Exploring Parenting as a Predictor of Criminogenic Thinking in College Students

Abstract

Crime-promoting cognitions and attitudes, globally labeled as criminogenic thinking, are shown to perpetuate maladaptive and antisocial behavior in criminals and nonoffenders. In the nonoffender population, these thinking patterns may not lead to illegal behavior, but can result in irresponsible or maladaptive behavioral consequences. Theories suggest that early childhood parent–child interactions may be partly responsible for the development of criminogenic thinking. While the relationship between parenting and antisocial behavior is well documented, the connection between parenting and the development of criminogenic thinking styles has not yet been explored. The current study examined the nature of the relationship between exposure to parenting behaviors and subsequent criminogenic thoughts in a nonoffender, college population. The sample included 119 undergraduates. Results indicate that parenting may affect general criminogenic thinking as well as specific types of criminogenic thinking styles. Relevance and importance of the findings with regard to clinical work and parenting are also discussed.

Keywords

parenting college criminogenic thinking delinquency

Maladaptive behavior in college students is a problem, with research indicating that large numbers of college students engage in illegal activities such as intoxicated driving, drug use, and interpersonal aggression (Fromme, Katz, & Rivet, 1997; Zimny, Robertson, & Bartoszek, 2008). Beyond the prevalence of illegal behavior, other maladaptive behaviors are also problematic in college student populations, such as academic and personal dishonesty (Zimny et al., 2008) and risky sexual behavior (Fromme et al., 1997). For example, almost 25% of college students indicate experiencing a significantly negative consequence in the past year due to either substance use, poor academic performance, suicide ideation, or personal injury (Carey, Scott-Sheldon, Carey, & DeMartini, 2007). Research has also shown that college students report engaging in criminal behavior, ranging from minor infractions (i.e., drinking alcohol underage) to violent offenses, such as stalking, rape, and assault (McCoy et al., 2006). For the purposes of this article, maladaptive behavior is defined as any behavior that potentially results in negative consequences, such as academic penalties, legal, or health problems (i.e., sexually transmitted diseases), as delineated in literature relevant to college student samples (Carey et al., 2007; Fromme et al., 1997; McCoy et al., 2006; Zimny et al., 2008).

Much research has been done in an attempt to understand possible causes of maladaptive behavior in adolescents and young adults. For example, research studies have indicated links between temperament (e.g., Kingston & Prior, 1995), impulsivity (e.g., Loeber, 1990), cognitive biases (Liau, Barriga, & Gibbs, 1998), cognitive attribution errors (Dodge, 1991), and familial variables (Dembo et al., 2007) to undesired behavioral outcomes in young adults, such as rule breaking. Liau et al. (1998) examined self-serving cognitive distortions in relation to adolescent misconduct, and found that a strong link exists between these cognitions and maladaptive behavior. Although research has connected cognitive errors with maladaptive behavior (Liau et al., 1998), little research has examined cognition in the framework of criminogenic thinking theories for the college population.

Criminogenic thinking is defined as those thinking styles or belief systems that precede criminal and other maladaptive behavior (Walters, 1990). According to Walters (2006), the objective for studying these thinking styles is to understand how criminogenic thinking predicts maladaptive behavior and whether an alteration in criminogenic thoughts results in a change in that behavior. Although criminogenic thinking is a concern among college students due to its established connection to maladaptive behavior (Walters, 1990), most research pertaining to criminogenic thoughts is founded in the research of Yochelson and Samenow (1976), which focused on the prison population. Walter’s criminal lifestyle theory defines the construct of criminal lifestyle as the development and crystallization of criminogenic beliefs, which form a cognitive system that is focused on increasing criminogenic thinking during adolescence. This cognitive system subsequently influences behavior during the course of one’s life (Walters, 1990). Walters identified eight thinking styles that comprise the criminal lifestyle, which include Mollification (Mo; blaming external factors for acts), Cutoff (Co; mental elimination of crime deterrents), Entitlement (En; idea of deserving ownership), Power Orientation (desire to seek control on the external environment), Sentimentality (negating negative behavior by expounding on other good acts), Superoptimism (overly positive attitude about avoiding crime consequences), Cognitive Indolence (using cognitive short cuts), and Discontinuity (inability to follow through on one’s goals; Walters, 1990).

Patterns of criminogenic thinking may differ based on a person’s status, according to research, which has shown demographic differences in these criminogenic patterns. Specifically, criminogenic thinking has been found to be higher among women and younger adults than men and older adults, respectively (Morgan et al., 2008; Walters, 2002). Also, findings from Dembo et al. (2007) and Walters (2002) have evidenced that significant differences in criminogenic thinking exist across ethnicities. For example, some studies have shown that African Americans and Latinos report higher levels of criminogenic thinking than Caucasians (Walters & Geyer, 2004). Because these demographic characteristics are often found to correlate with criminogenic thinking, they have been commonly used as control variables in previous studies (Butler, Fearon, Atkinson, & Parker, 2007).

Most studies of criminogenic thinking have focused on offender populations. Walters, however, posited that although criminogenic thinking is found in all criminals, it is observable in nonoffenders as well. Walters differentiates nonoffenders from offenders in that even though nonoffenders experience varying levels of criminogenic thinking, they have not “erected a lifestyle around these characteristics” (Walters, 1990, p. 130). While Walters stated that the difference between nonoffender criminogenic thinking and offender criminogenic thinking is a matter of severity, less severe cognitive patterns can be problematic when manifested in nonoffenders because they still can lead to behavior that is unhealthy and problematic (Walters, 1990).

