Hofstra Horizons Research

Language Development in Children Adopted From China

Jenny Roberts
Assistant Professor, Speech-Language-Hearing Sciences

The author’s niece Claire (shown on right) and her best friend, Grace.This spring, I attended my youngest sister’s wedding. As we stood posing for a family picture, I realized that with my older sister’s adoption of my niece Claire several years ago, my family has joined a growing number of families in the United States. We are considered a blended family. Many of us know someone who has adopted a child from abroad, and it is now entirely common to see mixed-race and blended families throughout the country. In fact, in recent years, the number of Americans who have adopted children from other countries has exploded. Since 1966, when just over a thousand children were adopted from abroad, more than a quarter of a million adopted children have arrived in the United States (U.S. Department of State).

Are Internationally Adopted Children at Risk for Language and Other Delays?

For roughly three decades, following the Korean War, the largest number of children available for international adoption came from South Korea. Children arriving from Korea were largely well cared for prior to adoption, many living with foster families, and arrived with few medical and other developmental concerns (Pertman, 2000). By the late 1980s, adoptions from Korea into the United States were waning and, as shown in Figure 1, remain at their current low levels. Following the Romanian revolution and overthrow of the Ceausescu regime, a large number of children were abandoned in Romania, and resided in state-run organizations that were documented as having extremely depriving conditions (Rutter et al., 1998). Media outlets such as The New York Times (Hunt, 1991) ran stories documenting the appalling conditions in which many of these children were cared for, triggering a massive surge in international adoptions, with many of the children adopted by Westerners, particularly citizens of Canada, Great Britain and the United States. Also shown in Figure 1, this resulted in a large spike of adoptions in 1991 of Romanian children.

In contrast to the circumstances surrounding adoption in these countries, however, children are released for adoption from China for very different reasons. In China, widespread abandonment of infants has resulted primarily from its one-child policy, an attempt by the government to reduce massive over-population and related poverty. In particular, female infants represent the vast majority of adoptees from China, due in part to prior male inheritance laws and historical preference for sons to be caretakers (Cecere, 2001; Rojewski, Shapiro and Shapiro, 2000). Currently, children from China represent the largest group of international adoptees arriving into the United States, with more than 5,000 children arriving annually (see Figure 2). This figure has been rising since 1991, when, in an effort to deal with increasing numbers of children living in orphanages, adoption laws in China were changed to allow foreigners to adopt Chinese children (Cecere, 2001). But how do these children fare considering their early life experiences?

Children who have spent time living in institutional settings are at particular risk for an assortment of problems, including cognitive, linguistic, emotional, social and other developmental delays. Clearly, orphanages are not ideal places to grow up. Although institutional settings today do not resemble the appalling conditions documented in Romania in the early 1990s, they nevertheless are frequently both understaffed and underfunded (Mason and Narrad, 2005). Researchers Gunnar, Bruce and Grotevant (2000) summarized three levels of deprivation that can occur in institutional settings: a lack of basic nutrition, hygiene and medical care, a lack of stimulation and opportunity to interact with the environment, and a lack of stable interpersonal relationships and opportunity to develop an attachment relationship with a consistent caregiver. Improvements have occurred worldwide in hygiene and other basic conditions, and more stimulation is provided in many institutions, including in China, where orphanage donations from adoptive families are now mandatory, increasing the likelihood of a more stimulating environment (Cecere, 2001). However, a lack of stable interpersonal relationships, with opportunities to develop meaningful relationships with consistent caretakers, remains a problem in many orphanages. For example, caretakers in orphanages in China frequently feed infants and toddlers in high chairs arranged in a row, moving from child to child, while other children sit for periods of time awaiting their turn. In many households in America, mealtime represents a prime opportunity for social interaction and language learning. It is clear that such opportunities are significantly reduced in an orphanage setting.

Because the vast majority of children adopted worldwide reside in orphanage settings prior to adoption (Johnson and Cole, 1999), including in China, there is a concern about the quality of life prior to adoption. A great deal of research now shows that the longer the amount of time spent in institutional settings, the poorer the outcomes, particularly for children residing in extremely depriving conditions (e.g., Miller and Hendrie, 2000; Rutter et al., 1998), with longer periods of institutionalization associated with greater general developmental delays (Miller and Hendrie, 2000; Rutter et al., 1998).

