Test Anxiety: A Cross-Cultural Perspective

C 2005) Clinical Child and Family Psychology Review, Vol. 8, No. 1, March 2005 (
DOI: 10.1007/s10567-005-2342-x

Test Anxiety: A Cross-Cultural Perspective Jaee Bodas1 and Thomas H. Ollendick1,2

The present paper examines test anxiety from a cross-cultural perspective with specific reference to the Indian and American cultures. The construct of test anxiety has been examined in many cultures all over the world. In this review, the importance of understanding and incorporating contextual factors in cross-cultural research is emphasized. Moreover, some of the methodological issues related to investigating culture-behavior relationship are discussed. Specifically, the derived-etic approach for conducting cross-cultural research is espoused. Then, research findings from western, cross-cultural, and Indian studies on test anxiety are reviewed. Consistent with the individualistic orientation of the western society, much of the research in the western world has adopted a de-contextualized approach. Inasmuch as many of the cross-cultural and Indian studies on test anxiety have their roots in western research, they have ignored the cultural context as well. To address this void, contextual variables relevant to test anxiety in the Indian setting are examined and hypotheses regarding the nature of test anxiety in Indian children are proposed. Finally, a research agenda is presented to examine these hypotheses using a derived-etic approach. KEY WORDS: test anxiety; cross-cultural; de-contextualized approach; children.

The primary purpose of this review is to examine the construct of test anxiety from a cross-cultural viewpoint. The concept of test anxiety has been much researched in many countries including Africa (e.g., Mwamwenda, 1994), America (e.g., Sarason, 1984), China (e.g., Yue, 1994), Czechoslovakia (e.g., Man, Budejovice, & Hosek, 1989), Egypt (e.g., El-Safty, 1995; Hocevar & El Zahhar, 1985), Germany (e.g., Hodapp & Benson, 1997), Holland (e.g., Van der Ploeg, 1983), India (e.g., Sud, 2001), Israel (e.g., Peleg-Popko & Klingman, 2002; Zeidner, 1993), Italy (e.g., Comunian, 1989), Japan (e.g., Araki, 1992), Jordan (e.g., Ahlawat, 1989), Korea (e.g., Schwarzer & Kim, 1984), Saudi Arabia (e.g., El-Safty, 1995), Turkey (e.g., Aysan, Thompson, & Hamarat, 2001),

and Urugway (e.g., Richmond, Rodrigo, & Lusiardo, 1989), suggesting that this problem is pervasive across geographic and cultural boundaries. However, most of the studies have not examined systematically the relationship between cultural factors and the nature of test anxiety. The objective of this review then is to better understand test anxiety in Indian and Western cultures so that better assessment and treatment can be developed and evaluated. The following questions are posed: (1) what is the nature of test anxiety among Indian children? (2) Is the manifestation of test anxiety in Indian children different from that in American (Western) children? (3) How do the cultural, psychosocial, and economic conditions in India influence test-related behavior of Indian children? (4) How should test anxiety be assessed in India? And (5) what are the research and clinical implications for these findings? It is theorized that given the large sociocultural differences in the environments of children from India and America, the nature and manifestation of test anxiety will be different in these two countries and differential implications for assessment and intervention will be evident.

1 Child

Study Center, Department of Psychology, Virginia Polytechnic Institute and State University, Blacksburg, Virginia. 2 Address all correspondence to Thomas H. Ollendick, Child Study Center, Department of Psychology, Virginia Polytechnic Institute and State University, Blacksburg, Virginia; e-mail: tho@ vt.edu.

65 C 2005 Springer Science+Business Media, Inc. 1096-4037/05/0300-0065/0


66 CROSS-CULTURAL RESEARCH The field of cross-cultural research, although a relatively new branch of psychology, is steadily gaining prominence in the field of psychology and the behavioral sciences more broadly. Indeed, crosscultural studies broaden the scope of research by allowing researchers to examine constructs in different cultural contexts and thereby determining the generalizability of their findings. Furthermore, such an approach leads to an enriched appreciation of the behavior-culture interactional processes.