There is little research on criminogenic thinking in nonoffenders, even though theory and research findings indicate that criminogenic thinking is prevalent in this population. In 2007, Walters studied criminogenic thinking in college students to examine the extent of criminogenic thinking found in a nonoffender population and to validate his measure of criminogenic thinking (i.e., the Psychological Inventory of Criminal Thinking–Layperson [PICTS-L]) for nonoffenders. Walters postulated two reasons why criminogenic thinking is important to explore in the nonoffender population. First, the study aids in understanding and identifying groups at high risk for engaging in criminal behavior. For example, McCoy et al. (2006) found that criminogenic thinking predicts aggression and property crimes among college students. Second, in understanding criminogenic thinking in nonoffenders more fully, preventive measures can be taken with the hopes of reducing the occurrence of criminal behavior (Walters, 2007). Furthermore, other researchers have implicated criminogenic thinking in other issues besides criminal behavior, including mental health treatment engagement (e.g., low levels of treatment retention and motivation; Hughes, 2009), antisocial acts (e.g., adolescent bullying, aggression; Ragatz, Anderson, Fremouw, & Schwartz, 2001), and general irresponsible thinking (Ragatz et al., 2001). Therefore, in addition to the prediction and prevention of crime, examining criminogenic thinking in nonoffenders is important to gain insight related to the more common maladaptive behavior.

In examining variables related to criminogenic thinking, some researchers have postulated that circumstantial variables, such as family relationships, are related to criminogenic thinking and maladaptive conduct. Previous research indicates that parenting is related to antisocial attitudes, a construct similar to criminogenic thinking, and behaviors as well as internalizing problems (Gendreau, Andrews, Goggin, & Chanteloupe, 1992; Hoeve et al., 2009). For example, Gendreau et al. (1992) suggested that low levels of parental warmth and poor parental supervision are risk factors for antisocial behavior. Although researchers have implicated the importance of experiences in childhood on later antisocial behavior, there has been little exploration connecting criminogenic thinking, specifically, with parenting variables.

A dominant theoretical framework of examining the quality of parenting implements a dimensional approach to parenting behaviors (i.e., Care and Protection; Maccoby, & Martin, 1983). This approach conceptualizes parenting behaviors as two separate constructs on a continuum of intensity. Protection measures the degree of psychological and behavior control parents exert over their offspring (Hoeve et al., 2009; Parker, Tupling, & Brown, 1979). Low levels of the care dimension of parenting are characterized as a rejecting approach to parenting and are related to delinquency. Conversely, high levels of care, indicative of a supportive approach to parenting, are inversely related to delinquency (Barnes & Farrell, 1992). Low levels of care and high levels of protection have shown to be associated with adolescent drinking, illicit drug use, and academic misconduct (Barnes & Farrell, 1992).

Literature has also drawn a connection between the protection dimension of parenting and negative outcomes. For example, high levels of protection, which often involves psychological and behavioral control as well as punitive treatment, are related to aggressive dispositions in children and antisociality in young adults (Schaffer, Clark, & Jeglic, 2009; Zhou, Eisenberg, Wang, & Reiser, 2004). Furthermore, high levels of behavioral control (i.e., rule setting, monitoring) have been found to be more strongly related to externalizing problems in adolescents, while psychological control (i.e., use of guilt, manipulation) has been linked to internalizing problems. Conversely, extremely low levels of behavioral control, also called protection, are characterized as permissive and disengaged (Barber, Olsen, & Shagle, 1994).

Another way to conceptualize parenting using the Protection and Care dimensions involves grouping parenting styles into quadrants that include (a) authoritarian, characterized by low levels of care and high levels of protection; (b) authoritative, characterized by a high level of care and protection; (c) permissive, characterized by high amounts of care and a low level of protection; and (d) neglecting, characterized by a low levels of care and protection (Baumrind, 1971). At the crux of this typology approach, parenting that involves extremes on either dimension (i.e., care and protection) is not ideal, and that authoritative parenting strikes a balance in demanding order while maintaining comfort and emotional intimacy (Baumrind, 1971).

According to Maccoby and Martin (1983) and Steinberg, Blatt-Eisengart, and Cauffman (2006), neglectful parenting is linked to delinquent behavior in particular; however, any parenting type except authoritative can be detrimental to child adaptation. For example, neglectful parenting is linked to adolescent’s low resistance to peer pressure, potentially predisposing these individuals to engage in risky or delinquent behaviors (Steinberg et al., 2006). As such, an authoritative style of parenting, described as democratic and rewarding, is a protective factor against negative behavioral outcomes and is related to higher levels of psychological health and personal responsibility for actions (Hoeve et al., 2009; Steinberg et al., 2006).

Research has also been done to examine the impact of specific parenting behaviors on childhood antisocial behavior. For example, Bowman, Prelow, and Weaver (2007) examined maternal monitoring and involvement, and found that these behaviors increase healthy psychosocial outcomes and decrease involvement in deviant peer groups among adolescents. Also, poor parental supervision and deficient caretaking is predictive of adolescent misconduct (Knutson, DeGarmo, & Reid, 2004). A meta-analysis conducted by Hoeve et al. (2009) indicated that parental monitoring, psychological control, and negative support account for approximately 11% of the variance in adolescent criminal behavior. Overall, a substantial amount of research has been done on discrete parenting behaviors and delinquent behavioral outcomes. However, the literature is inconsistent about the extent of the effect of parenting dimensions of care and protection on antisocial outcomes, with previous research providing various effect sizes (Hoeve et al., 2009).

While there is a large body of literature connecting poor parenting behaviors and maladaptive behavior, few studies have focused on connecting parenting to criminogenic thinking. Cuadra (2007) found that criminogenic thinking acts as a mediator in the relationship between childhood abuse and criminal behavior. The study indicated that the type of child maltreatment (i.e., physical, emotional abuse, neglect) is related to specific types of criminogenic thinking styles in adult male offenders. Cuadra (2007) found that physical and emotional abuse and neglect in one’s childhood are positively correlated with En, Mo, and Co criminogenic thinking styles among offenders. These findings highlight the importance of the type of criminogenic thinking as it relates to childhood treatment. Also, Dembo et al. (2007) found a strong positive correlation between poor family functioning, involving generally poor family cohesion and conflict, and overall criminogenic thinking.

