = .78). The risk aversion score is a summated average index of all responses; higher values indicate increased risk aversion. We also include an array of background characteristics, including age (in years), sex (male = 1; female = 0), race/ethnicity (Black = 1, Hispanic = 2, or White = 3; mutually exclusive categories), whether the respondent had a previous arrest ( = 1; no = 0), whether a family member or close friend had a previous arrest (family arrest = 1, no = 0) and whether the respondent had any previous experiences being offered a plea in a criminal case (previous plea = 1; no = 0).

Analytic Plan

As previously described, we utilized a 2 × 3 (innocent/guilty x good/average/bad attorney evaluation) fully-crossed factorial design, with 6 randomized conditions.⁶ There is evidence that randomization was successful, given that no significant differences (p < .05) are observed across treatment conditions with respect to age, sex, income, risk preference, previous arrest, family arrest, or previous plea experience. We analyzed our continuous dependent variables using Ordinary Least Squares (OLS) regression. We evaluated the data for outliers and identified 8 individuals with the potential to bias results. Sensitivity analysis indicated that our substantive results were not sensitive to the removal of these individuals, so we elected to retain them in the sample. We use robust standard errors throughout analysis because White's test provided evidence of heteroskedasticity (χ² = 162.20, p < .05). We found no evidence of multicollinearity given that the correlations between our variables were typically low-to-moderate (ranging from 0 to + /- 0.40),⁷ and variance inflation factors (VIFs) did not exceed 1.63 (where a VIF of 10 is often considered a threshold for deeper investigation).

Results

Initial Willingness to Accept a Plea (WTAP_i)

Within Table 2, we conduct F-tests and chi-squared tests to evaluate whether there are differences in WTAP_i and its theoretical predictors by race and ethnicity (RQ1). Table 2 further describes which racial/ethnic group comparisons are significant in the last column. Contrary to expectations, findings suggest that Black, Hispanic, and White individuals are not statistically distinguishable from one another in WTAP_i (F = 0.81, p = .45). A Kruskal-Wallis H test, a non-parametric analog to the ANOVA F-test, also detects no differences in the underlying distributions of WTAP_i by race and ethnicity (χ² = 1.23, p = .54).

Table 2.

Descriptive Statistics: by Race & Ethnicity (N = 593).

	White		Black		Hispanic
Variable	Mean	SD	Mean	SD	Mean	SD	F or χ²	Significant Differences
WTAP_i	32.42	33.68	35.61	35.66	36.45	35.13	F = 0.81
WTAP_e	32.33	32.25	36.53	35.62	36.70	33.97	F = 1.19
Guilt	0.50	–	0.45	–	0.43	–	χ² = 2.58¹
Attorney Recommendation
Good	0.33	–	0.36	–	0.33	^–	χ² = 0.61¹
Average	0.35	–	0.35	–	0.34	–	–
Bad	0.32	–	0.29	–	0.33	–	–
Probability of Conviction	49.60	27.94	58.62	31.28	52.75	28.21	F = 5.14**	B/W
Severity Discount	21.33	23.04	16.86	23.17	21.91	25.34	F = 2.42
Defense Attorney Satisfaction	0.07	0.93	-0.03	1.01	0.01	1.00	F = 0.54
Court Legitimacy (CL)	0.22	0.97	-0.32	0.97	0.03	0.88	F = 15.84***	All
Private Attorney	0.60	–	0.46	–	0.46	–	χ² = 11.74**	B/W; H/W
Age	49.36	16.39	41.17	17.00	39.46	15.62	χ² = 22.53***	B/W; H/W
Male	0.73	0.44	0.31	0.46	0.32	0.47	χ² = 103.90***	B/W; H/W
Household Income	7.47	5.11	5.39	3.94	6.57	4.31	χ² = 54.02**	All (H/W^†)
Risk Aversion	4.39	0.55	4.46	0.68	4.53	0.48	F = 2.76	All
Prior Arrest	0.16	0.37	0.10	0.31	0.09	0.29	χ² = 5.36	(B/W^†); H/W
Family Prior Arrest	0.27	0.44	0.46	0.50	0.38	0.49	χ² = 17.09***	B/W; H/W
Previous Plea	0.12	0.33	0.03	0.18	0.05	0.22	χ² = 14.04***	B/W; H/W
n		275		173		145

Notes: *p < .05; **p < .01; ***p < .001; SD: Standard Deviation; (1) This is consistent with expectations given that these conditions were randomized.

There are, however, some differences across racial/ethnic groups regarding the theoretical antecedents of WTAP_i. For instance, on average, Black individuals have a higher perceived probability of conviction than Whites (p < .01). Black individuals also perceive a smaller severity discount than White and Hispanic individuals, though the difference is not statistically significant. While there are no significant differences across race/ethnicity in perceived DAS, on average, minority individuals perceive the court system to be less legitimate, with Black individuals viewing the courts as the least legitimate (p < .001). Black and Hispanic individuals were also less likely than Whites to anticipate that they would be able to hire a private attorney (p < .01). Relative to Whites, Black and Hispanic individuals were more likely to have a family member with a prior arrest (p < .001 Black, p < .05 Hispanic), were less likely to have experienced a prior plea (p < .01), and Black individuals reported a significantly lower household income (p < .05). There were also demographic differences across racial/ethnic groups such that Black and Hispanic individuals were younger than Whites and less likely to be male (p < .001). Hispanic individuals were the most risk averse (p < .001), and White individuals were the least risk averse (p < .001).

In continuing to answer our first research question (RQ1), Table 3 provides OLS models evaluating case legal and extralegal factors as well as the defendant's prior experiences and perceptions that may be associated with WTAP_i, with a specific focus on perceptions of defense attorneys and court legitimacy. Consistent with bivariate tests, race and ethnicity are not significantly associated with WTAP_i. Perceived satisfaction with a defense attorney (DAS) is also not associated with WTAP_i, nor is having private attorney. However, an individual's perception of court legitimacy is positively associated with WTAP_i (b = 3.48, p < .05).

Table 3.

Predicting Initial Willingness to Accept a Plea (WTAP_i) (N = 593).

Variable	B	SE
Guilt	18.89***	2.66
Probability of Conviction	0.40***	0.05
Severity Discount	0.14*	0.06
Defense Attorney Satisfaction (DAS)	0.73	1.70
Court Legitimacy (CL)	3.48*	1.62
Private Attorney	−3.69	2.93
Black	−0.39	3.40
Hispanic	−0.22	3.19
Age	−0.26***	0.08
Male	−2.12	2.74
Household Income	0.53	0.33
Risk Aversion	5.85*	2.25
Prior Arrest	−5.71	5.00
Family Prior Arrest	1.76	2.78
Previous Plea	−1.01	5.68
Constant	−13.15	10.94
R-Squared	0.29
F-Test	18.29***
Root MSE	29.58

Notes: *p < .05; **p < .01; ***p < .001; SE: Standard Error

There are several other factors recognized in prior literature that we find to be associated with one's WTAP_i. For instance, guilt is associated with WTAP_i such that those who are guilty (relative to innocent) are about 19 percentage points more willing to accept a plea (b = 18.89, p < .001). A higher perceived probability of conviction (b = 0.40, p < .001) and perceived sentence severity discount (b = 0.14, p < .05) are also positively associated with WTAP_i. In addition, risk preferences are associated with WTAP_i; those with higher risk aversion are more likely to plead guilty (b = 5.85, p < .05). Age is the only demographic factor associated with WTAP_i, as those who are older are less likely to accept the plea offer (b = -0.26, p < .001). Sex, household income, previous arrest, prior family arrest, and previous plea experience have no significant association with WTAP_i.

