J Adult Dev (2011) 18:204–213 DOI 10.1007/s10804-011-9129-8
Relationship of Adult Representations of Childhood Parenting and Personality Tendencies to Adult Stressors and Political Ideology Carolyn A. Blondin • Jeff L. Cochran • Eun Jung Oh • Cora M. Taylor • Robert L. Williams
Published online: 12 October 2011 Ó Springer Science+Business Media, LLC 2011
Abstract Undergraduate students in a Southeastern US University (n = 232) responded to an inventory that included retrospective measures of their parents’ style of parenting (authoritarian vs. authoritative) and their own childhood psychological tendencies (insecurity vs. confidence), as well as their adult stressors and political orientation. Authoritative parenting positively correlated with childhood confidence and negatively correlated with both childhood insecurity and adult stressors. Conversely, authoritarian parenting was positively associated with childhood insecurity and adult stressors but was not significantly correlated with childhood confidence. For the most part, parenting styles, early childhood tendencies, and adult stressors were unrelated to adult political ideology, contrary to previous longitudinal research reporting these connections. Keywords Retrospective research Parenting Childhood security Stressors Political ideology
the relationship between childhood parenting and both childhood and adult characteristics. However, observing parenting behavior and children’s tendencies may produce more reliable and valid data than asking adults to describe their parent(s)’ behavior and report their own tendencies as children. One’s recall of childhood parenting and personal characteristics could be affected by numerous experiences between childhood and adulthood, as well as by personality tendencies and beliefs during adulthood. On the other hand, there are practical disadvantages of longitudinal research, including sample access, plus time and resources to track persons and data trends across significant portions of the lifespan. The retrospective approach provides a viable alternative to understanding how persons develop across time. This approach can serve to confirm longitudinal findings and build a reliable knowledge base through larger samples and the possibilities of numerous studies across similar variables and similar or varied populations. Retrospective Research Findings
Introduction There are two major ways to study the relationship between childhood experiences and adult characteristics: (a) longitudinally tracking the same individuals from childhood to adulthood and (b) asking adults to report memories of their childhood experiences and their current adult characteristics. The former is a more highly regarded way to examine
C. A. Blondin J. L. Cochran E. J. Oh C. M. Taylor R. L. Williams (&) Educational Psychology and Counseling, The University of Tennessee, Knoxville, TN 37996-3452, USA e-mail:
[email protected]
123
Adult reflections on childhood may not be entirely objective but those reflections may profoundly affect other aspects of adult lives. Asking adults to retrospectively report events from their childhood gives researchers access to a multitude of childhood experiences that would otherwise be lost if only longitudinal data were used. The current research is based on the assumption that retrospective research is a reliable and valuable way to determine how adults experienced childhood. Consistent with this assumption, MacDonald et al. (2009) found that lifestyle characteristics and physical activity could be accurately recalled for a period of 25 years in samples of both younger and older adults.
Relationship of Adult Representations
A sampling of studies that have used the retrospective approach to investigate the linkage between childhood variables and various adult characteristics/experiences includes Alea et al. (2004), Dolbin-MacNab et al. (2009), Faith et al. (2008), Gumpel and Ish-Shalom (2003), and Knabb et al. (2009). Alea et al. concluded from their research that theories of emotion and aging could be informed by memories of negative life events. DolbinMacNab et al. reported that adults who had been reared by grandparents evidenced intense emotional bonds with their grandparents and regarded those relationships as equivalent to parent–child relationships. Faith et al. found that reported teasing during childhood was linked to later symptoms of psychological distress. Gumpel and Ish-Shalom studied the long-term effects of peer rejection in childhood and adolescence, as well as protective factors that promoted remission of earlier peer rejection. Knabb et al. investigated retrospective meaning-making of Protestant adults whose parents had divorced during the adolescence of the participants. The retrospective studies most directly linked to the focus of the current study include Heer (2008), Klein et al. (1996), and Renk et al. (2005). One of the salient findings of the Heer study was that authoritative parenting from both mothers and fathers predicted secure attachments in adult relationships. Klein et al. reported that an authoritative style of parenting in childhood generally predicted positive self-perceptions in adulthood, whereas an authoritarian style generally predicted negative self-perceptions. Renk et al. found that college students and their parents moderately agreed in their recollections of the college students’ externalizing and internalizing problems during childhood. Overall, these retrospective studies suggest that adult ratings of the parenting received as children predict self-concept and social attachments in adulthood. Additionally, parents and college students generally agreed in their reporting of college students’ problem-coping tendencies as children. Related Longitudinal Research Findings The rather limited retrospective literature on childhood parenting and adult psychological tendencies is complemented by one major longitudinal study (Block and Block 2006) of the personality characteristics of nursery-school children and their personality and political ideology as young adults. Particularly unique about the Block and Block study is the linkage between childhood personality characteristics and adult political ideology. Early childhood descriptors reflective of insecurity (e.g., fearful, rigid, vulnerable) were positively associated with adult conservatism, whereas childhood descriptors conveying confidence (e.g., self-reliant, resilient, close peer relationships)
