USF Libraries
USF Digital Collections

Character strengths and virtues of young internationally adopted Chinese children


Material Information

Character strengths and virtues of young internationally adopted Chinese children a longitudinal study from preschool to school age
Physical Description:
Loker, Troy
University of South Florida
Place of Publication:
Tampa, Fla
Publication Date:


Subjects / Keywords:
Values in action classification
Positive psychology
Child development
International adoption
Preschool age
School age
Dissertations, Academic -- Psychological and Social Foundations -- Specialist -- USF   ( lcsh )
non-fiction   ( marcgt )


ABSTRACT: Shifting from traditional deficit-based psychological research, the current study aimed to broaden the understanding of post-adoption development through a strength-based approach and further explore the recently developed Values in Action (VIA) Classification of Character Strengths among a particularly resilient population of young children-internationally adopted Chinese children. Archival longitudinal data of parents' descriptions about their adopted Chinese children's positive characteristics were analyzed from two time points two years apart. Data on 179 children ages 4 - 5 years old (M = 59.67 months SD = 6.60 months) in Time 1 from 172 families were analyzed with content analysis coding procedures.Overall, the profile of character strengths among young Chinese adoptees was very comparable to that of a general sample of young children assessed in a previous research study: Both samples had 11 of the 24 character strengths from the VIA Classification represented among 10% or more of the children, while the remaining character strengths were rarely represented in the children's data. The five most prevalent character strengths for Chinese adoptees were Love, Kindness, Humor, Zest, and Social Intelligence. The biggest difference between adopted Chinese children from this study and non-adopted children was that Zest and Social Intelligence were represented at much higher rates. There were no significant changes over time in all but one of the prevalence rates for character strengths (i.e., Love decreased from Time 1 to Time 2) and for the more broadly categorized virtues (i.e., Courage increased from Time 1 to Time 2).The two most prevalent virtues, Humanity and Courage, were associated with lower levels of externalizing and internalizing problems, respectively, which may point to the positive traits particularly related to this population's marked resilience. Results serve to provide a broader understanding of post-adoption development and offer the first longitudinal data on character strengths among young children.
Thesis (Ed.S.)--University of South Florida, 2009.
Includes bibliographical references.
System Details:
Mode of access: World Wide Web.
System Details:
System requirements: World Wide Web browser and PDF reader.
Statement of Responsibility:
by Troy Loker.
General Note:
Title from PDF of title page.
General Note:
Document formatted into pages; contains 90 pages.

Record Information

Source Institution:
University of South Florida Library
Holding Location:
University of South Florida
Rights Management:
All applicable rights reserved by the source institution and holding location.
Resource Identifier:
aleph - 002069476
oclc - 608503290
usfldc doi - E14-SFE0003270
usfldc handle - e14.3270
System ID:

This item is only available as the following downloads:

Full Text
xml version 1.0 encoding UTF-8 standalone no
record xmlns http:www.loc.govMARC21slim xmlns:xsi http:www.w3.org2001XMLSchema-instance xsi:schemaLocation http:www.loc.govstandardsmarcxmlschemaMARC21slim.xsd
leader nam 2200397Ka 4500
controlfield tag 001 002069476
005 20101105165951.0
007 cr bnu|||uuuuu
008 100422s2009 flu s 000 0 eng d
datafield ind1 8 ind2 024
subfield code a E14-SFE0003270
LB1051 (Online)
1 100
Loker, Troy.
0 245
Character strengths and virtues of young internationally adopted Chinese children :
b a longitudinal study from preschool to school age
h [electronic resource] /
by Troy Loker.
[Tampa, Fla] :
University of South Florida,
Title from PDF of title page.
Document formatted into pages; contains 90 pages.
Thesis (Ed.S.)--University of South Florida, 2009.
Includes bibliographical references.
Text (Electronic thesis) in PDF format.
ABSTRACT: Shifting from traditional deficit-based psychological research, the current study aimed to broaden the understanding of post-adoption development through a strength-based approach and further explore the recently developed Values in Action (VIA) Classification of Character Strengths among a particularly resilient population of young children-internationally adopted Chinese children. Archival longitudinal data of parents' descriptions about their adopted Chinese children's positive characteristics were analyzed from two time points two years apart. Data on 179 children ages 4 5 years old (M = 59.67 months SD = 6.60 months) in Time 1 from 172 families were analyzed with content analysis coding procedures.Overall, the profile of character strengths among young Chinese adoptees was very comparable to that of a general sample of young children assessed in a previous research study: Both samples had 11 of the 24 character strengths from the VIA Classification represented among 10% or more of the children, while the remaining character strengths were rarely represented in the children's data. The five most prevalent character strengths for Chinese adoptees were Love, Kindness, Humor, Zest, and Social Intelligence. The biggest difference between adopted Chinese children from this study and non-adopted children was that Zest and Social Intelligence were represented at much higher rates. There were no significant changes over time in all but one of the prevalence rates for character strengths (i.e., Love decreased from Time 1 to Time 2) and for the more broadly categorized virtues (i.e., Courage increased from Time 1 to Time 2).The two most prevalent virtues, Humanity and Courage, were associated with lower levels of externalizing and internalizing problems, respectively, which may point to the positive traits particularly related to this population's marked resilience. Results serve to provide a broader understanding of post-adoption development and offer the first longitudinal data on character strengths among young children.
Mode of access: World Wide Web.
System requirements: World Wide Web browser and PDF reader.
Advisor: Linda Raffaele Mendez, Ph.D.
Values in action classification
Positive psychology
Child development
International adoption
Preschool age
School age
Dissertations, Academic
x Psychological and Social Foundations
t USF Electronic Theses and Dissertations.
4 856


Character Strengths and Virtues of Young Internationally Adopted Chinese Children: A Longitudinal Study f rom Preschool to School Age by Troy Loker A thesis submitted in partial fulfillment of the requirements for the degree of Education Specialist Department of Psychological and Social Foundations College of Education University of South Florida Major Professor: Linda Raffaele Mendez, Ph.D. Tony Tan, Ed.D. Robert Dedrick, Ph.D. Date of Approval: November 16 2009 Keywords: values in action classification, positive psychology, child development, international adoption, preschool age, school age Copyright 2009 Troy Loker


i T able of Contents List of Tables i ii Abstract iv Chapter One : Introduction 1 Character Strengths in Youth 2 A Resilient Population of Young Children: Internationally Adopted Children 5 Are Chinese adoptees different from adop tees from other birth origins? 6 The Current Study 9 Research questions 12 Chapter Two : Literature Review 14 Internationally Adopted Children 15 Internationally Adopted Children from China 1 8 Abandonment of Chinese children 19 Behavioral o utcomes 21 Shift in Resea rch: A New Focus on Strengths 24 Cha racter Strengths and Virtues 28 Criteria defining the strengths 30 VIA Classification applied to youth populations 36 Bridging Two Fields of Research 41 Chapter Three : Method 43 Research Design 4 3 Archival Data Source 44 Current Study 47 Sample 47 Measures and procedure s 48 Coding procedures 50 Statistical analyses 55 Chapter Four : Results 5 8 Frequency of Codes per Case 58 Frequency of Ch aracter Strengths and Vi rtues 58 Virtue o f Justice 61


ii Virtue of Temperance 61 Virtue of Transcendence 61 V irtue of Wisdom and Knowledge 62 Virtue of Courage 63 Virtue of Humanity 64 Change in Prevalence from Preschool to School Age 65 Variables Related to Virtues 67 Chapter Five : Discussion 72 Rarely Described Character Strengths in Young Children 73 Most Frequently Described Character Strengths in Young Children 74 Character and Age 76 R elations with Virtues 7 8 Limitations 80 Directions for Future Research and Practice 82 Concl us ion 84 References 85


iii List of Tables Table 1 VIA Classification of Character Strengths and Virtues 3 4 Table 2 Parent and Child Demographic C haracteristics ( N = 179) at Time 1 48 Table 3 Abbreviated Character Strength C odebook 53 Table 4 Fr equency of Character S trengths at Time 1 and Time 2 59 Table 5 Frequency of V irtues at Time 1 and Time 2 60 Table 6 Logistic Regression of V irtues at Time 1 6 8 Table 7 Logistic Regression of V irtues at Time 2 69


iv Character Stren gths and Virtues of Young Internationally Adopted Chinese Children: A Longitudinal Study f rom Preschool to School Age Troy Loker ABSTRACT Shifting from traditional deficit based psychological research, the current study aimed to broaden the understanding o f post adoption development through a strength based approach and further explore the recently developed Values in Action (VIA) Classification of Character Strengths among a particularly resilient population of young children internationally adopted Chines e children. Archival longitudinal data of were analyzed from two time points two years apart. Data on 179 children ages 4 5 years old ( M = 59.67 months SD = 6.60 mont hs) in Time 1 from 172 families were analyzed with content analysis coding procedures. Overall, the profile of character strengths among young Chinese adoptees was very comparable to that of a general sample of young children assess ed in a previous resear ch study: B oth samples had 11 of the 24 character strengths from the VIA Classification represented among 10% or more of the children, The five most prevalent characte r strengths for Chinese adoptees were Love, Kindness, Humor, Zest, and Social Intelligence. The biggest difference between adopted Chinese children from this study and non adopted children was that Zest and Social Intelligence


v were represented at much hig her rates. There were no significant changes over time in all but one of the prevalence rates for character str engths (i.e., Love decreased fro m Time 1 to Time 2) and for the more broadly categorized virtues (i.e., Courage increased from Time 1 to Time 2) The two most prevalent virtues, Humanity and Courage, were associated with lower levels of externalizing and internalizing problems, respectively, which may point to the positive traits particularly related to this population s marked resilience. Resul ts serve to provide a broader understanding of post adoption development and offer the first longitudinal data on character strengths among young children.


1 Chapter One Introduction For decades research and practice within psychology have involved the identification and treatment of distress and disorder. From these approach es well being is determined by the absence of these negative conditio ns. Some children are exposed to risk factors that increase the likelihood of distress and disorder, but c hildren who overcome e xposure to identifiable risk factors and avoid the development of various psychological disturbances are described to be resilie nt (Hauser, Viery a, Jacobson, & Wertreib, 1985). Promoting the absence of disturbance and the recovery from adversity are certainly important objectives; however, this deficit based appr oach may be missing l being. The positive psychology movement within recent years has offered a shift within psychological research to complement the traditional deficit based approach throug h the exploration and promotion of what is going right in the lives of individuals (Seligman & Csikszentmihalyi, 2000). Within positive psychology, the effort to explore positive individual traits has been one of the fields primary focuses, as Seligman and Csikszentmihaly i have whose 7). Studying positive individual traits among young children is valuable in its own right as they help broaden our understanding of human development and well b eing but these


2 traits may also have important pragmatic implications. Learning about positive individual traits may help us learn how to buffer against or prevent psychological disturbances (Seligman & Csikszentmihalyi, 2000) and parents, educators or me ntal health practitioners could learn how to use these trait s as assets to be capitalized on to help resolve psychological problems (Park & Peterson, 2008). Advancing our understanding of positive traits among young children in a way that adds to the kno wledge ascertained from deficit based approaches can be done by exploring the traits of children who have been particularly resilient. Exploring the development of positive traits among resilient children can not only broaden the understanding of their ov erall development beyond their absence of distress, but also can provide insight into the types of strengths that could be fostered among general populations of young children. The current study, therefore, aimed to explore positive traits within a sample of internationally adopted Chinese children, as research to be described on these children have de monstrated particularly favorable adjustment and resiliency from their early experiences of pre adoption adversity. Character Strengths in Youth The field of positive psychology is still in its early stages, particularly with young er children. However, there is a real demand to explore the development of positive traits among youth Perhaps foremost the goals for raising children are broader th an what get s captured within a deficit based goal of seeking an absence of disturbance. P arents, educators, and society ultimately want their children to develop into happy, healthy, and morally good individuals (Park & Peterson, 2008). The demand for


3 understanding can also be seen by the fact that positive character building is very frequently identified as a central goal in youth development programs (Roth & Brooks Gunn, 2003), and by the fact that there has been an expansio n of character education movements specifically designed to help teach various traits of good character (e.g., Character Education Partnership, Character Education Network, the Aspen Declaration of Character Education, and Character Counts; Park 2004). While there is a demand for empirical understanding of good character, only recently has a framework been proposed to scientifically conceptualize the broad family of morally valued individual traits that comprise good character. The research endeavor to create a science of positive individual traits evolved into the creation of the Values in Action Classification of Character Strengths and Virtues (Peterson & Seligman, 2004), likened to a strengths based version of the deficit based Diagnostic and Statis tical Manual of Mental Disorders (DSM) or the International Classification of Diseases (ICD) Created through a process of brainstorming among leading positive psychology scholars and broad based inventories of various cultural and historic depictions of good character and moral virtues, a classification system of 24 individual character strengths was created Each of the character strengths are also classified within six more broadly described virtues that represent core characteristics valued across mor al philosophers and religious thinkers. These virtues (and their composite character strengths) are Wisdom and Knowledge (including Creativity, Curiosity, Open Mindedness, Love of Learning, and Perspective), Courage (including Authenticity, Bravery, Perse verance, and Zest),


4 Humanity (including Kindness, Love, and Social Intelligence), Justice (including Fairness, Teamwork, and Leadership), Temperance (including Forgiveness, Modesty, Prudence, and Self Regulation), and Transcendence (including Appreciation of Beauty and Excellence, Gratitude, Hope, Humor, and Religiousness). Exploring how these character strengths manifest among young Chinese adoptees could advance research on the VIA Classification in its application to a unique and resilient population. Thus far, there have been limited peer reviewed studies on the assessment of character strengths in general populations of youth only One study has been conducted through the use of a 189 item self report questionnaire for children designed for childr en ages 10 to 17 years (Park & Peterson, 2006a) and one has been conducted through the content analysis of parental descriptions for children between the ages of 3 and 9 (Park & Peterson, 2006b). While different methods were used based on developmental ap propriateness (i.e., completing self reports demands levels of cognitive maturation greater than that of young children), both studies yielded results regarding the degree to which character strengths are prevalent among youth samples. In the older sampl e (10 and 13 year olds), among the most prevalent character strengths (i.e., those with higher mean ratings) were Love, Creativity, Teamwork, Gratitude, and Humor while among the least prevalent were Prudence, Self Regulation, Forgiveness, Authenticity, Ap preciation of Beauty, Modesty, and Leadership (Park & Peterson, 2006a). With younger children, the character strengths of Love, Creativity, and Humor were among the most prevalent characteristics described by parents, however, Curiosity and Kindness were also among the most described strengths (Park & Peterson,