Language Development

As Dr. Rena Krakow, Sharon Tao and I have described (Krakow, Tao and Roberts, 2005, p. 34), language delays are considered “part and parcel” of a pattern of developmental delays that often occur due to these environmental effects of institutional living. Researcher Sharon Judge suggests that among the potential delays with which internationally adopted children might present, language delays are the most serious, as language “mediates the whole process of adjustment to a new life” (Judge, 2004, p. 44).

Language development problems may be particularly difficult to sort out from the effects of early orphanage life because the length of time spent in these settings is intertwined with the amount of time spent hearing the native language. In addition to delays resulting from lack of interpersonal relationships and quality language interaction opportunities, children who are adopted internationally represent a unique group of language learners, due to the fact that they are typically adopted in the first two years of life, at a point in their development in which rapid language acquisition is typically occurring. Children who are adopted from China, in particular, experience a dramatic change in linguistic environment given the structural differences that exist between English and Chinese languages. Chinese languages, unlike English, are composed mostly of one-syllable words with no consonant clusters, with distinctions in meaning marked by changes in vocal pitch (Crystal, 1987; Pollock and Price, 2005). Most children who are adopted from China into the United States come to live with families who do not speak the language that the children were exposed to previously – they are “monolingual” (one language) families, and they thus experience an abrupt language switch. Given that the vast majority of children no longer receive exposure to the native language following adoption, some researchers have claimed that the term “bilingual” is inappropriate for this population (Glennen, 2002). Thus, my colleagues and I have come to use the term “second first language” to describe the language experience of this population. This process may constitute a unique language learning experience.

Given the potential for the language problems described above, what have we discovered about the language development of children adopted from China? It’s important to first note that most children adopted from China are relatively healthy upon arrival. Despite an increased risk of intestinal parasites, elevated blood levels, hepatitis B and other conditions, most children adopted from China progress nicely in stature and other indices of growth. For example, in a study that my colleagues and I conducted of 55 preschool-aged children adopted from China as infants and toddlers, Dr. Paul Wang performed medical examinations of the children and compared these to their early medical histories. He found that although medical measurements obtained shortly after arrival indicated variability in weight, length and head circumference, the majority of children were classified as healthy upon arrival, and none had any serious medical conditions when seen as preschoolers (Roberts et al., 2005a).

Typically, relatively little is known about communication and language abilities prior to adoption, and few language assessments are conducted in the native language prior to adoption. In my own work and that of my colleagues, several approaches have been taken to evaluate language abilities of children adopted from China. First, a number of parent-reported instruments of vocabulary development are available, which require parents to check off words from word lists that they believe their child says and/or understands. Instruments of this type are useful for evaluating early-developing English language skills in children during their first year or two post-adoption, because parents have been found to be reliable reporters of this aspect of language development. A number of studies now document that children adopted from China perform well on such instruments after periods of variable growth (Krakow and Roberts, 2003; Krakow, Tau and Roberts, 2005; Pollock, 2005; Tan and Yang, 2005). For example, researchers Tan and Yang (2005) found for childrenarriving between 18 and 25 months of age, after living with their adoptive families for about a year and a half, most were performing at age level on a parent-reported measurement of early language for number of words and phrases that they were saying. Ultimately, many adoptees actually surpassed the national normative sample on these measures for monolingual English-speaking children of the same age and gender.

Another approach to assessing children’s language is to examine their spontaneous language production during play sessions with their parents or others. For example, using such a method, Drs. Karen Pollock and Johanna Price examined the phonological (speech-sound) abilities of eight adopted children in the first few years following adoption. They found that although there was great variability in the development of speech-sound inventories and babbling when the children were toddlers, by 3 years of age nearly all were performing at the level of their non-adopted peers in phonological development (Pollock and Price, 2005).