205
were positively associated with adult liberalism. Additionally, childhood personality tendencies predicted adult personality characteristics similar to the childhood patterns: Adult liberals were rated as being bright, having wide interests, and tending toward non-conformity, whereas conservatives were rated as uneasy with ambiguity, traditionally sex-typed, and moralistic. Although their findings present clear trends regarding early childhood and adult personality characteristics of liberals and conservatives, Block and Block (2006) acknowledged that aspects of their available sample may limit the generalizability of their findings. The parent sample was diverse with respect to social class and education level but relatively small (n = 128 children) and skewed toward liberalism. The authors also noted that the study was conducted in one location (Berkeley and Oakland, California) widely recognized for its liberalism. Additionally, the authors acknowledged that the correlations between nursery school descriptors and adult liberalism– conservatism measures may be considered unimpressive (correlations mainly in the 0.20s to 0.40s). Other than these acknowledged limitations, Block and Block’s (2006) methods appear sound and comprehensive. All the ratings were done independently by individuals or sets of individuals without knowledge of the overall design and the expectations of the study. Furthermore, the ratings were based on extensive contact with the participants. For example, the two sets of three teachers who judged the personality characteristics of the youngsters had been with the children daily over a 7-month period. The six psychologists who rated the adult characteristics had done extensive interviewing or observing of the participants. The psychologists did their ratings independently of and prior to the assessment of liberalism and conservatism. Assessment of political ideology included self-assessment, agreement with positions usually distinguishing liberal from conservative views, and other scales related to political rights, political activism, conservatism, and liberalism. Certainly, the assessment of childhood personality tendencies, adult personality characteristics, and adult political ideology was broadly based, greatly decreasing the probability of bias resulting from singular measures of variables. To our knowledge, the Block and Block (2006) study is the only published longitudinal investigation of how childhood characteristics translate into adult political ideology. The study was conducted over approximately a 20-year period. Such studies, particularly with the methodology used by Block and Block, are labor-intensive. For that reason in particular, the study deserves much attention in the empirical literature on the development of conservatism and liberalism. Nonetheless, their skewed sample and the small to medium correlations between childhood
123
206
characteristics and adult political ideology suggest the need for both replication and extension of selected aspects of the Block and Block study. An issue needing attention in the Block and Block (2006) study is what contributed to the correlation between childhood personality tendencies and adult political values. One possibility would be a difference in parenting experienced by young adult liberals and conservatives. Drawing from the same longitudinal data set used in the Block and Block (2006) study, Feld (1995) earlier reported that the parenting behavior experienced by adult liberals during childhood had been generally more supportive than that experienced by adult conservatives. Parents of adult liberals were generally described as tender, playful, affectionate, easy going, and supportive, whereas parents of adult conservatives were described more as cold, tense, punitive, and authoritarian. One exception to this overall trend was fathers’ treatment of their daughters: Fathers of conservative daughters were inclined to be easy going and firm, whereas fathers of liberal daughters were described as unsupportive and distant.
Parenting Styles and Children’s Psychological Tendencies Parenting styles have often been classified as authoritarian, authoritative, and permissive (Burl 1991; Robinson et al. 1995). The major distinctions between these categories relate to the source and nature of parental control in children’s lives (Baumrind 1971). The principal source of control in both authoritarian and authoritative parenting rests with the parent(s), although the nature of control differs between these two categories. Authoritarian parents’ directives are final and are not to be questioned by their children. Authoritative parents also set limits for their children but allow them to question and sometimes negotiate those limits. Permissive parents attempt to follow their children’s lead and otherwise set minimal limits on their children’s choices and behavior. Although parents and children can reciprocally affect each other, most findings suggest that parenting style influences the development of children’s behavior and traits, rather than children’s psychological tendencies affecting parenting style (e.g., Aguilar 1998; Kitamura et al. 2009; Lytton 1990; Patterson 1982). In this vein, Bornstein’s (1992) review of developmental, social, and clinical studies of dependency concluded that overprotective, authoritarian parenting leads to child dependency and eventually to a variety of problems (e.g., depression, alcohol use and dependence, tobacco dependence, obesity, and eating disorders). Soenens et al. (2005) reported links from mothers’ and fathers’ psychological control to
123
C. A. Blondin et al.
daughters’ maladaptive perfectionism. Haycraft and Blissett (2010) found authoritarian and permissive parenting styles, but not authoritative, to be related to higher eating disorder symptoms. Berge et al. (2010) also found authoritarian parenting to be associated with higher overweight status among teenage sons.
