Measuring Collateral Consequences Among Individuals Registered for a Sexual Offense: Development of the Sexual Offender Collateral Consequences Measure

Abstract

Collateral consequences faced by individuals convicted of a sexual offense have been widely referenced in the literature. There is yet to be a systematic examination of collateral consequences affecting individuals, however, due to measurement inconsistencies and the absence of a psychometrically validated instrument. The current study developed and validated a measure of collateral consequences faced by individuals convicted of a sexual offense. Specifically, this study investigated (a) the underlying factor structure of collateral consequences commonly endorsed by individuals convicted of a sexual offense through Exploratory Factor Analysis (EFA) procedures and (b) reliability and validity indicators of the aforementioned scale. Participants were 218 individuals convicted of and registered for a sexual offense in the state of Texas. Study measures included a pool of 66 collateral consequences items in addition to psychological self-report instruments addressing hopelessness, shame, social well-being, and discrimination. EFA results revealed a two-dimensional construct representing collateral consequences affecting areas of social and psychological well-being. The current measure demonstrated adequate reliability and validity. Limitations and future directions of findings are addressed.

Keywords

sexual offender collateral consequences scale development measurement

Although any criminal conviction can carry with it certain sanctions that persist well after the punishment phase, collateral consequences are a stronghold of sexual offense convictions in particular. For individuals convicted of a sexual offense, collateral consequences extend not only from conviction but also from one’s status on the national sex offender registry and subjection to community notification laws, as the public is more frequently made aware of the sexual offender’s denigrated place in society (Tewksbury, 2005).

Collateral consequences have been conceptualized as generally arising from via two disparate mechanisms: formal policy or sanction and informal social control. Formal consequences are more frequently highlighted in law literature, defined as penalties handed directly from the court or sentencing body that serve to limit the civic and social interactions of former offenders through direct law (Logan, 2013; Roberts, 2008; Wheelock, 2005). These include policies limiting housing or employment opportunities, loss of the right to vote, travel restrictions, and any policies impacting public assistance, funding, or grants on account of one’s offender status (Wheelock, 2005). Informal collateral consequences, conversely, are often social in nature. Informal consequences similarly serve to limit civic and social participation for offenders, but are often executed by community members, neighbors, employers, and family members and friends out of self-protection or a desire to distance themselves from the “deviant” individual (Kruttschnitt et al., 2000; Logan, 2013; Uggen & Stewart, 2014).

Formal definitions of collateral consequences are uniquely presented in the law literature (Chin, 2017; Logan, 2013; Pinard, 2010; Pinard & Thompson, 2005). The general consensus of law definitions holds that while ancillary to direct punishment (i.e., prison or jail terms), collateral consequences are upheld by federal, state, or local governments through informal policy (Chin, 2017; Wheelock, 2005). This definition is of particular interest to law scholars, as it illustrates that certain collateral consequences may be protected by law or statute though relatively untouchable by constitutional provisions regulating criminal proceedings. Thus, the definitions of collateral consequences that are available in the literature tend to emphasize formal collateral consequences (e.g., loss of civil status, benefits ineligibility, registration, community notification, residence, and employment restrictions), with fewer examinations into informal sanctions (Logan, 2013). Yet defining collateral consequences in this way can be problematic, as formal collateral consequences are easily conflated with direct consequences. Formal consequences tend to arise through the operation of law and/or accompany direct punishment. Thus, the essence of collateral consequences as unintended, enduring, and social in nature is better encapsulated within the definition of informal collateral consequences.

As defined in the current study, collateral consequences affecting individuals convicted of a sexual offense include socially imposed (rather than court-sanctioned) barriers. These can include discrete experiences of loss or difficulty related to employment (Tewksbury & Zgoba, 2010; Vuolo, 2017), housing (Levenson, 2008; Tewksbury, 2007; Tolson & Klein, 2015), and financial opportunities (Tewksbury & Zgoba, 2010). Other experiences of harassment, discrimination, property damage, and assault have been documented in the literature (Lasher & McGrath, 2012; Levenson & Cotter, 2005a). Similarly, individuals convicted of a sexual offense can experience social exclusion (ten Bensel & Sample, 2019; Tewksbury & Lees, 2006), psychological turmoil (e.g., depression, shame, and hopelessness; Jeglic et al., 2012; Robbers, 2009), isolation and/or interpersonal avoidance (Bailey & Klein, 2018; Burchfield & Mingus, 2008a), and platonic and romantic relationship difficulties (Klein & Bailey, 2016; Winters et al., 2017a). Collateral consequences also encompass victimization of family members of individuals convicted of a sexual offense who may experience secondary housing and employment difficulties, harassment, or emotional problems related to their relative’s membership on the registry (Bailey & Sample, 2017; Levenson & Tewksbury, 2009).

Measurement Considerations

Accounts of collateral consequences faced by individuals convicted of a sexual offense have primarily been documented in descriptive and qualitative formats (Lasher & McGrath, 2012). Beginning with seminal research by Tewksbury (2005), existing collateral consequences measures appear to be derivative of a handful of established items that consistently appear in subsequent measures (e.g., Call, 2017, 2018; Frenzel et al., 2014; Levenson & Tewksbury, 2009). These “original” collateral consequences include challenges to employment, housing, and relationships and discrete experiences of harassment and discrimination in public settings. In addition, typically included are attitudinal items (e.g., negative affect and perceived stigma) surrounding one’s status as a sexual offender (e.g., “I feel like I am being unfairly punished by being on the sex offender registry”; Klein & Bailey, 2016).

A systematic review of the collateral consequences measurement literature, recently conducted by Hamilton (2020), found that studies have primarily reported descriptive data intending to demonstrate the impact of collateral consequences by using sheer numbers (Hamilton, 2020). That is, items tend to broadly target housing restrictions (e.g., “I had to move out of a home I owned”; Levenson, 2008), employment difficulties (e.g., “I lost a job because my boss or coworkers found out I am a sex offender”; Levenson & Cotter, 2005a), interpersonal problems (e.g., “I feel that my close relationships [friendships, family ties] have suffered because I am listed on the sex offender registry”; Klein & Bailey, 2016), and psychological consequences (e.g., “I feel alone and isolated because of Megan’s Law; Levenson & Cotter, 2005a). While the majority of measures/items appeared to emphasize experience with informal consequences in the aforementioned areas, other instruments specifically address formal collateral consequences (i.e., residence and employment restrictions and policy-imposed sanctions; Levenson, 2008; Levenson & Cotter, 2005b; Levenson & Tewksbury, 2009; Tewksbury & Levenson, 2009). Such formal consequences align with legal conceptualizations of collateral consequences (e.g., “an expression of law or legal sanction”; Logan, 2013) though it may be incongruous with the psychological literature.

Currently, there are no established collateral consequences measures that have been validated on a sample of sexual offenders (see Hamilton, 2020). Few studies assessing collateral consequences report scale reliabilities (e.g., Call, 2017, 2018; Harris et al., 2015), whereas no studies referenced indicators of validity or how validity may have been assessed in the process of scale development. Similar limitations noted in the current state of collateral consequences measurement included broad/limited response scales (sacrificing nuance), limited information regarding item development, and little discussion regarding definitional or conceptual underpinnings of collateral consequences (Hamilton, 2020). Results from the systematic review of collateral consequences measures highlighted the need not only for psychometrically valid measurement but for a unified conceptualization of collateral consequences with which to guide instrument development. Specifically, the author recommended maintaining the integrity of the definition of collateral consequences as (a) collateral, rather than direct; (b) unintended, rather than anticipated through policy; and (c) psychosocial in nature (Hamilton, 2020).

In addition, the measurement of collateral consequences has been hindered by lack of operational definition. Legal conceptualizations tend to highlight sanctions arising from policy or law as collateral consequences, which has remained the norm even in the psychological literature (Kirk & Wakefield, 2018). Thus, a more psychosocially oriented operational definition, building off themes from prior research, defined collateral consequences operationally as (a) experiences of discrimination, loss, suppression, or negative affect (e.g., Frenzel et al., 2014); (b) originating societally, interpersonally, or intrapersonally (Logan, 2013); (c) arising as a result of the sexual offender status or policies though independent of law or legal statute (would not be present without subjection to sexual offender policies or label; O’Reilly, 2018); and (4) unintended either in its basic existence or in terms of duration (lasting past the punishment phase; Uggen & Stewart, 2014) or severity (Hamilton, 2020). For example, having one’s personal information posted by a third-party website may be categorized as an incident of discrimination, arising societally/interpersonally, as a result of one’s sexual offender status through independent of legal statute and unintended in its basic existence (not identified as an intended consequence of punishment at the time of conviction). These definitions guided the current study in developing a validated measure of collateral consequences faced by individuals convicted of a sexual offense.