Although research has connected parenting to antisocial behavior outcomes, no research has been done to make a definitive connection between the parenting dimensions of care and protection and criminogenic thinking (Hoeve et al., 2009). Cuadra’s (2007) finding that Co, Mo, and En were significantly related to childhood maltreatment is partially consistent with Walter’s (2002) theory, which reasons that En and Mo criminogenic thinking styles may develop in part due to parental involvement in early childhood. According to Walters’s theory, Mo involves self-justification and rationalizations based on the inequities of personal life circumstances or the larger society. This criminogenic thinking style justifies one’s actions because of personal circumstances and perceived unfairness of the world, which originates in the immaturity of adolescence. Walters proposed that En involves the belief that societal rules do not apply because of personal privilege, and is common when parents send the message of unsurpassed worth, uniqueness, and entitlement. Therefore, those who engage in En criminogenic thinking often consider themselves as above societal rules and often misidentify wants as needs (Walters, 1990).

Based on the previous research that has connected childhood abuse to Mo, Co, and En (Cuadra, 2007), and Walters’s (2002) theoretical grounding of Mo and En, the current study will attempt to determine the relationship between those specific criminogenic thinking styles (i.e., Mo, Co, and En) and experienced parenting approaches. While Walters (2002) has delineated a total of eight criminogenic thinking styles in his theory, only Mo, Co, and En are used in this study as these three have empirical and/or theoretical support for a relationship with parenting experiences.

Hypotheses

Although research has been done to determine the relationship between parenting behaviors and subsequent criminal behavior (Bowman et al., 2007; Hoeve et al., 2009), researchers indicate that the relationship between parenting approaches and criminogenic thinking is still unclear (Dembo et al., 2007), highlighting the need for investigation. We explored the role of Care and Protection parenting dimensions in criminogenic thinking in an exploratory analysis to ascertain the overall relationship between parenting and criminogenic thinking, broadly defined. Furthermore, we examined the ability of Care and Protection to predict Mo, Co, and En. We hypothesized that these parenting variables would be significant predictors to all four types of criminogenic thinking.

Method

Procedure and Participants

Data were collected from 155 college students at a large southeastern U.S. university. Participants seeking research credits for undergraduate courses in psychology were directed to an online research website where the project was posted. Once self-selecting for the study from a list of all available research studies, they consented to the research and completed the consent form and all instruments. Participants completed measures examining demographics, experienced parenting behaviors, and criminogenic thinking, which are described in detail below. The three study instruments were presented in random order to control for order effects. Exclusion criteria used to determine the final sample included the three validity scales of the PICTS-L, which are the Cannot Say, Confusion, and Defensiveness scales. The Cannot Say validity scale cutoff is more than five responses left blank. In this sample, 3.8% (n = 6) left more than five items blank on the PICTS and were removed from the final sample. The Confusion validity scale uses a T-score of 81 as a cutoff for exclusion. Of the respondents in this sample, 1.3% (n = 2) violated this cutoff score and was removed from the final sample. Similarly, the Defensiveness validity scale cutoff score is 65 T, and 18.1% (n = 28) of the sample exceeded this cutoff score and were removed.

The final sample included 119 participants. The sample was predominately female (83.2%) and White (46.5%). Furthermore, the majority of the sample indicated their mothers as primary caregivers (81.5%). With regard to maladaptive behavior, the most prevalent behaviors reported are as follows: 21% (n = 25) of the sample reported cheating on a test, 9% (n = 11) reported shoplifting, 54% (n = 46) reported drinking alcohol as a minor, 30% (n = 35) reported drinking and driving, and 19% (n = 22) reported having unprotected sex. Demographic characteristics are presented in Table 1, and all reported behaviors determined by the demographic questionnaire are reported in Table 2.

Table 1.

Demographic Characteristics of the Sample.

Characteristic	n	%
Gender
Male	20	16.8
Female	99	83.2
Year in school (current)
Freshman	47	39.5
Sophomore	28	23.5
Junior	19	16.0
Senior	24	20.2
Primary caregiver
Mother	97	81.5
Father	8	6.7
Grandmother	6	5.0
Aunt	4	3.4
Other	4	3.4
Ethnicity
White	72	46.5
Black	71	45.8
Other	12	7.7
Relationship status
Single	89	74.8
Partnered/common law	15	12.6
Divorced	5	4.2
Married	10	8.4
Spiritual/religious identification or denomination
Agnostic/atheist	4	3.4
Catholic	14	11.8
Methodist	11	9.2
Baptist (i.e., southern, missionary, national Baptist convention)	65	54.5
Presbyterian	1	0.8
Church of God in Christ	8	6.7
Christian Episcopal	2	1.7
Unitarian-Universalist	1	0.8
Other	13	10.9
Secondary caregiver
Mother	12	10.1
Father	64	53.8
Grandparent	28	23.6
Aunt/uncle	6	5.1
Other	7	5.9
Received counseling/therapy services
Yes	33	27.7
No	86	72.3

Table 2.

Reported Behaviors.

Characteristics	n	%
Committed a crime in past 6 months
Yes	4	3.4
No	115	96.6
Arrested in past 6 months
Yes	2	1.7
No	117	98.3
Been incarcerated for a crime
Yes	3	2.5
No	116	97.5
Looked at someone’s test during an examination
Yes	25	21
No	94	79
Used someone else’s answer on an examination
Yes	21	17.6
No	98	82.4
Wrote down information secretly and used it during test
Yes	19	16
No	100	84
Used someone else’s paper and turned it in as yours
Yes	2	1.7
No	117	98.3
Stopped attending school without an excuse for more than 10 days
Yes	7	5.9
No	112	94.1
Took something from a store or a person without paying for it
Yes	11	9.2
No	108	90.8
Been in a gang fight
Yes	3	2.5
No	115	97.5
Consumed alcohol underage
Yes	64	53.8
No	55	46.2
Driven a car while drunk, buzzed
Yes	35	29.4
No	84	70.6
Used someone else’s credit cards/checks without asking permission
Yes	5	4.2
No	114	95.8
Unprotected sex with someone you weren’t in a relationship with
Yes	22	18.5
No	97	81.5
Followed someone when they didn’t want you to
Yes	9	7.6
No	110	92.4
Bought goods that may have been stolen
Yes	8	6.7
No	111	93.3
Given someone marijuana in return for goods/money
Yes	6	5
No	113	95
Used prescription drugs in any way other than those directed
Yes	11	9.2
No	108	90.8
Hurt or tried to hurt someone on purpose
Yes	16	13.4
No	103	86.6
Lied about age to buy cigarettes/alcohol
Yes	18	15.3
No	100	84.7
Intentionally set fire to another’s property
Yes	2	1.7
No	117	98.3