Contextualism With increasing globalization and the ensuing multicultural society in which we live, the need for understanding individuals as “embedded” in their familial and cultural context has become particularly relevant (Szapocznik & Kurtines, 1993). Much of the research on child development in Western culture has its foundation in theories that have considered individuals as being independent, or at least largely so, from their social and cultural environments. This approach, referred to as a de-contextualized approach, has been questioned by some who enlist Eastern theories and philosophies and who espouse contextualist approaches (e.g., Markus & Kitayama, 1991; Sinha, 1997). The contextualist approach is at the core of cross-cultural psychology in which researchers believe that culture itself has an impact on behavior and that the effects of culture must be systematically examined in research (Berry, 1980; Berry, Poortinga, Segall, & Dasen, 1992; Segall, Dasen, Berry, & Poortinga, 1990). Contextualist influences in psychology have come from many sources. However, Urie Bronfenbrenner’s ecological systems approach has been the major contextualist influence in psychology (1993; Bronfenbrenner & Morris, 1998). His ecological-systems approach conceptualizes context as four interacting levels of environment, namely: microsystem, mesosystem, exosystem, and macrosystem. The microsystem is the immediate context of development which involves “activities, roles, and interpersonal relations” of a child in a “face-to-face setting” that consists of interactions with other people such as peers, teachers, and parents as well as physical settings such as the school and the home (Bronfenbrenner, 1993). The mesosystem involves an interaction of two or more

Bodas and Ollendick microsystems. Thus, it comprises “the linkages and processes taking place between two or more settings that contain the developing person,” such as between peer groups and parents or between home and school (Bronfenbrenner, 1993). The exosystem involves “linkages and processes between two or more settings,” one of which does not involve the child such as the relation between home and parents’ workplace (Bronfenbrenner, 1993). This level includes major social institutions such as the government, the school system, and mass media (Miller, 2000). Finally, a macrosystem is composed of an “overarching pattern of micro, meso, and exosystems” characteristic of a given culture, particularly with reference to beliefsystems, lifestyles, and opportunity structures that characterize that culture. These levels interact with each other such that what impacts one level affects the other levels as well. Temperamental differences result in different children seeking out different environments. The concept of goodness of fit is applicable here, such that the same environmental conditions may not be suited to all children (Miller, 2000). As will be evident in the following sections, such a conceptualization of the cultural context in terms of multiple levels of the environment nested within each other is particularly relevant to this review.

Methodological Issues Although there is little dispute regarding whether or not the contextual or cultural factors play a role in shaping behavior, there are differing views regarding the nature of this role. The relationship between culture and behavior may be studied from different perspectives. For instance, culture-behavior relationship may be examined from within-culture or across-cultures perspectives. The within-culture approach of cultural psychology, considers culture as an integral part of the individual (Bruner, 1993; Greenfield, 2000; Jahoda, 1992). In this sense, culture does not exist in a vacuum outside the individual; rather, culture and behavior are inseparable. A cross-cultural psychology approach, on the other hand, primarily involves comparative research and considers culture to be a contextual factor outside of individuals that impacts their behavior (Lonner & Adamopoulos, 1997). Both these approaches may provide important information for understanding behavior. In 1967, Pike coined the terms “emic” and “etic” in describing the above mentioned perspectives to

Test Anxiety: A Cross-Cultural Perspective understanding behavior from a cultural standpoint. Specifically, the etic approach involves examining a behavior from “outside of a particular system,” entailing a comparative approach for studying a behavior across different cultures (Pike, 1967). An emic approach, in contrast, examines a phenomenon “from inside the system” (Pike, 1967) and involves exploring and understanding the behavior within the specific cultural framework of the individual. However, according to Pike, rather than being mutually incompatible or dichotomous, the emic and etic approaches can provide an understanding of the same phenomenon just from different perspectives. Berry (1969, 1989) offers a three step “imposed etic – emic – derived etic” procedure for using these concepts in cross-cultural research. Specifically, according to Berry, research begins with using a concept from the researcher’s culture; assuming it to be a valid basis for examining the concept in another culture and the researcher then compares it across the two cultures. Based on the assumptions related to this concept, Berry (1989) referred to this methodology as the “imposed etic” approach. That is, the researcher imposes his or her outlook on that of the to-be studied culture. The next step involves thoroughly examining the concept within the new cultural context, which is the emic approach and which (hopefully) will lead the researcher to the last “derived etic” stage. This last stage involves deducing features that are common to both cultures and that have been derived, in the first place, by emic research conducted in each culture. Basically, the emic approach allows one to “derive” a view based on the to-be studied culture. Emphasizing the necessity of such a three-step process, Berry (1969) has criticized the widespread use of the imposed etic approach because of the risks involved in making erroneous assumptions that one’s own culture provides an adequate basis for studying the culture of another. According to Berry (1969), sole reliance upon the imposed etic approach has generated largely uninterpretable cross-cultural findings. Conversely, however, the derived etic notion allows the researcher to make potentially well-founded comparisons between the two cultures. Thus, culture-behavior relationships may be examined from within-culture or across-cultures perspectives. However, Berry (1969, 1999) assumes an “inclusive” position in his “imposed etic – emic – derived etic” approach stating that a “symbiosis” between these apparently contrasting approaches is possible. Furthermore, he considers both etic and emic approaches to be “neces-