Continuity of Psychological Patterns Across the Lifespan Existing research suggests that childhood personality characteristics may predict adult tendencies, including stress level or accumulated stressors. A longitudinal study with 104 participants linked key personality traits in children as young as 3–4 years old to personality traits or tendencies 12–20 years later (Block 1993). Block concluded that children’s tendencies toward ego-control (a capacity to control impulses, desires, or actions) are identifiable from early childhood and continue to distinguish individuals at least into early adulthood. Similarly, differences in ego-resilience (the ability to cope with or adapt to stress or change) were identifiable in young children and tended to remain stable for boys from childhood to young adulthood. From the same longitudinal data set, Block et al. (1988) found tendencies within the traits of ego-control and ego-resiliency at preschool age to be related to drug-use patterns in adolescence. Further, Block et al. (1991) linked ego-control and egoresilience in early childhood to depressive tendencies in late adolescents. Ego-under-control from early childhood in boys predicted later tendencies toward dysthymia, while ego-over-control in young girls predicted later depressive tendencies. From her review of related research, Hampson (2008) concluded that personality traits have an enduring effect on well-being and that children’s reactions to stress may play a role in adult stress. As Hampson illustrates, a child who reacts to stress with hostility may become entrapped in a cycle of reciprocal hostility, resulting in chronic stress. Along that same line, a 19-year longitudinal study (Dennissen et al. 2008) found that less resilient personality types in children predict later difficulties in young adult roles, including difficulties with leaving the parental home, establishing a first romantic relationship, and getting a part-time job. The less resilient personality types included both under-controlling (a lack of ego-control or restraint of impulses, which linked to aggression) and over-controlling (excessive ego-control, which linked to shyness). With respect to the current study, we predicted that children prone toward insecurity would experience a greater magnitude of stressors in their adulthood than individuals who were confident as children.
Relationship of Adult Representations
Research Connecting Psychological Tendencies to Political Leanings The final link in the framework of Block and Block’s (2006) longitudinal study was the relationship of childhood and adult personality characteristics to adult political leanings. Recent studies seem to concur with parts of Block and Block’s findings. A series of three studies conducted by Jost et al. (2007) showed that ‘‘uncertainty avoidance (e.g., need for order, intolerance of ambiguity, and lack of openness to experience) and threat management (e.g., death anxiety, system threat, and perceptions of a dangerous world) each contributed independently to conservatism (vs. liberalism)’’ (p. 989). In a meta-analysis of 88 samples across 12 countries and 22,818 cases, Jost et al. (2003) concluded that several psychological variables predict political conservatism: death anxiety, perceived threats to social stability, intolerance of ambiguity, openness to experience (negative relationship), fear of threat and loss, and self-esteem (negative relationship). In another series of three different studies that included multiple domains and measurement techniques, Carney et al. (2008) claimed ‘‘consistent and converging evidence that personality differences between liberals and conservatives are robust, replicable, and behaviorally significant’’ (p. 807). Recently, Inbar et al. (2009) added to the body of literature linking psychological tendencies to political orientations, finding that a predisposition to feel disgust is associated with more conservative political attitudes and that this relationship is strongest for purity-related issues— specifically, abortion and gay marriage.
Framework of the Current Study Our study investigated whether the principal relationships reported in the Block and Block (2006) and Feld (1995) studies would be supported in a retrospective study of the linkage between childhood and adult experiences. However, we added parenting style as a way to possibly account for some potential linkages between childhood and adult characteristics. As opposed to incorporating a general measure of adult personality, as was done by Block and Block, we used a more focused measure of adult stressors. We predicted that authoritarian parenting would be positively associated with childhood insecurity, adult stressors, and adult conservatism but negatively related to childhood confidence and adult liberalism. In contrast, authoritative parenting style was expected to be positively correlated with childhood confidence and adult liberalism but negatively associated with childhood insecurity, adult stressors, and adult conservatism. Following from Block and Block’s (2006) findings, we also expected early childhood
207
psychological tendencies recalled by college students to predict adult stressors and political ideology.
Method Participants Students in the teacher-education program of a Southeastern American University (N = 232) responded to an inventory that included all the measures targeted in this study. In this sample, 168 of the participants were women and 64 were men. With respect to academic classification, most of the participants were 2nd- and 3rd-year students: 1st year (n = 18), 2nd year (n = 113), 3rd year (n = 65), 4th year (n = 27), and graduate students (n = 9). Most of the participants had been reared in two-parent homes: two parents (n = 128), single parent (n = 20), and not indicated (n = 84). Procedures Students responded on an out-of-class basis to an inventory containing all the measures included in the study. This inventory was posted at the course website at the beginning of the semester. Students were given scan forms on which they indicated their answers to the items on a 5-point scale spanning strongly representative to strongly non-representative, very often to never, or strongly support to strongly oppose. Item responses to the combined inventory were electronically scanned into a comprehensive data file. With the exception of an integrated conservative-liberal scale, high scores on each scale represented support for the concept assessed (i.e., authoritative parenting, authoritarian parenting, childhood insecurity, childhood confidence, and adult stressors). An integrated conservative-liberal scale was scored in the direction of a liberal political orientation, with higher scores representing liberalism and lower scores representing conservatism. Targeted Measures Items on the authoritarian and authoritative parenting subscales were adapted from a larger instrument that also included a measure of permissive parenting (Burl 1991). Burl reported that the test–retest reliability of the authoritarian and authoritative scales over a 2-week period ranged from 0.78 to 0.92 and their Cronbach’s alphas ranged from 0.82 to 0.87. In our adaptation of the scale, we used the term ‘‘caretaker(s)’’ to represent the primary person(s) (mother, father, grandmother, grandfather, other relative, and/or acquaintance) who had reared the student. We assumed the person(s) portrayed as the caretaker(s) in