Measures

Demographic Questionnaire

Participants completed a general demographic questionnaire created by the researcher for the purposes of the study. Items included questions concerning gender, age, ethnicity, and maladaptive/criminal behavior history. Maladaptive/criminal behavior history items examined for a history of a variety of maladaptive behaviors, including academic dishonesty, plagiarism, truancy, petty theft, gang involvement, underage drinking, drinking and driving, risky sexual behaviors, illicit/prescription drug use, drug selling, assault, and larceny. Other general criminal history items were also presented, examining a history of committing an unspecified crime or being arrested in past 6 months, and being incarcerated.

Parental Behavior

Parental behavior was assessed using the Parental Bonding Instrument (PBI), a retrospective self-report measure developed by Parker et al. (1979) to measure recollected parenting behaviors and parenting styles of respondents 16 years old and above. Answers on this 25-item measure were reported on a 4-point Likert-type scale from very like (0) to not very like (3), and the instrument has a range of 0 to 74. The instrument measures two variables: overprotection and care, in which higher scores indicate higher levels of measured constructs. Previous research has indicated thorough evidence of satisfactory reliability and validity (Parker et al., 1979; Gerlsma, Arrindell, van der Veen, & Emmelkamp, 1991).

Criminogenic Thinking

Criminogenic thinking was assessed with the PICTS-L edition (Walters, 1995). The 80-item PICTS-L is a self-report questionnaire designed to determine the presence of criminogenic thinking styles among nonoffenders. It is derived and developed by Walters based on the original PICTS measure, in which the wording and content of some items was modified to make the item content more applicable to nonoffenders. The PICTS-L was not altered for the current study.

The PICTS assesses eight criminogenic thinking styles (i.e., Mo, Cutoff, En, Power Orientation, Sentimentality, Superoptimism, Cognitive Indolence, and Discontinuity) in addition to two content scales (Current and Historical), three validity scales (i.e., Cannot Say, Confusion, Defensiveness), two composite criminogenic thinking scales (i.e., Proactive and Reactive), and four factor scales (i.e., Problem Avoidance, Infrequency, Self-Assertion, Denial of Harm). Also, a General Criminal Thinking (GCT) Scale measures overall criminogenic thinking. Answers are reported on a 4-point Likert-type scale, ranging from disagree (1) to strongly agree (4). For the purpose of this study, three criminogenic thinking style subscales (i.e., Mo, Co, and En) were used as well as the Criminogenic Thinking Total Scale (i.e., GCT). Three validity scales were used as exclusion criterion to ensure participants with invalid response styles. The Cannot Say Scale examines the number of items that were left blank. If five or more items were left unanswered on the instrument, the participant was excluded from the sample, as five or more missing items raises concerns of validity according to Walters (2010). The Confusion–Revised (Cf-r) Scale examines random responding, potentially indicating reading problems, faking bad, severe emotional disturbances, or careless responding. A T-score of 81 T or more on this scale rendered the test invalid, and the participant was excluded as indicated by Walters (2010). Finally, the Defensiveness–Revised (Df-r) Scale tests for “defensive or guardedness in responding” (Walters, 2010, p. 44). For this scale, a cutoff of 65 T was used to determine invalidity as indicated by Walters, because this score demonstrates significantly defensive responding across all thinking styles, factor, content, and composite scales (Walters, 2010).

Evidence of internal consistency, test–retest reliability, and preliminary validity for the PICTS-L has been found to be comparable with the original PICTS (Walters, 2002). Furthermore, other studies (Carr, Rosenfeld, Magyar, & Rotter, 2009; Magyar, Carr, Rosenfeld, & Rotter, 2010; Walters, Felix, & Reinoehl, 2009) provide evidence of internal consistency, test–retest reliability, internal reliability, and preliminary validity of the PICTS-L that is comparable with the original PICTS. Walters (2002) indicated test–retest reliability of the PICTS that ranges from .73 (Sentimentality) to .85 (Discontinuity) using a 2-week test interval. Internal consistency for the eight thinking styles was found to range from .55 to .79. Strongest evidence of internal consistency is found in the GCT score, which is .93. Extensive evidence of concurrent, predictive, and construct validity has also been reported (Walters, 2002). In the current sample, internal consistency for the GCT Scale (r = .95) was found to be strong, and internal consistency for the subscales used ranged from .54 to .78 (Walters, 2002).

Results

Means, standard deviations, and correlation coefficients for the instruments used are presented in Table 3. Overall, scores on the PICTS-L were within a standard deviation of those means reported in college samples (Walters et al., 2009). For the current sample, the average Care parenting subscale scores were within the normal range of scores reported in other studies; however, Protection parenting scores were slightly higher than those reported in studies using the measure (Pelcovitz et al., 2000) indicating that the current sample may have experienced a more protective and controlling parenting approach. The measures demonstrated good internal consistency (see Table 3).

Table 3.

Cronbach’s Alpha Coefficients of Measures Used.

Variable	R
1. PBI Care	.93
2. PBI Protection	.81
3. PICTS-L GCT	.95
4. PICTS-L Co	.77
5. PICTS-L Mo	.76
6. PICTS-L En	.81

Note. PBI = Parental Bonding Instrument; PICTS-L = Psychological Inventory of Criminal Thinking–Layperson; GCT = general criminal thinking; Co = Cutoff; Mo = Mollification; En = Entitlement.