5 2006b). Gratitude, which was one of highest rated strengths in the study with older children, was only described for 2% of the young children. Another strength that was much less prevalent in the younger sample was Hope, which was also found among only 2% of the sample with young children. Differences between these studies in terms of the character strength profile indicate developmental differences, but these conclusions are cross sectional in na ture. Park and Peterson (2006a; 2006b) called for further longitudinal studies in order to more closely examine the development and correlates of character strengths over time. A Resilient Population of Young Children: Internationally Adopted Children T he annual number of international adoptions in the U nited S tates has approximately tripled over a 15 year time period and s ince 2000, the leading source for adoptions into the United States has been China (USDS, 2007) As the population of internationall y adopted children has increased including those adopted from China, so has the need to explore their post adoption development as these children experience a unique set of early adversit ies Varying degrees of adversity are experienced by children prio r to their adoption placements including inadequate prenatal and perinatal medical care, maternal separation, psychological deprivation, insufficient health services, neglect, abuse, and malnutrition (Juffer & van IJzendoorn, 2005). Discovering how these children fare following placement into their adoptive families is significant for those with parents, adoption agencies, service providers with internationally adopted cl ients). Additionally, however, studying the post adoption development of this unique population


6 can provide important implications to developmental research in general. Investigations for issues of child resilience related to exposure to early adverse conditions to be A most predominate objective addressed within international adoption research of risk following their adoptive placements which parallels the deficit based inquiry found across traditional psychological research M uch of the literature has looked at behavioral and developmental outcomes for international adoptees originally born i n Romania, Russia, or other Eastern European countries (e.g., Albers et al., 1997; Chisholm 1998; since China only beca me a leading source for adoptions within the past few years A meta analys behavior problems and use of mental health services showed that international adoptees did exhibit more internalizing, externalizing, and total problem behaviors when compared to non adopted children; howev er, the overall effect sizes were small and international adoptees were still well within normal ranges of problem behaviors (Juffer & van IJzendoorn 2005). Therefore, despite the fact that results indicate d small differences from the norm, the current bo dy of evidence suggests that international adoptees as a whole are a relatively resilient and well adjusted population when examining vario us types of problem behaviors. Are Chinese adoptees d ifferent from adoptees from other birth origins ? E vidence sugge st s there may be some important differences between Chinese adoptees (who have


7 comprised roughly one fifth to one third of all international adoptions into the U.S. in recent years; USDS, 2009), and international adoptees from other birth origins Gunnar, Dulmen, and the International Adoption Project Team (2007) collected parent ratings data on the behavioral adjustment of 1,948 internationally adopted children representing birth origins from varying regions of the world including China (55% were from As ian countries; however, exact percent from China specifically was not reported) Their analysis indicated adoptees born in Russia and other eastern European countries exhibit ed higher levels of behavior problems than all other regions (i.e., Asia Latin Am erica/Caribbean, and O ther). Based on this finding, birth location appears to have an association with later outcomes. Moderating variables tied to birth location are likely factors that account for differences in post adoption outcomes For example, h igher rates of prenatal alcohol exposure among Russian children may contribute to the higher levels of problem behaviors, as the prevalence of Fetal Alcohol Syndrome among Russian orphans has been estimated to be approximately 14% (Warren et al., 2001). N otably, children adopted from China are characterized by a fairly unique set of pre adoption conditions. Specifically, large numbers of infants are abandoned in China, with estimates ranging between 100,000 to 160,000 per year, with the majority of these children being healthy, non disabled girls (Johnson, Huang, & Wang, 1998). Johnson and colleagues found the parents who abandon these young girls were typically married with occupations involving agricultural work in rural areas of the country. The prima ry reason for their abandonment was based on population control laws that have put restrictions on the


8 culture places preference on sons rather than daughters based on the practice of having males carry on the family name and becoming responsible for the care of elderly parents. The combination of the preference for sons and the strict birth planning laws has led to otherwise healthy daughters being abandoned by their p arents. Although some of these believe may decide to care for the children, many end up in child welfare institutions. Abandonment and institutionalized care provide a dverse environmental conditions prior these Chinese children do not suffer from adverse prenatal care conditions Although research on Chinese adoptees is not extens ive, results appear very positive thus far. For instance, while statistically significant differences were found to indicate slightly higher levels of problem behaviors among international adoptees in comparison to non adopted peers (Juffer & van IJzendoo rn 2005), results on Chinese adoptees suggest levels of internalizing, externalizing, and total behavior problems to be similar to or lower than those for non adopted peers (Tan & Marfo, 2006). Additionally, while older age at adoption has been correlated to poorer behavioral outcomes in international adoptees (e.g., Meese, 2005; Gunnar, Van Dulmen and the IAPT, 2007), age at adoption does not appear to be a related factor to behavioral adjustment samples of Chinese adoptees (e.g., Rojewski, Shapiro, & Sha piro, 2000, Tan & Marfo, 2006 ). Ado ptees from China, therefore, appear to be particularly resilient. Identifying a group to be resilient in overcoming the impact of adverse conditions provides an opportunity to


9 examine the positive characteristics of tha t group which may help complete the puzzle as to how children may thrive and flourish. The Current Study The objectives of the current study sought t o extend two fields of research. First, in recognition of the favorable behavioral and emotional adjus tment determined through deficit based assessments of internationally adopted Chinese children this study sought to advance the adoption development through a strengths based approach that explore s the development o f character among young children of this population. The second aim stems in part from the fact that positive psychology is geared towards learning from groups of individuals who thrive so that we can learn how to foster their strengths in other young peo ple (Seligman & Csikszentmihaly i 2000). Because international adoptees from China appear to being doing well despite their early adversity, examination into this unique character strengths and virtues as opposed to those of the general popu may offer distinctive, albeit preliminary implications regarding how children be come resilient and experience optimal human functioning. The methodology used to analyze character strengths among the young adopted children in the current study closely followed the methodology used in the only other study of the VIA Classification of Character Strengths among young children by Park and Peterson (2006b) While Park and Peterson specific ally ask ed parents to provide characteristics and individual qualities, the current study capitalized on and used an existing question within a widely used questionnaire, the Child


10 Behavior Checklist ( CBCL; Achenbach & Rescorla, 2000 describe the best thi The focus of both prompts is to elicit parents spontaneous descriptions of positive characteristics regarding a child. An archival longitudinal data set on a large sample of internationally adopted Chinese children already had bee n collected and included the CBCL as one major instrument Rather than collecting new data, this source was selected for secondary data analysis. The responses to the CBCL prompt were coded through content analysis procedures that were also closely align ed with the procedures described by Park and Peterson. While only a single open ended question was used to elicit descriptions of imilar methodology of using open ended parental descriptions and content analysis has been used successfully to verify Big Five personality traits among young children (e.g., Kohnstamm, Halverson, Mervield, & Havill, 1998). In a review of studies investigating the role of parents in the detection of child problems, Glascoe and Dworkin (199 concerns offered reliable insight into emotional, behavioral and developmental problems. Moreover, their findings suggested that gathering parent concerns was one of the best ons and understanding of their children. Because parents descriptions can be used to effectively identify personality traits and emotional, behavioral and developmental problems, it can be reasonably expected that parents are reliable informants of their In order to gain an understanding of the development of character strengths over time, parent descriptions from the archival data source were analyzed from when the


11 children were of preschool ages (i.e. 4 and 5 years old) to when they were of early elementary school ages (i.e. 6 and 7). These two developmental time periods were selected primarily based on previous data indicating higher levels of behavior problems among school age adopted Chinese children in comparison to prescho ol age adopted Chinese children (Tan & Marfo, 2006). Considering that there is evidence of behavioral change within this population that occurs between these developmental ages, it would be beneficial to explore whether this population also undergoes a ch anges in the development of character strengths and virtues across these time points. While differences in the prevalence of character strengths and virtues were examined longitudinally, logistic regression analysis at each time point was also used to de termine how child, parent, and family factors may relate to the likelihood that certain characteristics were described. Serving as a proxy for the extent of time spent in adverse pre adoption conditions and length of time away, age at adoption has been a factor correlated to differential behavioral outcomes in international adoptees ( e.g., Meese, 2005; Gunnar, Van Dulmen and the IAPT, 2007), though t his finding was not found within samples of Chinese adoptees (e.g., Rojewski, Shapiro, & Shapiro, 2000). Ag e at adoption is therefore an important variable to attend to within adoption research. Chronological age of the child was also selected for analysis at each time point in order to further detect possible developmental differences. Parents age, househol d income, single versus two parent household, and the presence of other siblings in the home were also selected as a parent and family variables, with the possibility that these factors could


12 strengths or the t ype of character traits that are valued and reinforced within their families. Additionally, child internalizing and externalizing behaviors, which have been the traditional focus in much of international adoption research were also examined in relation to character strengths. It is valuable in its own right to explore constructs of positive psychology like character strengths and virtues. However, linking the assessment of positive strengths to traditional deficit based constructs within a single study can help bridge the gap between deficit based and strengths based approach es to predictive relationships between certain problem behaviors and particular elements of good charac ter can be established, implications could be made with respect to the prevention of behavioral and emotional disturbances. Research q uestions In order to achieve the aims of the current study, the following research questions were used to guide data an alysis: 1. What character strengths and virtues do parents perceive in their internationally adopted Chinese children when their children are a) preschool age and b) early school age? 2. st rengths and virtues change over time from preschool age to early school age?


13 3. What parent, child, and/or family variables hold significant associations with at preschool age and early s chool age ?


14 Chapter Two Literature Review The following chapter begins with an overview of the developmental outcomes of international adoptees. After establishing this broader context, the relatively unique pre adoption experiences of international adoptees from China will be discussed as well as behavioral findings specific to this population. Despite early adversity, the relatively positive adjustment of international adoptees in general, and particularly those from China, that will be highlighted in this literature review provides an impetus for moving beyond the deficit based approach to exploring strengths of international adoptees. One study on Romanian adoptees that investigated behavioral and emotional strengths will be discussed in detail a strengths. This will be followed by a discussion of the emerging field of positive psychology and a significant project within the field involving the development of the Values in Action (VIA) Classification of Strengths and Virtues. With a new classification system for studying character strengths across the lifespan, investigations of the VIA classification with populations of youth will be thoroughly reviewed, and an argument will be p resented for why this classification system should be investigated further with children adopted from China


15 Internationally Adopted Children According to the U S sources of adoptees into the United States included Korea, Russia, and Romania. Other top countries of origin for internationally adopted children in the United States include Guatemala, Ukraine, Ethiopia, Colombia, Kazakhstan, Vietnam, India, Paraguay and Peru. Beyond obvious geogr aphic racial, ethnic, and cultural differences, these countries differ with respect to the other pre adoption environmental contexts (Pomerleau et al., 2005). For instance, many children who become wards of the state in South Korea are placed into foster care prior to being adopted, which constitutes a very different pre adoption experience than placement into an orphanage (Miller, 2000; Tarullo, Bruce & Gunnar, 2007). This family based system contrasts the institutional settings found in many other coun tries like Russia Romania and China where infant to caregiver ratios are much higher. While the particular pre adoption conditions across countries may differ, varying degrees of adverse factors characterize the experiences prior to ad option placements, including inadequate prenatal and perinatal medical care, maternal separation, psychological deprivation, insufficient health services, neglect, abuse, and malnutrition (Juffer & van IJzendoorn, 2005). With regard to post adopti on developmental outcomes, many studies have focused on behavioral and emotional problems of this population. Juffer and van IJzendoorn (2005) conducted a meta analysis of research published between 1950 and and use of mental health services. In the analysis of international adoptees compared to non adopted controls, a total of 47


16 studies on total behavior problems, 29 studies on externalizing problems (e.g., aggression, delinquency, hyperactivity), 30 studie s on internalizing problems (e.g., withdrawn, anxious, depressed), and 7 studies on mental health referrals (e.g., receiving clinical psychiatric treatment, placement into a residential setting or mental health facility) of international adoptees were revi ewed. Important to note, however, only one small scale study on children adopted from China was included (Rojewski, Shapiro, & Shapiro, 2000), making the results of this study not representative of this specific population. Across the studies reviewed, in ternational adoptees exhibited more internalizing, externalizing and total problem behaviors than those of non adopted controls, though degree of problem behaviors in most international adoptees was within the normal range. Overall, from a traditional def icit based model, international adoptees from countries other than China appear to have overcome their early adversity to become relatively well adjusted. Juffer and van IJzendoorn (2005) found a moderate effect size ( d = 0.37) for the overrepresentation of international adoptees with mental health referrals. These authors believed that the higher rates of referrals may have served to prevent behavior problems among international adoptees from increasing to above average rates. It may be that a more proa ctive approach is taken with these children, giving families and their children greater levels of skills early on to counteract problems related to early adversity. Little data to date, however, exist to examine the levels and types of strengths that these children have to support this conjecture.