Utilizing a third methodology, my colleagues and I collected comprehensive measures of language comprehension and production using a large battery of standardized tests (as well as spontaneous language samples), to examine a broad range of languageskills of preschool-aged children adopted from China as infants and toddlers. For example, we measured how many words children recognize when given a label and an array of pictures to choose from, and how many pictures they could identify when asked to label them. Vocabulary was further assessed by asking children to locate pictures that differed in size, quantity, location and direction. We examined children’s speech sound production using a normative measure by asking them to label a set of pictures or repeat words and then determined whether speech sound errors were made. Sentence comprehension was measured by asking children to look at two sets of pictures while listening to sentences, and then having them point to the correct pictures that corresponded to what was being said. Other language tests examined whether children knew English grammatical morphemes such as those that mark past tense (–ed), plural (–s), and present progressive (–ing), and whether they could repeat sentences that varied in length and complexity.

Across all of these measures, we found that the vast majority of the children that participated in our study performed in the average range of ability, and a significant number (27 percent) scored in the above average range. It is important to note that we examined these children using measures that were originally created to assess the language abilities of monolingual English-speaking children. We used these tests because normative measures of language ability have not yet been created for assessing the language of adopted children, and perhaps never will be.

A few of the children whom we examined performed less proficiently. We felt that it was important to continue to follow their language development for some period of time. Two years after our original study, we re-examined the language abilities of 10 of the lowest-performing children, to see how they were doing after additional English language exposure, and in most cases, speech and language therapy. We found that, on the whole, most of the children had made what is considered “clinically significant gain” on two or more language measures. However, for a few children, although performance improved, they continued to have difficulties in language development compared with both their adopted and non-adopted peers (Roberts et al., 2005b).

What about more advanced language skills? As children reach school age, the demands of the environment for language expectations greatly increase, in terms of reading, writing and other academic language skills. There is relatively little known about such advanced language skills in internationally adopted children in general, although studies conducted in Scandinavia suggest that internationally adopted children may experience difficulty as they face the challenges of school-age language (Dalen, 1995; Saetersdal and Dalen, 1991). In those studies, using teacher reports of language abilities, Dr. Monica Dalen and her colleagues found that although internationally adopted children appeared to do well on measures of so-called “everyday” or “conversational” language, they did less well on measures of “school-related” language in comparison to non-adopted peers. Her work suggested that such difficulties do not emerge until children are faced with more demanding language environments, such as those found in school settings. In the first comprehensive study of the oral language and literacy skills of children adopted from China, Dr. Kathy Scott directly assessed these skills in school-age children adopted as infants and toddlers. Rather than using parent- or teacher-reported measures, she administered an extensive battery of reading and language tests. She found that the majority of children performed in the average (58 percent) or above average (29 percent) range on a wide variety of measures (Scott, 2006). Children who were below average on two or more composite test measures presented with discrete profiles of performance, suggesting that there is not a single underlying cause for their problems.

In summary, researchers using a variety of methods to examine the language abilities of hundreds of children adopted from China have now overwhelmingly found good or even better-than-average language performance on measures of English language ability. We think that many complex factors may play a role in the relative language success of children adopted from China. First, they arrive in the United States in good health and appear to thrive once they arrive, as indexed by their ongoing good growth and medical outcomes some number of years later. Second, we can’t discount the relatively enriching environments into which they are adopted. The parents of children adopted from China are, by necessity, an older group of individuals who are usually eager and ready to parent (Cecere, 2001; Hollingsworth, 2000). According to Chinese laws, parents must be a minimum age of 35 to adopt. In addition, adoption is an expensive and time-consuming process, requiring a long period of paperwork, home study, waiting for referral and travel to China. We have found in our studies that parents are not only considerably older than the national average for parents of preschool-aged children (Fields, 2003), but are also far more educated. In ourpreschool sample, 85 percent of parents held a college degree, and of those, 57 percent held an advanced degree.

Although a small minority of children have a harder time acquiring English language skills, and they should be regularly monitored and offered speech-language and educational resources as needed, we now know that across numerous studies, the vast majority of children adopted from China do well in English language acquisition, even excelling in many cases. Considering their early life circumstances, we believe that this speaks not only to the loving, nurturing and linguistically stimulating environments into which they arrive, but also to the resiliency and robustness of the capacity for language acquisition.


Cecere, L.A. (2001). The Children Can’t Wait: China’s Emerging Model for Intercountry Adoption. Cambridge, MA: China Seas.