The Current Study

Research attempting to offer a blanketed examination of consequences affecting sexual offenders may be too broad—examinations often conflate direct and collateral consequences, and lack of definitional agreement may muddle the robustness of findings (Kirk & Wakefield, 2018). Although formal consequences may be considered as separate from direct punishment, for purposes of the current study, they were not included under the definition of collateral consequences. This is due to the fact that formal consequences are often conflated with direct punishment, as they tend to be expressed under policy or law, are implemented as a direct consequence of conviction, and are generally considered intended (i.e., made known to offenders during the sentencing and/or punishment phase; Hamilton, 2020). Informal consequences represent a fairly understudied aspect of collateral consequences although it may be most proximal to negative social and psychological outcomes for individuals convicted of a sexual offense (Harris et al., 2015), therefore warranting targeted exploration.

Furthermore, although collateral consequences appear to be implicated in important individual and criminogenic outcomes for sexual offenders (Kruttschnitt et al., 2000; Robbers, 2009; Tewksbury, 2007; Wakefield, 2006), they are yet to be systematically measured in the literature. In sum, much of the extant research reports a variety of collateral consequences endorsed by sexual offenders in frequency alone, without the use of a developed measure. In response to limitations of extant definitions and measurements, this study sought to develop a generalizable measure of collateral consequences faced by individuals convicted of a sexual offense. The current study is exploratory in nature and builds upon what has been documented in the literature by examining more nuanced interpersonal and intrapsychic aspects of collateral consequences. Distinct experiences of collateral consequences were elicited both through a systematic review of the literature and through qualitative interviews with registered individuals convicted of a sexual offense as part of a pilot study to the current research. Finally, the current study developed and validated a comprehensive measure of collateral consequences. Findings from this study have important implications in more cohesively conceptualizing the experience of collateral consequences to eventually determine how these experiences ultimately affect reintegration and desistance from crime, with the overarching goal of informing public safety.

Method

Participants and Procedure

Institutional Review Board (IRB) approval for the current study was granted by the researching university’s IRB in February 2019. We report how we determined our sample size, all data exclusions (if any), all manipulations, and all measures in the study. Participants were 218 individuals convicted of a sexual offense who were registered in the state of Texas. A goal of 200 participants was set, considering the accessibility of potential participants and minimum recommended sample size for factor analysis (Hair et al., 1998). Participants for the current study were recruited via sexual offender support and advocacy group in Texas as well as various sexual offender treatment providers around the central Texas area.

All participants were male, and they ranged in age from 19 to 81 years (M age = 49.18, SD = 13.92). Participants were predominantly White (69.4%). With regard to employment, 67.4% of the sample reported that they were currently employed full time, 11.4% reported part-time employment, and 21.2% reported unemployment at the time of assessment. Please see Table 1 for other specific demographic information. In terms of offense-specific demographic information of the current sample, a range of index offenses was represented. The mean length of time registered was 7.96 years (SD = 7.14). Of note, similar if not lower sexual offender re-offense rates were demonstrated in the current sample in comparison to previous research (e.g., Langan et al., 2003; see Table 2).

Table 1.

Description of the Sample.

Variable	N	%	Valid percentage
Total sample	193
Paper-and-pencil participation	131	68	68
Online participation	62	32	32
Race/ethnicity (n = 190)
Caucasian	134	69.4	70.5
Black/African American	20	10.4	10.5
Hispanic/Latino	25	13.0	13.2
Other	11	5.7	5.9
Educational level (n = 178)
Less than high school	15	7.8	8.4
High school or equivalent	67	34.8	37
Some college	34	17.6	19.1
College degree (Bachelors and Associates)	40	20.7	22.5
Graduate degree	22	11.4	12.4
Marital status
Single, never married	55	28.5	28.9
Married or in domestic partnership	73	37.8	38.4
Divorced	45	23.3	23.7
Widowed	7	3.6	3.7
Separated	9	4.7	4.7
Mean household income (dollars)
Prior to conviction	$68,440
Post-conviction	$48,806

Table 2.

Offense-Specific Characteristics of the Sample.

Variable	N	%	Valid percentage
Index offense type (n = 179)
Sexual assault of a child	12	6.2	6.7
Continuous sexual abuse of a child	1	0.5	0.6
Aggravated sexual assault of a child	16	8.3	8.9
Online solicitation of a minor	7	3.6	3.9
Indecency with a child by contact	29	15.0	16.2
Indecency with a child by exposure	6	3.1	3.4
Invasive visual recording/improper photography	1	0.5	0.6
Indecent exposure	3	1.6	1.7
Sexual assault	16	8.3	8.9
Aggravated sexual assault	18	9.3	10.1
Possession of child pornography	45	23.3	25.1
Distribution of child pornography	5	2.6	2.8
Two or more distinct offenses	13	6.7	7.3
Other	7	3.6	3.9
Assigned level of risk (n = 180)
Low	95	49.2	52.8
Moderate	56	29.0	31.1
High	5	2.6	2.8
Never assigned/unsure	24	12.4	13.3
Sex offender treatment status
In treatment	131	68	68
Completed treatment	62	32	32
Re-offense (n = 180)
Arrested since index offense	46	23.8	25.6
Arrested for sexual offense since index offense	7	3.6
Arrested or charged with probation/parole violation	51	26.4	28.3

Measures

A demographic questionnaire, collateral consequences item pool, and several related measures were included in the current study. Measures of general discrimination, social inclusion, shame, and hopelessness were included primarily to establish convergent validity with the collateral consequences measure and to preliminarily demonstrate empirical links between collateral consequences and psychosocial variables of interest.

Demographic questionnaire

Demographic data of the participants were elicited regarding age, race/ethnicity, level of education, employment status, marital status, current and past family income, offense name/description, assigned level of risk, treatment status, and the number of years spent on the registry.

Collateral consequences

An item pool assessing collateral consequences was developed for psychometric testing in the current study. The definition that guided scale development follows the operational definition proposed by Hamilton (2020) that collateral consequences are (a) experiences of discrimination, loss, suppression, or negative affect; (b) originating societally, interpersonally, or intrapersonally; (c) arising as a result of the sexual offender status or policies though independent of law or legal statute (would not be present without subjection to sexual offender policies or label); and (d) unintended either in its basic existence or in terms of duration (lasting past the punishment phase) or severity.

Items for the current measure were gathered in four primary ways. First, a systematic review of existing collateral consequences questionnaires used within a sexual offender population was conducted with the purpose of collating existing scale items (Hamilton, 2020). Second, relevant theoretical and law literature addressing collateral consequences was informally reviewed by the authors for additional discussion of collateral consequences. Third, experts in the field of collateral consequences and the sexual offender registry were consulted. Specifically, two experts who have published extensively in the field of collateral consequences were contacted via email in 2018 to provide feedback on the item pool.

Pilot study

In addition to the systematic review, review of theoretical and law literature, and consultation with experts, a pilot study was conducted in 2018 to include a small sample of registered sexual offenders separate from the current participant pool. A total of N = 38 registered individuals in Texas participated, which allowed for pilot testing of collateral consequences items as well as the collection of qualitative data related to other experienced collateral consequences of registration. Participants were recruited from an outpatient sexual offender treatment program in Texas. Participants were all men, with a mean age of 44.2 years. The average time spent on the registry was 11.7 years. An initial 41-item questionnaire, which was developed based on the aforementioned literature reviews and common items from existing sources, was disseminated to participants. Participants were asked to provide feedback on the questionnaire and were encouraged to provide narrative responses with regard to any additional experiences with collateral consequences. Based on the data gathered from this study, items assessing situational/behavioral avoidance were added as well as items addressing exploitation, traveling/visiting with family, various aspects of community involvement, and “positive” collateral consequences (e.g., motivated to prove character due to registry).

From the aforementioned sources, a 66-item pool was generated to assess participants’ experiences with collateral consequences. Items were written to emphasize collateral consequences experienced secondary to formal sanctions and to highlight the psychosocial factors that may trigger experience with collateral consequences (e.g., social pressure, publicization of status, fear of legal repercussion, fear of being found out, and functional avoidance). Responses were rated along a 5-point Likert-type scale, with 1 = “Never/0 times,” 2 = “Rarely/1–3 times,” 3 = “Sometimes/4–6 times,” 4 = “Often/7–9 times,” and 5 = “Always/10+ times.” This response scale was chosen (a) based on pilot data and the frequency with which individuals reported experiencing collateral consequences (i.e., multiple, rather than singular, experiences of discrimination, harassment, and loss), (b) based on conferring with experts, (c) based on response scales used in development studies of similar scales (e.g., assessing discrimination experiences; Nadal, 2011), and (d) to provide enough responses to be able to capture a range of frequency responses (5–9 being optimal; Spector, 1992). See Appendix A for the full item pool.