17 C hildren reared in deprived institutional settings appear to present with delays in physical development social, cognitive, motor and language skills (Gunnar, Bruce, & Grotevant, 2000; Meese, 2005). However, f ollowing arrival into their new environments, investigations of internationally adopted children have provided evidence that they experience accelerated rates of growth so that these developmental dimensions reach approximately average rates. Meese (2005 ) commented that this catch up across skill areas typically occurs by 2 to 4 years after adoption for children who are adopted before the age of 2. Children whose age at adoption is greater than 2 years have been shown to be at greater risk in these areas However, these conclusions were presented with a on longitudinal data collected on internationally adopted children originating from Romania. While Romanian adopt ions once represented a large percentage of internationally adopted children in the early 1990s, these adoptions have steadily decreased and it is Chinese adoptions that now comprise a larger percentage of U.S. adoptions (USDS, 2007); thus there is a grea ter need to focus on Chinese adoptees as they have increased in numbers. In a recently published large scale study of behavioral adjustment among internationally adopted children, Gunnar, Van Dulmen and the Internation al Adoption Project Team (IAPT; 200 7) collected data on 1,948 internationally adopted children between the ages of 4 and 18 years using the Child Behavior Checklist to address questions regarding the impact of institutionalization and age at adoption. Fifty five percent of children in the sample were from Asian countries, but a further breakdown


18 describing how many of those children were from China versus other countries included in the sa mple (e.g., Japan, Philippines, India) was not provided. In comparing results between post institutiona lized (PI) and comparison (CO) children, Gunnar Van Dulmen and the IAPT indicated that early institutional care was correlated with increased risk of behavior problems O lder age at adoption was also correlated with an increased risk of behavior problems In particular, e levated scores were reported for attention problems among the internationally adopted children. Unfortunately, descriptive statistics regarding chronological age were not provided by Gunnar, Van Dulmen and the IAPT (2007). Time in family, which could serve as a proxy for chronological age, was shown to correlate with higher levels of behavior problems. Therefore, the results indicating older age at adoption to be associated with increased risk for behavior problems may have been co nfounded by the ages within each category. For example, if the children adopted after 24 months were systematically older than the children adopted before 24 months in their sample, the results indicating more behavior problems for the children adopted at older ages could be adoption. Controlling for these age variables is important for adoption researchers to consider in their analyses in order to reduce the possibility of conflated results and imprecise conclusions. Internationally Adopted Children from China Literature on international adoptions across various origins of birth provides an overall awareness of how international adoptees from particular countries, like Ch ina,


19 fare following adoption. However, the findings from this literature might be misleading in some cases as the results are not necessarily generalizable to those adopted from China. In the recent study by Gunnar, Dulmen, and the IAPT (2007), analysis by region of birth provided evidence that children adopted from Russia/Eastern Europe were more likely to exhibit behavior problems than children adopted from elsewhere. Many factors could result in such a difference. One potential difference that was ci ted may be that children from these countries experience higher levels or prenatal toxins and alcohol. Many possibilities are worthy of exploration of why such a difference occurs, but this difference itself has made an impact on the field. Though the ar gument may have been inferred from logic, this study provides empirical evidence within a single research design (i.e. comparisons were made across the same study, using a uniform set of procedures and data collection) to suggest that outcomes of internat ional adoptions from one birth origin do not necessarily align with those from other birth origins. With China leading all other nations in birth origin of international adoptions in the United States for several years (USDS, 2007) efforts to study adop tees from China are becoming more and more essential. The amount of literature on international adoptions from China has only recently been developing and makes up a more limited fraction of the entire international adoption literature. Nevertheless, the body of work is emerging, with similar types of investigations being conducted with this particular population as those conducted with populations from other countries of origin. Abandonment of Chinese children. An important difference between Chinese adoptees and those from other countries is that most infants who are adopted from China


20 were abandoned shortly after their birth. The number of Chinese infants who are abandoned in China every year is not clear, but estimates from government officials an d civil affair publications reported rates ranging between 100,000 to 160,000 (Johnson, Huang, & Wang, 1998). Abandonment has historical roots in Chinese culture, but the most recent rise of abandonment began in the early 1980 s with the advent of birth p place to reduce population growth. The one child policy states that families are limited to having only one child born into their family, but this policy has not been uniform across regions of China. In many rural areas of China where agricultural work is the primary occupation, the policy in place is a one son or two child policy. This policy allows couples to have two children but only if their first child was femal e. Enforcement of this policy is rather stringent with heavy fines and sterilization serving as consequences for births outside of this rule (Johnson, Huang & Wang, 1998). Johnson, Huang, and Wang (1998) located and collected information from 237 Chi nese families who had abandoned a child to gather evidence regarding who abandons children and why children are abandoned. The details of family recruitment for the study were limited to stating that the participants were found through informal networks an d word of mouth, so the conclusions drawn from this study should be considered cautiously as the sample may not be representative of all families who have abandoned a child Nevertheless, the situational factors surrounding early abandonments appear to be quite different from those that might occur in other countries. In their sample, 99% of birth


21 ranging from mid to late 20s to late 30s. Eighty eight percent of birth par ents were from rural areas, level of education was average for their age and region (i.e. primary school or junior middle school education), and economic conditions were approximately average for their region. ce for sons, nine out of ten children abandoned were girls (Johnson, Huang & Wang, 1998). In China, sons remain y following marriage and must This cultural preference for sons in combination with the one son or two child policy has led to the high ratio of girls being abandoned. However, because these factors ar e contributing reasons for child abandonment, the girls who are abandoned are typically healthy and without known disabilities (86% of the sample). Most of these children are abandoned within their first six months, with a third being abandoned within the first two months. Overall, these results show that the typical child abandoned is a healthy girl who is less than six months old. Behavioral outcomes. Of the literature on internationally adopted children from China, behavior problems have received the greatest focus. Most of these studies have used standardized behavior rating scales for parents to report their perceptions of problem behaviors. The earliest known study that focused specifically on adoptees from China used the Behavior Assessment Scal e for Children (BASC) Parent Rating Scale as a measure of behavioral adjustment with a sample of 45 international adoptees from China


22 (Rojewski, Shapiro, & Shapiro, 2000). The children in the sample ranged in age from 22 months to 9 years, 8 months, with a mean age of 46.9 months. Ninety eight percent of the internationally adopted Chinese children in the sample scored within the normal range for all 9 of the behavior constructs of the BASC behavior checklist (i.e., Hyperactivity, Aggression, Conduct prob lems, Anxiety, Depression, Somatization, Atypicality, Withdrawal, and Attention problems). Age at adoption was also found to be insignificant in relation to behavior scores, with children adopted before and after 18 months having similar scores in the nor mal ranges. Therefore, this initial study indicated positive results for this population. Within the group, however, behavioral profiles differed by chronological age. Adopted girls ages 3 years and older ( n = 22) were reported to have higher levels of hyperactivity and aggression than the adopted girls under age 3 years ( n = 17). Conversely, the younger girls were reported to have higher levels of withdrawal behavior than their older peers. Because of the larger range of ages in the older subsample, it is difficult to interpret when the behavioral adjustment shifted to become less withdrawn but more aggressive and hyperactive. Of note, the BASC has items that assess some areas of strength to produce subscales scores beyond deficit based clinical proble ms in the areas of Adaptability, Social Skills, Leadership, and Study Skills. Had outcomes could have been explored as data on strengths could have been analyzed. St udies conducted by Tan and colleagues ( e.g., 2006; 2007) utilized the Child Behavior Checklist (CBCL) with much larger sample sizes. Unlike the BASC, the CBCL reports a summary scale score to provide a measure of an overall behavioral profile for


23 children Data for these studies were drawn from data sets of 695 ( Tan & Marfo, 2006) and 1122 ( Tan, Dedrick, & Marfo 2007) Chinese adoptees. With the larger samples Tan has been able to not only conduct data analysis at the whole group but also divide his sam ple into groups based on variables such as chronological age, pre adoption adversity, and age at adoption without losing power since these groups still have large numbers. In the first data set, parent ratings of 695 adopted Chinese girls ranging in age f rom 1.5 to 11 showed that 92% of these children scored within the normal range (Tan & Marfo, 2006). This finding is very similar to the findings of Rojewski, Shapiro and Shapiro (2000). When this sample was broken down into preschool age girls ( M = 43.7 months, SD = 15.2) and school age girls ( M = 89.8 months, SD = 16.5), both groups scored significantly lower on internalizing, externalizing and total problems behavior scales when compared to the CBCL normative reference group (Tan & Marfo, 2006). Internalizing behaviors include feelings of anxiety and worry, depression, and withdrawal. Eternalizing behaviors include outwardly aggressive and delinquent behaviors. The total problems construct is a combination of internalizing externalizing and oth er problems (e.g., thought problems) and provides a general indication of behavior problems present. Interestingly, the school age group had higher total problem scores than the preschool age group. This same pattern of older children having greater lev els of behavior problems was seen again with an even larger sample ( N = 1095) of Chinese adopted children in both special needs and non special needs adoptees (Tan, Marfo, & Dedrick, 2007). As a


24 group special needs adoptees (i.e. adopted at older ages an d disabled) had statistically more total behavior problems than non special needs adoptees for the preschool age sample only, but the larger difference was shown between preschool age and school age samples For the preschool age sample ( n = 754 M = 43.7 months ) 91% of special needs child ren scored in the normal range, and 94% for non special needs children. For the school age sample ( n = 341 M = 89.8 months ), only 80% of special needs child ren scored in the normal range and 83% for non special needs c hildren. While these findings lack the strength of a longitudinal design, the repeated findings across cross sectional comparisons show ing older children scored lower than younger children implies that a ral adjustment with age. N = 193 55% from China 45% from Korea), Johnston and colleagues (2 007) found more cultural fewer externalizing behaviors. With a greater focus on socialization, it may be that parents promote particular strengths within their chi ldren that serve as a buffer towards externalizing problem behaviors. Much remains to be explored regarding the types of Shift in Research: A New Focus on Strengths The fo cus of research efforts on internationally adopted children from China and other countries has primarily used a deficit based approach. This approach has actually revealed that the adoptee s are relatively well adjusted after a period of time in their


25 adop tive families. In recognition of this general finding in the literature, one recent study Ryan, Johnson, and Groza (2008) argued that while children do experience unique difficulties as they relocate from deprived pre adoption conditions to improved settings with their adoptive families, the challenges they endure do not characterize the majority of adopted children and their families, and they do not serve as a fu ll representation of strengths and an exploration of variables associated with those strengths, implications could begin to be drawn regarding how adoptive families and adoption workers could better use these strengths to the benefit of the children. A total of 91 surveys collected in 1995 on children adopted from Romania from a pattern of s trengths, the associations between strengths and pre adoption variables, and predictive factors of this population s strengths (Pearlmutter, Ryan, Johnson, & Groza, 2008). The adoptive families had high household incomes ( M = $118,514) and an average of 1 .47 other children living in the home, and half of the children were female, between the ages of 5.75 and 16.83 years old ( M = 10.06 years), and adopted between the ages of less than a month old to 10.5 years old ( M = 21.24 months). The method used to ass Scale (BERS; Epstein & Sharma, 1998). The BERS was designed to measure positive behavioral and emotional skills and competencies through response items linked to five dim ensions of strengths Interpersonal Strength (i.e., ability to control emotions or


26 behaviors within social settings), Intrapersonal Strength (i.e., views of the child on personal competence and accomplishments), Family Involvement (i.e., engagement and rela tionships of the child with the family), School Functioning (i.e., competence with school and classroom tasks), and Affective Strength (i.e., ability to accept affection and express emotions towards others). As a group, the adoptive children scored stat istically similar to the normative sample of non emotionally/behaviorally disturbed children in terms of Interpersonal Strength and School Functioning while scoring statistically higher than the normative sample in terms of Family Involvement, Intrapersona l Strength, and Affective Strength (Pearlmutter, Ryan, Johnson, & Groza, 2008). In examining how institutionalization was related to strengths within their sample, children placed into family settings between birth to six months were reported to have simi lar levels of strength to those still in institutional settings during that time. Children still in institutional settings at ages six to 12 months, however, scored significantly lower in Family Involvement than children in family settings by that age, wh ile children in institutional settings at ages 12 to 24 months scored significantly lower in Interpersonal and Affective Strengths than children in family settings by that age. For children not placed in family settings until after 24 months of age, their scores were lower than those in family settings by that age on all five strength subscales. Adoption into family settings at later ages appears to negatively influence the level of strengths. In conducting regression analysis, the strength of the parent child relationship (as measured by a summative score of seven four point Likert scale items on how well


27 parents and children get along, communicate, feel close to one another; the parents respect and trust for the child; the relative impact the relationshi p has had on the family; and the amount of time they spend together) was significant ly correlated in a positive direction with all five subscales of the BERS (Pearlmutter, Ryan, Johnson, & Groza, 2008) This finding suggests that a more positively perceiv ed parent child relationship is a significant predictor of higher levels of child strengths. Regression also indicated that older children were more likely to score lower than younger children on each subscale. Interestingly, girls were less likely to sc ore lower only on the School Functioning scale, which may indicate the difference in Interpersonal Strength based on gender shown through the t test may have been confounded by other variables not taken into account (i.e., age, parent child relationship an d pre adoption history). Unlike other assessment measures that may include items or some scales on competencies or strength areas, the BERS is an empirically validated measure (e.g., Epstein, 1999; Epstein, Cullinan, Ryser, & Pearson, 2002; Trout, Ryan, La Vigne, & Epstein, 2003) with a sole focus on strengths. The BERS was developed to reflect behavioral and emotional skills identified within the behavioral and emotional research literature, with the practical purpose of developing and validating a meas urement tool However, strengths can be and have been conceptualized in different ways other than as behavioral and emotional skills. Within the emerging field of positive psychology, research efforts have focused on scientifically identifying, measuring, and exploring good character. As opposed to competencies and strengths of skills, strengths of character are


28 considered a broad family of individual traits that are moral ly valued and that manifest across thoughts, feelings and behaviors. The following section will explore in greater depth the emergence of positive psychology and the development of the classification system for character strengths. Character Strengths a nd Virtues Deficit based approaches are not only found in adoption research but traditional psychological research in general. However, Seligman and Csikszentmihalyi (2000) advanced an argument for expanding the field of psychology to focus on three prima ry topics -positive subjective experiences, positive individual traits, and positive institutions that serve to improve qualities of life and prevent pathologies. Rather than concentrating on a deficit based model of human functioning focused on how indiv iduals can heal themselves, positive psychology shifts away from traditional psychological research with an aim at understanding positive human functioning among those who are thriving and use that knowledge to help individuals flourish. Recognizing the advantages that psychologists have experienced with the existence of empirically based classification schemes for understanding deficit based mental health pathologies like the Diagnostic and Statistical Manual of Mental Disorders (DSM) and the Internatio nal Classification of Diseases (ICD) Peterson and Seligman (2004) sought to advance a similar classification for the positive characteristics of individuals by more scientifically exploring what goes into good character Such a classification was propose d to help in achieving understanding and promoting positive individual traits through the establishment of a


29 common vocabulary that could be explored empirically an d applied to clinical practice. In developing such a classif ication system that would identify measurable positive traits, Peterson and Seligman noted that they needed to address several important issues such as how character is defined; whether there are universal traits of character or whether it is primarily soc ially constructed with distinctive cultural values; and whether character is a singular or multidimensional trait. The process with which the classification system for character developed began with brainstorming among a group of lead ing scholars associat ed with aspects of positive psychology research (Peterson & Seligman, 2004). Additionally, inventories of virtues and strengths were drawn from an incredibly diverse range of sources across time in order to identify all potential strengths of character, w hich ended up including such things as writings from historical figures like Benjamin Franklin, statements from Boy Scouts of America, goals from character education programs ( e.g., M.W. Berkowitz, 2000), messages found in Hallmark greeting cards, popular song lyrics, and profiles of Pokmon characters. In examining this wide array of sources, the brainstorm ed positive characteristics began falling into groups of similar traits that were described to capture tial groupings fell within what became the intermediate level of the classification system, developing into what are now referred to as character strengths The classification system developed to include a hierarchical depiction of different levels of abst raction, however, to include levels of virtues, or broader categories representing core characteristics that have been valued by moral philosophers and religious thinkers; character strengths, or the intermediate level of