To determine if race or ethnicity moderate these findings (RQ2), we run separate models for White, Black, and Hispanic individuals, and then conduct difference in coefficients Z-tests between groups (Paternoster et al., 1998).⁸ These findings indicate that the effects of these antecedents of WTAP_i are markedly consistent across groups, with one exception: probability of conviction (see Table 4). More specifically, there is a significant stronger positive association between the probability of conviction and WTAP_i among Hispanic relative to Black individuals (p < .05), but no significant association emerges between Blacks and Whites, or Hispanics and Whites. It is most notable that we do not find a moderating effect of race/ethnicity on our key constructs: defense attorney satisfaction, court legitimacy, and private attorney.

Table 4.

Predicting Initial Willingness to Accept a Plea (WTAP_i): by Race & Ethnicity (N = 593).

	White (n = 275)		Black (n = 173)		Hispanic (n = 145)
Variable	B	SE	B	SE	B	SE	Z_B/W	Z_H/W	Z_B/H
Guilt	16.13***	3.71	23.65***	5.67	16.62**	5.35	1.11	0.07	−0.90
Prob. of Conv.	0.41***	0.07	0.26**	0.09	0.61***	0.09	−1.33	1.72	2.64**
Severity Discount	0.14	0.08	0.21	0.12	0.06	0.09	0.48	−0.66	−0.97
DAS	−1.43	3.02	2.21	2.60	3.06	3.40	0.91	0.99	0.20
CL	3.69	2.53	3.55	3.04	0.92	3.29	−0.03	−0.67	−0.59
Private Attorney	1.90	4.49	−8.26	5.43	−7.75	5.58	−1.44	−1.35	0.07
Age	−0.20	0.12	−0.09	0.16	−0.48**	0.16	0.54	−1.37	−1.70
Male	−5.46	4.37	1.6	5.45	1.3	5.15	1.01	1.00	−0.04
Income	0.62	0.46	0.51	0.64	0.16	0.71	−0.15	−0.54	−0.36
Risk	6.81*	3.19	6.25	3.75	−2.69	5.49	−0.12	−1.50	−1.34
Prior Arrest	−9.02	6.34	8.34	9.62	−17.76	12.03	1.51	−0.64	−1.69
Family Arrest	1.03	4.51	0.65	4.99	1.54	5.20	−0.06	0.08	0.12
Previous Plea	−0.53	7.20	−6.64	16.58	10.4	12.52	−0.34	0.76	0.82
Constant	−20.29	16.73	−16.68	17.42	29.69	24.70	0.15	1.68	1.53
R-Squared	0.28		0.28		0.43
F-Test	8.96***		6.82***		9.50***
Root MSE	29.29		31.39		27.90

Notes: *p < .05; **p < .01; ***p < .001; SE: Standard Error; DAS: Defense Attorney Satisfaction; CL: Court Legitimacy.

Willingness to Accept a Plea Following an Attorney Evaluation of the Offer (WTAP_e)

We now transition to RQ3, evaluating the effect of an attorney evaluation on WTAP_e, and assessing whether attorney effects are moderated by defendant race and ethnicity. Findings indicate that an attorney's evaluation of a plea offer as “good,” “average,” or “bad,” indeed alters one's WTAP_e. To illustrate, Figure 1 plots the mean differences in WTAP before (WTAP_i) and after an attorney evaluation (WTAP_e). While there are no significant differences in WTAP_i across conditions (as expected because we randomized the attorney evaluations), receiving both a “good” (p < .01) or an “average” (p < .001) evaluation significantly increases an individual's WTAP_e. Receiving a “bad” evaluation, however, more drastically changes one's WTAP – decreasing it by about 15 percentage points (b = -16.5, p < .001). While these results indicate that attorney evaluations influence WTAP_e, on average, it is also notable that a fairly large percentage of the sample (39%) had no change in their WTAP. However, a significant proportion of that 39% (42%, n = 93) were respondents who were told their offer was “average,” which we expected to elicit less, if any, change. Supplemental analysis (available upon request) suggests that an “average” evaluation (relative to both a “good” and a “bad” evaluation) was indeed less likely to result in a change in WTAP (p < .01, p < .05, respectively).

Within Table 5, Model 1 uses regression analysis to evaluate the effect of an attorney evaluation. Here, there is no significant difference in WTAP_e between those who receive a “good” (b = 4.45, p = .16) and “average” evaluation. Those who received a “bad” evaluation, however, were about 15 percentage points less likely to accept the plea offer than those who received an “average” evaluation” (b = -15.15, p < .001), and 20 percentage points less willing to accept the offer than those who received a “good” evaluation (b = -19.60, p < .001),⁹ on average. Factual guilt continues to demonstrate a positive association with WTAP_e (b = 18.03, p < .001).

Table 5.

Effect of Attorney Evaluation on Willingness to Plea (WTAP_e): by Guilt Status and Race/Ethnicity (N = 593).

	Model 1		Model 2		Model 3
	Main Effects		2-way Interaction		3-way Interaction
	B	SE	B	SE	B	SE
Attorney Recommendation
Good	4.45	3.16	2.36	4.05	3.32	5.91
Bad	−15.15***	3.02	−8.17*	4.01	−8.89	5.16
Guilty	18.03***	2.63	20.89***	4.46	16.81	6.54
Good*Guilty			4.79	6.44	8.56	9.61
Bad*Guilty			−14.57*	6.11	−5.23	8.27
Black					−1.81	6.36
Good *Black					13.12	9.40
Bad *Black					10.84	9.34
Guilty*Black					19.30*	9.59
GoodGuiltyBlack					−27.95	14.58
BadGuiltyBlack					−34.86*	13.98
Hispanic
Good * Hispanic					−19.67	10.24
Bad * Hispanic					−11.74	9.98
Guilty* Hispanic					−5.98	11.51
GoodGuilty Hispanic					17.24	16.01
BadGuilty Hispanic					−0.10	15.18
Constant	29.00	2.32	27.35	2.73	7.88	11.85
R-Squared	0.20		0.21		0.23
F-Test	10.65***		9.56***		7.79***
Root MSE	30.59		30.38		30.26

Notes: *p < .05; **p < .01; ***p < .001; SE: Standard Error; Control variables included in analyses but omitted from output; full results available upon request

In Model 2, Table 5, we assess whether the effect of an attorney evaluation is moderated by guilt. This is important methodologically, as guilt was manipulated within the vignette and prior research consistently indicates that guilt has a significant influence on WTAP (e.g., Henderson & Levett, 2018; Wilford et al., 2020). Our findings suggest there is indeed moderation, particularly for a “bad” evaluation (p < .01). Figure 2 depicts these results by plotting mean WTAP_i and WTAP_e across guilt and evaluation conditions. First, there is not a significant difference in the effect of a “good” (relative to “average”) evaluation among either the guilty or the innocent. For example, receiving a “good” evaluation resulted in an 11 percentage point increase (on average)¹⁰ in WTAP among the guilty, and a 10 percentage point increase among the innocent: a difference of only 1 percentage point between guilt conditions. Those with an “average” evaluation have a 3 percentage point increase in WTAP_e among the guilty, and a 4 percentage points increase among the innocent: again, only a 1 percentage point difference in the effect of this evaluation between guilt conditions.