General discrimination

The 9-item Experiences of Discrimination scale (EOD; Krieger et al., 2005) measures self-reported experiences with discrimination. Participants were asked to indicate whether they have ever experienced discrimination in nine domains (e.g., “Getting hired or getting a job?” and “On the street or in a public setting?”). Responses were rated along a 4-point scale, where 0 = “Never,” 1 = “Once,” 2 = “Two or three times,” and 3 = “Four or more times.” Both a frequency and situational score were computed. The EOD has demonstrated good internal consistency across populations as well as high test–retest reliability and construct and concurrent validity (Krieger et al., 2005). The EOD references discrimination due to race, ethnicity, gender, sexual orientation, and social class. The instructions in the current study asked participants to complete the items based on experiences of discrimination due to their status on the sexual offender registry. Cronbach’s alpha for the current study was .81.

Social inclusion

The Social Inclusion Scale (SIS; Secker et al., 2009) is a 22-item measure that assesses perceived social inclusion across three domains: Social Isolation (five items; e.g., “I have felt terribly alone and isolated”), Social Relations (nine items; e.g., “I have been to new places”), and Social Acceptance (five items; e.g., “I have felt accepted by my neighbors”). The SIS has been used with diverse populations including individuals with psychiatric diagnoses and students (Secker et al., 2009; Wilson & Secker, 2015). Responses were rated along a 4-point scale, where 1 = “Not at all” and 4 = “Yes definitely.” The full-scale SIS and its subscales have demonstrated good internal consistency, construct validity, and predictive validity, (Margrove et al., 2013; Secker et al., 2009). For use in the proposed study, items were modified slightly to reference individuals convicted of a sexual offense rather than psychiatric patients (e.g., “I have felt some people look down on me because of my sexual offender status”). Cronbach’s alpha was .73 for the Social Isolation subscale, .66 for the Social Acceptance subscale, and .78 for the Social Relations subscale. Cronbach’s alpha for the total scale was .89.

Shame

Shame was measured using the Test of Self-Conscious Affect–Socially Deviant Version (TOSCA-SD; Hanson & Tangney, 1996). The TOSCA-SD consists of 13 brief scenarios, followed by four responses rated on a 5-point scale, measuring an individual’s shame-proneness, guilt-proneness, and externalization (i.e., attributing blame to others) reactions. The TOSCA-SD is comprised of five subscales to represent disparate shame responses: Constructive Guilt (13 response items), Externalization (13 response items), Detachment (13 response items), Shame–Negative Self-Appraisal (five response items), and Shame–Behavioral Avoidance (eight response items). The TOSCA-SD has demonstrated adequate reliability and validity among incarcerated sexual offenders (Hanson, 1996; Tangney et al., 2011) as well as criterion validity among offender populations in its relationship with recidivism (Tangney et al., 2014). Cronbach’s alpha was .81 for the Constructive Guilt subscale, .81 for the Externalization subscale, .76 for the Detachment subscale, .54 for the Shame–NA subscale, and .67 for the Shame–BA subscale. These reliabilities (particularly lowered reliabilities in the Shame–NA and Shame–BA subscales) parallel those encountered by Tangney et al. (2011).

Hopelessness

The Beck Hopelessness Scale (BHS; Beck et al., 1974) is a 20-item self-report measure that assesses negative expectations about one’s future. Responses were scored in a true/false format, with a total score ranging from 0 to 20. Higher scores indicated a greater degree of hopelessness, with a score of 9–14 indicating moderate hopelessness, and scores >15 indicating severe hopelessness. Half of the scale items are negatively worded (e.g., “I might as well give up because I can’t make things better for myself”), while half are positively worded and reverse coded (e.g., “I look forward to the future with hope and enthusiasm”). This measure has demonstrated content validity, concurrent validity with other measures of negative attitudes, construct validity, and internal consistency in both clinical and nonclinical samples (Beck et al., 1974; Steed, 2001). Cronbach’s alpha for the current study was .93.

Results

Preliminary Analyses

Preliminary analyses were conducted to scan the data set for input errors, missing data, outliers, and normality. Prior to assessing for normality, one collateral consequences item was removed from the data set due to a large amount of missing data (32.1%; item 42, “I have been denied housing due to my status on the registry”). Thus, normality and missing data analyses were conducted with the remaining 65 items.

Missing data were dealt with using listwise deletion and available item analysis (AIA). Listwise deletion was first carried out by removing from the data set cases missing more than 20% of overall data (an appropriate level of tolerance in accordance with AIA recommendations; Parent, 2013) and those that did not meet minimum requirements for participation. With the resulting N = 193 data sets, cases were inspected for rates of missing data on both collateral consequences items and included measures. The level of missing data for the collateral consequences items was minimal (see Supplementary Table 1). Further, missing data among the collateral consequences items were judged to be missing at random [MAR] according to Little’s test (χ² = 771.00, p = .432). Thus, AIA (i.e., pairwise deletion/inclusion) was appropriate for the current data set, and available data (rather than imputation) were used to create mean scores.

The level of missing data for the five included measures was also minimal. Little’s MCAR (missing completely at random) test was then conducted among the five included measures. The majority of the measures demonstrated nonsignificant chi-square tests, though the BHS and the TOSCA-SD Constructive Guilt subscales suggested a pattern to the missing data (p = .03). In addition to examining results from Little’s MCAR test, data on the two potentially MNAR (missing not at random) scales were inspected for obvious patterns of missing data, such as abnormally high rates of missing data on a few select items. Although Little’s test lends support as to whether data are MAR (missing at random)/MCAR, data may also be assumed missing at random if there is no clear pattern to the missingness (Parent, 2013). No clear bias in the missingness was apparent for the BHS or TOSCA-SD Constructive Guilt scales; the greatest proportion of missingness was 1.6% on a single item. Thus, the data were assumed missing at random and were judged appropriate for further analysis.

Finally, normality was assessed through an examination of skewness and kurtosis for the 65 collateral consequences items. To address non-normality, six collateral consequences items were removed from further analyses due to skewness/kurtosis levels well more than the recommended ±3.29 cutoff (Ghasemi & Zahediasl, 2012). See Supplementary Table 2 for normality data of the 65 collateral consequences items. In addition, the results of tests of non-normality for the five included measures are presented in Table 3.

Table 3.

Tests of Normality on Included Measures.

	Skewness	Kurtosis	Shapiro–Wilk
Measure	Skewness	Kurtosis	Statistic	df	Sig.
EOD
Frequency	0.953	0.573	.913	193	<.001
Situation	0.460	−0.383	.941	193	<.001
SIS
Social isolation	0.085	−0.534	.986	193	.045
Social relations	−0.182	−0.163	.984	193	.026
Social acceptance	−0.012	−0.144	.996	193	.940
TOSCA-SD
Constructive guilt	−2.077	6.653	.815	193	< .001
Externalization	0.626	−0.312	.951	193	<.001
Detachment	0.097	−0.339	.993	193	.439
Shame–Negative Self Appraisal	−0.232	−0.550	.978	193	.003
Shame–Behavioral Avoidance	0.676	0.477	.958	193	<.001
BHS	0.913	−0.251	.874	193	<.001

Note. BHS = Beck Hopelessness Scale; EOD = Experiences of Discrimination; SIS = Social Inclusion Scale; TOSCA-SD = Test of Self-Conscious Affect–Socially Deviant.

Exploratory Factor Analysis

Exploratory factor analysis (EFA) was conducted to determine the factor structure of the collateral consequences items. EFA assumes that underlying or latent constructs exist within the data that, through a number of statistical rotation techniques, give rise to manifest factors (Osborne, 2015). Principal axis factoring (PAF) was chosen as the factor extraction method for use with the current data. Comparisons of PAF and other extraction methods, such as maximum likelihood, found that PAF generally produces similar results to other extraction methods, may perform best with non-normal data, and provides reliable factor results (Osborne, 2015).