67 sary but not sufficient” for understanding the relationship between culture and behavior. In fact, such a symbiosis brings into being a broader discipline of cross-cultural psychology that is essential for achieving a “universal psychology” and understanding what has become known as “pan-human” behavior. An early study by Davidson, Jaccard, Triandis, Morales, and Diaz-Guerrero (1976) demonstrated how such a convergence between the emic and etic approaches can occur in research. Additionally, Berry (1999) has described how this approach has been used in the field of intelligence and cognitive styles. However, this approach has not been used sufficiently, especially in clinical psychology. Indeed, early on, Jahoda (1977) expressed reservations regarding the validity and applicability of this approach in crosscultural psychology. Specifically, Jahoda raised concerns about the approach being vague and abstract and about whether a phenomenon can truly be discovered in a new culture because of the structure, implicit or otherwise, imposed by the researcher. Empirically, the validity of this approach needs to be examined more extensively. Nevertheless, at the conceptual level, this approach attempts to overcome methodological limitations of the purely emic or etic approaches. On the one hand, it acknowledges the achievements of the dominant western science, unlike a purely emic approach; yet on the other hand, it does not assume that the western theories are universal as in a purely etic approach. It is thus an integrated approach that endeavors to be contextsensitive in its pursuit of universal psychological principles. This approach will be espoused in this review.

INDIA AND AMERICA: A CULTURAL COMPARISON A considerable amount of cross-cultural research has focused on comparisons between two or more countries with reference to specific psychological variables, where the country is assumed to represent a particular culture. Furthermore, studies have often adopted broad dimensions such as individualism and collectivism (Triandis, 1995) to explain crosscultural differences in a construct. Use of such broad and general classifications of the construct of culture (such as nation and individualism-collectivism) has been criticized primarily because of the difficulties related to meaningful interpretation of individual differences based on these classifications (van de Vijver & Leung, 2000). Thus, specific factors within these

68 broad classification systems need to be delineated for a better understanding of the relationship between culture and the psychological construct of interest. Triandis (1999) has outlined four attributes that serve as dimensions for distinguishing between individualistic and collectivistic cultures: definition of self, structure of goals, emphasis on norms versus attitudes, and emphasis on relatedness versus rationality. In other words, it is expected that people from these two types of cultures will differ on whether the self is viewed as interdependent or autonomous, whether in-group or personal goals are perceived as important, whether norms and duties or personal needs and attitudes determine social behavior, and whether relatedness or rationality is emphasized in relationships. Individualistic and collectivistic cultures are further distinguished on the dimension of horizontal-vertical relationships (Triandis, 1999). Thus, horizontal cultures believe in “equality” of people in which people are basically similar. In contrast, vertical cultures consider “hierarchy” in relationships as natural and assume that people are basically different. Numerous studies have shown that these dimensions are indeed useful in distinguishing between people from individualistic and collectivist cultures. For example, Verma and Triandis (1999) found that as compared to Indians, Americans were significantly higher on horizontal individualism but were significantly lower on vertical collectivism. Furthermore, among Indians, the definition of selfconcept as well as a perception of others depend on the context of behavior; whereas, Americans tend to have more de-contextualized judgments (Shweder & Bourne, 1984). Thus, when describing a person, Indians are more likely to use contextually qualified descriptions, while Americans tend to use contextfree descriptions more frequently. Consistent with this observation, Wang and Ollendick (2001) noted that the Chinese language does not have a comparable expression for “self-esteem.” Indeed, in the Confucianism philosophy, construal of self is interdependent such that self is conceived as being embedded in the context of relationships with significant others. Likewise, even within the American population, Asian American students are more likely to sacrifice personal goals, especially for their parents and family, as compared to European Americans (Suzuki & Greenfield, 2002). However, it is important to bear in mind that collectivism and individualism are “polythetic” constructs implying that there may be significant differences even within these cultures (Triandis, 1995). However, as the studies mentioned