With regard to the level of criminogenic thinking, this sample of college students reported meaningful levels of criminogenic thinking (m = 120.94; see Table 4). For the current sample, GCT scores averaged at 50 T, which is compared with prison population norms, in which a T-score above 60 indicates a significantly elevated criminogenic thinking level (Walters, 2010). Therefore, it is likely that college students do, in fact, engage in some criminogenic thinking, which may contribute to irresponsible or illegal behavior (i.e., drunk driving, drinking underage, academic dishonesty; Fromme et al., 1997; Zimny et al., 2008).

Table 4.

Means, Standard Deviations, and Correlation Coefficients for Study Measures.

	1	2	3	4	5	6
1. PBI Care	1	−.548^**	−.373^**	−.365^**	−.292^**	−.305^**
2. PBI Overprotection		1	.34^**	.252^**	.213*	.296^**
3. PICTS-L GCT			1	.87^**	.817^**	.842^**
4. PICTS-L Co				1	−.623^**	.69^**
5. PICTS-L En					1	.729^**
6. PICTS-L Mo						1
M	33.28	21.75	120.94	15.27	14.01	13.73
SD	8.33	6.73	28.96	4.79	4.69	4.39

Note. PBI = parental bonding instrument; PICTS-L GCT = Psychological Inventory of Criminal Thinking–Layperson general criminal thinking; PICTS-L Co = Psychological Inventory of Criminal Thinking–Layperson Cutoff subscale; PICTS-L En = Psychological Inventory of Criminal Thinking–Layperson Entitlement subscale; PICTS-L Mo = Psychological Inventory of Criminal Thinking–Layperson Mollification subscale.

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

Four hierarchical linear regression (HLR) analyses were conducted to evaluate the ability of parenting variables to predict levels of criminogenic thinking. One analysis was conducted for each of the three PICTS-L subscales of interest (i.e., Co, En, Mo), and for the total PICTS-L score (i.e., PICTS-L GCT). In each analysis, ethnicity and gender were used as control variables, and thus entered into the first block, as previous research has provided evidence about their unique impact on criminal thinking (Butler et al., 2007; Dembo et al., 2009; Walters, 2002). Gender was dichotomized (i.e., 1 = male, 0 = female) for the analyses. Ethnicity was dummy coded such that Black and Other were each contrasted with White (i.e., 0 = White, 1 = Black, 2 = Other). In the second block, parenting variables of Care and Protection were entered as continuous variables. In all analyses, the dependent variable was one of the four criminogenic thinking variables.

For the first hypothesis, which stated that Care and Overprotection will predict GCT scores, a HLR analysis was used to predict the PICTS-L GCT score. Ethnicity and gender were entered into the first block of predictors in the model, and Care and Overprotection were added in the second block. Care and Overprotection accounted for an additional 16.5% of the variance in PICTS-L GCT scores, which was statistically significant, ΔF(2, 111) = 10.995, p < .001. Individually, Care predicted PICTS-L GCT scores (p = .013), whereas Overprotection did not (p = .060). Therefore, the hypothesis was partially supported (see Table 5).

Table 5.

Summary of HLR Predicting PICTS-L GCT Scores From Demographic Variables and PBI Score Variables.

Predictor variable	B	SE B	β	R ²	ΔR²
Step 1				.003	.003
Gender, 1 = male	2.814	7.635	.037
Black ethnicity	2.627	5.711	.045
Other ethnicity	3.747	12.094	.031
Step 2				.168	.165^***
PBI Care	−0.916	0.362	−.263*
PBI Overprotection	0.865	0.455	.200

Note. HLR = hierarchical linear regression; PICTS-L = Psychological Inventory of Criminal Thinking–Layperson general criminal thinking; PBI = parental bonding instrument.

p < .05. ***p < .001.

For the second hypothesis, which stated that Co scores on the PICTS-L will be predicted by Care and Overprotection scores on the PBI, a HLR model was used to predict PICTS-L Co scores. Ethnicity and gender were entered into the first block of predictors in the model. When added to the analysis in the second block, Care and Overprotection accounted for an additional 13.6% of the variance in PICTS-L Co scores, which was statistically significant, ΔF(5, 111) = 8.741, p < .001. Individually, Care subscale scores predicted PICTS-L Co scores (p = .002), whereas Overprotection subscale scores did not (p = .535; see Table 6).

Table 6.

Summary of HLR Predicting PICTS-L Co From Demographic Variables and PBI Score Variables.

Predictor variable	B	SE B	β	R ²	ΔR²
Step 1				.003	.003
Gender, 1 = male	−0.668	1.249	−.053
Black ethnicity	0.108	0.934	.011
Other ethnicity	0.758	1.979	.038
Step 2				.139	.136^***
PBI Care	−.187	0.060	−.328**
PBI Overprotection	0.047	0.076	.067

Note. HLR = hierarchical linear regression; PICTS-L = Psychological Inventory of Criminal Thinking–Layperson; Co = Cutoff; PBI = Parental Bonding Instrument.

p < .01. ***p < .001.

For the third hypothesis, which stated that En scores on the PICTS-L will be predicted by Care and Overprotection scores on the PBI, a HLR model was used to predict PICTS-L En scores. Ethnicity and gender were entered into the first block of predictors in the model. The second block, consisting of Care and Overprotection accounted for 8.9% of the variance in PICTS-L En scores, which was significant, ΔF(2, 111) = 5.498, p = .05. Individually, Care predicted PICTS-L En scores (p = .023), whereas Overprotection did not (p = .467; see Table 7).

Table 7.

Summary of HLR Predicting PICTS-L En From Demographic Variables and PBI Score Variables.

Predictor variable	B	SE B	β	R ²	ΔR²
Step 1				.014	.014
Gender, 1 = male	1.067	1.228	.086
Black ethnicity	0.210	0.919	.022
Other ethnicity	1.392	1.945	.071
Step 2				.103	.089*
PBI Care	−0.140	0.061	−.248*
PBI Overprotection	0.056	0.076	.080

Note. HLR = hierarchical linear regression; PICTS-L = Psychological Inventory of Criminal Thinking–Layperson; En = Entitlement; PBI = Parental Bonding Instrument.

p < .05.