30 psychological processes or mechani ( p. 13) ; and situational themes or the more specific habits that manifest character strengths in given situations. C riteria defining the strengths In clarifying how character strengths would be determined in this process, a series of 10 criteria were developed to more clearly describe the qualities of a character strength (Peterson & Seligman, 2004). The first criteri on illments that constitute the good life, for 17). This criter ion emphasizes that character strengths add to emphasizes the factors that lead individuals to cope with adversity and only resolve distress or disorder. Another important distinction of this first criteri on is that a strength contributes rather than causes fulfillment. The example used to depict this line of reasoning was that do ing a favor for someone is an act of kindness that involves an inherent satisfaction of being helpful, rather than causing a person to feel satisfied with himself /herself at a later point in time. and do produce desirable outcomes, each strength is morally valued in its own right, even in the absence of obvious on establishes that while character strengths could have significant c orrelations or causal relationships to outcomes such as reduc ed likelihood of mental health challenges, the outcomes are not a defining feature of character strengths. It also emphasizes that a strength is a quality that is morally valued, one that all pe ople can aspire to develop. Building off this point, it


31 was contended that people tend to admire the display of character strengths and become elevated i n witnessing them, rather than feel jealous or experience negativity in their presence. With this in mind, the third criteri on 21). More linguistically based, the fourth criteri on th in a felicitous way counts against regarding it as a 22). The premise behind this criterion was that a character strength should have a clear antonym that has negative connotations, rather than antonym s that could be more positive. For example, flexibility was given as an example of potential strength, but its opposites could be inflexibility (negative) or steadfastness (positive). ange of an thoughts feelings, and/or actions in such a way that it can be assessed. It should be trait like in the sense of having a degree of generality across 23) In other words, strengths need to be exhibited in more than just one type of situation or point in time; however, the criterion is flexible enough to include strengths that could be described as increasing or decreasing in intensity based on relevant settings; e.g., bravery) The sixth criterion relates to keeping the character strengths sufficiently distinct from one another, as Peterson and Seligman (2006


32 24). This also relates to their efforts to keep the character strengths similar horizontally and leaving the categories of virtu es as a vertical level at which a given virtue could be p otential character strength, but because its definition consumed aspects of other character strengths of open mindedne ss and fairness, it did not meet this criteria and was not included in the classification. check, in which there should be real, apocryphal, or mythic models that exemplify the character strength in order for it to be included in the classification system. While many of the criteria did not specifically mention children in the development of the classification, this criterion did in describing that children grow up learning about character through paragon role models. Additionally, the authors pointed out that a character strength is one in which people can easily identify someone who truly exem plifies that strength from within their own social networks. That is, if a person was asked to identify a person who typified a given strength in the classification system, such as humor, or kindness, or hope, they would be able to identify someone withou t any significant trouble. This speaks to the fact that individuals are described to have signature strengths, ones that an individual feels ownership towards and conveys frequently. Somewhat related is criterion eight, in which Peterson and Seligman ( 2004) 25). Similar to how children can b e mentally advanced with


33 regard to specific domains like music or athletics, the same was argue d to be plausible, though admittedly based on anecdotal evidence due to a lack of empirical evidence, for character strengths. Similar to the issues discussed in relation to criterion seven regarding whether paragons of some strengths are more prevalent th an others, the same was unknown regarding which strengths are more apt be displayed precociously in children. For both issues, there is also little known about the developmental course of character strengths during childhood, and more is needed to begin a nswering these types of questions. On the opposite end of the spectrum from criterion seven and eight, criterion nine selectively the total absence of a given s another aspect that defines character st rength s (Peterson & Seligman, 2004, p. 26). This extends the notion that character is plural and consists of a family of traits by establishing that people can have some character strengths but be lacking one or more other strengths. The last of the cr iteria was borrowed from the theory of Eriks psychosocial stages. C associated rituals for cultivating strengths and virtues and then for sustaining their & Seligman, 2004, p. 27). According to this final criterion, social agents (e.g., p arents, school programs, Little League baseball team) help in the development of character strengths, but the social context are also important in maintaining character str engths. In returning to the issue of tonic versus phasic strengths, tonic strengths were described as being able to adequately carry on in a person as long as


34 they were not punished in by others as these have more stability across situations. For phasic strengths, like teamwork, appropriate ways for enacting this strength may need more explicit communication or instruction and reinforcement to develop. Taken together, these criteria helped create the classification to include 24 character strengths and s erves to depict what the key components are to good character in psychological terms. The resulting character strengths are presented in Table 1, along with their corresponding definitions Table 1 VIA Classification of Character Strengths and Virtues (ad apted from Park, Peterson, & Seligman, 2004 p.606; Park & Peterson, 2005 p. 21 22) 1. Wisdom and knowledge cognitive strengths that entail the acquisition and use of knowledge. Creativity: Thinking of novel and productive ways to do things; includes art istic achievement but is not limited to it. Curiosity: Taking an interest in all of ongoing experience; finding all subjects and topics fascinating; exploring and discovering. Open mindedness: Thinking things through and examining them from all sides; no t jumping to conclusions; being able to change one's mind in light of evidence; weighing all evidence fairly Love of learning: Mastering new skills, topics, and bodies of knowledge, whether on one's own or formally; obviously related to the strength of cu riosity but goes beyond it to describe the tendency to add systematically to what one knows. Perspective: Being able to provide wise counsel to others; having ways of looking at the world that make sense to the self and to other people. 2. Courage emotio nal strengths that involve the exercise of will to accomplish goals in the face of opposition, external or internal. Bravery: Not shrinking from threat, challenge, difficulty, or pain; speaking up for what is right even if there is opposition; acting on c onvictions even if unpopular; includes physical bravery but is not limited to it. P erseverance: Finishing what one starts; persisting in a course of action in spite of obstacles; "getting it out the door"; taking pleasure in completing tasks.


35 Authenticit y: Speaking the truth but, more broadly, presenting oneself in a genuine way; being without pretense; taking responsibility for one's feelings and actions. Zest: Approaching life with excitement and energy; not doing things halfway or halfheartedly; livin g life as an adventure; feeling alive and activated. 3. Humanity interpersonal strengths that involve "tending" and "befriending" others (Taylor et al., 2000). Kindness: Doing favors and good deeds for others; helping them; taking care of them. Love/int imacy: Valuing close relations with others, in those in which sharing and caring are reciprocated; being close to people. Social intelligence: Being aware of the motives and feelings of other people and the self; knowing what to do to fit into different s ocial situations; knowing what makes other people tick 4. Justice civic strengths that underlie healthy community life. Teamwork: working well as member of a group or team; being loyal to the group; doing one's share. Fairness: Treating all people the s ame according to notions of fairness and justice; not letting personal feelings bias decisions about others; giving everyone a fair chance. Leadership: Encouraging one's group to get things done and, at the same time, encouraging good relations within the group; organizing group activities and seeing that they happen. 5. Temperance strengths that protect against excess. Forgiveness/mercy: Forgiving those who have done wrong; giving people a second chance; not being vengeful. Modesty: Letting one's accom plishments speak for themselves; not seeking the spotlight; not regarding oneself as more special than one is. Prudence: Being careful about one's choices; not taking undue risks; not saying or doing things that might later be regretted. S elf reg u lation: Regulating what one feels and does; being disciplined; controlling one's appetites and emotions. 6. Transcendence strengths that forge connections to the larger universe and provide meaning. A ppreciation of beauty and excellence: Noticing and appreciati ng beauty, excellence, and/or skilled performance in all domains of life, from nature to art to mathematics to science to everyday experience. Gratitude: Being aware of and thankful for the good things that happen; taking time to express thanks. Hope: Ex pecting the best in the future and working to achieve it; believing that a good future is something that can be brought about.


36 Humor : Liking to laugh and tease; bringing smiles to other people; seeing the light side; making (not necessarily telling) jokes Religiousness : Having coherent beliefs about the higher purpose and meaning of the universe; knowing where one fits within the larger scheme; having beliefs about the meaning of life that shape conduct and provide comfort. VIA Classification applied t o youth p opulations To assess character strengths in youth populations (i.e., ages 10 to 17 years ), the Values in Action Inventory of Strengths for Youth (VIA Youth) was developed. The VIA questionnaire underwent revisions over a 3 year time period that was influenced by focus groups with 459 high school students (Steen, Kachorek, & Peterson, 2003), guidance from developmental psychologists and teachers, and psychometric adequacy data on versions of the measure (Park & Peterson, 2006 a ). The resulting mea sure included 189 items using 5 point scale responses, with seven to nine items corresponding with each of the 24 character strengths. Following the determination of substantial test retest reliability of the measure over a period of 6 months with a samp le of 250 youth (119 youth were 10 years of age, 131 youth were 13 years of age), Park and Peterson (2006 a ) explored the prevalence of each strength and several demographic correlates. Overall, the average scores for all character strengths were rated abo ve the median rating of 3 (i.e. ranging between 3.29 and 4.02. Those with higher mean scores were considered more prevalent than those with lower mean scores. The least prevalent character strengths (in ascending order) were Prudenc e, Self Regulation, Forgiveness, Authenticity, Appreciation of Beauty, Modesty, and Leadership, which had average scores ranging from 3.29 and 3.42. The most prevalent character strengths (in ascending order) were


37 Love, Creativity, Teamwork, Gratitude, Hu mor, which had average scores ranging from 3.77 and 4.02. The remaining character strengths had average scores ranging from 3.50 to 3.71. When compared to a sample of adults (Park, Peterson, & Seligman 2006) to determine some developmental profile dif ferences in the prevalence of character strengths, Hope, Teamwork, and Zest were relatively more common within the youth sample than the adult sample (Park & Peterson, 2006 a ). On the other hand, Appreciation of Beauty, Authenticity, Leadership and Open mi ndedness were relatively less common within the youth sample than the adult sample These results provide evidence for differential profil es of character based on age which the authors suggested could be arguably attributed to levels of maturation. When compared within the sample, however, the average character strength profile (i.e., the order of least prevalent to most prevalent character strength) was highly comparable between 10 year old and 13 year old youth ( = .82, p < .001) indicating little de velopmental change within a three year period. Nevertheless, these results are based on cross sectional data, not longitudinal data, which leaves the finding open to limitations such as cohort effects. Park and Peterson acknowledged that longitudinal res earch could bring needed clarity to the change and stability of character strengths and their manifestations over time. The average character strength profile was also found to be highly comparable on the other demographic variables assessed, though thes e only included gender and Whites versus non whites. When comparing specific mean scores, males scored lower on scales of beauty, fairness, kindness, and perspective, and Whites scored lower than non Whites


38 for religiousness. Further exploration on variou s demographic variables is need ed to determine the extent to understand which factors may influence the development of character strengths and virtues. Park and Peterson (2006 a ) also explored the relation between character strengths and various other psy chological constructs. In particular, a different sample of 134 middle school students who completed an earlier version of the VIA Youth also completed a Child Behavior Checklist (CBCL), from which the authors were able to compare character strengths to i nternalizing and externalizing problems. With regard to internalizing problems, moderate negative correlations were found for hope, zest, and leadership. With regard to externalizing problems, moderate negative correlations were found for perseverance, a uthenticity, prudence, and love. While research has explored these character strengths in relation to older children, only a single study to date has applied the VIA Classification to young children. Park and Peterson (2006 b ) explored character strength s manifested within children between the ages of three and nine years Recognizing that young children would not be able to provide self report ratings to measure character strengths in the manner that is used to assess older children and adults, Park and Peterson relied on free parental descriptions as their data source. Using such a data source has been carried out previously in personality research, where the analysis of parent descriptions confirmed Big Five personality traits among children ranging f rom two to thirteen years old (Kohnstamm et al. 1998). characteristics and individual qualities in an effort to allow the researchers know their


39 child well. These responses s erved as the data for the subsequent content analysis in which a coding framework based on the VIA Classification of Character Strengths was used. Park and Peterson (2006b) obtained descriptions for 680 children between the ages of 3 and 9 years from a convenience sample of parents recruited through notices distributed electronically and posted in toy stores, daycare centers, and pediatricia n offices. Gender was evenly distributed (51% boys, 49% girls) and the ethnic composition was mostly White (85%). About half (49%) of the sample was reported to be upper or upper middle class, with another 40% reporting to be middle class, and only 10% reporting lower or lower middle class. Mean age was not reported, but the frequency of each of the seven ages were relatively equivalent, ranging between 12 % and 16% of the sample. Result of the content analysis demonstrated wide variability in the prevalence of character strengths among this sample of young children (Park & Peterson, 2006b) Thirteen of the 24 ch aracter strengths were described for 8% or less of the sample Authenticity, Gratitude, Modesty, Forgiveness Open Mindedness, Hope, Appreciation of Beauty, Perspective, Religiousness, Fairness, Leadership, Bravery, and Prudence. The authors noted that amo ng the more rarely described character strengths was theoretically aligned with the developmental notion that the degree of cognitive maturation differed across the various character strengths. However, with the exception of gratitude which the authors ex plored based on its empirically known relationship with life satisfaction, the authors did not explore if there were higher rates of these less prevalent character


40 strengths in the older versus younger children in the sample. Such an analysis of age diffe rences could serve to elucidate when such cognitive maturation, or moral development, begins to emerge. The 11 character strengths that were prevalent in 10% or more of children were Zest, Teamwork, Social Intelligence, Self Regulation, Perseverance, L ove of learning, Curiosity, Humor, Creativity, Kindness, and Love (Park & Peterson, 2006b) The authors kind, creative, humorous, and curious, as those characteristics represented the five most prevalent character strengths in the sample. In addition to the coding for character strengths Park and Peterson (2006b) also coded each case on a happiness sca le, ranging from 1 to 7, based on the degree to which Of the 24 character strengths, Love, Hope, and Zest were the three strengths found to have positive correlatio ns, with Love and Zest exhibiting a more moderate relationship ( r = .31 for both strengths) and Hope exhibiting a smaller relationship ( r = .12). However, when separate analyses were run for children seven years and older in the sample, Gratitude also had a modest positive relationship with happiness ( r = .16). Two other positive characteristics were coded for within the parent description data, Intelligence and Attractiveness, which were found to be prevalent among 49% and 8% of the sample, respectivel y (Park & Peterson, 2006 b ). For the most part, there was no relationship between either of these two positive characteristics and the character


41 strengths with exception of a reportedly modest relationship between Love of Learning and Intelligence. While the authors noted such a relationship was understandably interpretable, similar significant relationships that could also have seemingly understandable relationships with intelligence, such as Creativity, Perspective or other strengths within the virtue of Wisdom and Knowledge, were not found. Beyond the correlation analyses with Happiness, Intelligence, and Attractiveness, no further analyses were explored in relation to the character strengths Instead, relationships with other child and family demogr aphic variables were only tested for Happiness, not character strengths Delving into family and child variables in relation to character strengths could bring further insight into possible why certain strengths are perceived to develop more frequently fo r some children but not others. Additionally, exploring the connection between character strengths and negative indicators of mental health, such as problem behaviors, rather than positive indicators of mental health, such as happiness, could expand our u nderstanding of how positive character strengths may Bridging Two Fields of Research The literature on the international adoptee population s demonstrates their overall resilience in terms of beha vioral adjustment within a deficit based approach of assessment with children from China faring particularly well Thus far, only one known study within international adoption research moved from the traditional approach of assessing problem behaviors to a strengths based approach assessing behavioral and emotional skills and competencies Within the positive psychology movement, however,


42 a call has been made to build a science of human strengths. As a result, the VIA Classification of Character Strength s was created to reflect universally accepted positive traits of good character that exist across the lifespan; however, limited research have assessed character strengths among youth and only among general populations. The current study seeks to advance development through a strengths based approach and advance the research on the development of character strengths in young children by assessing children known to be particularly resilient.


