The Positive Effects of Parentification : An Exploratory Study among Students

The present study was designed to examine the relationship between parentification and choice of education: Psychology versus humanities. Additionally, we investigated the association between parentification, on the one hand, and cognitive and affective empathy as well as resilience, on the other. The rational for this study was the increasing evidence that parentification may not only induce several possible adverse effects but that it can also facilitate the development of some specific positive abilities (e.g., higher empathic skills and resilience). We compared 265 psychology students with 51 humanities students on the variables parentification, empathy, and resilience. Within the group of psychology students, we conducted hierarchical regression analyses on cognitive empathy, affective empathy, and resilience with parentification and possible confounders as predictors. Psychology students reported more parentification experiences in their families than the humanities students, but they did not score higher on empathy and resilience. Among psychology students, parentification was associated with higher resilience and higher cognitive empathy, while there was no connection with affective empathy. These findings partially support the hypothesis of specific mental growth in parentified children.


Introduction
Adequate and sensitive parenting is a prerequisite for the healthy development of children (Thompson, 2008).This not only holds true for early childhood, but also in adolescence an enmeshed parent-child dyad can compromise intrafamilial roles and interpersonal boundaries and seriously interfere with a normal healthy development (Garber, 2011).The term parentification refers to a form of role-reversal within families.More precisely, children take on responsibilities and roles that are usually taken by parents and adult caregivers (Boszormenyi-Nagy & Spark, 1973).This may occur when parents fail to accomplish the adult role adequately, such as in cases of parental abuse of alcohol or drugs (Godsall, Jurkovic, Emshoff, Anderson, & Stanwyck, 2004), in one-parent households (Jurkovic, Thirkield, & Morrel, 2001), when one of the parents is militarily deployed (Harrison & Albanese, 2012), within immigrant families (Titzman, 2012), but also if one of the parents is a workaholic (Carroll & Robinson, 2000).In all these examples, the parents create an environment in which caring behavior by the child is promoted (Hooper, 2008).Jurkovic (1997) distinguishes between instrumental and emotional parentification.Instrumental parentification consists of practical help, like grocery shopping, cooking, and taking care of parents or siblings, whereas emotional parentification refers to emotionally supporting family members.This may include taking the role of confidant for specific family members or being a mediator during conflicts.
However, this is just one side of the coin.More recently, there is increasing suggestive evidence that parentification may have positive effects on the child (Hooper, 2008;Jurkovic, 1997) and can function as a buffer as well as a mediator for negative consequences (Hooper, Doehler, Jankowski, & Tomek, 2012).For example, Kuperminc, Jurkovic, and Casey (2009) found that parentification was positively related to interpersonal competence.Furthermore, relationships have been shown between parentification with adaptive coping skills (Stein, Rotheram-Borus, & Lester, 2007), self-efficacy (Titzman, 2012), and social and interpersonal competence (Champion et al., 2009).
Parentification may also promote personality development.For example, a salient feature of the parentified child is exceptional sensitivity to the wishes and needs of others (DiCaccavo, 2006), because it learned to anticipate the emotional state of parents and siblings and, therefore developed a "powerful emotional antenna" (Glickauf-Hughes & Mehlman, 1995, p. 213).This effect may manifest itself as increased empathy, an interpersonal skill with both a cognitive and an affective component (Strayer, 1987).Cognitive empathy is the ability to understand emotions of others, whereas affective empathy refers to the capacity to experience emotions of others (Jolliffe & Farrington, 2006).Whether parentification specifically promotes one kind of empathy or both forms to the same degree has yet to be determined.
Resilience, another possible positive outcome of parentification (Ungar, Theron, & Didkowsky, 2011), refers to the ability to reduce the negative effects of stressors and to bring about positive change in negative circumstances (Wagnild & Young, 1993).A child's responsibility for family tasks when the family is under stress may facilitate the development of coping and problem-solving skills (Hooper, 2008;Stein et al., 2007;Walsh, Shulman, Bar-On, & Tsur, 2006).Overcoming hardship at an early age may result in a higher resilience, much the same as has been described for the exposure to stressors in general, which may "inoculate" individuals and foster their resistance (Meichenbaum, 2007).
Parentification can further be a motivating factor for the choice of caring professions (DiCaccavo, 2002).Nikcevic, Kramolisova-Advani, and Spada (2007) demonstrated that psychology students reported more parentification than business students.Earlier, Fussell and Bonney (1990) showed that psychotherapists reported more child-parent role reversal, more neglect and having taken on a caring role in their families more often than physicists.
The present study examines whether there are differences in the degree of parentification between psychology students and humanities students.We further explored the relationship between parentification on the one hand, and cognitive empathy, affective empathy, and resilience on the other, in the psychology students.The level of neuroticism in parents can be accounted as both a possible cause of the development of instability in the family (Jardine, Martin, Henderson, & Rao, 1984) and a higher degree of neuroticism in the child (Viken, Rose, Kaprio, & Koskenvuo, 1994).The degree of neuroticism of the parentified person is therefore taken into account as a possible confounding factor.
We hypothesize that psychology students will report more parentification than the humanities students.Furthermore, we anticipate that childhood parentification is associated with increased levels of empathy and resilience.