The last hypothesis stated that Mo scores on the PICTS-L will be predicted by Care and Overprotection scores on the PBI. To test this hypothesis, a HLR model was used. Ethnicity and gender were entered into the first block of predictors in the model. Care and Overprotection were then entered in the second block. These subscale score predictors accounted for 13.1% of the variance in PICTS-L Mo scores, which was significant, ΔF(2, 111) = 8.633, p < .001. Individually, Care subscale scores was not predictive (p = .077), whereas Overprotection subscale scores significantly predicted PICTS-L Mo scores (p = .034; see Table 8).

Table 8.

Summary of Hierarchical Multiple Regression Predicting PICTS-L Mollification From Demographic Variables and PBI Score Variables.

Predictor variable	B	SE B	β	R ²	ΔR²
Step 1				.004	.004
Gender, 1 = male	1.888	1.148	.161
Black ethnicity	−0.435	0.858	−.049
Other ethnicity	−1.490	1.818	−.080
Step 2				.161	.157^***
PBI Care	−0.099	0.055	−.186
PBI Overprotection	0.150	0.070	.227*

Note. PICTS-L = Psychological Inventory of Criminal Thinking–Layperson; PBI = Parental Bonding Instrument.

p < .05. ***p < .001.

Discussion

Results indicated that each criminogenic thinking score was related to only one of the parenting variables, but never both. Therefore, all four hypotheses were only partially supported. Overall criminogenic thinking, as well as Co and En thinking styles, were related to the care dimension of parenting, meaning that experiencing high levels of warm and emotionally close parenting is related to lower levels of these criminogenic thinking scores. Finally, the parenting dimension of Overprotection was significantly related to Mo criminogenic thinking, meaning that as parenting becomes more restrictive and controlling, self-justification for maladaptive behavior increases.

Beyond finding a link between parenting and criminogenic thinking, the findings from the current study are important because they shed light on the relationship between specific dimensions of parenting and criminogenic thinking in young adulthood. Examining this relationship further highlights the impact of childhood experiences on adulthood adjustment. According to Walters (2010), Co is often indicative of behavior that is impulsive or self-defeating (e.g., drug and alcohol abuse), as well as uncontrollable emotions. Because of the inverse relationship between Care parenting and Co, it is likely that parenting high in warmth may aid in the development of appropriate emotional expressions and healthy decision making. In regard to En criminogenic thinking, this form of thinking involves a sense of ownership in violating societal constraints as well as a tendency to identify wants as needs. Because of the relationship between care-based parenting and En, it seems that warm parenting may buffer against this thinking style, perhaps by boosting social responsibility. Parenting that lacks emotional warmth may be a risk factor for later Co and En thinking patterns and related problems.

Finally, Mo reflects neglecting personal responsibility for ones behavior. Because the results indicate that overprotective parenting is related to increases in Mo, it seems that parenting that is highly controlling of a child may negatively affect their ability to develop an appropriate sense of responsibility. Therefore, less controlling, and autonomy-boosting, parenting approaches may aid in a child developing a sense of personal responsibility for behavior. Overall, the relationships examined in the current study establish the nature of criminogenic thinking patterns that develop after childhood experiences of parenting.

Current findings seem to have support of previous literature stating that the three criminogenic thinking styles of En, Mo, and Co are related to parenting experiences (Cuadra, 2007). Although Cuadra’s (2005) work focused on extreme parental mistreatment in relation to criminogenic thinking, current findings shed light on the phenomenon of possibly harmful dimensions of parenting (e.g., high levels of psychological and behavioral control) and subsequent criminogenic thinking. It appears that more common detrimental aspects of parenting, and not just parental abuse, may be related to later criminogenic thinking. Also, the current results corroborate previous findings that lack of parental care is related to maladaptive outcomes (Barnes & Farrell, 1992) and that overprotection is related to antisocial attitudes in young adulthood (Zhou et al., 2004).

This study helps establish a link between parenting and criminogenic thinking. Thus, the findings from this study may aid adolescent counselors or university counselors who treat clients with such thinking styles, while considering that other studies evidence a connection between criminogenic thinking and problematic behaviors (e.g., Ragatz et al., 2011). Focusing treatment on altering criminogenic thinking styles and processing negative parent/child interactions may be beneficial for patients. Specifically, asking questions regarding Care and Protective facets of parenting may uncover risk factors for criminogenic thinking patterns. Furthermore, a measure of criminogenic thinking seems to be a fruitful tool to uncover specific criminogenic thinking styles to inform treatment aimed at reducing these types of thinking (PICTS, Walters, 2010; Measure of Offender Thinking Styles–Revised [MOTS-R], Mandracchia, Morgan, Garos, & Garland, 2007). Overall, the findings not only highlight the connection between criminogenic thinking and parental behaviors but also inform psychological treatment of criminogenic thinking and what potential behaviors are related to such thinking.

Results from this study have implications for the population represented in the current sample. Extensive research concerning variables of maladaptivity in college students focuses on behavior (e.g., elicit drug use), and often fails to examine the related constructs of criminogenic thinking (Fromme et al., 1997; Zimny et al., 2008). Therefore, the current study provides the levels of criminogenic thinking that a largely nondelinquent sample of college students engages in. The current study supports Walters et al.’s (2009) previous finding that college students do, in fact, engage in moderate levels of criminogenic thinking. A sizable portion of the present sample reported numerous unhealthy or irresponsible behaviors (i.e., underage alcohol consumption and drinking and driving), while some respondents reported cheating on examinations and having unprotected sex, respectively.