43 Chapter Th ree Method The current study examine d their internationally adopted Chinese children over time as a means to further explore sou ght to explore how the positive characteristics of children may differ over time from preschool age to early school age, an d determine how potential child, parent and family variables relate to the prevalence of character strengths and virtues The purpos e of this chapter is to describe the methods selected to accomplish the goals of this study. The chapter opens with a description of the research design. This is followed by a description of the data collection procedures for the archival data source that was used within the current study. Sample characteristics are then provided, followed by a description of the specific measures used in this study Finally there is a description of the steps taken to code the qualitative data and a discussion of the st atistical analysis used to further explore the coded content. Research Design The current study falls within a larger program of research on the long term development of children adopted from China. The main goal of the current study is to explore the development of young internationally adopted Chinese children from a strength based perspective using the VIA Classification of Strengths and Virtues. This


44 study aims to not only focus on a resilient population with a unique set of early childhood experie nce s but to extend the limited research on character strengths and virtues with young children across developmental time points. To achieve these goals, data from a longitudinal study were analyzed to ( a ) to understand the strengths that adoptive parents perceive in their children and ( b from preschool age to early school age Data were collected from the same adoptive parents when their adopted children were between 4 and 5 years of age (i e. pres chool ages) and then when the chi ldren were between 6 and 7 years of age (i.e. early school ages). Using a longitudinal panel design (i.e., one that follows the same cohort of participants over time) provided the benefit of being able to draw better conc lusions about change and consistency over time since data reflect change in p erceptions of the same children and are less affected by confounding variables associated with cross sectional designs. Archival Data Source The original procedures used to col lect the archival data source used in the cu rrent study began in early 2005 when participant recruitment was conducted through adoption related internet discussion groups and adoption agencies. For the internet discussion groups, a group moderator was con tacted and information regarding the research project was provided. The moderator was then asked to write a short introduction to the research project and post that introduction with a recruitment letter for the members of their respective groups. Direct ors of 10 adoption agencies in the U.S. (e.g., Chinese


45 also were contacted and asked to disseminate the same recruitment letter. The study was endorsed by at least 120 internet discussion groups and 6 adoption agencies. The focuses of the internet discussion groups were varied some were for adoptive families in general (e.g., Families with Children from Chin a, Raising China Children), others were for adoptive families with children from specific regions of China, and others were for adoptive families to discuss specific aspects of development (e.g., attachment, special needs, identity, general adjustment). M any families belonged to multiple discussion groups and reported receiving information regarding the study more than once. If parents were interested in participating in the study, they were instructed to contact the research program via email with inform ation regarding the number of biological children, number of children adopted from China, ages of each child, and their mailing address. Important to note is that the nature of the sampling procedure involved volunteer participants rather than random sam pling. This must be taken into account when interpreting the results, as these volunteer participants may be distinctly different from participants who did not volunteer to participate. Additionally, although two methods were employed to reach potential parents of adopted Chinese children, the use of technology may have biased the sampling by excluding those who do not use internet discussion groups. An envelope containing a family background survey, child background survey, the Child Behavior Checklist (CBCL), instruction sheet, consent form, and stamped


46 return envelope was mailed to each family that expressed interest in the study (families with more than one adopted and/or biological child in the home received a CBCL and child background survey for eac estimated length of time for completing the surveys (i.e., 20 to 30 minutes), the instruction sheet assured the parents of confidentiality for the information they provided. The parents were then i nstructed to complete the background surveys first, followed by the CBCL assessment. Of the 1,092 families who requested surveys (1001 from United States, and 91 from Canada, Australia, and the UK), 853 families returned the surveys. This resulted in a considerably high return rate for survey research of approximately 78.1 % Of the 853 families who returned surveys, data were provided for a total of 1,122 children adopted from China. Each of those children was assigned a family and child code number t hat served as their unique identifiers. Following data collection in 2005, periodic thank you letters, newsletters, and updates on international adoption research were sent out to the families through email. This served both as a way to give back to th e participants and as a means to keep participants' contact information up to date as the primary investigator of that study request ed that families inform him of any changes to facilitate communication. In the spring of 2007, the same families who comple ted surveys in the 2005 study were asked to participate in a follow up study to collect the first wave of longitudinal data. Contact was first attempted via email. In cases when there were no responses to the emails or when email addresses were no longer correct or in use, the investigator sent letters ou t to


47 families using the most up to date mailing addresses on file. A majority of families were reached, with a majority also expressing interest in continuing to participate. The materials parents recei ved in this second wave of the study included an instruction sheet, CBCL, Social Skills Rating System (SSRS) questionnaire, child background survey for those families with any new children, and stamped return envelope. The surveys were marked with the sam e code numbers that were used to identify the families and children in the 2005 data collection. This coding allowed for data to be matched across data collections. Current Study Sample Following approval from University of South Florida's Institutiona l Review Board, the current researcher obtained access to the parent surveys and data files from Dr. Tony Tan on the adopted Chinese children from the 2005 and 2007 data collections. Separate files that included positive characteristics, CBCL scores, and d emographic data of interest were merged into a single spreadsheet. Children whose parents only participated during the first wave of data collection were removed from the data set, along with children who were either younger than 4 or older than 5 years o ld. The resulting sample included a total of 172 parents who reported on 179 children (i.e., seven parents reported on two children) who were preschool ages at Time 1 (i.e., 48 to 71 months; M = 59.57 months, SD = 6.60 at Time 1) and who were school ages at Time 2 (i.e., 72 to 95 months; M = 83.57 months SD = 6.60). The ages of adoption for the children ranged from 7 to 55 months, with a median age of 13 months ( M = 15.66, SD = 9.16). The mean age of parent respondents at Time 1 was 45.18 ( SD = 5.12), an d


48 ranged from 34 to 58 and the same parent respondents completed the questionnaires at both time points. Additional demographic data are presented in Table 2. Table 2 Parent and Child Demographic Characteristics (N=179) at Time 1 Characteristic Frequenc y Percentage Gender Male 7 3.9 Female 172 96.1 Income ($) < 29,999 4 2.2 30,000 59,999 33 18.4 60,000 89,999 55 30.7 90,000 119,999 38 21.2 120,000 149,999 12 6.7 >150,000 37 20.7 Marital Status Married 10 8 60.3 Living with S pouse /P artner 9 5.0 Separated or Divorced 16 8.9 Widowed 1 0.6 Never Married 44 24.6 Highest Educational Attainment (Respondent) High School 2 1.1 Some College 16 8.9 College 57 31.8 Masters 7 3 40. 8 Doctorate 27 15.1 Post Doctoral 4 2.2 Special Needs Adoption No 15 8 8 8.3 Yes 21 11.7 Other Children in Home Only Child 53 29.6 Child has Siblings 126 70.4 Measures and p rocedures. P ortions of the archival data source were us ed in the current study In the 2005 data collection, demographic variables were gathered from the


49 child and family background survey s For the current study, items regarding the adoptive married, separated, divorced, widowed, living with same sex partner, living with opposite sex partner), whether other biological or adoptive children were living in the home household income (1 = under $19,999 to 15 = over $150,000), and e selected as variables in regression analyses Demographic information about the adopted child was also collected from the parents. Relevant information for the current study ata collection, and ag e at adoption date of birth and date of data collection were used to calculate chronological age in months. The Child Behavior Checklist (Achenbach & Rescorla, 2000 ; 2001 ), a widely administered in both 2005 and 2007 data collections. The Child Behavior Checklist (CBCL) is the parent form measure from the Achenbach System of Empirically Based Assessment (ASEBA) and is used to maladaptive functioning. Two versions were used to collect data from parents including one designed for preschool ages (CBCL/1.5 5) and one for school ages (CBCL/6 18). Studies have frequently use this measure to obtain quantitative results regarding inte rnalizing, externalizing and total problem behaviors as parents are asked to rate a series of behaviors (i.e., 99 behaviors on the CBCL/1.5 5 and 118 behaviors on CBCL/6 18). For more exploration of the results from the CBCL with this sample of adopted C hinese children, see Tan, Marfo and


50 Dedrick (2007) For the purposes of this study, the internalizing and externalizing scores serve d as predictor variables in the logistic regression analyses In addition to these questions, both versions of the CBCL include open ended items. The item was used to elicit free parental descriptions of the positive characteristics of their children from which the current study coded in terms of character strengths and virtues ask parents to, The CBCL provides a space of about 1.5 by 8.5 inches for parents to respond. This prompt has not specifically been used in the past to assess character strengths; however, a prompt expressing a similar sentiment, though more detaile d, was used in the only other study assessing charact er strengths in young children: qualities. What can you tell us so that we might know your child well? Even small details are of about spelling or grammar. An answer of at least several hundred of words would be most useful to us, but you can write as much as you wish. If you want, you can share a story about your chil d that captu (Park & Peterson, 2006b p. 327 329) Free parental descriptions have also been used to assess personality traits in samples of young children, serving to successfully verify Big Five personality traits (Kohns tamm, Halverson, Mervield, & Havill, 1998). Coding p rocedures. Coding responses through a series of steps. T he qualitative dat a were first read and re read multiple times to obtain a sense of the scope of positive characteristics that the adoptive parents reported. A content analysis procedure was then selected to systematically analyze the


51 data. Bazeley (2003) identified conte nt analysis as a method of analysis for narrative data that involves categorizing segments of the text so that segments within each category are similar to each other and different from segments in other categories. The specific coding framework that was applied to the content analysis procedure was informed by positive psychology literature on character strengths. Specifically, Peterson and Seligman (2004) put forth the Values in Action (VIA) Classification purport ing a set of 24 positive traits of good character that have been grouped into six more broadly defined core virtues courage, justice, humanity, temperance, transcendence, and wisdom. These character strengths have been empirically validated through self assessment measures for adults and youth (Peterson, Park & Seligman, 2005; Park & Peterson 2006a). For preschool age children, self assessment would not be strength by Park and Peterson (2006b) provides support for the application of the VIA classification with this population Specifically, Park and Peterson (2006b) analyzed written narratives from parents ( N = 680 ) about their children. The two authors independently coded each description for the presen ce or absence of each of the 24 character strengths. Because the character strengths were initially developed for adult populations, the VIA classification was elaborated on to include ways in which parents might describe those strengths for t heir young c hildren (i.e. for Zest, parents might kappa reliability between the two authors was .70 or higher for all cases. In t he present study a similar framework was used, which included making adapt ations to the original VIA classification by including


52 behavioral indicators of strengths that were developmental ly appropriate for this age group. A stringent coding procedure employing three independent raters (i.e., the author, a graduate student and a faculty member) was then used to code a part of the data. While conventionally two raters are required in coding qualitative data, three raters were used in the current analysis because 1) data were collected with one o pen ended questions, thus leaving more room for different interpretations in some cases; 2) three rater agreement, instead of inter rater agreement, can potentially further validate the themes emerged from the data. The graduate student and the faculty member were trained by the current researcher with the codebook, guided discussions, and an on site tryout. First each code was reviewed separately by describing the title of the code, the meaning of the code, and the defining features and examples of the code An abbreviated version of the codebook is presented in Table 3, in which examples of behaviors and traits are presented. These examples include ones described in Park and Peterson (2006b), as well as examples developed by the current research er based on preliminary readings of th also incorporated the descriptions of the character strengths that are included in Table 1 of Chapter Two which were drawn from definitions presented in Park, Pe terson, and Seligman (2004) and Park and Pete rson (2004). intelligence, physical or athletic talents, physical attractiveness, and level of happiness were not to be coded as character strengths, as these characteristics do not represent aspects of good character but other positive attributes or subjective well being.