First, to ensure that the data were suitable for EFA, sampling adequacy was tested via three means: Bartlett’s (1950) test of sphericity, the Kaiser–Meyer–Olkin (KMO) measure of sampling adequacy, and by examining item intercorrelations. First, Bartlett’s test of sphericity provides a measure of scale factorability and is used to estimate the probability that the matrix correlations are zero. Bartlett’s test indicated the data were suitable for reduction (χ² = 8591.93, p < .001). Second, the KMO measure of sampling adequacy indicates the extent to which the correlation matrix between a subset of variables contains factors versus chance correlations. KMO values of .60 or higher indicate good factorability (Tabachnick & Fidell, 2001). The KMO value obtained with the current data was .89, indicating sampling adequacy. Last, the item correlation matrix was then examined to ensure that each of the items correlated with at least one other item; it is suggested that items that do not moderately correlate (≥.3) with any other item included in the factor analysis may not represent well the latent construct being measured (Tabachnick & Fidell, 2001). All 59 items correlated with at least one other item at .3 or above.

Once the data were found suitable for reduction, PAF was performed with the remaining 59 collateral consequence items to determine the number of meaningful factors to retain. Four methods were used to determine factor selection: eigenvalues, scree test, parallel analysis, and the proportion of variance accounted for by each factor. It is generally best to use multiple factor selection criteria to support decision-making in EFA, as the process in itself is subjective by nature (Costello & Osborne, 2005; Worthington & Whittaker, 2006). Research has recommended that all rationale and decisions made during EFA are clearly documented and supported for the purpose of transparency (Worthington & Whittaker, 2006).

Initial eigenvalues were examined prior to rotation. Potential factors are indicated by eigenvalues >1.0 (Kaiser’s rule; Kaiser, 1960), though it is highly suggested that multiple methods are used for factor selection (e.g., scree test and variance thresholds). The initial model suggested 13 factors with eigenvalues >1.0, explaining 71.76% of the total variance. Next, a scree plot was inspected (scree test; Cattell, 1966), particularly the point of inflection where eigenvalues “drop off,” indicating the number of ideal factors to retain (the point at which variance accounted for by each factor loading becomes negligible). The scree plot indicated a plateau effect after Factor 3, suggesting a parsimonious three-factor solution. A shortcoming to relying on scree tests in factor retention decisions is its ambiguity and subjectivity, particularly with smaller sample sizes and a low variable to factor ratio (Gorsuch, 1983; Tabachnick et al., 2007). Furthermore, the scree tests, like Kaiser’s rule, may tend to overestimate the number of factors to extract (Henson & Roberts, 2006).

Parallel analysis is touted as one of the more accurate methods in determining the number of factors to retain though it is often underreported in the literature (Henson & Roberts, 2006; Zwick & Velicer, 1986). Parallel analysis was conducted in SPSS to reveal an optimal factor solution (O’Connor, 2000); however, results from parallel analysis supported a 7-factor model. Parallel analysis, similar to other selection methods, and when conducted with principal axis factoring, tends to over-extract factors (Buja & Eyuboglu, 1992). O’Connor (2000) suggests considering the results of the parallel analysis in conjunction with other factor selection methods, including interpretability, to trim trivial factors.

Finally, the variance accounted for by each factor was examined. It has been recommended that the proportion of variance accounted for by the factor be at least 5% to 10% to justify retention (Suhr, 2006). Prior to rotation, Factor 1 accounted for 32.41% of the variance, Factor 2 accounted for 7.80% of the variance, and Factor 3 accounted for 4.70% of the variance. Using the lower threshold suggested by Suhr (2006), factors accounting for >5% of the variance were retained. Because (a) eigenvalue analysis, scree tests, and parallel analysis tend to overestimate the number of factors to be extracted and (b) the central tenet of factor analysis is to explain the greatest amount of variance in the construct with the fewest number of factors possible, it was decided that the lower estimate of a two-factor model was most parsimonious.

Once a two-factor model was selected, a factor rotation method was chosen. An oblique rotation was chosen due to the hypothesized overlap in the variance among factors, given the nature of the construct of collateral consequences (a constellation of discrimination and disenfranchisement experiences associated with negative behavioral and psychological consequences; Henson & Roberts, 2006; Tabachnick & Fidell, 2001). Promax was chosen as the specific oblique rotation method; researchers generally recommend rotating with Promax if using an oblique rotation method (Gorsuch, 1983; Kim & Mueller, 1978). Item-specific data (i.e., communalities, loadings, and cross-loadings) were then examined to determine the need for deletion. Item communalities were examined to ensure that each item (a) was related to the other items and (b) did not represent an ill-fitting factor that should not be explored. Four items were deleted during EFA due to communalities <.40 (Costello & Osborne, 2005).

Next, factor loadings, or correlations between the variable and factor, were analyzed. It is suggested that variables with loadings >0.32 are suitable for retention (Tabachnick & Fidell, 2001). All items in the two-factor model demonstrated loadings >0.32; no items were removed due to low factor loadings. In addition, cross-loadings were examined to ensure that factors did not share items that cross-load too highly on more than one factor. Two items were removed from analysis during EFA for demonstrating high (≥.32) cross-loadings on multiple factors (Items 64 and 61; Costello & Osborne, 2005; Tabachnick & Fidell, 2001).

Finally, interpretability was considered when deciding whether to retain or remove variables from a factor, that is, whether or not the item was meaningful and made conceptual sense as part of the factor. In interpreting the factors, Factor 1 appeared to represent psychological consequences of sexual offense conviction/registration, such as hopelessness, avoidance, and perceived isolation. Factor 2 appeared to represent distinct incidents of discrimination and harassment due to sexual offense conviction/registration. A two-factor, Promax-rotated solution with 20 items was ultimately chosen. The two-factor model explained the greatest amount of variance (41.99%) with the fewest items. Factor 1 was labeled “Psychological Consequences” (13 items), and Factor 2 was labeled “Social Consequences” (seven items; See Table 4).

Table 4.

Pattern and Structure Coefficients, Communality Estimates, Alpha Coefficients, and Means/Standard Deviations.

20 items (α = .94)	Pattern coefficient		Structure coefficient		h ²	M	SD
20 items (α = .94)	F1	F2	F1	F2	h ²	M	SD
Factor 1: Psychological Consequences (13 items; α = .94, 33.47% variance)
65. I am constantly afraid someone will find out I am on the registry.	1.068	−.111	.815	.363	.742	3.60	1.48
66. I feel shame due to my status on the registry.	0.887	−.053	.599	.238	.496	4.04	1.33
55. I have felt it is difficult to move on with my life because being on the registry frequently reminds me of my offense.	0.827	−.086	.707	.351	.567	3.55	1.39
47. I have avoided making plans for the future because I feel I will be stigmatized wherever I go.	0.766	−.042	.816	.423	.717	3.22	1.37
46. I have avoided things I used to enjoy because I feel ashamed or embarrassed being on the registry.	0.747	.026	.804	.446	.692	3.21	1.38
50. I have felt like an outcast like I’m not a part of a community or society at large.	0.716	.003	.835	.467	.736	3.66	1.43
52. I have felt that I had to hide parts of myself in order to not be rejected by others.	0.713	−.047	.752	.399	.656	3.51	1.43
49. I have avoided participating in things I used to enjoy because I am tired of looking over my shoulder and worrying that someone will find out about me.	0.620	.085	.799	.528	.690	3.13	1.35
59. I have isolated myself for fear of being found out as a sex offender.	0.598	−.013	.769	.401	.688	3.18	1.28
51. I have felt unable to create meaning in my life, because of my status on the registry.	0.585	−.070	.740	.391	.656	3.24	1.45
57. I have avoided social situations for fear of my safety.	0.452	.149	.613	.478	.523	2.46	1.30
45. I have avoided social situations for fear of my family's safety.	0.450	.168	.679	.545	.641	2.26	1.31
48. I have felt helpless like, “no one believes I can change so why even try?”	0.389	−.033	.599	.378	.497	2.70	1.46
Factor 2: Social Consequences (7 items; α = .90, 8.22% variance)
26. My family member or cohabitant was treated rudely/harassed because I am on the registry.	0.021	.876	.436	.801	.654	1.86	1.15
38. I was called out in public because someone recognized me as a sex offender.	−0.127	.861	.298	.694	.524	1.49	0.96
25. I received harassing telephone calls/mail/flyers/notes because someone found out I am a sex offender.	−0.005	.839	.320	.747	.637	1.61	1.10
23. I was harassed in person because someone found out I am a sex offender.	0.004	.806	.432	.817	.716	2.22	1.31
22. I was treated rudely or snubbed because someone found out I am a sex offender.	0.030	.771	.513	.808	.730	2.85	1.31
29. Neighbors treated me poorly or expressed concern about me because I am on the registry.	–0.123	.709	.354	.742	.605	1.87	1.15
33. I was asked to leave a business, restaurant, or public place because I am on the registry.	–0.165	.580	.257	.587	.430	1.41	0.87

Note. h² = communality estimates. Factor loadings greater than 0.32 are in boldface.