Bodas and Ollendick above suggest, some generic attributes can be applied to define the relative or general nature of these societies. Thus, at a broader level, the individualismcollectivism dimension may provide a basis to compare and contrast cultures in the two countries. Another conceptual basis for cross-cultural comparison has been provided by Berry’s (1976; 2001) ecocultural framework. This theory attempts to explain diversity in psychological constructs (including similarities and differences in individuals and groups) using ecological variables and the cultural and biological variables that mediate this relationship. Ecosocial dimensions for categorizing nations include variables such as religion, education, economy, mass communications, and population; cultural and biological factors include variables such as enculturation, socialization, and genetics (Georgas & Berry, 1995). In a recent study, Georgas, van de Vijver, and Berry (2004) investigated the relationship between these ecosocial factors and psychological variables across different nations and found that ecological and sociopolitical indices such as religion and affluence that characterize nations were related in a systematic way to psychological factors such as autonomy, hierarchy, individualism, well-being, and commitment. This line of research has thus marked out specific contextual variables that can be used in cross-cultural research in place of the more generic variables such as “nation” or culture. On the basis of the studies reviewed above, three factors can be deduced for the comparison of cultures from different countries, namely: intraindividual variables, interpersonal variables, and institutional or ecological indices. These factors are generally consistent with Bronfenbrenner’s theoretical framework consisting of multiple levels of the cultural context. Individual level variables refer to individual differences obtained on the constructs of interest. More specifically, for our purposes here, individual level variables would consist of individual differences in factors such as test anxiety scores, pattern of cognitions, frequency of off-task thoughts, or task performance. Interpersonal variables include interpersonal relations between individuals and their parents or family, peers, teachers, and colleagues. In terms of test anxiety research, significant interpersonal variables might include parental expectations, family values, and parental education. Finally, the two cultures may be compared on the basis of ecological variables such as population, economy, and education. These ecocultural variables provide a framework for interpretation of similarities and differences

Test Anxiety: A Cross-Cultural Perspective in the psychological constructs that are investigated cross-culturally (Georgas et al., 2004).

TEST ANXIETY AND WESTERN CULTURE Much of what we know and understand of test anxiety today is based on research undertaken in the western world. The first systematic investigation of test anxiety was undertaken at Yale University in 1952 by Sarason and Mandler. Specifically, they reported that individuals low in test anxiety outperformed those high in test anxiety on intelligence tests (Mandler & Sarason, 1952). The construct of test anxiety has been extensively researched since then, leading to its further development in terms of conceptualization, assessment, and treatment.

Theory and Conceptualization An overall consistent finding related to test anxiety has been its relation to poor performance on tests and achievement tasks. A number of theories have been proposed to explain these results. One line of reasoning attributes poor performance of high test-anxious individuals to the interfering effects of anxiety in evaluative situations. This perspective, known as the interference model, assumes that test anxiety interferes with retrieval of previously learned information in test situations by producing task-irrelevant responses (Alpert & Haber, 1960; Hembree, 1988; Liebert & Morris, 1967; Sarason, 1984; Wine, 1971). A number of theorists have built upon and expanded these early notions. However, the validity of these theories came into question when treatments designed to reduce test anxiety were successful in reducing anxiety but failed to show a corresponding and significant impact on academic performance (Tyron, 1980). In an attempt to explain these seemingly discrepant findings, an alternative deficits model of test anxiety was proposed by a number of researchers (e.g., Culler & Hollahan, 1980; Tobias, 1985), wherein poor performance of high test-anxious individuals was attributed to lower ability and deficient study habits (Hembree, 1988). In contrast to these previous two sets of theories, Benjamin, McKeachie, Lin, and Holinger (1981) provided support for an information-processing model of test anxiety. This model is best viewed as a combination or synthesis of the interference and deficit models (Naveh-Benjamin, 1991; Naveh-Benjamin,

69 McKeachie, & Lin, 1987). According to this model, poor performance of test- anxious individuals can be attributed to a number of cognitive deficits at all stages of information processing including encoding and organization deficits reflected in poor study skills, as well as problems with retrieving information (due to interference). Finally, in their review article on test anxiety in children and adolescents, King, Ollendick, and Gullone (1991) enumerated other potential theories attempting to explain the development of test anxiety. Specifically, they reported upon early studies that attributed test anxiety to unrealistic parental expectations, as well as school factors such as experiences of success and failure. Although King et al. (1991) recommended inclusion of parental and systemic variables they reported that such research was difficult to conduct and that much of the current research on test anxiety focused on the role of cognitive-attentional deficits. Thus, a host of theories from the western perspective have been put forth to account for the relationship between test anxiety and lower performance, most of them endorsing an individualistic philosophy and approach. Although the theories of test anxiety provide important insights regarding the process (interference model, deficits model, or information processing model) by which test anxiety affects performance, much of the recent research on test anxiety has focused on the differential impact of emotionality and worry factors of test anxiety on performance (Hembree, 1988). Emotionality refers to the physiological reactions such as arousal, trembling, sweating, etc. that are experienced in an evaluative situation. Worry, on the other hand is the cognitive manifestation of test anxiety. Studies examining the impact of the emotionality factor on performance, however, suggest a lack of consistency in findings. Somewhat unexpectedly, the majority of the studies have found that the emotionality factor (i.e., physiological arousal) has weak or insignificant effects on performance (Hembree, 1988). According to the Yerkes-Dodson law (1908), an inverted U-shaped function relates performance to arousal, with the peak of performance occurring at some intermediate level of arousal. Consistent with this theory, Cassady and Johnson (2002) provided evidence that moderate, but not low or high, levels of physiological arousal were related to higher exam performance. Most other studies, however, have failed to support this finding (Hembree, 1988; Hong, 1999; King, Ollendick, & Prins, 2000; Sarason, 1984).