53 Table 3 Abbreviated C haracter S trength C odebook Character Strengths (by Virtues) Example codebook descriptions Wisdom & Knowledge Creativity imaginative; talented at music, arts, dance, writing, etc. Curiosity inquisitive; always asks questions; is interested in everything Open Mindedness always considers all the angles; is a critical thinker Love of Learning loves to read all the time; love s school (or a particular subject) Perspective has a good awareness of the world, settles disputes among friends Courage Authenticity always tells the truth, takes responsibility for her or his actions Bravery not afraid to do thi ngs; not afraid to try new things Perseverance persistent; works hard at things; never gives up Zest full of energy or life; spirited; outgoing; enjoys physical activity Humanity Kindness helps out around the house; good natured; wants to please others Love loving; affectionate; has close friends Social Intelligence understands and is sensitive to others emotions; good social skills Justice Fairness insists on equal treatment Leadership followed by other children; is an alpha child Teamwork cooperates well with playmates Temperance Forgiveness never holds a grudge Modesty lets others shine Prudence cautious; careful about her decisions Self Regulation follow s rules well; is easy going and flexible with change Transcendence Appreciation of Beauty loves to look at paintings, listen to classical music Gratitude always says thank you Hope always looks on the bright side Humor p layful; funny; good sense of humor; makes me laugh Religiousness enjoys bible stories; reminds the family to pray


54 Communication included more than simply didactic training as the author encouraged the raters to ask questions when explanations se emed unclear These discussion s helped clarify feat ures for different codes. For example, it was not initially readings of the data) would be coded, but through group discussion, it was determined enthusiastic approach to daily life. After going through each code and having such discussions, approximately 15 sample cases of pare nt description data were analyzed by the three raters. After the author read aloud each data entry, the raters looked through the codebook independently to determine which code(s) applied to the data. After coming to independent decisions, the raters sha red their coding decisions with each other and discussed the reasons for their choices. Discussing the reasoning was particularly important in refining the codes and establishing reliability as this provided an opportunity to identify differences of inte rpretation and understanding This was especially important given that data were collected using one open ended question and some responses were somewhat ambiguous. Through this collaborative reviewing process, the three raters clarified the original codi ng framework to more suitably address descriptors that potentially could be given different codes or were ambiguous For example, one parent described his/her child as as sociated with pleasing others, the codes of K indness and Social I ntelligence were discussed as being possible codes. Through this discussion, it was agreed upon that


55 si tuations and that one may display charming behaviors for purposes other than trying to be kind Following this training session, the three raters independently coded 50 randomly selected cases The three raters codes were then compared with each other, as well as against the codebook, to determine agreement across raters. When initially combining the codings, a total of 181 codes (3.62 codes per child) were endorsed by a t least one rater within the data. Agreement on each code between any of the two of the three raters was high ( i.e., 75.1 %), although the three rater agree ments were understandably lower ( i.e., 58.0 %). Instances in which raters had questions or in whic h there were discrepancies across raters were then discussed in person similar to the initial try out discussions described above Through these discussions, further clarity was brought to the types of characteristics and behaviors that fall into each ca tegory of codes. The raters discussed the codes until mutual agreement across the three participants was achieved. As a result, the final number of codes endorsed was 144 (2.88 codes per child). Guided by the insights gained from the discussions, the aut hor coded the rest of the dataset. Statistical a nalyses. The frequency counts of each coded character strength provided data used to determine the percent each characteristic was described within the sample at Time 1 and Time 2. Nonparametric statistica l testing was used to determine if significant differences exist ed between prevalence rates at Time 1 and Time 2. Because data were collected on the same children at two different times, the data were dependent. Therefore, the McNemar test of marginal ho mogeneity was used as this statistical test is


56 used for nominal data with matched pairs of subjects. The McNemar test utilizes a 2 x 2 classification table to determine whether a significant difference exists between the presence of a particular character istic in children from preschool to school age. This test assess ed whether a significant change over time occ urred in the proportion of character strengths over time. Recognizing that the likelihood for a Type I error would have been relatively high if 2 4 separate tests were conducted for each of the character s trengths with in the VIA Classification, the character s trengths were re coded into smaller categories prior to running further analyses to determine significant predictive relationships with child, parent, and/or family variables. Specifically, the 24 character strengths were grouped in Wisdom and Knowledge, Courage, Humanity, Justice, Temperance, and Transcendence (see Table 1 or Ta ble 3 for inf ormation on which character s trengths comprise each of these virtues). Character s treng ths data were re coded for each v irtue based on the presence or absence of any one or more character strengths associated with a virtue. When none of the character str engths within a v irtue catego ry was coded for a child, the v irtue was coded 0 (not pre sent), but when one or more character s trengths within a v irtue category were coded for a child, the virtue category was coded a 1 (present). For example, if a child rece ived codes for Kindness a nd Love, that child received a code of 1 for Humanity. A child who received a code for Love only was given a code of 1 for Humanity. Similar to the character strengths, v irtues became dichotomous variables, not summative scores. Frequency counts, percentages of prevalence, McNemar tests, and percentages of


57 consistency from Time 1 to Time 2 were calculated for v irtues in the same manner as was done for character s trengths. With a decreased likelihood for Type I errors, virtues (as opposed to character strengths ) were selected as criterion variables for logistic regression analysis. Logistic regression rather than multiple regression was used due to the fact that dichotomous criterion variables (i.e. presence or absence of virtues) were analyzed rather than continuous variables. Predictor variables included in the analyses were selected to reflect child, parent, and family variables. Specifically, internalizing problems, CBCL externalizing proble ms, family income, single or two variables of interest. Correlation analyses were run to evaluate multicolinearity among the predictor variables Correlations among t he majority of these variables at Time 1 were very low, ranging from .152 to .187. While the correlations between CBCL internalizing problems and CBCL externalizing problems at both time points were not as weak as the associations between other variables ( r = .631 at Time 1, r = .625 at Time 2), the relationship between these problem behavior scores in this sample indicate them to be sufficiently independent for logistic regression analyses.


58 Chapter Four Results Frequency of Codes per Case At Time 1 there were an average of 2.92 ( SD = 1.62) character strengths ( Range = 0 8); in Time 2, the average was 2.94 ( SD = 1.25; Range = 1 6). (Strengths data were missing for two children in Time 1 and for eight children in Time 2 Cases with missing data were excluded in relevant subsequent analyses.) The number of character strengths per case between the two time points was moderate ly correlated ( r = .38; p < .05), suggesting that parents provided descriptions with a similar number of character strength s at both time points. Frequency of Character Strengths and Virtues Table 4 provides a summary of the frequency of character strengths at Time 1 and Time 2, with the character strengths organized in ascending order based on the prevalence of the charact er strengths at Time 1. The number of children with the same character strength endorsed at both time points is also presented to denote the consistency in parent descriptions. The percentage of consistency represents the percent of children with the str ength at both time points from those who had that strength at Time 1. It should be noted that cases in which there were data missing at Time 1 or Time 2 were removed in calculating these percentages.


59 Table 4 Frequency of C haracter S trengths at Time 1 an d Time 2 Character Strength ( Virtue ) Time 1 ( N =177) Time 2 ( N =171) In Both Times N (%) N (%) N (%) 1. Gratitude (TR) 0 (0) 0 (0) 0 (0) 2. Forgiveness (TE) 0 (0) 1 (0.6) 0 (0) 3. Modesty (TE) 0 (0) 1 (0.6) 0 (0) 4. Prudence (TE) 0 (0) 1 (0.6) 0 (0) 5. Apprecia tion of Beauty (TR) 1(0.6) 1 (0.6) 1 (100.0) 6. Perspective (WK) 2 (1.1) 7 (4.1) 1 (50.0) 7. Fairness (J) 3 (1.7) 0 (0) 0 (0) 8. Leadership (J) 3 (1.7) 2 (1.2) 0 (0) 9. Religiousness (TR) 3 (1.7) 2 (1.2) 0 (0) 10. Hope (TR) 3 (1.7) 5 (2.8) 2 (66.7) 11. Open Mindedness (TR) 3 (1.7) 5 (2.8) 0 (0) 12. Authenticity (C) 3 (1.7) 8 (4.7) 1 (33.3) 13. Teamwork (J) 7 (4.0) 6 (3.5) 1 (16.7) 14. Bravery (C) 20 (11.3) 20 (11.7) 5 (29.4) 15. Love of Learning (WK) 23 (13.0) 15 (8.8) 3 (14.3) 16. Self Regulation (TE) 23 (13.0) 24 (14.0) 5 (22.7) 17. Pe rseverance (C) 26 (14.7) 29 (17.0) 10 (41.7) 18. Curiosity (WK) 27 (15.3) 25 (14.6) 9 (33.3) 19. Creativity (WK) 36 (20.3) 39 (22.8) 14 (40.0) 20. Social Intelligence (H) 41 (23.2) 38 (22.2) 9 (23.1) 21. Zest (C) 58 (32.7) 56 (32.7) 23 (42.6) 22. Humor (TR) 62 (35.0) 65 (36.7) 40 (65.6) 23. Kindness (H) 69 (38.9) 83 (48.5) 48 (69.6) 24. Love* (H) 103 (58.2) 68 (39.8) 47 (47.5) NOTE: WK=Wisdom and Knowledge; C= Courage; H=Humanity; J=Justice; TE=Temperance; and TR= Transcendence *Significant difference between Preschool Age and 2 ; p < .05) As seen in Table 4, considerable variability existed concerning the pre valence at which each character strength was represented in the sample. Thirteen of the 24 character strengths at both time points were found in less than 5% of the sample. The remaining 11 character strengths ranged in prevalence from 11.3% to 58.2% at Time 1 and a slightly smaller range of 8.8% to 48.5% at Time 2.


60 As described earlier in the data analysis section, data were re coded so that if any one or more character strengths of a virtue category were coded present, then that virtue received a single frequency count (i.e., A child with three character strength codes of Bravery, Perseverance, and Love would be given virtue codes of Courage an d Humanity). Table 5 displays the prevalence of the six virtues from the VIA Classification, which are organized in ascending order based on prevalence at Time 1, just as the character strengths were presented. This table also includes the number of chil dren whose descriptions continued to include the same virtue at Time 1 and Time 2 along with a percentage of consistency over time, which was calculated by dividing the number of children who continued to have the same virtue at Time 2 by the total number of children who had that virtue at Time 1. Further description of the individual character strengths are provided within the descriptions of their associated virtue categories to which they belong within the VIA Classification. Table 5 Frequency of V irtue s at Time 1 and Time 2 Virtue Time 1 ( N =177) Time 2 ( N =171) In Both Times (%) N (%) N (%) N (%) 1. Justice 12 (6.8) 8 (4.7) 1 (10.0) 2. Temperance 23 (11.9) 26 (15.2) 5 (22.7) 3. Transcendence 65 (36.7) 72 (42.1) 43 (70.3) 4. Wisdom 69 (39.0) 72 (40.1) 38 (57 .6) 5. Courage* 79 (44.6) 89 (52.0) 51 (70.8) 6. Humanity 137 (77.4) 125 (73.1) 101 (75.9) 2 ; p < .05)


61 Virtue of Justice. As seen in Table 5 the least prevalent virtue at both time po ints was Justice. The three character strengths that comprise this virtue, Fairness, Leadership and Teamwork, ranged in prevalence from 0% to 4.0% in the sample. Fairness emerged ncluded the children negotiating with others to make sure no one was left out, having a strong sense of justice, and trusting. The few who received codes for Leadership were described e coded to exhibit Teamwork, the most prevalent character strengths of the virtue of Justice, were described as being cooperative, sharing, and playing well with others. Virtue of Temperance. The second least prevalent virtue was Temperance, which is co mprised of Forgiveness, Modesty, Prudence, and Self Regulation within the VIA Classification. However, Forgiveness, Modesty, and Prudence were only endorsed for one child each during Time 2 and no children during Time 1. Self Regulation, however, was end orsed for 13.0 % and 14.0 % of the sample at Time 1 and Time 2, respectively. Self Regulation in the sample was mostly characterized to include children being able to adjust, adapt, and maintain good behavior in new situations or conditions. One child who e a good eater and fairly well behaved. She sleeps well and maintains a regu lar routine, but is fairly flex ible too Virtue of Transcendence The virtue of Transcendence and the virtue of Wi sdom and Knowledge differed in rank ordering of prevalence between time points. At Time 1, Transcendence was less frequently described than Wisdom and Knowledge, but at Time 2 it was the opposite; however, the difference in prevalence that separated these two


62 virtues at both time points was approximately two percent, making these two virtues represented at about the same frequency. Five character strengths comprise Transcendence in the VIA Classification. No children were described as being appreciative or thankful, making Gratitude the only character strength to be not present at either time point in this sample of young children. A single child was described to have an Appreciation of Beauty at both time points, going from a simple description of loving music and ballet at Time 1 to a description of being ballet, Irish dance, music, playing violin, drawing and coloring. She loves to go to symphony concerts, dance performances and plays. Religiousness was also rarely described, and was manifested in parents descriptions as also was rare but manifested in parents descriptions as children having a positive outlook or attitude. Humor was present in a little more than a third of the children at both time points, making this the third most prevalent character strength overall at both time points. Cases that were coded for Humor mostly involved the parent describing their child as being funny or hav ing a good sense of humor. Virtue of Wisdom and Knowledge. Compared to Transcendence, the virtue of Wisdom and Knowledge was slightly more prevalent than Transcendence at Time 1 and slightly less prevalent at Time 2. This category was comprised of five character strengths Curiosity, Creativity, Open Mindedness, Love of Learning, and Perspective. Least prevalent at Time 1 and represented in less than 5% of the sample were Perspective, followed by Open Mindedness, while at Time 2 it was Open Mindedness,


63 followed by Perspective. Descriptions of Perspective for this sample ranged from children having common sense, to holding unique views on the world, to having as Open M inded was described as works through i ssues over time and changes her reactions le other descriptions included being good problem solvers, deep thinkers, and rational. Love of Learning was the third most prevalent character strength in Wisdom and Knowledge, with most descriptions explici tly stating a child to either have a strong interest, enjoyment, or love for learning, while the next most modal description was a love for books or reading. Curiosity, the second most prevalent in this virtue, was typically described as being curious, in quisitive, or full of questions, but also included descriptions of a general interest in the world or desire to try new things. The most prevalent in this virtue and the sixth most prevalent overall was Creativity. Many of these children were described s imply as creative or imaginative, but also there were many instances of children being described to enjoy engaging in the arts (e.g., singing, dancing, playing musical instruments), which were considered to be behaviors indicative of creativity and thus co ded under this strength. Virtue of Courage The second most prevalent virtue in the sample was Courage. descriptions as being truthful, unafraid to express oneself and o pinions, and taking responsibility. Bravery was seen in a little more than one out of every ten children, and included various descriptions of not shying away from difficult situations. Discussions during the coding process led to a broad definition of th is category, and included


64 descriptions of being proud of Chinese background and of being independent as examples of how bravery could manifest in this unique population. Descriptions that were coded as Perseverance were mostly related to the child being s elf motivated, focused, and determined, as well as trying hard and being persistent even when things become difficult. Embodying the strength of Perseverance o what she wants and does her best to get it. Zest was depicte d the most, occurring in about one third of all descriptions. Many Zest descriptions stated that the child was full of life, energy, or enthusiasm. Several which was deemed as a manifestation of a how a child enjoyed being active, such as enjoying sports, were also coded as characteristics of Zest. Virtue of Humanity. By far the most prevalent virtue in the sample was Humanity, w ith this virtue being represented in approximately three fourths of child descriptions at both time points. Social Intelligence was the least prevalent, though overall it was the fifth and sixth most prevalent character strength at Time 1 and Time 2, resp ectively. Descriptions coded as Social Intelligence generally described children as enjoying being social, making friends easily, or being empathetic. Kindness was the second most prevalent character strength at Time 1 and the most prevalent at Time 2. Fi nally, Love was overall the most prevalent character strength at Time 1 and second most prevalent at Time 2. The majority of descriptions that were coded as Love involved descriptions of their child being loving or expressing affection.