123
208
an individual’s inventory responses represented the adult(s) with whom the respondent spent the most time during childhood and who had the greatest influence on the respondent. Students responded to the items in these subscales on a (1) strongly representative to (5) strongly nonrepresentative basis. The authoritative and authoritarian themes reflected conventional interpretations of these parenting styles. Authoritative parenting represented a caretaker–child relationship in which caretaker(s) established limits for their children but not without input from the children. Children could question those limitations, with the possibility that their questions might cause caretaker(s) to alter the limitations. On the other hand, authoritarian caretaker(s) set limits for their children without regard to their children’s input. Authoritarian caretaker(s) demanded respect and obedience from their children under all circumstances. The measures of childhood personal tendencies (i.e., insecurity versus confidence) were adapted largely from Block and Block (2006), who used a Q-set scale to evaluate the behavioral tendencies of youngsters in kindergarten. Three teachers in two different nursery schools independently identified several patterns that represented what we referred to in our study as childhood insecurity (e.g., feeling unworthy, easily offended, being anxious about uncertainties) and confidence (e.g., being resourceful, behaving autonomously, reflecting pride in their accomplishments). Our 8-item measure of these two scales included 4 items that reflected childhood insecurity and 4 items that represented childhood confidence. Students indicated how often they felt a particular way as a child on a 5-point scale from 1 (very often) to 5 (never). Our measure of adult psychological characteristics assessed the magnitude of stressors for college students, in contrast to Block and Block’s (2006) broader measure of adult personality characteristics. Our measure of adult stressors was adapted from Feldt’s (2008) College Student Stress Scale. This scale includes 11 items, all of which represent potential stressors for college students (e.g., personal relationships, family matters, financial matters, academic matters). Items were answered on a 5-point continuum from 1 (very often) to 5 (never) indicating how frequently students felt distressed about each domain. Feldt reported the internal consistency (coefficient alpha) for the stressor scale to be 0.87 and the test–retest reliability over a 5-week period to range from 0.62 to 0.86 for individual items, with a mean of 0.73 for the total scale. The final measure in our study was an integrated measure of political liberalism and conservatism. We adapted one of several measures used by Block and Block (2006) to assess political conservatism and liberalism. They used the Kerlinger Liberalism Scale (1984) that contained 15 items
123
C. A. Blondin et al.
indicating issues on which liberals and conservatives tend to disagree: liberal issues—civil rights, racial equality, socialized medicine, the United Nations; conservative issues—patriotism, capitalism, law and order, moral standards. Our adaptation included eight of the Kerlinger items plus three others more reflective of contemporary issues (e.g., same-sex marriage, civil rights for suspected terrorist, higher taxes for the rich). Political ideology was assessed on one dimension because both conservatism and liberalism are determined not only by what individuals favor but also by what they oppose. Students responded to the political issues on a 1 (strongly support) to 5 (strongly oppose) scale. Our adapted and expanded version of the Kerlinger scale yielded a lower alpha (0.66) than that reported by Block and Block (0.87) for the original scale.
Results The primary questions addressed in the study were (a) whether retrospective assessment of parenting styles would complement the relationships highlighted in the Block and Block (2006) longitudinal study and (b) whether the relationships reported by Block and Block between childhood tendencies and political ideology would be found in a retrospective investigation with a Southeastern U.S. college sample. Whereas the Block and Block study was a prospective longitudinal study conducted over a period of 20 years, our study first assessed college students’ retrospective perceptions of treatment from their parents and their own levels of insecurity vs. confidence during childhood and then assessed their contemporaneous perspectives of their stressors and political ideology as adults. Parenting Styles and Psychological Measures in Childhood and Adulthood Although Feld (1995) suggested that parent–child relationships were linked to childhood personality characteristics, as well as adult personality and political orientations, his study did not explicitly assess authoritarian versus authoritative parenting styles. Table 1 indicates that authoritative parenting in our study was negatively correlated with childhood insecurity and positively correlated with childhood confidence (p \ .001). On the other hand, authoritarian parenting was positively related to childhood insecurity (p \ .001) but not significantly related to childhood confidence. The two parenting styles evidenced opposite relationships with adult stressors: authoritative parenting was negatively related to magnitude of adult stressors (p \ .001), whereas authoritarian parenting was
Relationship of Adult Representations
209
Table 1 Overview of correlations between parenting styles, childhood psychological tendencies, and adult psychological tendencies (N = 232) Parenting styles
Authoritative Authoritarian
Childhood tendencies
Adult variables Stress
Political
-0.21**
-0.18*
Authoritative
Authoritarian
Insecurity
–
-0.25**
-0.43**
–
Child insecurity Child confidence Adult stress
Confidence 0.39**
0.19**
-0.11
0.20**
-0.05
–
-0.50**
0.34**
0.09
–
-0.24**
0.00
–
0.04
* p \ .01; ** p \ .001