Among those who received a “bad” evaluation, however, there is a larger decrease in WTAP_e among those who were guilty (p < .01). This is notable in Figure 2, as receiving a “bad” evaluation resulted in about a 6 percentage point decrease in WTAP among the innocent, and a 23.5 percentage-point decrease among the guilty; this shows a difference in effect of this evaluation of about 17.5 percentage points between guilt and innocence.

Given the guilt moderation finding, Model 3 within Table 5 evaluates whether there is a three-way interaction between attorney evaluation, guilt, and race/ethnicity. Here, because race and ethnicity are not randomized and because some factors may be associated with both race/ethnicity and WTAP_e, we include our full set of theoretical variables in the model.¹¹ Results indicate that there are significant three-way interactive effects on WTAP_e.¹² For ease of interpretation, Figure 3 plots average predicted values for WTAP_e based on attorney evaluation, guilt, and race/ethnicity.

Among those who are White (Figure 3, top graph), guilty individuals (relative to those who are innocent) are more likely to accept a guilty plea across all attorney evaluations (25, 16, and 11 percentage points more likely for a “good,” “average,” or “bad” evaluation, respectively), though the difference is only significant with a “good” evaluation (p < .05). Relative to a “good” attorney evaluation, a “bad” evaluation among those who are White results in a lower WTAP_e for both those who are guilty (from 54 to 28%, a total of 26 percentage points lower, p < .001) and innocent (from 28 to 17%, a total of 12 percentage points lower, p < .05).

Among those who are Black (Figure 3, middle graph), there is only a significant difference between guilty conditions for an “average” evaluation, such that those who are guilty are about 36 percentage points more willing to accept a plea (p < .001). Among those who are Black and innocent, there is only a significant difference between a “good” and “average” evaluation (p < .05), where those with a “good” evaluation were about 16 percentage points more willing to accept a plea than those with an “average” evaluation. However, among those who are Black and guilty, both “good” and “average” evaluations resulted in a significantly higher WTAP_e relative to a bad evaluation (p < .01, p < .05, respectively), where those with a “bad” evaluation were 38 percentage points less willing to accept a plea than an “average” evaluation, and 35 percentage points less willing to accept than a “good” evaluation. There was only a non-significant three 3 percentage point difference between a “good” and “average” evaluation among black individuals who are guilty.

Among those who are Hispanic (Figure 3, bottom graph), the innocent were less likely to accept a plea than those who were guilty (WTAP_e percentage point difference: 37-“good”; 11-“average”; 6-“bad”) across all conditions, though only the difference within the “good” evaluation is significant (p < .05). This finding is consistent with Whites. Three additional interesting patterns emerged with Hispanic respondents. First, while guilty Hispanics had a 35 percentage point decreased WTAP_e for a “bad” relative to “good” evaluation (like Black and White individuals), they (like Black individuals) only had a non-significant 4 percentage point difference between a “good” and “bad” evaluation when innocent. Second, innocent Hispanic individuals who received an “average” evaluation had a 16 percentage point higher WTAP_e than those who received a “good” evaluation, which is unexpected. Though this finding was not significant, it differs in direction from both White (who had a non-significant decrease of 3percentage points) and Black individuals (who had a significant decrease of 16 percentage points, p < .05). Third, regardless of guilt status, Hispanic individuals with a “bad” (relative to “average”) evaluation had a reduced WTAP_e: a reduction of 26 percentage points among the guilty and 21 percentage points among the innocent (p < .001).

Discussion

This paper sought to improve upon calls to understand the guilty plea process (Ulmer, 2012) and specifically, perceptions of this process through the eyes of defendants (Clair, 2021). Due to demonstrated racial and ethnic disparities in trial rates relative to pleas (Lee & Richardson, 2020; Testa & Johnson, 2020) and in final sentence severity (King & Light, 2019; Mitchell, 2005), we were first interested in exploring racial and ethnic differences in willingness to accept a guilty plea (WTAP_i), as observed differences in this initial stage of decision-making may explain subsequent sentencing disparities. Further, if we observed differences, we were interested in whether factors such as views on court legitimacy, perceived satisfaction with defense attorneys, or being able to afford a private attorney explained the findings. We further sought to understand how race and ethnicity may moderate the effect of theoretical correlates of plea decision-making. Finally, we aimed to explore how an attorney evaluation of an offer influences WTAP_e and whether this effect is consistent across racial/ethnic groups.

No Direct Effects of Race and Ethnicity on Initial WTAP

Contrary to our expectations, race and ethnicity were not significant predictors of initial WTAP (see Table 2). There were, however, differences in important theoretical antecedents of WTAP_i by race and ethnicity. For instance, relative to White individuals, Black and Hispanic individuals had a higher perceived probability of conviction (though this was only significant for Black individuals) after viewing the crime vignette. Furthermore, Black and Hispanic individuals saw the court system as less legitimate relative to those who were White, with Black individuals having the lowest perceptions of legitimacy. This finding is consistent with prior research findings that minorities are less likely to feel they should trust and cooperate with criminal justice institutions, and that Black individuals have even lower perceptions than Hispanic individuals (Bolger & Walters, 2019; Higgins et al., 2009; Madon et al., 2017; Ra & Kim, 2019; Sun & Wu, 2006; Wu & Cao, 2018). As we mention above, this variation in perception is likely because racial and ethnic minorities are treated more harshly throughout the criminal justice system, as there is evidence of racial/ethnic disparities at nearly every stage of the system including police stops, arrest, detention, and sentencing (Braga et al., 2019; King & Light, 2019; Kutateladze et al., 2014, 2016; Mitchell, 2005; Piquero, 2008).

While the finding that WTAP_i does not vary by race or ethnicity is surprising, particularly since it is somewhat inconsistent with previous research indicating that racial and ethnic minorities are less likely to plead guilty than Whites (e.g., Lee & Richardson, 2020; Testa & Johnson, 2020), we posit a few potential explanations. First, we held the value of the plea offer value constant for all participants and an emerging body of work indicates that racial and ethnic minorities may be offered harsher terms at the start of negotiations than non-minorities (Kutateladze et al., 2016), which may lead to a lower rate of plea acceptance. Unfortunately, prior plea offers are rarely recorded and/or shared with researchers, so while this is a plausible explanation, it is challenging to verify with court data (Johnson et al., 2016). Research also suggests that defense attorneys perceive differences in the quality of a plea “bargain” they believe they could elicit for their clients based on their client's race, which supports the idea that initial offers may vary based on extralegal defendant characteristics (Edkins, 2011).