65 Change i n Prevalence from Preschool to School Age McNemar tests of homogeneity were used to determine whether there were statistically significant differences in overall preva lence rates of each individual character strength and each v irtue. While a trend was not ed that Kindness had a fairly large increase in prevalence from Time 1 to Time 2, the increase was not statistically significant at the .0 5 significance level 2 (1, N = 169) = 2.62, p = .106. The only character strength that did have a significant change over time was Love, which decreased significantly from being represented among 58.2% of children at Time 1 to 39.8% of children at Time 2 2 (1, N = 169) = 12. 33, p < 0.05 In terms of virtues, Courage was the only virtue that had a significant change over time, increasing from 44.6% of children at Time 1 to 52.0% of children at Time 2 (1, N = 169) = 4.82, p < 0.05 In Table s 4 and Table 5 consistency from Tim e 1 to Time 2 is also depicted. At both the character s trength and v irtue levels, considerable variability existed with regard to how consistent the children continued to be described with the same characteristic from Time 1 as in Time 2. While the perce nt at which there was consistency in the character strengths that were represented in less than 5% of the sample (e.g., Gratitude, Appreciation of Beauty) ranged from 0 to 100%, the low rates at which these strengths were represented in the sample makes th ese results less reliable. The percents at which there was consistency in the character strengths among the more prevalent character strengths ranged from 14.3 % to 66.7% While there was not a definitive criteri on to determine the strength at which thes e percentages of consistency


66 could be determined low or high, the rates did cluster together into relative categories of lower and higher co nsistency. Love of Learning had the lowest rate of consistency, with only 14.3% of children who were described to l ove learning at Time 1 continuing to be described to love learning at Time 2. Other character strengths that had relatively low consistency of less than a quarter of the children being described with the same characteristic were Self Regulation and Social Intelligence. Slightly greater rates of consistency were found for Bravery and Curiosity, in which approximately three out of ten children who were described with these character strengths at Time 1 continuing to be described with those characteristics a t Time 2. More moderate rates of consistency were found for Perseverance, Creativity, Zest, and Love, which ranged between 40.0% and 47.5%. Finally, the highest rates of consistency were found for Humor (65.6%) and Kindness (69.6%). Interestingly, the m ost prevalent four character strengths at both ages (i.e., Zest, Humor, Kindness, and Love) also had the highest consistency over time. The range of consistency was greater when data were re coded for virtues The least consistent virtue was Justice, wi th only one of ten children who were given a code of Justice at Time 1 continuing to be described as such at Time 2. Also with a rather low rate was Temperance, which was based on the low rate of its character strength of Self Regulation, since the other three character strengths of that virtue were not present in any of the preschool age children. A large jump was found for the next most consistently coded virtue of Wisdom and Knowledge, where 57.6% of children who were described with this virtue at Time 1 continued to be described this way at Time 2. The remaining


67 three virtues of Transcendence, Courage, and Humanity had similar high rates of consistency with each other of 70.3%, 70.8%, and 75.9%, respectively. Variables Related to Virtues For each virtue, logistic regression was carried out to determine whether the predictor variables age, age at adoption, internalizing problems, externalizing problems, family income, single or two parent household, and presence or absence of other siblings, significantly predicted the presence of each virtue at Time 1 and at Time 2. Results are displayed below in Table 6 and Table 7.


68 Table 6 Logistic Regression of V irtues at Time 1 Wisdom and Knowledge Courage Humanity Justice Temperance Transcendence Predictors SE SE SE SE SE SE Child's Age .023 .025 .040 .025 .033 .029 .034 .050 .043 .038 .033 .025 Age at Adoption .003 .018 .024 .018 .017 .023 .006 .033 .005 .024 .001 .018 Externalizing Problems .025 .029 013 .028 .025 .035 .056 .057 .046 .048 .008 .029 Internalizing Problems .099* .041 .027 .040 .091 .051 .129 .090 .018 .064 .007 .041 M .002 .033 .019 .032 .018 .038 .098 .065 .084 .051 .054 .034 Two Parents (1=yes, 0=no) .270 .40 0 .386 .391 1.122* .512 .577 .822 .479 .597 .276 .400 Household Income .024 .045 .046 .044 .007 .053 .047 .089 .201* .080 .026 .045 Siblings (1= yes, 0 = no) .387 .380 .224 .363 .421 .429 .661 .687 .571 .611 .334 .381 Note: A ll variables measure d at Time 1. p < .05


69 Table 7 Logistic Regression of Virtues at Time 2 Wisdom and Knowledge Courage Humanity Justice Temperance Transcendence Predictors SE SE SE SE SE SE Child's Age .054* .026 .000 .025 .020 .029 .034 .065 .032 .034 .027 .026 Age at Adoption .004 .018 .000 .018 .015 .020 .233 .164 .021 .029 .008 .018 Externalizing Problems .001 .020 .034 .020 .052* .023 .104* .053 .066* .029 .025 .020 Internalizing Problems .013 .021 .054* .021 .040 .024 .068 .049 .007 .028 .020 .021 .024 .035 .014 .034 .026 .038 .072 .088 .030 .047 .036 .035 Two Parents (1=yes, 0=no) .241 .409 .067 .407 .891 .487 .121 .974 .485 .582 .601 .410 Household Income .040 .046 .058 .046 .010 .051 .007 .117 .010 .062 .020 046 Siblings (1= yes, 0 = no) .730 .395 .520 .382 .108 .410 1.209 1.216 .011 .529 .657 .392 Note: All variables measured at Time 2. p < .05


70 or absence of siblings were not significantly related to any of the virtue s at either Time 1 or Time 2. Family factors of household income and two parent households were only significant during Time 1. Specifically, children within two parent households were more likely to be described with characteristics of Humanity than those children within single parent households, and children from families with higher household incomes were significantly re lated to the virtue of Wisdom and Knowledge at Time 2 only, where older children were more likely to be described with characteristics of that virtue than younger children. Problem behaviors were the only factors that were significant predictors for mor e than one virtue though there were more significant relationships during Time 2 than Time 1. At Time 1, a significant positive relationship was found between internalizing problems and Wisdom and Knowledge, indicating that children with greater internal izing problems were more likely to be described with characteristics of Wisdom and Knowledge. This relationship was not found at Time 2. Instead, a significant negative relationship was found between internalizing problems and Courage, indicating that ch ildren with fewer internalizing problems were more likely to be described with characteristics of Courage. Externalizing problems were not a significant predictor at Time 1 but w ere at Time 2 fo r three virtues Humanity, Temperance, and Justice all had significant negative


71 relationships with externalizing problems, indicating that children with greater levels of externalizing problems were less likely to have characteristics of those three virtue s.


72 Chapter Five Discussion This study utilized a st rength based approach to analyze parental descriptions of the positive characteristics regarding young internationally adopted children from China at ages 4 5 and ages 6 7. Drawing upon recent work from positive psychology, the Values in Action (VIA) Clas sification of Character Strengths and Virtues was used as the coding framework for content analysis The methodology chosen for assessing character strengths was closely aligned with the methodology used by lead researchers of the VIA project Park and Pet erson (2006b) in their study of non adopted sample of young children. As the first longitudinal study of VIA character strengths among young children, and the second study on internationally adopted children to primarily assess strengths as opposed to beh avioral or emotional problems, the current study served as a replication study with independent researchers (i.e. those not involved with the team who worked in the development of the VIA Classification). The current study revealed several similarities a nd differences in comparison to (Park & Peterson, 2006b ). Methodologically, the prompt used in the current study elicited responses containing an average of 2.92 and 2 .94 codes at Time 1 and Time 2 respectively Beyond being considera bly similar within this study, these averages were also quite similar to the average number of strengths (3.09) found in the non adopted


73 children of Park and Peterson's (2006b) study. Su ch consistency in the average number of codes elicited provides some evidence on the dependability of using spontaneous parental descriptions as a method to identify top strengths in young children. Rarely Described Character Strengths in Young Children Of the 24 character strengths, Gratitude, Forgiveness, Modesty, and Prudence were not observed at Time 1 when the children were 4 5 years old, while Gratitude and Fairness were not observed at Time 2 when the children were 6 7 years old. Gratitude was the refore the only character strength not represented at either time points; however, many character strengths were not observed much more frequently. Thirteen character strengths occurred in less than 5% of children at both time points In the study of n on adopted young boys and girls by Park and Peterson (2006b) the prevalence for the character strengths showed a similar pattern to the current study Specifically, 12 of the 13 rarely described character strengths overlapped in the two studies, al though the overall range of frequency for less prevalent character strengths were wider from 1% to 9%, instead of 0% to 5%). Therefore, the strengths that appeare d to be more rarely described among young Chinese adoptees appear to be the same as those among non adopted children. The replication of these findings argues for a systematic explanation as to why these character strengths appear at such low rates. Pa rk and Peterson (2006b) proposed that some character strengths, like Gratitude, were described as more sophisticated or requiring greater psychosocial development and maturation (Peterson & Seligman, 2004).


74 The fact that the same rarely described characte r strengths were represented similarly in with differences in adoption status between samples seemingly not account ing for much variation in terms of when character strengths begin to manifest The wider range of frequency among rare character strengths found in Park and (2006b) sample may be a reflection of the fact that older children were included in their sample Gratitude became significantly related to ratings of happiness o nly within a subsample of children seven and older. While they did not report the change in frequency across ages, it could be that gratitude was more prevalent among the older children in the sample to allow for a significant relationship to emerge. The current these more rarely described strengths were actually beginning to be represented at higher levels among the oldest children in the sample. If so, this could exp lain the slightly wider range in prevalence, as they may have been slightly more frequently described among older children. Further longitudinal research is needed to continue exploring when certain character strengths begin emerging at greater levels Ne vertheless, the fact that a narrower age range of children was assessed in the current study should be taken into Most Frequency Described Character Strengths in Young Children Simi larly the more prevalent strengths among the Chinese adoptees were also Eleven character strengths were described in 10% or more of parental responses in the


75 non adop ted sample and the adopted Chinese sample at Time 1(Love of Learning fell to 8.8% at Time 2, but the other 10 strengths remained above 10%), and 10 of these 11 more prevalent character strengths were the same across studies. Another close similarity, seve adopted sample. Within these more prevalent character strengths, however, was greater variability i n rates, which makes the similarities and differences between samples in the rank ordering and actual rates of character strengths more interpretable than within the rare character strengths that varied considerably less in terms of frequencies within that group. The top two strengths at both time points, Love and Kindness, were the top two strengths among non adopted young children. Humor was the third and fourth most prevalent character strength among adopted and non adopted children, respectively. Whi le the character strengths of the Wisdom and Knowledge virtue (i.e., Love of Learning, Curiosity, and Creativity) were prevalent at lower rates among Chinese adoptees when compared to the non adopted children, the order of prevalence for these character s trengths within both samples was the same (i.e., Love of Learning was less than Curiosity, which was less than Creativity). The most dramatic difference appeared with respect to Zest and Social Intelligence, which were much more prevalent among the adopt ed children (i.e., 32.7% and 23.2% at Time 1, respectively) compared to nonadopted children (i.e., 10% and 12%, respectively) Zest and Social Intelligence were among the


76 top five most prevalent character strengths at both time points among the adopted ch ildren, while only the ninth and eleventh most prevalent among non adopted children. Character and Age With the benefit of the longitudinal design, insight can be drawn regarding the temporal sequencing of character development. Within the two year time period in which the children developed from preschool ages to early school ages, there were very few differences in the rates at which each character strength and virtue were present. The exceptions were the significant decrease in Love over time at the character strength level and the significant increase in Courage over time at the virtue level. In terms of Love, it is important to note that it while it decreased in frequency, it was still the second most prevalent character strength when the children were of school ages. One interpretation to why this character strength witnessed a greater decrease could relate to a greater sensitivity to establishing a loving relationship with an adopted child early in their post adoption lives. A concern for the dev elopment of an attachment disorder, of which an inability to give or receive love and affection is an associated outcome (e.g., Levy & Orlans, 2000) in an adoptive child could lead adoptive parents to being particularly sion of love early on but become less focused on this particular characteristic as children get older. The decrease in Love also makes sense from a developmental perspective, as the expression of love through affection for a caregiver is something that can be seen from infancy. With older age, and the cognitive maturation, more strengths can begin to develop, which could lead parents to describe these characteristics more and describe Love less.


77 Courage, a virtue comprised of Authenticity, Bravery, Persev erance, and Zest, became more prevalent among parents descriptions for children of school ages. Courage ongitudinal data provide evidence that of the six virtues, Courage may be expected at this age to develop The less prevalent virtues (i.e., Justice, Temperance, Transcendence) appear to increase at older ages not assessed in the current study. The longit udinal design of the current study allowed for the opportunity to school entry. Children from this population appear to hold some character strengths more consisten tly over time, while others are not as stable. Humor and Kindness were held the most consistently (of the 11 more prevalent strengths for which the frequency was high enough to be meaningfully interpreted), indicating a higher likelihood that humorous or kind children in their preschool years will continue to hold those traits in their early elementary school years. Preschool age children who were described in ways to indicate Self Regulation, Social Intelligence, and Love of Learning, however, were much less likely to be described similarly at school age. These three less consistently held character strengths are more context specific than Humor and Kindness, and it may be that while a child may have appeared to display characteristics indicating Self Re gulation earlier in life, the demands for associated behaviors increase as children must become more regulated in their academic and social behaviors within school classroom contexts. Similar statements could be made for Social Intelligence and Love of Le arning.


78 Another key age variable to consider in adoption research is the age at which the child is adopted. Among Romanian adoptees, children adopted after 24 months were at greater risk for higher levels of problem behaviors (Meese, 2005) and lower leve ls of behavioral and emotional strengths (Pearlmutter et al., 2008). Nevertheless, knowing a preschool or school age adoptees. This finding parallels deficit based find ings for Chinese adoptees that demonstrated no significant relations between age at adoption and behavior problems. This finding may have been influenced by the relatively young ages at adoption. Only 13% of the sample was adopted at ages above 24 months While there were children adopted at ages up to 55 months, the limited variation in ages at adoption may have reduced the likelihood of picking up any significant associations with the virtues. R elations with Virtues A few relations emerged from the l ogistic regression at each time point. Humanity, the most prevalent virtue in the sample, was related to lower rates of externalizing behaviors at Time 2. A similar finding in previous research with older children found Love, one of the character strengt h of Humanity, to be correlated with lower levels of externalizing behaviors (Park & Peterson, 2006a). Because Humanity was represented in nearly three quarters of the sample, this particular link to lower levels of externalizing problems is a salient fin ding to help account for the favorable levels of behavioral adjustment reported for this population in previous resear ch (e.g., Tan & Marfo, 2006).