positively related to magnitude of adult stressors (p \ .001). A major departure from the expected relationships between parenting styles and adult characteristics occurred in their relationships with adult political orientations. The only significant relationship (p \ .01) was between authoritative parenting and the liberal/conservative continuum. However, even that relationship was weak and opposite the expected direction (r = -0.18), suggesting that the authoritative parenting style was inversely related with liberalism. Consequently, although parenting style was predictably related to some elements of childhood and adult psychological characteristics delineated by Feld (1995), the negative relationship between authoritative parenting and political ideology was opposite what might have been expected from Feld’s research. Although one might be inclined to consider the relationship between authoritative parenting and political ideology an outlier, the magnitude of that relationship is not that different from other relationships involving parenting (e.g., -0.21 between authoritative parenting and magnitude of adult stressors, 0.20 between authoritarian parenting and magnitude of adult stressors, 0.19 between authoritarian parenting and childhood insecurity). However, using Howell’s (1992) procedure for testing the difference in two correlations computed from the same sample, we found the magnitude of authoritarian parenting relationships with childhood insecurity (0.19) and confidence (-0.11) to be significantly lower (p \ .05) than the correlations between authoritative parenting styles and childhood insecurity (-0.43) and childhood confidence (0.39). Childhood Insecurity/Confidence and Adult Measures Although the principal finding in the Block and Block (2006) research was the relationship between childhood personality tendencies and adult political ideology, they also found that childhood personality characteristics predicted adult psychological tendencies. Table 1 indicates that childhood insecurity and confidence in our study both
significantly predicted the magnitude of self-reported adult stressors. Childhood insecurity was positively related to adult stressors (r = 0.34), whereas childhood confidence was negatively related to adult stressors (r = -0.24). The difference between these two correlations proved statistically significant (p \ .001), based on Howell’s (1992) test of the difference between two correlations computed from the same sample. Correlations between childhood characteristics and adult political ideology were negligible and non-significant for both insecurity and confidence. Thus, although our findings were consistent with Block and Block’s (2006) proposed linkage between childhood and adult personality characteristics, our results did not confirm the relationship between childhood characteristics and adult political ideology—the most crucial and controversial finding in the Block and Block study. Instead, recalled childhood characteristics were unrelated to adult political perspectives. Group Differences Between Adult Conservatives and Liberals Because our measure of political orientation did not correlate substantially with any other variable in the study, we explored the possibility that the political spectrum distribution was too homogeneous to correlate significantly with the other variables. Indeed, we found that students tended to be centrist in their responses to political issues, obtaining a mean score (33.15) precisely in the middle of the political spectrum (ambivalent). The highest mean rating on any one item was 3.50 on ‘‘welfare for the poor,’’ and the lowest mean rating on an item was 2.40 for ‘‘display of the Ten Commandments on public property,’’ representing approximately a one-point difference between the most extreme item means. This limited variability in item means suggests that our sample was relatively homogenous with respect to specific political issues. Because the centrist response tendency to the political items may have limited the possibility of obtaining significant correlations between political perspectives and responses to the other measures, we then identified the top
123
210
C. A. Blondin et al.
Table 2 Political-ideology item means for political quartiles Quartile Abortion rights Government health insurance Welfare for the poor Military interventionsa Civil rights for terrorists Affirmative action Government gun control Environmental regulation Taxes for rich
N
Item means
SD
Sig.
1–5
-14.12
0.000
1–5
-14.02
0.000
1–5
-8.69
0.000
1–5
-5.75
0.000
1–5
-10.01
0.000
Conservative
55
1.56
1.05
61
4.16
0.93
Conservative
55
1.78
0.96
Liberal
61
4.05
0.78
Conservative
55
2.82
0.88
Liberal
61
4.07
0.66
Conservative
55
2.47
0.90
Liberal
61
3.51
1.02
Conservative
55
1.82
0.80
Liberal
61
3.43
0.92
Conservative Liberal
55 61
2.53 3.38
1.10 0.86
1–5
-4.65
0.000
1–5
-10.40
0.000
1–5
-9.54
0.000
1–5
-8.42
0.000
1–5
-11.42
0.000
1–5
-10.29
0.000
11–55
-29.92
0.000
Conservative
55
1.96
0.96
Liberal
61
3.82
0.96
Conservative
55
3.33
0.90
Liberal
61
4.66
0.57
Conservative
55
2.31
1.02
61
3.92
1.04
Same-sex marriage
Conservative
55
2.09
1.36
Liberal
61
4.48
0.85
Ten commandmentsa
Conservative
55
1.60
0.87
Liberal
61
3.41
1.00
a
t ratios
Liberal
Liberal
Political preference
Possible range
Conservative
55
24.27
3.39
Liberal
61
42.87
3.30
Reversed scored: disagreement weighted higher than agreement
and bottom quartiles of the political spectrum (top = liberal and bottom = conservative). Table 2 indicates that these two groups differed significantly on their total political spectrum means, as well as on every item on this scale. The bottom quartile obtained a mean of 24.25 on the combined political items, and the top quartile obtained a mean of 42.87 on the combined items. Analysis of differences in item means indicated that abortion rights [t(115) = -0.14.12, p \ .001] and government health insurance [t(115) = -14.02, p \ .001] were the two items on which our conservative and liberal groups differed the most. The major purpose of the group analyses was to determine whether the conservative and liberal groups would differ significantly on the other variables included in the study. Table 3 shows that out of five comparisons across the groups, only one comparison produced a significant difference: conservatives had a higher recollection of authoritative parenting than did liberals [t(231) = -2.71, p \ 0.01]. This finding is consistent with the weak, significant, and negative correlation between authoritative parenting and adult conservatism. Whether this finding is an outlier or a bona fide relationship that can be replicated in future research remains open to question. Otherwise, our
123
Table 3 Differences in comparison-variable means for conservative and liberal quartiles Conservative (n = 55)
Liberal (n = 61)
Sig.