Participants
Two-hundred and seventy-five psychology students (199 women, 76 men), aged 17-48 (M=20.2,SD=3.1) and 52 students from the Faculty of Humanities (19 females, 32 males), aged 17-27 (M=21.1 SD=2.3) took part in the study.Ten psychology participants were kept out of the analyses.Five were under 18, three were substantially older than the majority of the sample (36,43,48), one participant had a considerable amount of missing data, and one participant responded neutrally or extremely positive on most of the questionnaires.The analyses were thus performed on the responses of 265 psychology students, ranging in age from 18 to 28.The majority (95.1%) were Dutch, and 48.3% were involved in a romantic relationship.The majority (61%) of participants aimed to specialize in 'Psychology and Health,' whereas 22% expected to specialize in 'Psychology and Society.' Seventeen per cent had not yet decided about their future specialization.One humanities student was excluded from the analysis due to being under 18.The majority of humanities students (96.1%) had the Dutch nationality, and 60.8% were in a romantic relationship.Participants received course credits for participation.These were required to complete the first year of their studies.

Procedure
After registration, participants received an electronic link to the questionnaire, which took 50 to 60 minutes to complete.The data collection ran between January 2013 and April 2013.

Measures
Participants provided the following background information: age, gender, ethnicity, and whether they were in a romantic relationship (demographic variables).
The Parentification Questionnaire -Adult (PQ-A; Sessions & Jurkovic, 1986;Dutch translation: Truyens, 1998) was used to measure childhood parentification.It contains 42 statements (e.g., "There were times when I felt I was the only one who could support my mother or father") with a true/not true response format.Seventeen items were recoded, and the sum score was used as an indication of the degree of parentification.The original English questionnaire was found to have good internal (α=.83) and test-retest reliability (r=.86) (Sessions & Jurkovic, 1986).In the current sample, Cronbach's alpha was .84.
Empathy was assessed using the Basic Empathy Scale (BES; Jolliffe & Farrington, 2006;Dutch translation: Lehman, Huis in 't Veld, & Vingerhoets, 2013), which assesses cognitive and affective empathy.It contains eleven questions for affective empathy (e.g., "Other people's feelings don't bother me at all" and "I often get swept up in my friend's feelings") and nine questions for cognitive empathy (e.g., "When someone is feeling 'down' I can usually understand how they feel" and "I have trouble figuring out when my friends are happy").Answers are given on a 4-point Likert scale ranging from 'strongly agree' to 'strongly disagree.'The reliability of the entire scale was α=.87, of the cognitive subscale α=.74, and of the affective subscale α=.85 (Jolliffe & Farrington, 2006).Prior research with the Dutch translation found a reliability of α=.75 for both subscales (Lehman et al., 2013).
To measure neuroticism, we used the neuroticism subscale of the NEO Five-Factor Inventory (Costa & McCrae, 1992;Dutch translation: Hoekstra, Ormel, & De Fruyt, 2003).This trait is in particular characterized by upsetability and is the polar opposite of emotional stability.The 12 items are answered on a 5-point Likert scale, ranging from 'completely agree' to 'completely disagree'.The Dutch version has good internal and test-retest reliability (Hoekstra et al., 2003).In the current sample, we found a Cronbach's alpha of .86.