Second, given some research showing disparities in policing in minority communities (e.g., Gelman et al., 2007), racial and ethnic minorities in the criminal justice system may be less likely to be factually guilty and thus less likely to accept a plea offer. Though this is a difficult premise to verify empirically, research finding that Black individuals are overrepresented in exonerations lends credence to the idea (Gross, 2017; Olney & Bonn, 2015; Smith & Hattery, 2011; Taslitz, 2006). Given that guilt was a manipulated factor within our study, and not a natural product of the system, we cannot rule out this possibility in practice.

Moreover, race and ethnicity were only weak moderators of one antecedent of WTAP_{i –} probability of conviction. While all three groups showed a positive association between probability of conviction and WTAP_i, the association was stronger among Hispanic relative to Black (p < .01) individuals, though there was no significant difference in the association between White and Black individuals or White and Hispanic individuals. These findings cumulatively suggest that the effect of court legitimacy, defense attorney satisfaction, attorney type, and many other theoretical predictors of WTAP_i do not vary across race and ethnicity. This indicates that while there may be differences in the starting level of many of these factors, the way they influence decision making is largely consistent regardless of racial/ethnic background.

Attorney Evaluations of Plea Offers Significantly Influence Defendants’ WTAP_e

Next, we explored the effects of attorney evaluations on WTAP and whether race and ethnicity moderated these effects. Consistent with prior research (Henderson & Levett, 2018; Henderson & Shteynberg, 2020; Lee et al., 2021), an attorney evaluation of a plea influenced defendant decision-making – such that, on average, both a “good” or “average” evaluation increased one's WTAP_e, and a “bad” evaluation decreased WTAP_e (Figure 1). However, the effects of “good” and “average” evaluations were statistically indistinguishable from one another. While this result may seem somewhat counterintuitive at first, it might be because risk aversion causes people to be more afraid of accepting a “bad” offer, rather than not accepting a “good” one (Tversky & Kahneman, 1992). “Average” offers may also be interpreted fairly positively, as the effect did not vary from a “good” evaluation. In addition, consistent with prior research (Lee et al., 2021), guilt moderated the influence of attorney evaluations such that there was a larger effect of a “bad” evaluation among those who were guilty (Figure 2). This finding supports extant research suggesting that innocent and guilty individuals likely have variations in their decision-making processes (Henderson & Levett, 2018; Wilford et al., 2020). These results also indicate that attorney evaluations of plea offers influence defendants’ WTAP_e, it is also notable that 39% of the sample did not change their WTAP_e following an attorney evaluation (though a lack of change was strongly associated with receiving an “average” plea offer evaluation, indicating that “bad” and “good” evaluations are more likely to influence WTAP_e). Further understanding the factors that elicit (and do not elicit) within-individual change in WTAP is a promising avenue for future research.

Race and Ethnicity Moderate the Effect of Attorney Evaluation and Guilt on WTAP_e

Finally, our results indicate that there are important nuances in the effect of an attorney evaluation and guilt on WTAP_e by race and ethnicity (Table 5 and Figure 3). For example, when White or Hispanic individuals were told by their attorney that the plea offer was “average,” WTAP_e did not differ by guilt status. For Black individuals in the same condition, however, there was a significant difference; those who were innocent were 36 percentage points, on average, less likely to plea than those who were guilty. In addition, when told their offer was “good,” guilty White and Hispanic defendants varied in their WTAP_e relative to innocent defendants, but Black defendants did not. This may suggest that Black individuals interpret an attorney's evaluation of a plea offer differently depending on their guilt. For example, perhaps being innocent and charged with a crime is viewed particularly negatively among this minority group who already has, on average, the most negative perceptions of the criminal justice system's legitimacy. It is also possible that the attorney's evaluation of the offer altered perceptions of legitimacy after our legitimacy measures were collected which should be a key inquiry of future scholarship.

Summary

In sum, observed differences in guilty pleas among racial and ethnic minorities do not appear to be the direct result of initial differences in their WTAP_i nor differences in the effect of known antecedents of WTAP_i across race and ethnicity. Rather, differences in plea acceptance rates across race and ethnicity may emerge because of racialized differences in initial levels of perceptual correlates of WTAP_i (e.g., probability of conviction, court legitimacy), or differences in WTAP following an attorney evaluation that are also sensitive to a person's guilt. This is particularly notable because these differences were observed while holding constant many theoretical factors that may differ across groups, including perceptions of court legitimacy, ability to hire a private attorney, and expected satisfaction with their attorney.

Limitations, Future Directions, and Conclusion

Despite these contributions, there are limitations of our study which provide directions for future research. First, as with all vignette studies, this is a hypothetical situation. We asked respondents (prior to viewing the vignette, to avoid potential confounding of attorney quality with any other case characteristics such as plea offer quality) to imagine themselves being charged with a crime, to predict whether they would have a private or public defense attorney, and to imagine how satisfied they would be with their counsel. It is thus possible that individuals in a “real world” situation may react differently. However, our finding that attorneys can influence their clients’ WTAP_e is consistent with previous experimental work (Henderson & Levett, 2018; Henderson & Shteynberg, 2020; Lee et al., 2021). For ethical reasons, it would be impossible to randomize attorney quality or evaluation in real case settings. Thus, an improvement upon this issue would be to conduct observations of real court cases and negotiations. Additionally, pairing plea negotiation observations with interviews from defendants and defense attorneys would provide critical context for understanding the importance of the attorney/client relationship. Clair's (2021) analysis of disadvantaged defendants also provides a clear example of the critical insights gained from qualitative methods.

Along those lines, while we improved upon some prior studies (Henderson & Shteynberg, 2020; Lee et al., 2021) with a more racially and ethnically heterogenous sample, the present study did not include a large proportion of individuals with prior arrests or plea offer experience. Research on legitimacy has demonstrated that direct personal experience with police, courts, or the criminal justice system have effects on perceptions of legitimacy and willingness to cooperate (Brown & Benedict, 2002; Sun & Wu, 2006; Wolfe, Nix, Kaminski, & Rojek, 2016; Woolard et al., 2008). Therefore, additional experimental plea research is warranted, particularly with justice-involved populations. While we believe that our results would be predictive of actual decision-making in a case, it is likely that individuals with significant criminal history and experience interacting with criminal justice officials may have a different decision-making calculus than those who have none.

In sum, contrary to expectations, we found that race and ethnicity are not directly associated with one's initial WTAP. Instead, widely established factors such as guilt and perceptions of punishment were critical in predicting WTAP_i. Notably, several of these perceptual correlates of WTAP such as court legitimacy and probability of conviction were also racialized, in that racial and ethnic minorities reported different levels than their White counterparts; for example, in line with previous work, racial and ethnic minorities reported lower perceptions of court legitimacy. Race and ethnicity do not appear to have any meaningful moderating effects on these relationships, suggesting that the effect of theoretical antecedents of WTAP are consistent across race/ethnicity. Instead, the impact of race/ethnicity may be tied to differing baseline levels of these antecedents. Significant 3-way interactions also emerged between attorney evaluation, race/ethnicity, and guilt – suggesting that an attorney's influence on their client is likely based on legal characteristics of the case (e.g., guilt) as well as the defendant (e.g., race and ethnicity).