Authoritative
38.05
34.38
0.01
Authoritarian
34.42
32.95
ns
Child insecurity Child confidence
10.24 15.15
11.00 14.93
ns ns
Adult stress
32.73
33.00
ns
ns non-significant
group comparisons failed to show any differences between the political groups on the comparison variables. We concluded that, even at the extremes, political preferences were generally not related to the other variables in the study.
Discussion Our retrospective measures of childhood parenting and early childhood psychological tendencies proved to be significantly related to one another and predictive of
Relationship of Adult Representations
current adult stressors. These findings appear consistent with the longitudinal findings of Block and Block (2006) and Feld (1995), as well as findings in related retrospective studies (e.g., Heer 2008; Klein et al. 1996; Renk et al. 2005). Parenting styles appear linked to early childhood psychological tendencies in predictable directions: authoritative parenting facilitates childhood confidence but authoritarian contributes to childhood dependency (Bornstein 1992). Additionally, authoritative parenting predicts positive tendencies in adulthood: secure attachments in adult relationships (Heer 2008) and positive self-concept (Klein et al. 1996). In contrast, authoritarian parenting may lead to negative self-concept, depression, substance abuse, obesity, and eating disorders in adulthood (Bornstein 1992; Klein et al. 1996).
Predictive Potential of Parenting Styles and Early Childhood Characteristics Although our findings are consistent with several major findings regarding the psychological implications of parenting styles, our correlations between recalled authoritative parenting and early childhood measures were significantly stronger than those between recalled authoritarian parenting and childhood measures. In other words, relationships between authoritarianism and childhood characteristics appeared somewhat muted, compared with moderate relationships between authoritativeness and childhood characteristics. Perhaps certain aspects of authoritarian parenting can promote childhood security. For example, parental limits are clear and unequivocal in authoritarian parenting. Also, parents are often perceived as knowing best and their judgments regarded as absolutely dependable. Although authoritative parenting involves more bilateral communication with children than does authoritarian parenting, both parenting styles emphasize clear limits on a child’s behavior that must be respected. Certainly, the two approaches to parenting are not mutually exclusive. The small negative correlation between the two parenting styles leaves considerable potential for overlap between them. The effects of both recalled parenting and childhood characteristics appear somewhat enduring. By adulthood, recalled parenting and childhood characteristics were significantly correlated in opposite directions with adult stress. Recalling parents as being authoritative was negatively related to adult stress, whereas recalling parents as being authoritarian was positively related with adult stress. Although different in directionality, these correlations were similar in magnitude and relatively weak (Cohen 1988). Thus, the effects of recalled parenting endured into adulthood but not to a substantial degree. Similarly, both recalled childhood insecurity and confidence significantly
211
predicted adult stress. Both the directionality and magnitude of these correlations were different, with childhood insecurity more strongly linked in a positive direction with adult stress than childhood confidence was linked in a negative direction with adult stress. Prediction of Adult Political Perspectives To this point in our analysis, our findings would be substantially consistent with those reported by Block and Block (2006) and Feld (1995) regarding the inter-relationships between parenting styles, childhood well-being, and adult psychological characteristics. However, these previous researchers also found significant relationships between most of these variables and adult political ideology. Feld reported that parenting behavior experienced by adult liberals tended to be more supportive than parenting experienced by adult conservatives. Block and Block found that early childhood characteristics significantly predicted adult political ideology. Our study generally did not confirm the relationships of parenting and childhood characteristics with adult political ideology. Due to abundant differences in methodology and sample composition between our study and those of Block and Block (2006) and Feld (1995), the failure of retrospective measures of childhood parenting and psychological security to predict adult political ideology should not be considered a challenge to the earlier longitudinal findings. Without question, Block and Block’s methodology was more diverse and psychometrically stronger than our methodology. A precise replication of their methodology might have produced findings similar to theirs. However, two-decade follow-ups similar to those of the Block and Block and Feld studies will likely appear infrequently in the literature due to the practical challenges of the longitudinal approach (e.g., keeping track of participants living in diverse locations over a period of several years). What accounts for the differences between our findings and those of Block and Block (2006) is not entirely clear. One possibility is that regional differences contributed to differences in results. Our students generally came from a cultural background considered highly conservative in national voting patterns since 2000, whereas the Block and Block sample came from one of the most liberal regions in the country (Dave Leip’s Atlas of U.S. Presidential Elections - 2008 n.d.). A liberal political orientation would be largely incongruous with the cultural background of most of our students, whereas the liberal orientation in the Block and Block study would have been highly consistent with the cultural background of their participants. Being out of step with mainstream political values in one’s culture may contribute to tension in relationships with family members, friends, and acquaintances outside of