Reliability and Validity

Once the final factor solution was achieved, Cronbach’s alphas for each subscale were calculated as a measure of internal consistency. Cronbach’s alpha was .94 for the Psychological Consequences subscale, .90 for the Social Consequences subscale, and .94 for the total Sexual Offender Collateral Consequences Measure (SOCCM).

Content validity for the proposed measure was preliminarily established via consulting experts in the field of collateral consequences faced by individuals convicted of a sexual offense. A pilot study was also conducted to assess the relevance of the items and to gather qualitative information from individuals convicted of a sexual offense regarding common experiences with collateral consequences (see Method section for a full discussion of expert consultation and results from the pilot study).

Construct validity of the SOCCM was similarly assessed. A variety of measures were included that are thought to relate to both the Social Consequences and the Psychological Consequences subscales of the collateral consequences measure. The Social Consequences subscale was strongly, positively correlated with both the situation (r = .69, p < .001) and the frequency (r = .73, p < .001) scores of the EOD scale. The Psychological Consequences subscale was significantly correlated with the BHS (r = .46, p < .001) and the Social Isolation subscale of the SIS (r = −.50, p < .001). Subscales of the TOSCA-SD were used to demonstrate discriminant validity, as certain expressions of shame (e.g., detachment and externalization) were thought to be conceptually dissimilar from collateral consequences. The Social Consequences and Psychological Consequences subscales were not significantly related to the Detachment subscale of the TOSCA-SD (r = −.01, p = .89 and r = −.01, p = .93, respectively) and the Externalization subscale of the TOSCA-SD (r = .06, p = .41 and r = .11, p = .20, respectively). See Table 5 for scale intercorrelations.

Table 5.

Scale Intercorrelations.

Measure	1	2	3	4	5	6	7	8	9	10	11	12	13
1. CC-Psych	—
2. CC-Social	.48**	—
3. EOD-F	.53**	.73**	—
4. EOD-S	.54**	.69**	.92**	—
5. SIS-SI	−.50**	−.18*	−.22**	−.19**	—
6. SIS-SR	−.48**	−.19**	−.16*	−.14	.69**	—
7. SIS-SA	−.53**	−.37**	−.44**	−.40**	.64**	.50**	—
8. TOSCA-CG	.24**	.12	.17*	.14*	−.07	−.04	−.04	—
9. TOSCA-E	.06	.11	.08	.07	−.05	−.05	−.03	−.42**	—
10. TOSCA-D	−.01	−.01	−.02	−.01	.02	.01	.07	−.35**	.67**	—
11. TOSCA-NS	.26**	.17*	.27**	.26**	−.14	−.10	−.22**	.23**	.05	−.14	—
12. TOSCA-BA	.03	.09	−.05	.05	−.09	−.09	−.05	−.42**	.54**	.42**	.33**	—
13. BHS	.46**	.18*	.32**	.28**	−.40**	−.47**	−.46**	−.09	.14	.12	.18*	.23**	—

Note. BHS = Beck Hopelessness Scale; CC-Psych = Psychological Collateral Consequences; CC-Social = Social Collateral Consequences; EOD-F = Experiences of Discrimination-Frequency; EOD-S = Experiences of Discrimination-Situation; SIS-SI = Social Inclusion Scale-Social Isolation; SIS-SR = Social Inclusion Scale-Social Relations; SIS-SA = Social Inclusion Scale-Social Acceptance; TOSCA-CG = Test of Self-Conscious Affect–Constructive Guilt; TOSCA-E = Test of Self-Conscious Affect–Externalization; TOSCA-D = Test of Self-Conscious Affect–Detachment; TOSCA-NS = Test of Self-Conscious Affect–Negative Self-Appraisal; TOSCA-BA = Test of Self-Conscious Affect–Behavioral Avoidance.

p < .05. **p < .01.

Criterion validity indicates the extent to which a measure is related to a particular outcome and is represented through a combination of concurrent and predictive validity. Concurrent validity for the Social Consequences subscale (measuring discrimination experiences) was demonstrated via the robust, positive correlation with the EOD frequency and situation subscales. Concurrent validity for the Psychological Consequences subscale (measuring psychological distress in relation to an individual’s sexual offense) was demonstrated via a robust, positive correlation with the BHS. Because the sample was assessed at one time point, only concurrent validity was established. The full SOCCM is provided in Appendix B.

Discussion

Investigation into the factor structure of a measure of collateral consequences revealed a two-factor solution representing social (i.e., discrete experiences of harassment and discrimination related to one’s offense/status on the registry) and psychological (i.e., internal experiences of psychological distress related to one’s offense/status on the registry) collateral consequences. Results from the current study build upon previous literature that has typically defined collateral consequences bidimensionally (e.g., formal versus informal and direct versus indirect). In contrast to previous research, however, items assessing the experience of formal collateral consequences, such as subjection to registration, community notification, and probation stipulations, were excluded due to their likeness to direct punishment arising from the fact of conviction rather than collaterally. Rather, a psychosocial (rather than legal) framework, including measurement of informal (rather than formal) collateral consequences, was prioritized.

Results from the current study suggest that collateral consequences may be best understood as social and psychological in nature. These results parallel thematic findings of Lasher and McGrath (2012) who categorized the predominant areas of reintegration impacted by sexual offender policies as social (e.g., loss of social supports and social safety) and psychological (e.g., stress, shame, and hopelessness). Several other studies similarly emphasize the social and psychological aspects of collateral consequences (Brannon et al., 2007; Levenson & Cotter, 2005a; Levenson et al., 2007; Mercado et al., 2008; Robbers, 2009; Tewksbury, 2005). Thus, although the current study’s conceptualization was framed around bidimensional conceptualizations of formal and informal collateral consequences, the specific language of social and psychological collateral consequences both (a) accurately captured the essence of collateral versus direct consequences and (b) fit the data best.

As predicted, the social and psychological collateral consequences factors correlated strongly with one another so that greater reported discrimination and harassment experiences were related to psychological distress surrounding one’s offense/status on the registry. These findings were parallel to those of Jeglic et al. (2012) who found that individuals convicted of a sexual offense who experienced more harassment experiences also reported greater levels of depression and hopelessness than offenders who reported few to no negative experiences. Although directionality cannot be established from the current results, the association between social and psychological collateral consequences highlights the cruciality of discrete social exclusion experiences on psychological well-being in relation to one’s offender status.

Perhaps expectedly, items retained in the current collateral consequences measure mapped on adequately with themes from prior research. For example, included in the Social Consequences subscale were items assessing personal harassment, harassment of family members and cohabitants, and whether the participant had been asked to leave an establishment, each of which were highlighted in previous qualitative and descriptive research addressing collateral consequences (Frenzel et al., 2014; Tewksbury & Mustaine, 2009). Although collateral consequences have not previously been represented by a validated measure, it appears that typical experiences with external consequences of discrimination and harassment have been well represented in existing descriptive studies.

In contrast, the Psychological Consequences subscale included items addressing fear, shame, behavioral and psychological avoidance, isolation, stagnancy, and helplessness in relation to the “sexual offender status.” While psychological consequences associated with a sexual offense have been broached in the literature, seminal research has typically focused on discrete experiences of discrimination and harassment, particularly surrounding employment and housing (Frenzel et al., 2014; Tewksbury & Mustaine, 2009; Zevitz & Farkas, 2000). Current findings emphasizing a negative psychological component to collateral consequences both parallel and build upon previous research. Studies have shown that individuals subjected to sexual offender policies report greater negative mental health outcomes (Edwards & Hensley, 2001; Lasher & McGrath, 2012; Levenson & Cotter, 2005a). Prior measures of collateral consequences, however, have typically included one or two items assessing shame, hopelessness, and fear in particular (Harris et al., 2015; Levenson, 2008; Levenson & Cotter, 2005b).

In addition to assessing the aforementioned aspects of negative affect, a contribution of the current study was the inclusion of items assessing behavioral and psychological avoidance in response to the sexual offender status (e.g., “I have avoided making plans for the future because I feel I will be stigmatized wherever I go”). Another concept highlighted in the current subscale that has been relatively underrepresented in previous research is the perceived ability to create meaning in one’s life (e.g., “I have felt unable to create meaning in my life because of my status on the registry”). These intrapsychic experiences represent an empirically novel aspect of collateral consequences and may help explain the relationship between collateral consequences and other psychosocial factors implicated in rehabilitation and reintegration.