Though it has been documented in previous research that racial and ethnic minorities are more likely to pursue a trial rather than take a plea compared to their White counterparts, the causal mechanism behind this remains unclear. Our study attempted to dig deeper into this issue and from our results, it does not seem that there are general pre-existing differences by race or ethnicity in initial WTAP_i. We hypothesize that the difference in plea rates may instead be due to cumulative differences in initial levels of theoretical antecedents of WTAP, differences in the initial quality of an offer by a prosecutor, or delivery of offers by a defense attorney. Our finding that guilt status and race/ethnicity moderate the effect of an attorney evaluation could also be particularly consequential to disparities in plea rates if minority individuals are more likely to be stopped, arrested and/or charged with a crime when they are innocent. More work is necessary to determine if any of these speculations garner empirical support.

Footnotes

Funding

The author(s) received no financial support for the research, authorship, and/or publication of this article

ORCID iDs

Chae M. Jaynes

Heath N. Franks

Notes

Author Biographies

Chae M. Jaynes is an Assistant Professor at the University of South Florida (USF). She is also a Leadership Team member for the Center for Justice Research and Policy at USF. She earned her PhD from the University of Maryland. Her research interests include decision-making within several aspects of the criminal justice system, including the courts. She also studies labor market perceptions and the relationship between employment and crime.

Jacqueline G. Lee is an Assistant Professor at Boise State University. She earned her PhD from the University of Maryland. She has also earned a J.D. from the University of Oklahoma. Her research interests include courts and sentencing, criminal justice policy, integration of social science research with legal research, inequalities in the justice system, and the female experience of violence, offending, and punishment.

Heath N. Franks earned a master’s degree in criminal justice from the University of North Alabama. His research interests involve public perceptions of the criminal justice system and institutional legitimacy, particularly how these perceptions form and vary across race and ethnicity and their influence on conforming behaviors.

Appendix A. Scale Items

Defense Attorney Satisfaction (DAS) (alpha = .93)

(1) “Your defense attorney overall”

(2) “The amount of time your defense attorney would likely be able to spend focusing on your case “

(3) “Your defense attorney’s knowledge of the law”

(4) “Your defense attorney’s familiarity with the court system and procedures”

(5 )“Your defense attorney’s ability to convince other court actors (e.g., judge, prosecutor) to act in your favor”

Court Legitimacy (alpha = .88)

(1) “The courts use rules and procedures that are fair to everyone”

(2) “Courts provide opportunities for unfair decisions to be corrected”

(3) “Courts and their employees would treat you with respect if you had contact with them for any reason”

(4) “People often receive fair outcomes from the courts”

(5) “People usually receive the outcomes they deserve under the law”

(6) “When court actors (e.g. judge, prosecutors) deal with people like me, the court actors always behave according to the law”

(7) “The law represents the moral values of people like me”

(8) “Rehabilitates criminal offenders”

(9) “Keeps dangerous offenders off of the streets”

(10) “Makes sure an offender restores or repairs damages inflicted on the victim or society”

(11) “Punishes someone for a wrongful act”

(12) “Convicts individuals who are guilty”

(13) “Finds innocent people not guilty”

(14) “Completes court proceedings in a timely manner”

References

Atkins

B. M.

Boyle

E. W.

. 1976. “Prisoner Satisfaction with Defense Counsel.” Criminal Law Bulletin 12:427‐50.

Baker

Pickett

J. T.

Amin

D. M.

Golden

Dhungana

Gertz

Bedard

. 2015. “Shared Race/Ethnicity, Court Procedural Justice, and Self-Regulating Beliefs: A Study of Female Offenders.” Law & Society Review 49(2):433‐66. https://doi.org/10.1111/lasr.12137.

Baumgartner

F. R.

Epp

D. A.

Shoub

. 2018. Suspect Citizens: What 20 Million Traffic Stops Tell Us About Policing and Race. Cambridge, UK: Cambridge Univ. Press.

Baumgartner

F. R.

Epp

D. A.

Shoub

Love

. 2017. “Targeting Young men of Color for Search and Arrest During Traffic Stops: Evidence from North Carolina, 2002–2013.” Politics Groups Identities 5(1):107‐31.

Beck

Ajzen

. 1991. “Predicting Dishonest Actions Using the Theory of Planned Behavior.” Journal of Research in Personality 25(3):285‐301.

Bibas

2016. “Designing Plea Bargaining from the Ground up: Accuracy and Fairness Without Trials as Backstops.” William and Mary Law Review 57(4):1055‐82.

Boccaccini

M. T.

Boothby

J. L.

Brodsky

S. L.

. 2004. “Development and Effects of Client Trust in Criminal Defense Attorneys: Preliminary Examination of the Congruence Model of Trust Development.” Behavioral Sciences & the Law 22(2):197‐214. https://doi.org/10.1002/bsl.584.

Boccaccini

M. T.

Brodsky

S. L.

. 2002. “Attorney–Client Trust among Convicted Criminal Defendants: Preliminary Examination of the Attorney–Client Trust Scale.” Behavioral Sciences & the Law 20(1–2):69‐87. https://doi.org/10.1002/bsl.469.

Bolger

C. P.

Walters

G. D.

. 2019. “The Relationship Between Police Procedural Justice, Police Legitimacy, and People’s Willingness to Cooperate with law Enforcement: A Meta-Analysis.” Journal of Criminal Justice 60:93‐9. https://doi.org/10.1016/j.jcrimjus.2019.01.001.

10.

Bordens

K. S.

1984. “The Effects of Likelihood of Conviction, Threatened Punishment, and Assumed Role on Mock Plea-Bargaining Decisions.” Basic and Applied Social Psychology 5(1):59‐74. https://doi.org/10.1207/s15324834basp0501_4.

11.

Braga

A. A.

Brunson

R. K.

Drakulich

K. M.

. 2019. “Race, Place, and Effective Policing.” Annual Review of Sociology 45(1):535‐55.

12.

Brown

Benedict

W. R.

. 2002. “Perceptions of the Police.” Policing (Bradford, England) 25(3):543‐80. https://dx-doi-org.web.bisu.edu.cn/10.1108/13639510210437032.

13.

Bushman

B. J.

Moeller

S. J.

Konrath

Crocker

. 2012. “Investigating the Link Between Liking Versus Wanting Self-Esteem and Depression in a Nationally Representative Sample of American Adults.” Journal of Personality 80(5):1453‐69. https://doi.org/10.1111/j.1467-6494.2012.00781.x.

14.

Bushway

S. D.

Redlich

A. D.

. 2012. “Is Plea Bargaining in the “Shadow of the Trial” a Mirage?” Journal of Quantitative Criminology 28(3):437‐54. doi:10.1007/s10940-011-9147-5

15.

Clair

2021. “Being a Disadvantaged Criminal Defendant: Mistrust and Resistance in Attorney-Client Interactions.” Social Forces. 100(1):194-217. https://doi.org/10.1093/sf/soaa082.

16.

Cohen

T. H.

2014. “Who is Better at Defending Criminals? Does Type of Defense Attorney Matter in Terms of Producing Favorable Case Outcomes.” Criminal Justice Policy Review 25(1):29‐58. https://doi.org/10.1177/0887403412461149.

17.

Crutchfield

R. D.

Skinner

M. L.

Haggerty

K. P.

McGlynn

Catalano

R. F.

. 2012. “Racial Disparity in Police Contacts.” Race and Justice 2(3):179‐202.