Crystal, D. (1987). The Cambridge Encylopedia of Language. Cambridge: Cambridge University Press.

Dalen, M. (1995). Learning difficulties among inter-country adopted children. Nordisk pedagogikk, 15(4), 195-208.

Fields, J. (2003). Children’s characteristics and living arrangements: March 2002. Retrieved August 1, 2003, from http://www.census.gov/prod/2003pubs/ p20-547.

Glennen, S. (2002). Language development and delay in internationally adopted infants and toddlers: A review. American Journal of Speech Language Pathology, 11, 333-339.

Gunnar, M.R., Bruce, J., & Grotevant, H.D. (2000). International adoption of institutionally reared children: Research and policy. Development and Psychopathology, 12, 677-693.

Hollingsworth, L. D. (2000). Who seeks to adopt a child? Findings from the National Survey of Family Growth (1995). Adoption Quarterly, 3, 1-23.

Hunt, K. (1991, March 24). “The Romanian baby bazaar,” The New York Times Magazine, 24-29.

Johnson, D., & Dole, K. (1999). International adoptions: Implications for early intervention. Infants and Young Children, 11(4), 34-45.

Judge, S. (2004). The impact of early institutionalization on child and family outcomes. Adoption Quarterly, 7(3), 31-48.

Krakow, R., & Roberts, J.A. (2003). Acquisition of English vocabulary by young Chinese adoptees. Journal of Multilingual Communication Disorders, Volume 1(3), 169-176.

Krakow, R., Tao, S., & Roberts, J. (2005). Adoption age effects on English language acquisition: Infants and toddlers from China. Seminars in Speech and Language. Special issue titled “Internationally Adopted Children: Guidelines for Speech and Language,” 26(1), 33-43.

Mason, P., & Narad, C. (2005). International adoption: A health and developmental prospective. Seminars in Speech and Language, 26, 1-9.

Miller, L.C. & Hendrie, N.W. (2000). Health of children adopted from China. Pediatrics, 105(6), 1-6.

Pertman, A. (2000). Adoption Nation. NY: Basic Books.

Pollock, K. (2005). Early language growth in children adopted from China: Preliminary normative data. Seminars in Speech and Language, 26, 22-32.

Pollock, K., & Price, J.R. (2005). Phonological skills of children adopted from China: Implications for assessment. Seminars in Speech and Language, 26, 54-63.

Roberts, J.A., Krakow, R., Pollock, K., Price, J., Fulmer, K., & Wang, P. (2005a). Language development in preschool-aged children adopted from China. Journal of Speech, Language and Hearing Research, 48, 93-107.

Roberts, J., Pollock, K., & Krakow, R. (2005b). Continued Catch-Up and Language Delay in Children Adopted from China. Seminars in Speech and Language. Special issue titled “Internationally Adopted Children: Guidelines for Speech and Language,” 26(1), 76-85.

Rojewski, J.W., Shapiro, M.S., & Shapiro, M. (2000). Parental assessment of behavior in Chinese adoptees during early childhood. Journal of Child Psychiatry and Human Development, 31, 79-96.

Rutter, M., and The English and Romanian Adoptees Study Team. (1998). Developmental catch-up and deficit following adoption after severe global early privation. Journal of Child Psychology and Psychiatry, 39(4), 465-476.

Saetersdal, B., & Dalen, M. (1991). Norway: Intercountry adoptions in a homogeneous country. In H. Altstein & R. Simon (Eds.), Intercountry adoption: A multinational perspective (pp. 83-108). New York: Praeger.

Scott, K. (2006). The Written Language Development of Children Adopted From China. Unpublished doctoral dissertation, Temple University.

Tan, T.X., & Yang, Y. (2005). Language development of Chinese adoptees 18-35 months old. Early Childhood Research Quarterly, 20, 57-68.

U.S. Department of State, Immigrant Visa Control and Reporting Division, Washington, D.C.

Main photo copyright istockphoto.com. Photo: Geopaul.

Hofstra in the News

More In The News
Hofviews - See More Photos

Hofstra Weather

° F
Heat Index ° F / Wind Chill ° F
Humidity %
Wind mph / Direction °
Rain in
See More from the Project WX Weather Stations

Recent Faculty News