Study Limitations and Future Directions

Findings from the current study should be interpreted in light of a few notable limitations. First, the current study was exploratory in nature. Primarily, the purpose of the study was to develop a reliable and valid measure that could empirically explain the structure of collateral consequences as they impact individuals convicted of a sexual offense. Therefore, although the current study evidenced a preliminary organizational framework for collateral consequences, substantive conclusions regarding the factors underlying collateral consequences should not be drawn. Future research would benefit from executing confirmatory factor analysis to test hypotheses more systematically with the use of inferential statistical techniques.

Second, the robustness of the current findings was weakened due to (a) a moderate to small sample size and (b) lower explained variance. Although the sample size utilized in the current study was adequate for exploratory analyses based on guidelines provided by some researchers (Hair et al., 1998), other researchers have argued that larger sample sizes help to better determine the validity of factor structure and individual items (Costello & Osborne, 2005). As noted by Worthington and Navarro (2003), however, populations from which there are constraints on participant recruitment, or groups that represent a unique subpopulation wherein variance may be lower compared to a general representative sample, analyses may have to progress with an adequate though nonideal sample size. Furthermore, with regard to explained variance, lower variance is typically expected within research conducted in the social sciences and humanities (Hair et al., 1998; Williams et al., 2010).

Third, participant responses to the collateral consequences items were likely influenced by the city, state, or otherwise jurisdictional mandates surrounding housing and other restrictions. For example, items assessing housing and employment difficulties did not load strongly on either factor, contrary to prior research identifying these restrictions as the foremost collateral consequences (Levenson, 2008; Levenson & Tewksbury, 2009; Tewksbury, 2007).

Other items assessing social collateral consequences may be more sensitive to jurisdictional differences, which implicates the generalizability of findings to the broader population of individuals convicted of a sexual offense. This complicating factor speaks to the difficulty in drawing conclusions surrounding the experiences of individuals convicted of a sexual offense as a homogeneous unit. For example, city- and state-imposed restrictions tend to be poorly documented and can contradict one another (Monjeau, 2011). In Texas, for example, while state law includes child safety zone statutes, determined by a parole panel for offenders with victims under 17 years of age, residency restrictions are open to home rule and determined by the city rather than state government. In response, to capture the confusion reported by individuals convicted of a sexual offense regarding enforceable social restrictions (e.g., residence restrictions, buffer zones, and accessibility of public spaces), as well as to prioritize informal consequences, proxy items were included in the current study to instead assess avoidance of situations and events (e.g., due to fear of violating specific policy stipulations). Even so, threats to the widespread applicability of the current measure are present.

Other threats to generalizability include varying years spent on the registry and/or in treatment. Participants were in varying stages of the treatment process; treatment participation and/or completion may have affected participants’ ability to articulate their experiences with collateral consequences. Furthermore, time spent on the registry certainly influences one’s experience with collateral consequences due to sheer time. Future research should consider exploring treatment completion and years registered as mediators between collateral consequences and psychosocial outcomes. Finally, as highlighted by Craun and Bieri (2014), two major limitations to the current study are the lack of comparison groups and the priming effect inherent to self-report studies, each of which may serve to inflate conclusions regarding the harm of the registry.

Practical Implications

In developing and validating the current measure of collateral consequences, it is hoped that research in the area of outcomes for individuals convicted of a sexual offense will advance. Although actuarial and dynamic risk measures continue to make advancements for use with individuals who sexually offend (Harris & Hanson, 2010), instruments gauging the subjective offender experience have lagged in comparison. In fact, researchers have called for increased evidence-based tools and recommendations in addition to structured risk assessments, to advocate for policy reform (Levenson, 2018). In developing the current measure, examinations into the collateral consequences faced by individuals convicted of a sexual offense can more easily transition from purely theoretical and descriptive to an evidence-based approach. For example, research has stated that collateral consequences may impact recidivism as a function of social punishment (e.g., Prescott & Rockoff, 2011; Sandler et al., 2008; Wakefield, 2006; Zevitz, 2006). Ideally, the current measure may help bolster major arguments surrounding collateral consequences through more stringent empirical examination.

A bidimensional conceptualization of social and psychological collateral consequences also offers implications with regard to risk assessment and treatment. As researchers continue to empirically link collateral consequences and criminogenic outcomes, certain experiences with collateral consequences may emerge as dynamic risk factors implicated in sexual reoffending. Social collateral consequences in particular may contribute to risk of reoffending as a function of reduced social support, a risk factor included in existing risk assessments (e.g., LSI-R, Andrews & Bonta, 2001). Psychological consequences may also be implicated in risk. Negative/hopeless outlook and lack of plans are highlighted both in the SOCCM and in existing risk assessment instruments (e.g., SVR-20; Boer et al., 1997). Thus, measuring collateral consequences may both complement existing methods of risk assessment and identify additional social and psychological risk factors warranting inclusion in future instruments. With regard to both risk assessment and treatment, examining collateral consequences using the SOCCM lends itself well to the risk-needs-responsivity approach (Bonta & Andrews, 2010). Assessment-based identification of collateral consequences can more easily target social and psychological risks, needs, and individual attributes through intervention.

Conclusion

Overall, findings from this study suggest that individuals experience ongoing consequences, attached to their conviction, in both the social domain through discrete experiences of discrimination and harassment and the psychological domain through ongoing negative affect and poor self-appraisal. Subsequently, these findings suggest a two-dimensional structure underlying the often ambiguously defined term of collateral consequences, offering a guiding framework for future studies. The current study also lays the empirical groundwork in support of the humanistic argument against current policies affecting individuals convicted of a sexual offense, with an emphasis on demonstrating how social and psychological collateral consequences may further disenfranchisement among convicted individuals, families, and the public. Regardless of ideological camp, few researchers, policymakers, or legal professionals would argue the importance of promoting desistance among individuals convicted of a sexual offense with one overarching goal—to prevent future victims.

Supplemental Material

sj-docx-1-sax-10.1177_10790632211019733 – Supplemental material for Measuring Collateral Consequences Among Individuals Registered for a Sexual Offense: Development of the Sexual Offender Collateral Consequences Measure

Supplemental material, sj-docx-1-sax-10.1177_10790632211019733 for Measuring Collateral Consequences Among Individuals Registered for a Sexual Offense: Development of the Sexual Offender Collateral Consequences Measure by Emma Hamilton, Delida Sanchez and Matthew L. Ferrara in Sexual Abuse: A Journal of Research and Treatment

Supplemental Material

sj-docx-2-sax-10.1177_10790632211019733 – Supplemental material for Measuring Collateral Consequences Among Individuals Registered for a Sexual Offense: Development of the Sexual Offender Collateral Consequences Measure

Supplemental material, sj-docx-2-sax-10.1177_10790632211019733 for Measuring Collateral Consequences Among Individuals Registered for a Sexual Offense: Development of the Sexual Offender Collateral Consequences Measure by Emma Hamilton, Delida Sanchez and Matthew L. Ferrara in Sexual Abuse: A Journal of Research and Treatment

Footnotes

Appendix A

Appendix B

Authors’ Note

The authors take responsibility for the integrity of the data and the accuracy of the data analyses and have made every effort to avoid inflating statistically significant results.

Declaration of Conflicting Interests

The author(s) declared no potential conflicts of interest with respect to the research, authorship, and/or publication of this article.

Funding

The author(s) disclosed receipt of the following financial support for the research, authorship, and/or publication of this article: This research was supported in part by a dissertation grant from the University of Texas at Austin Graduate School.

ORCID iD

Emma Hamilton

Supplemental Material

Supplemental material for this article is available online.

References

Andrews

D. A.

Bonta

(2001). Level of Service Inventory–Revised (LSI-R): User’s manual. Multi-Health Systems.

Bailey

D. J.

Klein

J. L.

(2018). Ashamed and alone: Comparing offender and family member experiences with the sex offender registry. Criminal Justice Review, 43, 440–457.

Bailey

D. J.

Sample

L. L.

(2017). An examination of a cycle of coping with strain among registered citizens’ families. Criminal Justice Studies, 30, 158–180.

Bartlett

M. S.

(1950). Tests of significance in factor analysis. British Journal of Statistical Psychology, 3(2), 77–85. https://doi.org/10.1111/j.2044-8317.1950.tb00285.x

Beck

A. T.

Weissman

Lester

Trexler

(1974). The measurement of pessimism: The Hopelessness Scale. Journal of Consulting and Clinical Psychology, 42(6), 861–865. https://doi.org/10.1037/h0037562

Boer

D. P.

Hart

S. D.

Kropp

P. R.

Webster

C. D.

(1997). Manual for the sexual violence risk-20. The British Columbia Institute Against Family Violence.