70 In contrast to these mixed findings, the worry component of test anxiety has been shown consistently to have an inverse relationship with performance; a relationship that has been observed in children as well as adults (Cassady & Johnson, 2002; Hembree, 1988; Hong, 1999; King et al., 2000; McIlroy & Bunting, 2002; Morris, Smith, Andrews, & Morris, 1975; Sarason, 1984; Seipp, 1991; Zatz & Chassin, 1983), in both genders (Sowa & LaFleur, 1986), and in other cultures (O’Neil & Fukumura, 1992; Padua, 1993). Moreover, these studies have also reliably shown that worry is manifested as task debilitating cognitions, including more negative self-evaluations and off-task thoughts and fewer positive self-evaluations. Still other studies have investigated the relationship between test anxiety and psychopathology; specifically, anxiety disorders. For instance, elementary school children with severe test anxiety were assessed by Beidel and Turner (1988) using the Child Assessment Schedule (Hodges, McKnew, Cytryn, Stern, & Kline, 1982). Their findings indicated that high test-anxious children were more worried about academic performance, their popularity, and friendships as compared to non-test anxious children. Although none of the children from the nontest-anxious group were diagnosed with an anxiety disorder, about 60% of those in the high test-anxious group met criteria for an anxiety disorder. Similar results were obtained by King, Mietz, Tinney, and Ollendick (1995) using the Interview Schedule for Children (ISC; Kovacs, 1985) for a sample of adolescents from Australia. Thus, there seems to be considerable evidence indicating that many children who are anxious in testing situations also experience a diagnosable anxiety disorder, such that the presence of test anxiety in children may serve as an indicator of more pervasive psychological distress (Beidel & Turner, 1988; King et al., 1995, 2000; Warren, Ollendick, & King, 1996). Indeed, King et al. (1995) reported that the high test-anxious group endorsed significantly higher levels of psychopathology on various self report measures, perceiving themselves as physiologically anxious, prone to worry, and socially sensitive. Collectively, these theories provide a rich and broad account of test anxiety and its relation to more generalized anxiety. Summarizing the findings from the test anxiety literature, it is evident that there is less evidence for the role of emotionality in affecting performance of test-anxious children. In contrast, the worry component of test anxiety is consistently seen to be debili-

Bodas and Ollendick tative to performance, especially academic achievement. Furthermore, as compared to their low testanxious counterparts, high test-anxious children and adolescents report greater levels of psychopathology and are also more likely to be diagnosed with a clinically significant anxiety disorder. Assessment A variety of assessment tools have been developed in response to the evolving theories of test anxiety. For example, Mandler and Sarason (1952) developed the Test Anxiety Questionnaire (TAQ) which was the first scale for assessing individual differences in test anxiety. Subsequently, as the field of test anxiety advanced with more and more research being done in the area, more sophisticated tests comprising factor-analytically derived subscales were developed. The Test Anxiety Scale developed by Sarason (TAS; 1958) replaced the TAQ because it was easier to administer and score. In 1960, the Test Anxiety Scale for Children (TASC) was developed by Sarason, Davidson, Lighthall, Waite, and Ruebush (1960) for younger children. This, in turn, was followed by development of the Worry – Emotionality Questionnaire (WEQ) by Liebert and Morris (1967) based on their conceptualization of test-anxiety as comprising those two primary components. In contrast to the TAQ and the TAS, the WEQ consisted of two rationally derived subscales and conceptualized test anxiety as a multidimensional construct. The Test Anxiety Inventory (TAI; 1980) developed by Spielberger made an important contribution to the field by using a factor analytic approach to develop the worry and emotionality subscales. Subsequently, the WEQ was revised (Morris, Davis, & Hutchings, 1981), with the worry-emotionality subscales being derived factor-analytically as well. More recently, I.G. Sarason developed the Reactions to Tests scale (RTA; 1984), on the basis of his four-factor conceptualization of test anxiety, consisting of four distinct subscales, namely: worry, test irrelevant thoughts, tension, and bodily symptoms. Finally, Friedman and Bendas-Jacob (1997) developed a new measure of test anxiety comprising three factors, namely, cognitive obstruction, tenseness, and social derogation. Thus, over time, advances in research have led to the development of more sophisticated and psychometrically sound measures of test anxiety, which reflect, at least in part, further development of the theory and a more refined conceptualization of the construct of test anxiety.