Regression Analyses
To evaluate our hypotheses regarding the relationships between parentification and the dependent variables, we performed three hierarchical regression analyses.Tables 2 and 3 summarize the results of these analyses.Both BES subscales were regressed on the independent variables age, gender, relationship status, neuroticism, resilience, and parentification.In the final step, parentification was added.This model explained 12% of the variance in cognitive empathy (F(6,258)=5.75,p<.001).Controlled for the other variables in the model, we found positive relationships between both parentification and resilience, on the one hand, and cognitive empathy, on the other.No associations were found between cognitive empathy and the other predictors (i.e., age, gender, relationship status and neuroticism).
In the case of affective empathy, the addition of parentification failed to increase the amount of explained variance.We, therefore, chose the second model, in which the predictors explained 32% of the variance in affective empathy (F(6,258)=20.04,p<.001).Controlled for the other variables in the final model, women scored higher on affective empathy and also neuroticism was positively related to affective empathy.We found no significant association between affective empathy, on the one hand, and resilience and relationship status, on the other.The hypothesis of a positive linear relationship between resilience and parentification was examined in the final regression analysis.The dependent variable resilience was regressed on age, gender, relationship status, cognitive empathy, affective empathy, neuroticism, and parentification.The addition of parentification in the third step resulted in a significant increase of the explained variance.In this model, the predictors explained 35% of the variance in resilience (F(7,257)=22.29,p<.001).Controlled for the other variables, positive relationships were found between the dependent variable resilience and the predictors age, cognitive empathy, and parentification.In addition, a negative association was found between resilience and neuroticism.There were no significant associations between resilience and the other predictors.