18.

Denbeaux

Kearns

Ricciardelli

M. J.

. 2016. “Racial Profiling Report: Bloomfield Police and Bloomfield Municipal Court.” SSRN Electronic Journal, 1-27. https://doi.org/10.2139/ssrn.2760382.

19.

Dervan

L. E.

Edkins

V. A.

. 2013. “The Innocent Defendant’s Dilemma: An Innovative Empirical Study of Plea Bargaining’s Innocence Problem.” The Journal of Criminal Law & Criminology 103:1.

20.

Devers

2011. Plea and Charge Bargaining Research Summary. Washington, DC: Bureau of Justice Assistance.

21.

Edkins

2011. “Defense Attorney Plea Recommendations and Client Race: Does Zealous Representation Apply Equally to all?” Law and Human Behavior 35(5):413‐25. https://doi.org/10.1007/s10979-010-9254-0.

22.

Eisenstein

Jacob

. 1977. Felony Justice. Boston, MA: Little, Brown.

23.

Everett

R. S.

Wojtkiewicz

R. A.

. 2002. “Difference, Disparity, and Race/Ethnic Bias in Federal Sentencing.” Journal of Quantitative Criminology 18(2):189‐211.

24.

Exum

M. L.

Bouffard

J. A.

. 2010 “Testing Theories of Criminal Decision Making: Some Empirical Questions About Hypothetical Scenarios.” Pp. 581‐94 in Handbook of Quantitative Criminology, edited by Piquero

Weisburd

. New York, NY: Springer.

25.

Feeley

M. M.

. 1979. The process is the punishment: Handling cases in a lower criminal court. Russell Sage Foundation. New York, NY: Russell Sage Foundation.

26.

Fox

Moule

R. K.

Jaynes

C. M.

Parry

M. M.

. 2021. “Are the Effects of Legitimacy and Its Components Invariant? Operationalization and the Generality of Sunshine and Tyler’s Empowerment Hypothesis.” Journal of Research in Crime and Delinquency 58(1):3‐40. https://doi.org/10.1177/0022427820926228.

27.

Friedman

L. M.

1979. “Plea Bargaining in Historical Perspective.” Law & Society Review 13(2):247‐59. https://doi.org/10.2307/3053251.

28.

Gardner

Steinberg

. 2005. “Peer Influence on Risk Taking, Risk Preference, and Risky Decision Making in Adolescence and Adulthood: An Experimental Study.” Developmental Psychology 41(4):625‐35. https://doi.org/10.1037/0012-1649.41.4.625.

29.

Garnier-Dykstra

L. M.

Wilson

. 2019. “Behavioral Economics and Framing Effects in Guilty Pleas: A Defendant Decision Making Experiment.” Justice Quarterly: 38(2):224-248. https://doi.org/10.1080/07418825.2019.1614208

30.

Gelman

Fagan

Kiss

. 2007. “An Analysis of the New York City Police Department’s “Stop-and-Frisk” Policy in the Context of Claims of Racial Bias.” Journal of the American Statistical Association 102:813‐23.

31.

Gross

S. R.

2017. “What we Think, What we Know and What we Think we Know About False Conviction.” Ohio State Journal of Criminal Law 14(753):753‐86.

32.

Hamm

J. A.

. 2020. “An application of the integrated framework of legitimacy to the state courts context”. International Journal of Comparative and Applied Criminal Justice 44(1-2):85-102.

33.

Harlow

C. W.

(2000). Defense Counsel in Criminal Cases (NCJ 179023; p. 12). Bureau of Justice Statistics. https://bjs.ojp.gov/content/pub/pdf/dccc.pdf

34.

Heen

M. S.

Lieberman

J. D.

Miethe

T. D.

. 2014. “A Comparison of Different Online Sampling Approaches for Generating National Samples.” Center for Crime and Justice Policy 1(9):1‐8.

35.

Henderson

K. S.

Levett

L. M.

. 2018. “Investigating Predictors of True and False Guilty Pleas.” Law and Human Behavior 42(5):427‐41. https://doi.org/10.1037/lhb0000297.

36.

Henderson

Levett

. 2019 “Plea bargaining: The influence of counsel.” in Advances in psychology and law (pp. 73-100). Cham: Springer.

37.

Henderson

K. S.

Shteynberg

R. V.

. 2020. “Plea Decision-Making: The Influence of Attorney Expertise, Trustworthiness, and Recommendation.” Psychology, Crime & Law 26(6):527‐51. https://doi.org/10.1080/1068316X.2019.1696801.

38.

Higgins

G. E.

Wolfe

S. E.

Mahoney

Walters

N. M.

. 2009. “Race, Ethnicity, and Experience: Modeling the Public’s Perceptions of Justice, Satisfaction, and Attitude Toward the Courts.” Journal of Ethnicity in Criminal Justice 7(4):293‐310. https://doi.org/10.1080/15377930903382282.

39.

Horowitz

Utada

(2018). Community Supervision Marked by Racial and Gender Disparities . https://pew.org/2L4b6kQ

40.

Hussemann

Siegel

. 2021. “Pleading Guilty: Indigent Defendant Perceptions of the Plea Process.” Tennessee Journal of Law and Policy, 13(2):1-65. https://trace.tennessee.edu/tjlp/vol13/iss2/3

41.

Irizarry

May

D. C.

Davis

Wood

P. B.

. 2016. “Mass Incarceration Through a Different Lens: Race, Subcontext, and Perceptions of Punitiveness of Correctional Alternatives When Compared to Prison.” Race and Justice 6(3):236‐56. https://doi.org/10.1177/2153368715603103.

42.

Johnson

B. D.

2019. “Trials and Tribulations: The Trial Tax and the Process of Punishment.” Crime and Justice 48:313‐63. https://doi.org/10.1086/701713.

43.

Johnson

B. D.

King

R. D.

Spohn

. 2016. “Sociolegal Approaches to the Study of Guilty Pleas and Prosecution.” Annual Review of Law and Social Science 12(1):479‐95. https://doi.org/10.1146/annurev-lawsocsci-110615-084755.

44.

Johnson

B. D.

Richardson

. 2019 “Race and Plea Bargaining.” in A System of Pleas (pp. 83-106). Oxford University Press, New York, NY, USA. https://doi.org/10.1093/oso/9780190689247.003.0006.

45.

King

R. D.

Light

M. T.

. 2019. “Have Racial and Ethnic Disparities in Sentencing Declined?” Crime and Justice 48:365‐47. https://doi.org/10.1086/701505.

46.

Kovera

M. B.

2019. “Racial Disparities in the Criminal Justice System: Prevalence, Causes, and a Search for Solutions.” Journal of Social Issues 75(4):1139‐64. https://doi.org/10.1111/josi.12355.

47.

Kurlychek

M. C.

Johnson

B. D.

. 2019. “Cumulative Disadvantage in the American Criminal Justice System.” Annual Review of Criminology 2(1):291‐319. https://doi.org/10.1146/annurev-criminol-011518-024815.

48.