Given that the current study’s sample reported engaging in meaningful levels of criminogenic thinking as well as a variety of risky behaviors separately (see Table 2), using this information for college-based harm reduction programming or intervention efforts (e.g., alcohol/drug use prevention, safe sex interventions) may prove beneficial. Knowing what types of thinking errors to target with college students at risk for engaging in maladaptive behaviors (i.e., alcohol/drug use, unsafe sex) may aid helping professionals in developing and implementing these types of programs.

The current findings, which highlight the importance of parenting behaviors on college student’s thinking, provide important implications for parents and caregivers. Findings indicate that high levels of care are related to lower levels of criminogenic thinking. This type of parenting approach is considered a core component to secure parent–child attachments and is strongly tied to social competence and adjustment (Dekovic, Janssens, & Van As, 2003). Thus, parents are encouraged to increase their awareness about practices that facilitate a warm and nurturing home environment, in which a child’s potential is facilitated through support, empathy, and warmth. Research shows that specific parenting practices coupled with positive parent–child relationships, in which parents provide a caring and warm environment, result in the most positive behavioral outcomes in adolescents. For example, parenting characterized by proactive strategies, instead of solely reactive parenting strategies, results in positive behavioral outcomes in their adolescents, such as a reduction in externalizing disorders (Padilla-Walker, Christensen, & Day, 2011). These types of parenting practices include reasoned deference (i.e., discussing problems with child, and then allowing the child to make their decision) and pre-arming (i.e., providing a plan for action before a potential difficulty) skills (Padilla-Walker et al., 2011). Furthermore, parenting resources provide tips to engage in empathic and supportive parenting, such as listening to a child’s concerns, spending time with them engaging in their hobbies, and showing involvement in their interests (Center for Disease Control and Prevention, 2012).

The findings concerning parental protection can inform caregivers on healthy parenting behaviors to engage in regarding the issues of control and rule establishment. Previous literature has established that healthy control is characterized as consistent provision of appropriate rules and disciplinary boundaries (Dekovic et al., 2003). Parents are encouraged to avoid dysfunction responses to a child’s misbehavior or violation of rules, characterized by anger and overreactivity, and stringent disciplinary styles. Previous research has established that this type of stringent disciplinary style is contributory to internalizing problems (i.e., depressive symptoms) and externalizing problems (i.e., delinquency; Barber et al., 1994). Furthermore, healthy protective parenting involves the acknowledgement of a child’s individuality and a commitment to the development and promotion of healthy autonomy (Barber, 1996). Parents are encouraged to respect an older child’s need for privacy, and include the child when determining expectations and rules in a collaborative way (Center for Disease Control and Prevention, 2012).

Several limitations of the current study should be considered. The sample in the current study overwhelmingly identified as either Caucasian or Black, and the sample largely consisted of females. Therefore, caution should be taken in generalizing results to college students of other minority statuses and to males. The findings from the current study have implications for possible future studies. As mentioned previously, maladaptive behavior was not examined in the current study. Therefore, research should be conducted to determine which parenting dimensions and which criminogenic thinking styles are most related to subsequent maladaptive behavior in college students. Examining whether behavioral manifestations occur as a result of criminogenic thinking is an important relationship to explore because having a greater understanding of this relationship can aid in the prediction of maladaptive behaviors.

Also, a possibly fruitful avenue of research may involve examining parenting and maladaptive behavioral outcomes longitudinally. This research direction could result in a better understanding of the impact of previous parenting on young adults’ thinking and behavior. Overall, a better understanding of this population’s development of criminogenic thinking and related behavioral consequences (i.e., academic dishonesty, drinking under the influence, and underage drinking) is needed to ameliorate and reduce those harmful behaviors most commonly engaged in by college students. Future directions may also be aimed at determining the utility and effectiveness of criminogenic thinking-focused treatment for college-aged maladaptive behavior, as well as psychoeducational programs for parents to improve parenting strategies and caretaking behaviors to positively affect their children’s future behavior.

Footnotes

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) received no financial support for the research, authorship, and/or publication of this article.

References

Barber

(1996). Parental psychological control: Revisiting a neglected construct. Child Development, 67, 3296-3319.

Barber

Olsen

Shagle

. (1994). Associations between parental psychological and behavioral control and youth internalized and externalized behaviors. Child Development, 65, 1120-1136.

Barnes

Farrell

. (1992). Parental support and control as predictors of adolescent drinking, delinquency, and related problem behaviors. Journal of Marriage and Family, 54, 763-776.

Baumrind

(1971). Current patterns of parental authority. Developmental Psychology, 4(1, Pt. 2), 1-103.

Bowman

Prelow

Weaver

(2007). Parenting behaviors, association with deviant peers and delinquency in African American adolescences: A mediated-moderation model. Journal of Youth and Adolescence, 36, 517-527.

Butler

Fearon

Atkinson

Parker

(2007). Testing an interactive model of symptom severity in conduct disordered youth: Family relationships, antisocial cognitions, and social-contextual risks. Criminal Justice and Behavior, 34, 721-738.

Carey

Scott-Sheldon

Carey

DeMartini

(2007). Individual-level interventions to reduce college student drinking: A meta-analytic review. Addictive Behavior, 32(11), 2469-2494.

Carr

Rosenfeld

Magyar

Rotter

(2009). An exploration of criminal thinking styles among civil psychiatric patients. Criminal Behavior and Mental Health, 19, 334-346.

Center for Disease Control and Prevention. (2012). Child development, positive parenting tips. Retrieved from http://www.cdc.gov/ncbddd/childdevelopment/positiveparenting/index.html

10.

Cuadra

(2007). Child maltreatment and adult criminal behavior: Criminal thinking as the mediator (Doctoral dissertation). Retrieved from ProQuest Digital Dissertations. (UMI No. 3293888)

11.

Dekovic

Janssens

Van As

. (2003). Family predictors of antisocial behavior in adolescence. Family Process, 42, 223-235.

12.

Dembo

Jainchill

Turner

Fong

Farkas

Childs

(2007). Levels of psychopathy and its correlates: A study of incarcerated youths in three states. Behavioral Sciences & the Law, 25, 717-738.

13.