Test Anxiety: A Cross-Cultural Perspective Researchers have also attempted to understand test anxiety by exploring children’s cognitions. Based on Wine’s cognitive-attentional theory (1971) of test anxiety, the Children’s Cognitive Assessment Questionnaire (CCAQ; Zatz & Chassin, 1983) was developed to tap four types of cognitions, namely: off task thoughts, on task thoughts, positive evaluations, and negative evaluations. Likewise, Smith, Arnkoff, and Wright (1990), used the Checklist of Positive and Negative thoughts (CPNT; Galassi, Frierson, & Sharer, 1981) to tap the thoughts experienced by students during an exam. Assessment procedures, other than the conventional self-report questionnaires, have also been used in test anxiety research. For example, some researchers have employed the “thought listing technique” (Bruch, 1978), which requires participants in an evaluative situation to report their thoughts in a written or verbal format (e.g., Prins & Hanewald, 1997; Smith et al., 1990). Occasionally, physiological measures have been utilized in test anxiety research to tap the somatic changes that are associated with the phenomenological experience of test anxiety. Cardiovascular indices such as heart rate measures and blood pressure have been more commonly investigated in test anxiety research (e.g., Beidel, 1988; Morris & Liebert, 1970). For example, Beidel (1988) reported that testanxious children had significantly larger heart rate increases during social-evaluative tasks. A few other studies have employed palmar sweat indices (e.g., Cornish & Dilley, 1973) and salivary cortisol (e.g., Spangler, 1997) in assessment of physiological responses in test anxious persons. Although physiological measures appear to overcome some of the shortcomings of the self-report measures, a number of problems have been reported regarding their use, including a lack of normative data, questionable construct and ecological validity, and their reliability (see Zeidner, 1998 for a review). In addition to physiological assessments, behavioral observational measures have also been used in test anxiety research in which behaviors characterizing test anxiety are operationally defined and measured. Wine (1979), for example, used behavioral observation as a tool for assessing test anxiety in addition to administering the TASC (Sarason et al., 1960). In the study by Wine (1979), classroom behavior was videotaped and later coded in observational categories such as attending behaviors, taskrelated behaviors, communications, and interactional behaviors. Although behavioral observation measures provide rich data regarding the participants’ be-

71 haviors in evaluative situations, these measures have been used infrequently by researchers because of a number of difficulties related to their utilization in research, including, observer bias and reliability issues related to their reactivity, as well as their high costs in terms of time and finances (King et al., 1991). Based on the “three response systems” approach to assessment, anxiety is conceptualized in terms of three distinct response modes: cognitive, behavioral, and physiological (Lang, 1968). The measures of test anxiety that have been developed and used in research indeed reflect measurement of these three response modes. Moreover, multiple methods of measurement have been employed in research including self-reports, questionnaires, observational procedures, and physiological methods (Ollendick & Hersen, 1993). Interestingly, most of the measurement approaches have focused on assessment of the individual and have disregarded measurement of the environmental or sociocultural context. This approach is consistent with western, individualistic, notions, which consider test anxiety to be within the child.

Treatment Numerous investigations have been carried out to investigate the effectiveness of a variety of interventions on the deleterious effect of test anxiety on the academic performance of children and youth. Generally, behavioral interventions have focused on the treatment of the emotionality component of test anxiety, attempting to reduce the physiological arousal and reactions of test-anxious students (Hembree, 1988). Among the behavioral interventions used in the treatment of test anxiety are relaxation training (e.g., Bedell, 1976), systematic desensitization (e.g., Gonzalez, 1995), covert reinforcement (e.g., Wisocki, 1973) and biofeedback (e.g., Vagg & Papsdorf, 1995). Systematic desensitization has been the most commonly used treatment for test anxiety, followed by relaxation training (Hembree, 1988). Alternatively, the cognitive approaches, based on the assumption that cognitive processes are key factors underlying the test-anxious behavior of students, have attempted to manage the worry component of test anxiety (e.g., Bruch, 1978; Kreitler & Kreitler, 1987; see Zeidner, 1998 for review). Cognitive behavioral interventions have attended to both worry and emotionality components of test anxiety (Hembree, 1988). Additionally, treatments focusing