Kutateladze

B. L.

Andiloro

N. R.

Johnson

B. D.

. 2016. “Opening Pandora’s box: How Does Defendant Race Influence Plea Bargaining?” Justice Quarterly 33(3):398‐426.

49.

Kutateladze

B. L.

Andiloro

N. R.

Johnson

B. D.

Spohn

C. C.

. 2014. “Cumulative Disadvantage: Examining Racial and Ethnic Disparity in Prosecution and Sentencing.” Criminology; An Interdisciplinary Journal 52(3):514‐51. https://doi.org/10.1111/1745-9125.12047.

50.

LaFree

G. D.

1985. “Adversarial and Nonadversarial Justice: A Comparison of Guilty Pleas and Trials.” Criminology; An Interdisciplinary Journal 23(2):289‐312. https://doi.org/10.1111/j.1745-9125.1985.tb00338.x.

51.

Lee

J. G.

Jaynes

C. M.

Ropp

. 2021. “Satisfaction, Legitimacy, and Guilty Pleas: How Perceptions and Attorneys Affect Defendant Decision-Making.” Justice Quarterly 0(0):1‐33. https://doi.org/10.1080/07418825.2020.1786147.

52.

Lee

J. G.

Richardson

R. L.

. 2020. “Race, Ethnicity, and Trial Avoidance: A Multilevel Analysis.” Criminal Justice Policy Review 31(3):422‐51. https://doi.org/10.1177/0887403418812998.

53.

Lee

J. G.

Ropp

. 2020. “Sometimes I’m just wearing the prosecutor down:” An exploratory analysis of criminal defense attorneys in plea negotiations and client counseling. Forthcoming in Journal of Qualitative Criminal Justice and Criminology.

54.

Madon

N. S.

Murphy

Sargeant

. 2017. “Promoting Police Legitimacy among Disengaged Minority Groups: Does Procedural Justice Matter More?” Criminology & Criminal Justice 17(5):624‐42. https://doi.org/10.1177/1748895817692849.

55.

May

D. C.

Wood

P. B.

. 2005. “What Influences Offenders’ Willingness to Serve Alternative Sanctions?” The Prison Journal 85(2):145‐67. https://doi.org/10.1177/0032885505276988.

56.

Mazerolle

Antrobus

Bennett

Tyler

T. R.

. 2013a. “Shaping Citizen Perceptions of Police Legitimacy: A Randomized Field Trial of Procedural Justice.” Criminology; An Interdisciplinary Journal 51(1):33‐63. https://doi.org/10.1111/j.1745-9125.2012.00289.x.

57.

Mazerolle

Bennett

Davis

Sargeant

Manning

. 2013b. “Legitimacy in Policing: A Systematic Review.” Campbell Systematic Reviews 9(1):i‐147. https://doi.org/10.4073/csr.2013.1.

58.

Metcalfe

Chiricos

. 2018. “Race, Plea, and Charge Reduction: An Assessment of Racial Disparities in the Plea Process.” Justice Quarterly 35(2):223‐53. https://doi.org/10.1080/07418825.2017.1304564.

59.

Mitchell

2005. “A Meta-Analysis of Race and Sentencing Research: Explaining the Inconsistencies.” Journal of Quantitative Criminology 21(4):439‐66.

60.

Mitchell

Caudy

M. S.

. 2015. “Examining Racial Disparities in Drug Arrests.” Justice Quarterly 32(2):288‐313.

61.

Moule

R. K.

Burruss

G. W.

Parry

M. M.

Fox

. 2019. “Assessing the Direct and Indirect Effects of Legitimacy on Public Empowerment of Police: A Study of Public Support for Police Militarization in America.” Law & Society Review 53(1):77‐107. https://doi.org/10.1111/lasr.12379.

62.

O’Brien

Pheterson

Wright

Hostica

. 1977. “The Criminal Lawyer: The Defendant’s Perspective.” American Journal of Criminal Law 5:283.

63.

Olney

Bonn

. 2015. “An Exploratory Study of the Legal and non-Legal Actors Associated with Exoneration for Wrongful Conviction: The Power of DNA Evidence.” Criminal Justice Policy Review 26(4):400‐20. https://doi.org/ 10.1177/0887403414521461.

64.

Omori

Petersen

. 2020. “Institutionalizing Inequality in the Courts: Decomposing Racial and Ethnic Disparities in Detention, Conviction, and Sentencing*.” Criminology; An Interdisciplinary Journal 58(4):678‐713. https://doi.org/10.1111/1745-9125.12257.

65.

Parker

K. F.

Stults

B. J.

Rice

S. K.

. 2005. “Racial Threat, Concentrated Disadvantage and Social Control: Considering the Macro-Level Sources of Variation in Arrests*.” Criminology; An Interdisciplinary Journal 43(4):1111‐34. https://dx-doi-org.web.bisu.edu.cn/10.1111/j.1745-9125.2005.00034.x.

66.

Paternoster

Brame

Mazerolle

Piquero

. 1998. “Using the correct statistical test for the equality of regression coefficients”. Criminology 36(4):859-866.

67.

Piquero

A. R.

2008. “Disproportionate Minority Contact.” The Future of Children 18(2):59‐79.

68.

K. H.

Kim

. 2019. “Racialized Perceptions of the Police: Measuring the Dimensions of Juveniles’ Perceptions of the Police.” Policing: An International Journal 42(2):301‐15. https://doi.org/10.1108/PIJPSM-11-2017-0144.

69.

Reaves

B. A.

2013. Felony Defendants in Large Urban Counties, 2009-Statistical Tables. Washington, DC: US Department of Justice.

70.

Redlich

A. D.

Shteynberg

R. V.

. 2016. “To Plead or not to Plead: A Comparison of Juvenile and Adult True and False Plea Decisions.” Law and Human Behavior 40(6):611‐25. https://doi.org/10.1037/lhb0000205.

71.

Reisig

M. D.

Bratton

Gertz

M. G.

. 2007. “The Construct Validity and Refinement of Process-Based Policing Measures.” Criminal Justice and Behavior 34(8):1005‐28. https://doi.org/10.1177/0093854807301275.

72.

Rodriguez

2013. “Concentrated Disadvantage and the Incarceration of Youth: Examining how Context Affects Juvenile Justice.” Journal of Research in Crime and Delinquency 50(2):189‐215. https://doi.org/10.1177%2F0022427811425538.

73.

Rojek

Rosenfeld

Decker

. 2012. “Policing race: The racial stratification of searches in police traffic stops”. Criminology 50(4):993-1024.

74.

Seguino

Brooks

(2018). “A Deeper Dive into Racial Disparities in Policing in Vermont.” University of Vermont and Cornell University. https://hrc.vermont.gov/sites/hrc/files/publications/A%20Deeper%20Dive%20into%20Racial%20Disparities%20in%20Policing%20in%20Vermont_3.26_final.pdf

75.

Sheppard

B. H.

Hartwick

Warshaw

P. R.

. 1988. “The Theory of Reasoned Action: A Meta-Analysis of Past Research with Recommendations for Modifications and Future Research.” The Journal of Consumer Research 15(3):325‐43.

76.

Shook

J. J.

Goodkind

Kolivoski

K. M.

Ballentine

K. L.

. 2021. “Procedural Justice and Legal Socialization among Juvenile Offenders: The Role of Defense Attorneys.” Journal of Social Issues 77(2):484‐503. https://doi.org/10.1111/josi.12445.