Dodge

(1991). The structure and function of reactive and proactive aggression. In Pepler

D. J.

Rubin

K. H.

(Eds.), The development and treatment of childhood aggression (pp. 201-218). Hillsdale, NJ: Lawrence Erlbaum.

14.

Fromme

Katz

Rivet

(1997). Outcome expectancies and risk-taking behavior. Cognitive Therapy and Research, 21, 421-442.

15.

Gendreau

Andrews

Goggin

Chanteloupe

(1992). The development of clinical and policy guidelines for the prediction of criminal behaviour in criminal justice settings (Programs Branch user report). Ottawa, Ontario: Ministry of the Solicitor General of Canada.

16.

Hoeve

Dubas

Eichelsheim

van der Laan

Smeenk

Gerris

(2009). The relationship between parenting and delinquency: A meta-analysis. Journal of Abnormal Child Psychology, 37, 749-775.

17.

Hughes

A. L.

(2009). Criminal thinking and drug treatment entry among federal inmates. (Doctoral dissertation). Retrieved from ProQuest Digital Dissertations. (UMI No. 3401276)

18.

Kingston

Prior

(1995). The development of patterns of stable, transient, and school-age onset aggressive behavior in young children. Journal of the American Academy of Child & Adolescent Psychiatry, 34, 348-358.

19.

Knutson

DeGarmo

Reid

(2004). Social disadvantage and neglectful parenting as precursors to the development of antisocial and aggressive child behavioral: Testing a theoretical model. Aggressive Behavior, 30, 187-205.

20.

Liau

Barriga

Gibbs

(1998). Relations between self-serving cognitive distortions and overt vs. covert antisocial behavior in adolescents. Aggressive Behavior, 24, 335-346.

21.

Loeber

(1990). Development and risk factors of juvenile antisocial behavior and delinquency. Clinical Psychology Review, 10, 1-41.

22.

Maccoby

Martin

(1983). Socialization in the context of the family: Parent-child Interaction. In Mussen

P. H.

(Ed.), Handbook of child psychology: Socialization, personality and social development (Vol. IV, pp. 1-101.). New York, NY: John Wiley.

23.

Magyar

Carr

Rosenfeld

Rotter

(2010). An exploration of the relationship between criminal cognitions and psychopathy in a civil psychiatric sample. International Journal of Offender Therapy and Comparative Criminology, 54, 865-877.

24.

Mandracchia

J. T.

Morgan

Garos

Garland

J. T.

(2007). Inmate thinking patterns: An empirical investigation. Criminal Justice and Behavior, 34, 1029-1043.

25.

McCoy

Fremouw

Tyner

Clegg

Johanssson-Love

Strunk

(2006). Criminal-thinking styles and illegal behaviors among college students: Validation of the PICTS. Journal of Forensic Science, 51, 1174-1177.

26.

Morgan

R. D.

Bauer

Fisher

W. H.

Duan

Mandracchia

J. T.

Murray

(2008). What happens to offenders with mental illness during incarceration? Effects of imprisonment on mental health and criminal risk factors. Unpublished manuscript.

27.

Padilla-Walker

Christensen

Day

(2011). Proactive parenting practices during early adolescence: A cluster approach. Journal of Adolescence, 34, 203-214.

28.

Parker

Tupling

Brown

(1979). A parental bonding instrument. British Journal of Medical Psychology, 52, 1-10.

29.

Pelcovitz

Kaplan

Ellenberg

Labruna

Salzinger

Mandel

Weiner

(2000). Adolescent physical abuse: Age at time of abuse and adolescent perception of family functioning. Journal of Family Violence, 15, 375-389.

30.

Ragatz

Anderson

Fremouw

Schwartz

(2011). Criminal thinking patterns, aggression styles, and psychopathic traits of late high school bullies and bully-victims. Aggressive Behavior, 37, 145-160.

31.

Schaffer

Clark

Jeglic

(2009). The role of empathy and parenting style in the development of antisocial behaviors. Crime & Delinquency, 55, 586-599.

32.

Steinberg

Blatt-Eisengart

Cauffman

(2006). Patterns of competence and adjustment among adolescents from authoritative, authoritarian, indulgent, and neglectful homes: A replication in a sample of serious juvenile offenders. Journal of Research on Adolescence, 16, 47-58.

33.

Walters

(1990). The criminal lifestyle: Patterns of serious criminal conduct. Newbury Park, CA: Sage.

34.

Walters

(1995). Psychological inventory of criminal thinking styles: I. Reliability and preliminary validity. Criminal Justice and Behavior, 22, 307-325.

35.

Walters

(2002). The Psychological Inventory of Criminal Thinking Styles (PICTS): A review and meta-analysis. Assessment, 9, 278-291.

36.

Walters

(2006). Appraisal, researching and conceptualizing criminal thinking: A personal view. Criminal Behavior and Mental Health, 16, 87-99.

37.

Walters

(2007). Taxometric analysis of the Psychological Inventory of Criminal Thinking Styles in incarcerated offenders and college students. Criminal Justice and Behavior, 34, 781-793.

38.

Walters

(2010). The psychological inventory of criminal thinking styles professional manual. Allentown, PA: Center for Lifestyles Studio.

39.

Walters

Felix

Reinoehl

(2009). Replicability and cross-gender invariance of a two-dimensional model of antisociality in male and female college students. Personality and Individual Differences, 46, 704-708.

40.

Walters

Geyer

(2004). Criminal thinking and identity in male white-collar offenders. Criminal Justice and Behavior, 31, 263-281.

41.

Yochelson

Samenow

(1976). The criminal personality, Volume I: A profile for change. Northvale, NJ: Jason Aronson.

42.

Zhou

Eisenberg

Wang

Reiser

(2004). Chinese children’s effortful control and dispositional anger/frustration: Relations to parenting styles and children’s functioning. Developmental Psychology, 40, 352-366.

43.

Zimny

Robertson

Bartoszek

(2008). Academic and personal dishonesty in college students. North American Journal of Psychology, 10, 291-312.