72 on study skills training have concentrated on training students to study efficiently by managing their time effectively and by improving their information processing skills. (e.g., Hembree, 1988; see Zeidner, 1998 for review). “Testwiseness training” involves further teaching of test taking skills that help students become proficient in interpreting test questions accurately, retrieving information, and organizing and reporting their knowledge of the subject (Spielberger & Vagg, 1987). A comprehensive meta-analytical investigation by Hembree (1988) found the behavioral and cognitive behavioral interventions that address both the emotionality and worry components of test anxiety to be most effective in reducing test anxiety as well as improving performance. In addition, testwiseness training has been shown to be effective in reducing test anxiety for students with poor test taking skills, while study skills training was found to be effective when combined with cognitive behavioral treatments. In general, reduction in test anxiety was accompanied by improved test performance. Not inconsistent with the above analysis, a recent meta-analytic study reported that treatments combining skills training with cognitive or behavioral approaches were the most effective form of treatment (Ergene, 2003). Additionally, this review found that individual therapy combined with group intervention resulted in more improvement than either of these approaches used alone. In sum, integrated approaches to treatment (cognitive-behavioral and skills training; group and individual formats) appear to be more effective in reducing test anxiety. As with assessment, these interventions are all individually oriented and assume that test anxiety exists within the child.

TEST ANXIETY AND CROSS CULTURAL RESEARCH Conceptualization, Assessment, and Treatment As mentioned earlier, the construct of test anxiety has been investigated in a variety of cultures. Much of the cross-cultural research in this area has been focused on investigating the factor structure of test anxiety by adapting the existing “Western” assessment measures to the particular culture being studied. For instance, Spielberger’s TAI (1980) has been translated and adapted into numerous cross cultural editions including Chinese (Rocklin

Bodas and Ollendick & Ren-Min, 1983), Czech (Man, Budejovice, & Hosek, 1989), Dutch (Van der Ploeg, 1983), Hebrew (Zeidner & Nevo, 1993), Hindi, (Sharma, Sud, & Spielberger, 1983), Hungarian (Sipos, Sipos, & Spielberger, 1985), Italian (Comunian, 1985), Japanese (Araki, Iwawaki, & Spielberger, 1992), Jordanian (Ahlawat, 1989), Korean (Schwarzer & Kim, 1984), Norwegian (Hagtvet, 1984), Spanish (Bauermeister, Collazo, & Spielberber, 1983), and Turkish (Oner et al., 1987) languages. Other studies have compared two or more cultures in terms of their scores on equivalent measures of test anxiety, task performance, and other related constructs. For example, such comparisons have been made between youth in American and Chinese (Yue, 1994), American and Indian (Sud, 1991) children, Asian American and European American (Pang, 1991), Indian and Iranian (Sharma, Parnian, Spielberger, 1983), and Indian and Italian (Sud, 1991) cultures. Moreover, one such study compared high school and college students from four Asian (Indian, Jordanian, Chinese, and Korean) and five western (Hungarian, Turkish, Italian, German, and American) cultures (Sharma & Sud, 1990). Cross-cultural data obtained using the TAI versions is summarized in Tables I and II. A meta-analytic study conducted by Seipp and Schwarzer (1996) examined TAI (Spielberger, 1980) data obtained from 14 different nations. They reported that women in the majority of the cultural groups tended to have higher levels of test anxiety than men. This trend can be seen in the data summarized in Tables I and II. Moreover, this pattern is more prominent in the emotionality scores than in the worry scores. The role of socialization practices that encourage women to express their emotions and men to suppress their emotions including anxiety has been implicated in an attempt to explain these cross-cultural trends (Seipp & Schwarzer, 1996; Zeidner, 1998). Thus, gender differences may not be true differences in the level of test anxiety experienced by men and women, but may reflect the degree to which men and women are willing to admit their test anxiety. Among the various nations, the highest levels of test anxiety were noted for youth from Egypt, Jordan, and Hungary, while the lowest levels of test anxiety were reported for China, Italy, Japan, and the Netherlands. Overall, the Islamic countries as a group reported the highest levels of test anxiety while Western European and Asian countries reported the lowest levels of test anxiety. Seipp and Schwarzer (1996) attributed the high scores of Arabic students on the test anxiety scale to the severe

Test Anxiety: A Cross-Cultural Perspective

73

Table I. Means (Standard deviations) of Test Anxiety Total Scores for High School Children Males

Females

Country

Authors

M (SD)

N

M (SD)

N

China Czechoslovakia Germany Hungary Italy India Iran Jordan Korea Netherlands Turkey US

Rocklin and Ren-Min (1989) Man et al. (1989) Schwarzer and Kim (1984) Sipos et al. (1985) Comunian (1985) Sud and Sharma (1990) Sharma, Parnian, et al. (1983) Ahlawat (1989) Schwarzer and Kim (1984) Van der Ploeg (1982) Oner & Kaymak (1987) Spielberger (1980)