77.

Smith

Hattery

. 2011. “Race, Wrongful Conviction, and Exoneration.” Journal of African American Studies 15:74‐94. https://www-jstor-org-443.web.bisu.edu.cn/stable/43525410

78.

Steiner

P. M.

Atzmüller

. 2016. “Designing valid and reliable vignette experiments for survey research: A case study on the fair gender income gap”. Journal of Methods and Measurement in the Social Sciences 7(2):52-94.

79.

Sudnow

1965. “Sociological Features of the Penal Code in a Public Defender Office.” Social Problems 12:255‐76.

80.

Sun

I. Y.

. 2006. “Citizens’ Perceptions of the Courts: The Impact of Race, Gender, and Recent Experience.” Journal of Criminal Justice 34(5):457‐67. https://doi.org/10.1016/j.jcrimjus.2006.09.001.

81.

Sunshine

Tyler

T. R.

. 2003. “The role of procedural justice and legitimacy in shaping public support for policing”. Law and Society Review 37(3):513-548.

82.

Tankebe

2013. “Viewing Things Differently: The Dimensions of Public Perceptions of Police Legitimacy.” Criminology; An Interdisciplinary Journal 51(1):103‐35. https://doi.org/10.1111/j.1745-9125.2012.00291.x.

83.

Taslitz

A. E.

2006. “Wrongly Accused: Is Race a Factor in Convicting the Innocent?” Ohio State Journal of Criminal Law 4(1):121‐33.

84.

Testa

Johnson

B. D.

. 2020. “Paying the Trial Tax: Race, Guilty Pleas, and Disparity in Prosecution.” Criminal Justice Policy Review 31(4):500‐31. https://doi.org/10.1177/0887403419838025.

85.

Tor

Gazal-Ayal

Garcia

S. M.

. 2010. “Fairness and the Willingness to Accept Plea Bargain Offers.” Journal of Empirical Legal Studies 7(1):97‐116. https://doi.org/10.1111/j.1740-1461.2009.01171.x.

86.

Tversky

Kahneman

. 1992. “Advances in Prospect Theory: Cumulative Representation of Uncertainty.” Journal of Risk and Uncertainty 5(4):297‐323.

87.

Tyler

T. R.

Huo

Y. J.

. 2002. Trust in the Law: Encouraging Public Cooperation with the Police and Courts Through. Russell Sage Foundation. https://muse.jhu.edu/book/19346/.

88.

Tyler

T. R.

Jackson

. 2014. “Popular Legitimacy and the Exercise of Legal Authority: Motivating Compliance, Cooperation, and Engagement.” Psychology, Public Policy, and Law 20(1):78‐95. https://doi.org/10.1037/a0034514.

89.

Ulmer

J. T.

2012. “Recent Developments and New Directions in Sentencing Research.” Justice Quarterly 29(1):1‐40. https://doi.org/10.1080/07418825.2011.624115.

90.

Ulmer

J. T.

Eisenstein

Johnson

B. D.

. 2010. “Trial Penalties in Federal Sentencing: Extra-Guidelines Factors and District Variation.” Justice Quarterly 27(4):560‐92. https://doi.org/10.1080/07418820902998063.

91.

Weber

E. U.

Blais

A. R.

Betz

N. E.

. 2002. “A Domain-Specific Risk-Attitude Scale: Measuring Risk Perceptions and Risk Behaviors.” Journal of Behavioral Decision Making 15(4):263‐90. https://doi.org/10.1002/bdm.414.

92.

Wilford

W. M.

Wells

G. L.

Frazier

. 2020. “Plea-bargaining law: The Impact of Innocence, Trial Penalty, and Conviction Probability on Plea Outcomes.” American Journal of Criminal Justice 46:554‐75.

93.

Wilson

W. J.

1987. The Truly Disadvantaged: The Inner City, the Underclass, and Public Policy. University of Chicago Press, Chicago, IL, USA

94.

Wolfe

S. E.

Nix

Kaminski

Rojek

. 2016. “Is the Effect of Procedural Justice on Police Legitimacy Invariant? Testing the Generality of Procedural Justice and Competing Antecedents of Legitimacy.” Journal of Quantitative Criminology 32(2):253‐82. https://doi.org/10.1007/s10940-015-9263-8.

95.

Woolard

J. L.

Harvell

Graham

. 2008. “Anticipatory Injustice among Adolescents: Age and Racial/Ethnic Differences in Perceived Unfairness of the Justice System.” Behavioral Sciences & the Law 26(2):207‐26.

96.

Wright

D. B.

Skagerberg

E. M.

. 2012. “Measuring Empathizing and Systemizing with a Large US Sample.” PloS ONE 7(2):e31661. https://doi.org/10.1371/journal.pone.0031661.

97.

2016. “Racial/Ethnic Discrimination and Prosecution: A Meta-Analysis.” Criminal Justice and Behavior 43(4):437‐58. https://doi.org/10.1177/0093854815628026.

98.

Cao

. 2018. “Race/Ethnicity, Discrimination, and Confidence in Order Institutions.” Policing: An International Journal 41(6):704‐20. https://doi.org/10.1108/PIJPSM-03-2017-0031.

99.

Yan

Bushway

S. D.

. 2018. “Plea Discounts or Trial Penalties? Making Sense of the Trial-Plea Sentence Disparities.” Justice Quarterly 35(7):1226‐49. https://doi.org/10.1080/07418825.2018.1552715.

Evaluating Racial and Ethnic Invariance Among the Correlates of Guilty Pleas: A Focus on the Effect of Court Legitimacy,Attorney Type,Satisfaction,and Plea-Offer Evaluation

Abstract

Objectives

Methods

Results

Conclusions

Keywords

Racial and Ethnic Disparities in Court Processes and Outcomes

Differences in Legitimacy Across Race and Ethnicity

Defendants and Their Attorneys

The Current Study

Data

Methodological Design

Dependent Variable

Willingness to Accept a Plea (WTAP)

Independent Variables

Defense Attorney Satisfaction (DAS)

Court Legitimacy (CL)

Private Attorney

Additional Measures

Probability of Conviction

Severity Discount

Demographics and Additional Control Variables

Analytic Plan

Results

Initial Willingness to Accept a Plea (WTAPi)

Willingness to Accept a Plea Following an Attorney Evaluation of the Offer (WTAPe)

Discussion

No Direct Effects of Race and Ethnicity on Initial WTAP

Attorney Evaluations of Plea Offers Significantly Influence Defendants’ WTAPe

Race and Ethnicity Moderate the Effect of Attorney Evaluation and Guilt on WTAPe

Summary

Limitations, Future Directions, and Conclusion

Footnotes

Funding

ORCID iDs

Notes

Author Biographies

Appendix A. Scale Items

References

Initial Willingness to Accept a Plea (WTAP_i)

Willingness to Accept a Plea Following an Attorney Evaluation of the Offer (WTAP_e)

Attorney Evaluations of Plea Offers Significantly Influence Defendants’ WTAP_e

Race and Ethnicity Moderate the Effect of Attorney Evaluation and Guilt on WTAP_e