Bonta

Andrews

D. A.

(2010). Viewing offender assessment and rehabilitation through the lens of the risk-need-responsivity model. In McNeill

Raynor

Trotter

(Eds.), Offender supervision: New directions in theory, research and practice (pp. 19–40). Willan.

Brannon

Y. N.

Levenson

J. S.

Fortney

Baker

J. N.

(2007). Attitudes about community notification: A comparison of sexual offenders and the non-offending public. Sexual Abuse, 19(4), 369–379. https://doi.org/10.1177/107906320701900403

Buja

Eyuboglu

(1992). Remarks on parallel analysis. Multivariate Behavioral Research, 27(4), 509–540. https://doi.org/10.1207/s15327906mbr2704_2

10.

Burchfield

K. B.

Mingus

(2008). Not in my neighborhood: Assessing registered sex offenders' experiences with local social capital and social control. Criminal Justice and Behavior, 35(3), 356–374. https://doi.org/10.1177/0093854807311375

11.

Call

(2017). How acceptable are their obstacles to reentry? Attitudes of community corrections professionals toward collateral consequences for sex offenders. Corrections, 2, 237–257. https://doi.org/10.1080/23774657.2017.1305878

12.

Call

(2018). The community corrections perspective toward sex offender management policies and collateral consequences: Does contact with sex offenders matter? Criminal Justice Studies, 31, 1–17. https://doi.org/10.1080/1478601X.2017.1393418

13.

Cattell

R. B.

(1966). The scree test for the number of factors. Multivariate Behavioral Research, 1(2), 245–276. https://doi.org/10.1207/s15327906mbr0102_10

14.

Chin

(2017). Collateral consequences. In Luna

(Ed.), Reforming criminal justice: Punishment, incarceration, and release (Vol. 4, pp. 371–395). Arizona State University.

15.

Costello

A. B.

Osborne

J. W.

(2005). Best practices in exploratory factor analysis: Four recommendations for getting most from your analysis. Practical Assessment, Research & Evaluation, 10, 1–9. https://doi.org/10.7275/jyj1-4868

16.

Craun

S. W.

Bierie

D. M.

(2014). Are the collateral consequences of being a registered sex offender as bad as we think: A methodological research note. Federal Probation, 78, 28–31.

17.

Edwards

Hensley

(2001). Contextualizing sex offender management legislation and policy: Evaluating the problem of latent consequences in community notification laws. International Journal of Offender Therapy and Comparative Criminology, 45(1), 83–101. https://doi.org/10.1177/0306624X01451006

18.

Frenzel

E. D.

Bowen

K. N.

Spraitz

J. D.

Bowers

J. H.

Phaneuf

(2014). Understanding collateral consequences of registry laws: An examination of the perceptions of sex offender registrants. Justice Policy Journal, 11(2), 1–22.

19.

Ghasemi

Zahediasl

(2012). Normality tests for statistical analysis: A guide for non- statisticians. International Journal of Endocrinology and Metabolism, 10(2), 486–489. https://doi.org/10.5812/ijem.3505

20.

Gorsuch

R. L.

(1983). Factor analysis (2nd ed.). Lawrence Earlbaum Associates.

21.

Hair

Anderson

Tatham

Black

(1998). Multivariate data analysis (5th ed.). Prentice Hall.

22.

Hamilton

(2020). Toward a focused conceptualization and measurement of collateral consequences: A systematic review. Sexual Abuse: A Journal of Research and Treatment. Advance online publication. https://doi.org/10.1177/1079063220981906

23.

Hanson

R. K.

(1996, November). Coping with culpability [Paper presentation]. Meeting of the Association for the Treatment of Sexual Abusers, Chicago, IL, United States.

24.

Hanson

R. K.

Tangney

J. P.

(1996). The Test of Self-Conscious Affect—Socially Deviant populations (TOSCA-SD). Department of the Solicitor General of Canada, Corrections Research.

25.

Harris

A. J. R.

Hanson

R. K.

(2010). Clinical, actuarial and dynamic risk assessment of sexual offenders: Why do things keep changing? Journal of Sexual Aggression, 16(3), 296–310. https://doi.org/10.1080/13552600.2010.494772

26.

Harris

A. J. R.

Walfield

S. M.

Shields

R. T.

Letourneau

E. J.

(2015). Collateral consequences of juvenile sex offender registration and notification: Results from a survey of treatment providers. Sexual Abuse, 28(8), 770–790. https://doi.org/10.1177/1079063215574004

27.

Henson

R. K.

Roberts

J. K.

(2006). Use of exploratory factor analysis in published research: Common errors and some comment on improved practice. Educational and Psychological Measurement, 66(3), 393–416. https://doi.org/10.1177/0013164405282485

28.

Jeglic

E. L.

Mercado

C. C.

Levenson

J. S.

(2012). The prevalence and correlates of depression and hopelessness among sex offenders subject to community notification and residence restriction legislation. American Journal of Criminal Justice, 37(1), 46–59. https://doi.org/10.1007/s12103-010-9096-9

29.

Kaiser

H. F.

(1960). The application of electronic computers to factor analysis. Educational and Psychological Measurement, 20(1), 141–151. https://doi.org/10.1177/001316446002000116

30.

Kim

Mueller

C. W.

(1978). Factor analysis: Statistical methods and practical issues. SAGE.

31.

Kirk

D. S.

Wakefield

(2018). Collateral consequences of punishment: A critical review and path forward. Annual Review of Criminology, 1, 171–194. https://doi.org/10.1146/annurev-criminol-032317-092045

32.

Klein

Bailey

(2016). Technical report: The effects of living on the registry—Experiences of registrants and family members. https://doi.org/10.13140/RG.2.2.11563.85287

33.

Krieger

Smith

Naishadham

Hartman

Barbeau

E. M.

(2005). Experiences of discrimination: Validity and reliability of a self-report measure for population health research on racism and health. Social Science & Medicine, 61(7), 1576–1596. https://doi.org/10.1016/j.socscimed.2005.03.006

34.

Kruttschnitt

Uggen

Shelton

(2000). Predictors of desistance among sex offenders: The interaction of formal and informal social controls. Justice Quarterly, 17(1), 61–87. https://doi.org/10.1080/07418820000094481

35.

Langan

Schmitt

Durose

(2003). Recidivism of sex offenders released from prison in 1994. Bureau of Justice Statistics, Office of Justice Programs, U.S. Department of Justice.

36.

Lasher

M. P.

McGrath

R. J.

(2012). The impact of community notification on sex offender reintegration: A quantitative review of the research literature. International Journal of Offender Therapy and Comparative Criminology, 56(1), 6–28. https://doi.org/10.1177/0306624X10387524

37.

Levenson

J. S.

(2008). Collateral consequences of sex offender residence restrictions. Criminal Justice Studies, 21(2), 153–166. https://doi.org/10.1080/14786010802159822

38.

Levenson

J. S.

(2018). Sex offender management policies and evidence-based recommendations for registry reform. Current Psychiatry Reports, 20(21), 1–7. https://doi.org/10.1007/s11920-018-0884-0

39.

Levenson

J. S.

Cotter

L. P.

(2005a). The effect of Megan’s Law on sex offender reintegration. Journal of Contemporary Criminal Justice, 21(1), 49–66. https://doi.org/10.1177/1043986204271676

40.

Levenson

J. S.

Cotter

L. P.

(2005b). The impact of sex offender residence restrictions: 1,000 feet from danger or one step from absurd? International Journal of Offender Therapy and Comparative Criminology, 49, 168–178. https://doi.org/10.1177/0306624X04271304

41.

Levenson

J. S.

D’Amora

D. A.

Hern

A. L.

(2007). Megan’s law and its impact on community re-entry for sex offenders. Behavioral Sciences & the Law, 25(4), 587–602. https://doi.org/10.1002/bsl.770

42.

Levenson

J. S.

Tewksbury

(2009). Collateral damage: Family members of registered sex offenders. American Journal of Criminal Justice, 34(1–2), 54–68. https://doi.org/10.1007/s12103-008-9055-x

43.

Logan

W. A.

(2013). Informal collateral consequences. Washington Law Review, 88, 1103–1118.

44.

Margrove

K. L.

, SE-SURG (South Essex Service User Research Group), Heydinrych

Secker

(2013). Waiting list-controlled evaluation of a participatory arts course for people experiencing mental health problems. Perspectives in Public Health, 133(1), 28–35. https://doi.org/10.1177/1757913912461587

45.

Mercado

C. C.

Alvarez

Levenson

(2008). The impact of specialized sex offender legislation on community reentry. Sexual Abuse, 20(2), 188–205. https://doi.org/10.1177/1079063208317540

46.