123
212
a university setting. Coming from a conservative background and attending a university where liberal ideas are explored is a psychologically different experience than coming from a liberal background and attending a university where liberal ideas predominate. This incongruity for our liberal students with respect to their cultural background and their college experience may have blurred some relationships between political ideology and other variables in the study. Despite the surface appeal of the cultural-incongruity argument, our assessment of adult stressors did not support that notion. Specifically, if holding political views counter to the mainstream in one’s culture contributes to tension in relationships, that difference should have been evident in the respective stress levels of our conservative and liberal students. However, the stress means of the two political groups did not differ. The latter finding might be attributable to how conservatives and liberals handle incongruity between their beliefs and cultural influences. Previous research suggests that liberals may be more open to such incongruity than are conservatives. This tolerance of incongruity might temper the stress liberal students would otherwise experience from this incongruity (Jost et al. 2003, 2007; Sibley and Duckitt 2008). Another possible explanation for the differences in findings between our study and those of Block and Block (2006) was that our measure of political ideology did not adequately differentiate students’ political values. Students’ mean responses to political items tended to be centrist, with relatively low standard deviations. However, when we compared the political means of the top and bottom quartiles on the political measure, we found that the groups differed significantly on the overall political measure and on every item subsumed in that measure. But when we then compared these groups on all the other variables, we found that they differed significantly on only one variable, authoritative parenting, with the conservative group regarding their parents as having been more authoritative than did the liberal group. Therefore, our measure appears to have clearly differentiated the most conservative and the most liberal students with respect to specific political issues, even though those two groups did not differ significantly on most other variables. Concluding Perspectives Researchers who investigate highly sensitive sociopolitical issues in one region of the country should be cautious about generalizing their findings to other regions. However, it is important to underscore that virtually all of our relationships between parenting styles, childhood insecurity/confidence, and adult stressors would complement the Block and Block (2006) findings. Nonetheless, political ideology
123
C. A. Blondin et al.
was minimally related to any of these variables in our study. Current psychological tendencies are more likely to influence representations of childhood relationships and childhood psychological tendencies than current political ideology is to influence those representations of past experiences. Perhaps this notion helps to explain why adult stress was significantly related to perceptions of childhood relationships and personality characteristics but that adult political ideology generally was not. In general, our retrospective methodology appears not to have been a critical factor in differentiating our findings from the longitudinal findings of the Block and Block (2006) and Feld (1995) studies. Nonetheless, when the findings of a 20-year longitudinal study are contrasted with those of a retrospective study in which participants reconstruct relationships and personality tendencies of early childhood, several factors (e.g., the time lapse, self-reported versus observable responses, and influence of present experiences on memory of past experiences) all make retrospective data more questionable than longitudinal data. However, the potential accuracy of retrospective data should not be discounted. We noted earlier that MacDonald et al. (2009) reported evidence that recall of past experiences can be accurate over a 25-year period. Even if one considers retrospective data simply as recollection or reconstruction of past experiences, rather than as an accurate account of the past experiences, those retrospective data are still important. That being the case, the most defensible conclusion from our study is that what students reported about past and present experiences supported most of the Block and Block and Feld findings but did not support the reported relationships between childhood characteristics and adult political ideology.
References Aguilar, M. L. (1998). Parental socialization practices and personality traits associated with well-being: A longitudinal investigation of parenting, personality, and perceptions. Dissertation Abstracts International: Section B: The Sciences and Engineering, 58, 6276. Alea, N., Bluck, S., & Semegon, A. B. (2004). Young and older adults’ expression of emotional experience: Do autobiographical narratives tell a different story? Journal of Adult Development, 11, 235–250. Baumrind, D. (1971). Current patterns of parenting authority. Development psychology Monographs, 4, 1–103. Berge, J. M., Wall, M., Bauer, K. W., & Neumark-Sztainer, D. (2010). Parenting characteristics in the home environment and adolescent overweight: A latent class analysis. Obesity, 18, 818–825. Block, J. (1993). Studying personality the long way. In D. C. Funder, R. D. Parke, C. Tomlinson-Keasy, & K. Widaman (Eds.), Studying lives through time (pp. 9–41). Washington, DC: American Psychological Association.