Discussion
The finding that psychology students reported more childhood parentification than humanities students is consistent with previous reports of higher levels of parentification and trauma in mental health professionals and psychology students (Elliot & Guy, 1993;Fussell & Bonney, 1990;Nikcevic et al., 2007;  2002 for opposite findings).Our results thus further substantiate the role of parentification as a possible determinant of pursuing a career as a psychologist.
Remarkably, there were no significant differences in the levels of cognitive and affective empathy between both student groups.This is surprising as the work of mental health professionals requires highly developed empathic skills (Hassenstab, Dziobek, Rogers, Wolf, & Convit, 2007).One possible explanation is that empathic skills first increase as students reach more advanced levels and when they obtain more experience in the mental health field of work.This has been found earlier among physiotherapists (Thomson, Hassenkamp, & Mansbridge, 1997) and midwifery students (McKenna et al., 2011).
Our results further support the hypothesis of a positive linear relationship between parentification and, specifically, cognitive empathy.This finding is consistent with characteristic descriptions of parentified therapists which suggest that they have a better understanding of the feelings of others (Glickauf-Hughes & Mehlman, 1995;Miller, 1981).More generally, it corresponds with previous findings demonstrating that parentified individuals show higher competency on interpersonal skills (e.g., Kuperminc et al., 2009).
We did not find a similar positive linear relation between parentification and affective empathy, which suggests that parentified students are not necessarily more proficient in experiencing the emotions of others.This discrepancy in findings between affective empathy and cognitive empathy is surprising, as most researchers (e.g., Jolliffe & Farrington, 2006) consider affective empathy as a precursor to cognitive empathy.However, this might be a too simple representation of the facts.For example, alternatively, cognitive empathy and affective empathy might be regarded phenomena that develop independently in the course of development, and therefore should be considered separately.Indeed, some previous studies found a relative independence of cognitive and affective empathic skills (e.g., Dziobek et al., 2008;Harari, Shamay-Tsoory, Ravid, & Levkovic, 2010;Schechtman, 2002).
It is important to be aware that our cross-sectional research design does not allow drawing more definitive conclusions regarding the direction of the found relationship.For example, a more empathic child may more likely take on the parentified role.In addition, an interaction between aptitude and experience is not unlikely.It may require a certain degree of empathy to observe and understand the distress of parents or siblings.The child's skill to determine the emotional state of others may further develop, when, in the parentified role, it practices assessing the needs of other persons.However, this reasoning is just speculative and should be evaluated in future, preferably longitudinal research.
As anticipated, parentified psychology students wereat equal levels of empathy and neuroticismmore resilient than their non-parentified fellow students.This finding supports the theory of Kuperminc et al. (2009, p. 15), who describe the influence of parentification as "competence at a cost."The higher competency does not exclude a possible negative long-term influence of parentification on the mental health of the child (Hooper et al., 2011;Jurkovic, 1997).For example, according to Glickauf-Hughes and Mehlman (1995), parentified therapists might be more vulnerable to burnout, because they have trouble expressing their own wishes and needs, and fail to properly set their personal and professional boundaries.
The present results additionally raise some questions about the validity of the affective empathy subscale of the BES.Without controlling for neuroticism, we found a negative relationship between age and affective empathy.This does not correspond with Hoffman's (1976) notion that empathy increases during adolescence and college years.This surprising result might be explained by the scale's emphasis on negative emotions, such as fear, anger or sadness, whereas the sharing and coexperiencing of positive feelings are not addressed.This could also explain the moderate positive correlation (r=.39) between neuroticism and affective empathy.However, this unexpected finding may also result from the limited age range of the participants.
Some further limitations of the present study also deserve attention.The sample sizes of both student groups differ considerably.Moreover, as all variables were assessed with questionnaires, the results may be prone to self-report-bias.Although the recollection of childhood events is deemed fairly accurate (Brewin, Andrews, & Gotlib, 1993), social desirability, over-reporting, underreporting may all have had an influence on our findings.Finally, we measured self-reported empathy instead of other-observed empathy.More generally, there may be discrepancies between an individual's self-assessment of his/her empathic skills and his or her level of empathy as others experience it.
Future research on the effects of parentification among counselors and psychology students may benefit from distinguishing instrumental and emotional parentification.The effects of emotional parentification are likely to be more detrimental to mental health (Hooper et al., 2011) and it is possible that the effects of instrumental and emotional parentification on empathy also differ.Specifically, stronger effects of emotional parentification might be anticipated, as children who offer emotional support and advice to family members may even be more empathetic than children who mainly do household chores.Furthermore, it is interesting to connect our results to the treatment outcome of psychologists who are already employed in clinical practice.In most psychological and psychotherapeutic traditions, the therapist's empathy is considered among the most effective factors in the working alliance (Elliott, Bohart, Watson, & Greenberg, 2011;Steering Committee, 2001).
Of course, empathy is only one facet of the counselor's personality and other interpersonal factors, such as positive regard and congruence (Norcross, 2002), also likely affect the effectiveness of the counselor.Because parentification is common in this profession, more research in this area is needed to obtain insight into the interrelationships and its practical relevance for training and guidance of counselors and therapists.The present study emphasizes the importance of a better insight into both the weaknesses and the strengths of parentified therapists.
In conclusion, this study is among the first to show a positive relationship between parentification and cognitive empathy.We additionally found a positive association between resilience and childhood parentification.Finally, psychology students reported more parentification in their childhood than humanities students.For the training and supervision of psychologists, it is important to have a complete picture of the positive and negative effects of parentification.For these professions, empathy is a most important variable for its demonstrated positive effect on treatment outcome.The (future) therapist may -being aware of his strengths and weaknesses make use of past experiences to achieve a better treatment outcome.

Table 3 .
Regressions of ResilienceNote.Standardized regression weights from the final model are presented.