Monjeau

C. J.

(2011). All politics is local: State preemption and municipal sex offender residency restrictions in New York state. Boston University Law Review, 91, 1569–1596.

47.

Nadal

K. L.

(2011). The Racial and Ethnic Microaggressions Scale (REMS): Construction, reliability, and validity. Journal of Counseling Psychology, 58(4), 470–480. https://doi.org/10.1037/a0025193

48.

O’Connor

B. P.

(2000). SPSS and SAS programs for determining the number of components using parallel analysis and Velicer’s MAP test. Behavior Research Methods, Instruments, & Computers, 32(3), 396–402. https://doi.org/10.3758/BF03200807

49.

O’Reilly

M. F.

(2018). Collateral consequences of a conviction and circuits of exclusion. In O’Reilly

M. F.

(Ed.), Uses and consequences of a criminal conviction (pp. 203–255). Palgrave Macmillan.

50.

Osborne

J. W.

(2015). What is rotating in exploratory factor analysis? Practical Assessment, Research & Evaluation, 20, Article 2. https://doi.org/10.7275/hb2g-m060

51.

Parent

M. C.

(2013). Handling item-level missing data: Simpler is just as good. The Counseling Psychologist, 41(4), 568–600. https://doi.org/10.1177/0011000012445176

52.

Pinard

(2010). Collateral consequences of criminal convictions: Confronting issues of race and dignity. New York University Law Review, 85, 457–534.

53.

Pinard

Thompson

A. C.

(2005). Offender reentry and the collateral consequences of criminal convictions: An introduction. New York University Review of Law & Social Change, 30, 585–620.

54.

Prescott

J. J.

Rockoff

J. E.

(2011). Do sex offender registration and notification laws affect criminal behavior? The Journal of Law and Economics, 54(1), 161–206. https://doi.org/10.3386/w13803

55.

Robbers

M. L.

(2009). Lifers on the outside: Sex offenders and disintegrative shaming. International Journal of Offender Therapy and Comparative Criminology, 53(1), 5–28. https://doi.org/10.1177/0306624X07312953

56.

Roberts

(2008). The mythical divide between collateral and direct consequences of criminal convictions: Involuntary commitment of sexually violent predators. Minnesota Law Review, 93, 670–740.

57.

Sandler

J. C.

Freeman

N. J.

Socia

K. M.

(2008). Does a watched pot boil? A time-series analysis of New York State’s sex offender registration and notification law. Psychology, Public Policy, and Law, 14(4), 284–302. https://doi.org/10.1037/a0013881

58.

Secker

Hacking

Kent

Shenton

Spandler

(2009). Development of a measure of social inclusion for arts and mental health project participants. Journal of Mental Health, 18(1), 65–72. https://doi.org/10.1080/09638230701677803

59.

Spector

P. E.

(1992). Summated rating scale construction: An introduction. SAGE.

60.

Steed

(2001). Further validity and reliability evidence for Beck Hopelessness Scale scores in a nonclinical sample. Educational and Psychological Measurement, 61(2), 303–316. https://doi.org/10.1177/00131640121971121

61.

Suhr

D. D.

(2006, March). Exploratory or confirmatory factor analysis? [Paper presentation]. Annual meeting of the SAS Users Group International, San Francisco, CA, United States.

62.

Tabachnick

B. G.

Fidell

L. S.

(2001). Using multivariate statistics (4th ed.). Harper & Row.

63.

Tabachnick

B. G.

Fidell

L. S.

Ullman

J. B.

(2007). Using multivariate statistics (Vol. 5). Pearson.

64.

Tangney

J. P.

Stuewig

Martinez

A. G.

(2014). Two faces of shame: The roles of shame and guilt in predicting recidivism. Psychological Science, 25(3), 799–805. https://doi.org/10.1177/0956797613508790

65.

Tangney

J. P.

Stuewig

Mashek

Hastings

(2011). Assessing jail inmates’ proneness to shame and guilt: Feeling bad about the behavior or the self? Criminal Justice and Behavior, 38(7), 710–734. https://doi.org/10.1177/0093854811405762

66.

ten Bensel

Sample

L. L

. (2019). Social inclusion despite exclusionary sex offense laws: How registered citizens cope with loneliness. Criminal Justice Policy Review, 30, 274–292.

67.

Tewksbury

(2005). Collateral consequences of sex offender registration. Journal of Contemporary Criminal Justice, 21(1), 67–81. https://doi.org/10.1177/1043986204271704

68.

Tewksbury

(2007). Exile at home: The unintended collateral consequences of sex offender residency restrictions. Harvard Civil Rights-Civil Liberties Law Review, 42, 531–540.

69.

Tewksbury

Lees

(2006). Perceptions of sex offender registration: Collateral consequences and community experiences. Sociological Spectrum, 26(3), 309–334. https://doi.org/10.1080/02732170500524246

70.

Tewksbury

Levenson

(2009). Stress experiences of family members of registered sex offenders. Behavioral Sciences & the Law, 27, 611–626. https://doi.org/10.1002/bsl.878

71.

Tewksbury

Mustaine

E. E.

(2009). Stress and collateral consequences for registered sex offenders. Journal of Public Management & Social Policy, 15(2), 215–239.

72.

Tewksbury

Zgoba

K. M.

(2010). Perceptions and coping with punishment: How registered sex offenders respond to stress, internet restrictions, and the collateral consequences of registration. International Journal of Offender Therapy and Comparative Criminology, 54(4), 537–551. https://doi.org/10.1177/0306624X09339180

73.

Tolson

Klein

(2015). Registration, residency restrictions, and community notification: A social capital perspective on the isolation of registered sex offenders in our communities. Journal of Human Behavior in the Social Environment, 25, 375–390.

74.

Uggen

Stewart

(2014). Piling on: Collateral consequences and community supervision. Minnesota Law Review, 99, 1871–1912.

75.

Vuolo

(2017). Situating the prediction problem within collateral consequences: Communicating risk to employers. Criminology & Public Policy, 16(3), 959–961. https://doi.org/10.1111/1745-9133.12327

76.

Wakefield

(2006). The vilification of sex offenders: Do laws targeting sex offenders increase recidivism and sexual violence. Journal of Sexual Offender Civil Commitment: Science and the Law, 1, 141–149.

77.

Wheelock

(2005). Collateral consequences and racial inequality: Felon status restrictions as a system of disadvantage. Journal of Contemporary Criminal Justice, 21(1), 82–90. https://doi.org/10.1177/1043986204271702

78.

Williams

Onsman

Brown

(2010). Exploratory factor analysis: A five-step guide for novices. Australasian Journal of Paramedicine, 8(3), 1–13. http://ro.ecu.edu.au/jephc/vol8/iss3/1

79.

Wilson

Secker

(2015). Validation of the Social Inclusion Scale with students. Social Inclusion, 3(4), 52–62. https://doi.org/10.17645/si.v3i4.121

80.

Winters

G. M.

Jeglic

E. L.

Calkins

Blasko

B. L.

(2017). Sex offender legislation and social control: An examination of sex offenders’ expectations prior to release. Criminal Justice Studies, 30(2), 202–222. https://doi.org/10.1080/1478601X.2017.1299894

81.

Worthington

R. L.

Navarro

R. L.

(2003). Pathways to the future: Analyzing the contents of a content analysis. The Counseling Psychologist, 31(1), 85–92. https://doi.org/10.1177/0011000002239402

82.

Worthington

R. L.

Whittaker

T. A.

(2006). Scale development research: A content analysis and recommendations for best practices. The Counseling Psychologist, 34(6), 806–838. https://doi.org/10.1177/0011000006288127

83.

Zevitz

R. G.

(2006). Sex offender community notification: Its role in recidivism and offender reintegration. Criminal Justice Studies, 19(2), 193–208. https://doi.org/10.1080/14786010600764567

84.

Zevitz

R. G.

Farkas

M. A.

(2000). Sex offender community notification: Assessing the impact in Wisconsin. Office of Justice Programs, National Institute of Justice, U.S. Department of Justice.

85.

Zwick

Velicer

(1986). Comparison of five rules for determining the number of components to retain. Psychological Bulletin, 99, 432–442. https://doi.org/10.1037/0033-2909.99.3.432

Supplementary Material

Please find the following supplemental material available below.

For Open Access articles published under a Creative Commons License, all supplemental material carries the same license as the article it is associated with.

For non-Open Access articles published, all supplemental material carries a non-exclusive license, and permission requests for re-use of supplemental material or any part of supplemental material shall be sent directly to the copyright owner as specified in the copyright notice associated with the article.

0.00 MB

0.02 MB

0.04 MB