Relationship of Adult Representations Block, J., & Block, J. H. (2006). Nursery school personality and political orientation two decades later. Journal of Research in Personality, 40(5), 734–749. Block, J., Block, J. H., & Keyes, S. (1988). Longitudinally foretelling drug usage in adolescence: Early childhood personality and environmental precursors. Child Development, 59, 336–355. Block, J., Gjerde, P. F., & Block, J. H. (1991). Personality antecedents of depressive tendencies in 18-year-olds: A prospective study. Journal of Personality and Social Psychology, 60, 726–738. Bornstein, R. F. (1992). The ‘‘dependent personality’’ developmental, social, and clinical perspectives. Psychological Bulletin, 112, 3–23. Burl, J. R. (1991). Parental authority scale. Journal of Personality Assessment, 57, 110–119. Carney, D. R., Jost, J. T., Gosling, S. D., & Potter, J. (2008). The secret lives of liberals and conservatives: Personality profiles, interaction styles, and the things they leave behind. Political Psychology, 29, 807–840. Cohen, J. (1988). Statistical power analysis for the behavioral sciences (2nd ed.). Hillsdale, NJ: Erlbaum. Dave Leip’s Atlas of U.S. Presidential Elections - 2008 (n.d.). Election 2008. Retrieved October 3, 2011, from http://uselectionatlas. org/2008.php. Dennissen, J. J. A., Asendorph, J. B., & van Aken, M. A. G. (2008). Childhood personality predicts long-term trajectories of shyness and aggressiveness in the context of demographic transitions in emerging adulthood. Journal of Personality, 76, 67–99. Dolbin-MacNab, M. L., Rodgers, B. E., & Traylor, R. M. (2009). Bridging the generations: A retrospective examination of adults’ relationship with their kinship caregivers. Journal of Intergenerational Relationships, 7, 159–176. Faith, M. A., Storch, E. A., Roberts, J. W., & Ledley, D. R. (2008). Recalled childhood teasing among non-clinical, non-college adults. Journal of Psychopathology and Behavioral Assessment, 30(3), 171–179. Feld, P. J. (1995). A prospective longitudinal study of personality and political orientation. Unpublished doctoral dissertation at the University of California at Berkeley. Feldt, R. (2008). Development of a brief measure of college stress: The college student stress scale. Psychological Reports, 102, 855–860. Gumpel, T. P., & Ish-Shalom, K. V. (2003). How do young adults remember their childhood social status? A retrospective analysis of peer-rejection in childhood and adolescence, and protective factors predictive of its remission. Social Psychology of Education, 6(2), 129–157. Hampson, S. E. (2008). Mechanisms by which childhood personality traits influence adult well-being. Current Directions in Psychological Science, 17, 264–268. Haycraft, E., & Blissett, J. (2010). Eating disorder symptoms and parenting styles. Appetite, 54, 221–224. Heer, P. A. (2008). The relationship between college students’ retrospective accounts of parenting styles and self-reported adult
213 attachment styles. Dissertation Abstracts International: SectionB: The Sciences and Engineering, 69(3-B), 1955. Howell, D. C. (1992). Statistical methods for psychology (3rd ed.). Boston: Duxbury. Inbar, Y., Pizarro, D. A., & Bloom, P. (2009). Conservatives are more easily disgusted than liberals. Cognition and Emotion, 23, 714–725. Jost, J. T., Glaser, J., Kruglanski, A. W., & Sulloway, F. J. (2003). Political conservatism as motivated by social cognition. Psychological Bulletin, 129, 339–375. Jost, J. T., Napier, J. L., Thorisdottir, H., Gosling, S. D., Palfai, T. P., & Ostafin, B. (2007). Are needs to manage uncertainty and threat associated with political conservatism or ideological extremity? Personality and Social Psychology Bulletin, 33, 989–1007. Kerlinger, F. N. (1984). Liberalism and conservatism: The nature of structure and social attitudes. Hillsdale, NJ: Lawrence Erlbaum. Kitamura, T., Shikai, N., Uji, M., Hiramura, H., Tanaka, N., & Shono, M. (2009). Intergenerational transmission of parenting style and personality: Direct influence or mediation? Journal of Child and Family Studies, 18, 541–556. Klein, H. A., O’Bryant, K., & Hopkins, H. R. (1996). Recalled parental authority style and self-perceptions in college men and women. Journal of Genetic Psychology: Research and Theory on Human Development, 157(1), 5–17. Knabb, J. J., Brokaw, D. W., Reimer, K. S., & Welsh, R. K. (2009). Retrospective meaning-making in adulthood: A qualitative study of conservative Protestant adults who experienced parental divorce as adolescents. Journal of Psychology and Christianity, 28, 44–56. Lytton, H. (1990). Child effects—still unwelcome? Response to Dodge and Wahler. Developmental Psychology, 26, 705–709. MacDonald, D. J., Horton, S., Kraemer, K., Weir, P., Deakin, J. M., & Cote, J. (2009). Application and reliability of the retrospective interview procedure to trace physical activity patterns in master athletes and nonactive older adults. Educational Gerontology, 35(12), 1107–1122. Patterson, G. R. (1982). Coercive family process. Eugene, OR: Castilia. Renk, K., Roberts, R., Klein, J., Rojas-Vilches, A., & Sieger, K. (2005). Retrospective reports of college students’ childhood problems. Journal of Clinical Psychology, 61, 235–250. Robinson, C. C., Mandleco, B., Olsen, S. F., & Hart, C. H. (1995). Authoritative, authoritarian, and permissive parenting practices: Development of a new measure. Psychological Reports, 77, 819–830. Sibley, C. G., & Duckitt, J. (2008). Personality and prejudice: A metaanalysis and theoretical review. Personality and Social Psychology Review, 12, 248–279. Soenens, B., Elliot, A. J., Goosens, L., Vansteenkiste, M., Luyten, P., & Duriez, B. (2005). The intergenerational transmission of perfectionism: Parents’ psychological control as an intervening variable. Journal of Family Psychology, 19, 358–366.
123