79 Similarly, the second most prevalent virtue of Courage was related to lower levels of internalizing prob lems at Time 2. Zest, a character strength of Courage, has been found to be related to lower levels of internalizing problems among older children (Park and Peterson, 2006a). The strengths of the virtue Courage, which were represented highly in this sampl e due in large part to the relatively higher frequency of Zest, may be another Surprisingly, the character strengths associated with Wisdom and Knowledge were positively related to higher rates of internalizing problems. A positive relation between positive characteristics and negative behavior problems was not anticipated. However, as noted earlier, the character strengths of this virtue are represented at lower levels than are fo und within non adopted children. While a positive relation may exist with internalizing problems, not as many children in this sample were perceived to hold this virtue; thus, the relation has less of an impact on the overall levels of internalizing probl ems, which could help keep internalizing problems from being significantly elevated in comparison to non adopted peers. Justice and Temperance also had significant relations to lower levels of externalizing problems, but the rates at which these two virt ues were present in the sample were quite low. So while there may be a relationship, relatively few children in the sample would have been impacted. Nevertheless, these may be two rare virtues that if promoted and developed among young children could hel p prevent externalizing problems from developing.


80 For the most part, parent and family variables were not significantly related to virtues, and not at all related during school ages. When children were preschool ages, however, children in two parent fam ilies were more likely to be described with the character strengths of Humanity, and children in families of higher socioeconomic status were less likely to be described with the character strengths of Temperance. Taken together, parent and family variabl es were more related to virtues at preschool age while child variables were more related to virtues at school age, implying a potentially weaker influence of environmental contexts on character development as children get older. Importantly, however, caus al relationship cannot be assumed based on the non experimental design and regression results. Limitations Several limitations should be kept in mind when evaluating the conclusiveness of hrough non random, convenience sampling procedures. The parents self selected their participation in the study, and parents who chose not to participate may have systematically responded differently in describing their perceptions of strengths, held diffe rent demographic variables (e.g., children adopted at later ages), or reported different levels of problem behaviors). Given this design limitation, it should also be noted that of the parents who received study survey materials in the original data colle ction, there was a 78% response rate at Time 1. Of those who participated at Time 1, 87% of those parents continued to participate at Time 2. These are considerably high response rates for mail based survey research, and may help mitigate the potential bi as ing effects of self selection procedures.


81 Another limitation is the fact that the method of data collection used to gather parent descriptions of character strengths was a single item prompt with limited space for written response without follow up qu estioning. Had the area for written response been An answer of at least several hundred of words would be most useful to us, but you can write as much as you wish. If you want, you can share a story about your child have led parents to describe a greater range of character strengths per response. However, in comparison to the an average of approximately three codes per case, roughly the same average was found at the very short descri ptions were useful for our purposes because they invariably listed R ange = 3 1351 words), the length of responses was reported to be unrelated to any of their results. Given these findings, the relatively short responses in the current study may not have provided significantly different resulting codes. Nevertheless, while the current study may have yielded similar average number of codes, it was also similar in that only a sing le method was used to assess character strengths in a sample of young children. Observational, experimental, and ratings scale assessments could all be helpful tools in learning more about the manifestation of character in these young children as well as serve as a validation check for the method used in the current study. Additionally, while three raters were used in the refinement of


82 codes and building consensus, additional inter rater reliability checks for the final independent coding could further su pport the validity of the codes. Directions for Future Research and Practice While this study provides initial data on the development of character strengths overtime in young children, much more research is needed to understand when particular strength s tend to become more or less prevalent across ages of youth. The larger program of research from which the data for this study were taken is designed to follow the same children throughout their development every two years. The same procedures to analyz e parents descriptions should be carried out once again for the third wave of data collection when the children are 8 and 9 years old. By the fourth wave of data collection, these same children would be old enough to complete the VIA Youth self report me development would help pinpoint times at which certain aspects of character become more or less pronounced. Continuing to follow these children could help answer several importa nt research questions Do parents perceptions of character strengths during early childhood relate to self reported strengths during adolescence? Are some character strengths more consistently held than others (e.g., if Forgiveness, Kindness, and Creativi ty are each described for a preschool age child, are each of these character strengths equally likely to be scored highly when that child completes a VIA Youth measure during adolescence)? Do the variables related to virtues at early ages continue to have similar predictive relationships at later ages?


83 shared several similar characteristics. Both samples included families of mostly, Caucasian parents (i.e., 85% in Park and current study) of middle to high socioeconomic status. While the children in the current study are Chinese by birth, they were raised in families with similar demographic composition post adoption. Future res earch is needed to further explore the impact of culture on the development of character strengths among young children. How do children born and raised in China compare to children adopted from China? What character strengths are most prevalent among yo ung children raised in families of lower socioeconomic status or of other ethnic minority backgrounds? While the content analysis of free parental descriptions provided one means for assessing character strengths and virtues in young children, additiona l methods will need to be developed for assessing character in young children. The parental description while a parent or teacher rating scale would help assess the deg ree to which each character strength manifests for a child. For older children and adults, a self report rating scale measure already exists that assess the same 24 character strengths, but different forms are used in order to include age appropriate items A similar type of scale to be filled out by parents or teachers could be informed in part by the qualitative results from this study in order to write developmentally appropriate items that capture how character s trengths manifest at young ages. Rating scales would facilitate the ability of practitioners


84 inform interventions but also as an additional outcome measure following the implementation of interventions. Con clusion Examining positive characteristics of internationally adopted Chinese children broadens our understanding of their post adoption development. This unique population that appeared to be thriving from a deficit based perspective also appears to ho ld a very similar profile of character strengths in comparison to non adopted children. The findings of this current study helped to further validate the line of reasoning that certain character strengths do not manifest to a large extent at very young ag es. As this classification system was designed to be used across the lifespan, the assessment of character strengths across a two year time period and the analysis of age as a predictor variable provided results contributing to this aim. With its longitu dinal nature, the study found little overall change in character strength profiles from preschool to early school ages within this population. The most notable differences were found in the higher rates of Zest and Social Intelligence. As both of these strengths are part of the virtues Courage and Humanity, which were the most prevalent virtues and associated with lower levels of problem behaviors, Zest and Social Intelligence should be explored further as potential traits that could be promoted to foste r resilience.


85 References Achenbach, T. M., & Rescorla, L. A. (2000). Manual for ASEBA Preschool Forms & Profiles Burlington, VT: University of Vermont, Research Center for Children, Youth, & Families. Achenbach, T. M., & Rescorla, L. A. (2001). Manual for ASEBA School Age Forms & Profiles Burlington, VT: University of Vermont, Research Center for Children, Youth, & Families. Albers, L. H., Johnson, D. E., Hostetter, M. K., Iverson, S., & Miller, L. C. (1997). Health of children adopted from the former Soviet Union and eastern Europe: Comparison with preadoptive medical records. Journal of the American Medical Association, 278 (11), 922 924. Bazeley, P. (2003). Computerized data analysis for mixed methods research. In Handbook of mixed methods in social and behavioral research (pp. 385 422) Thousand Oaks, CA: Sage. Buckley, J. A. & Epstein, M. H. (2004). The Behavioral and Emotional Rating Scale 2 (BERS 2): Providing comprehensive approach to strengths based assessment. The California School P sychologist, 9, 21 27. Chisholm, K. (1998). A three year follow up of attachment and indiscriminate friendliness in children adopted from Romanian orphanages. Child Development, 69 (4), 1092 1106.


86 Epstein, M. H. (1999). The development and validation of a scale to assess the emotional and behavioral strengths of children and adolescents. Remedial and Special Education 20 258 262. Epstein, M., & Sharma, J. (1998). Behavioral and Emotional Rating Scale: A strengths based approach to assessment anual). Austin, TX: Pro Ed. Epstein, M. H., Cullinan, D., Ryser, G., & Pearson, N. (2002). Development of a scale to assess emotional disturbance. Behavioral Disorders, 28 5 22. Erikson, E. H. (1963). Childhood and Society (2 nd ed.). New York: Norton. Glascoe, F. P., & Dworkin, P. H. (1995) The role of parents in the detection of developmental and behavioral problems. Pediatrics, 95 (6), 829 836. follow up study of ch ildren adopted from Romania into the UK. Adoption Quarterly, 5(1), 5 22. Gunnar, M. G., Bruce, J., & Grotevant, H. D. (2000). International adoption of institutionally reared children: Research and policy. Development and Psychopathology 12, 677 693. Gun nar, M. G., Van Dulmen, M. H. M., & The International Adoption Project Team (2007). Behavior problems in postinstitutionalized internationally adopted children. Development and Psychopathology, 19, 129 148. Hauser, S. T., Vieyra, M. A., Jacobson, A. M., & Wertreib, D. (1985). Vulnerability and resilience in adolescence: Views from the family. Journal of Early Adolescence 5 (1), 81 100.


87 J ohnson, K., Huang, B, & Wang, L (1998). Infant abandonment and adoption in China. Population and Development Review, 2 4 (3), 469 510. Johnston, K. E., Swim, J. K., Saltsman, B. M., Deater Deckard, K., & Petreill, S. A. Asian children. Family Relations, 56 (Oct.) 390 402. Juffer, F., & van IJzendoorn, M. H. (2005). Behavior problems and mental health referrals of international adoptees: A meta analysis. Journal of American Medical Association, 293 (20), 2501 2515. Kohnstamm, G., Halverson, C., Mervielde, I., & Havill, V. (1998). Pa rental descriptions of child personality: Developmental antecedents of the Big Five? Mahwah, NJ: Lawrence Erlbaum Associates. Levy, T., & Orlans, M. (2000). Attachment disorder as an antecedent to violence and antisocial patterns in children. In T. Levy (E d.), Handbook of attachment interventions (pp. 1 26). San Diego, CA: Academic Press. Meese, R. L. (2005). A few new children: Postinstitutionalized children of intercountry adoption. Journal of Special Education, 39 (3), 157 167. M., Beckett, C., Keaveney, L., Kreppner, J. M., & the English and Romanian Adoptees Study Team. (2000). The effects of global severe privation on cognitive competence: Extension and longitudinal follow up. Child Development, 71 (2), 376 390. O'Connor, T. G (2003). Natural experiments to study the effects of early experience: Progress and limitations. Development and Psychopathology 15 (4), 837 852


88 Park, N. (2004). Character strengths and positive youth development. Annals of the American Academy of Polit ical and Social Science, 591 40 54. Park, N., Peterson, C., & Seligman, M. E. P. (2004). Strengths of character and well being. Journal of Social and Clinical Psychology, 23 (5), 603 619. Park, N., Peterson, C., & Seligman, M. E. P. (2006). Character s trengths in fifty four nations and the fifty US states. The Journal of Positive Psychology, 1 (3), 118 129. Park, N., & Peterson, C. (2005). The Values in Action Inventory of Character Strengths in youth. In Eds. K.A. Moore & L.H. Lippman, What do Child ren Need to Flourish? Conceptualizing and Measuring Indicators of Positive Development. New York: Springer. Park, N., & Peterson, C. (2006a). Moral competence and character strengths among adolescents: The development and validation of the Values in Act ion Inventory of Strengths for Youth. Journal of Adolescence, 29, 891 909. Park, N., & Peterson, C. (2006b). Character strengths and happiness among young children: Content analysis of parental descriptions. Journal of Happiness Studies, 7 323 341. Park, N. & Peterson, C. (2008). Positive psychology and chara cter strengths: Application to s trengths based school counseling. Professional School Counseling 12 (2), 85 92.


89 Pearlmutter, S., Ryan, S.D., Johnson, L.B. & Groza, V. (2008). Romanian adoptees and pre adoptive care: A strengths perspective. Child & Adolescent Social Work Journal, 25 (2), 139 156. Peterson, C. & Seligman, M. E. P. (2004). Character Strengths and Virtues : A Handbook and Classification. Washington, DC: American Psychological Ass ociation. Pomerleau A., Malcuit, G., Chicoine, J. F., Seguin, R., Belhumeur, C., Germain, P., Amyot, I. & Jeliu, G. (2005). Health status, cognitive and motor development of young children adopted from China, East Asia, and Russia across the first 6 month s after adoption. International Journal of Behavioral Development, 29 (5), 445 457. Rojewski, J. W., Shapiro, M. S., & Shapiro, M. (2000). Parental assessment of behavior in Chinese adoptees during early childhood. Child Psychiatry and Human Development, 31 (1), 79 96. Roth, J. L., & Brooks Gunn, J. (2003) Youth development progra ms: Risk, prevention, and policy Journal of Adolescent Health 32 170 82. Seligman, M. E. P. & Csikszentmihalyi, M. (2000). Positive psychology: An introduction [Special Issu e]. American Psychologist, 55, 5 14 Steen, T.A., Kachorek, L.V., & Peterson, C. (2003). Character strengths among youth. Journal of Youth and Adolescence, 32 (1), 5 16.


90 Tan, T. X., & Marfo, K. (2006). Parental ratings of behavioral adjustment in two sa mples of adopted Chinese girls: Age related versus socio emotional correlates and predictors. Applied Developmental Psychology, 27, 14 30 Tan, T. X., Marfo, K., & Dedrick, R. F. (2007). Special needs adoption from China: Exploring child level indicators, adoptive family characteristics, and correlates of behavioral adjustment. Children and Youth Services Review 29, 1269 1285. Tarullo, A. R., Bruce, J., & Gunnar, M. R. (2007). False b elief and emotion understanding in post institutionalized children. Social Development, 16 (1), 57 78. Trout, A., Ryan, J., La Vigne, S., & Epstein, M. (2003). Behavioral and Emotional Rating Scale: Two studies of convergent validity. Journal of Child and Family Studies 12 (4), 399 410. United States Department of State (2007) Immigrant vi sas issued to orphans coming to the US Retrieved April 22, 2007, from /family/adoption/stats/stats_451.html U.S. Department of State (2009). Intercountry Adoption. Office of Children's Issues Retrieved April 26, 2009, from chart.html Warren, K. R., Calhoun, F. J., May, P. A. Viljoen, D. L., Li, T. K., Tanaka, H., Marinicheva, G. S., Robinson, L. K., & Mundle, G. (2001). Fetal alcohol syndrome: An international perspective. Alcoholic Clinical and Experimental Research, 25 (5) 202S 206S.