37.9 (8.8) 37.6 (8.9) 37.0 (9.3) 38.8 (8.4) 38.1 (9.6) 39.7 (10.4) 46.2 (10.7) 48.5 (11.1) 49.2 (13.3) 32.7 (10.5) 39.2 (9.8) 40.9 (12.8)

100 73 426 332 500 359 40 1014 367 57 197 527

38.6 (8.1) 41.9 (9.8) 40.6 (11.3) 41.2 (8.9) 42.4 (10.9) 42.1 (10.8) 51.8 (8.2) 50.4 (11.2) 51.1 (13.7) 37.4 (11.2) 44.6 (10.2) 45.7 (13.6)

105 81 473 368 500 465 40 839 389 97 134 591

Note. Adapted from Zeidner (1998).

negative consequences related to poor examination performance in these countries. In general, two primary hypotheses have been proposed to explain the cross-cultural differences in test anxiety that have been observed (Seipp & Schwarzer, 1996; Zeidner, 1998). First, it has been proposed that socialization practices and parental expectations in certain cultures may put excessive pressure on students to achieve. Second, the impact of high stake examination systems has been hypothesized to cause high levels of test anxiety. For instance, Schwarzer and Kim (1984) compared Korean youth with those from other countries, with reference to their scores on the TAI, and found that Korean students had higher means on test anxiety than students from other nations. The authors attributed the high scores of the Korean youth to authoritarian parenting and a high stress school and examination system. Similar explanations have been evoked in understanding relatively high levels of test anxiety among students in Jordan (Ahlawat, 1989). How-

ever, these hypotheses failed to hold true for students from Turkey (Oner & Kaymak, 1987) and Japan (Araki et al., 1992). Thus, Oner and Kaymak (1987) found that Turkish youth had lower or equal levels of test anxiety as compared to youth from Korea, India, America, Germany, and Hungary. Likewise, the Japanese youth had significantly lower levels of anxiety as compared to students from Czechoslovakia, Korea, Germany, Hungary, and the United States. Thus, the empirical evidence for these hypotheses is inconsistent. Finally, cross-cultural research examining the effectiveness of interventions for test anxiety has mirrored the research in western societies, and has been primarily cognitive-behavioral in orientation (e.g., Hwang, 1997; Sud & Prabha, 1995; Sud & Sharma, 1990). Interestingly, while western research in the field of test anxiety has assumed the cultural context in which test anxiety is studied, many of the cross-cultural studies cited above have ignored the

Table II. Means (Standard deviations) of Test Anxiety Total Scores for College Students Males

Females

Country

Authors

M (SD)

N

M (SD)

N

China Italy India Israel Iran Japan Jordan Netherlands Turkey

Rocklin and Ren-Min (1989) Comunian (1985) Sud and Sharma (1990) Zeidner and Nevo (1992) Sharma, Parnian, et al. (1983) Araki et al. (1992) Ahlawat (1989) Van der Ploeg (1982) Oner and Kaymak (1987)

36.7 (9.3) 36.4 (8.9) 38.5 (10.3) 38.7 (10.2) 39.8 (9.9) 36.3 (10.4) 44.6 (9.9) 34.1 (10.3) 34.5 (8.2)

167 304 454 245 40 308 768 116 130

38.4 (8.8) 35.4 (10.3) 42.5 (9.7) 41.8 (9.9) 46.9 (10.6) 37.1 (9.4) 47.8 (10.7) 36.3 (10.6) 37.5 (9.6)

163 169 446 283 40 451 951 68 143

Note. Adapted from Zeidner (1998).

74 contextual information in their investigation of crosscultural differences. Much of the cross-cultural research then, with a few exceptions, has adopted the “imposed etic” approach to investigate the nature of test anxiety in different cultures. As suggested by Berry (1989), although these types of investigations (imposed etic) are a good starting point, it is important for the researchers to advance their research using the “emic” approach for individual cultures and finally to adopt the “derived etic” approach to make meaningful inferences.

TEST ANXIETY AND INDIAN CULTURE The first attempt to investigate the phenomenon of test anxiety among school children in India was made in 1972 by Nijhawan. She translated the Test Anxiety Scale for Children (TASC; Sarason et al., 1960) into Hindi and Punjabi and explored the nature of test anxiety using a sample of school children in India. Her initial work demonstrated the negative impact of test anxiety on task performance and prompted further research in the field of test anxiety in India.

Theory and Conceptualization Several preliminary studies have attempted to investigate the nature of test anxiety in India. For example, A. Sud and H. Sharma (1989) examined the role of cognitive interference in understanding test anxiety. Specifically, 180 students with high, medium, and low levels of worry on the Hindi version (TAI-H, Sharma, Sud, et al., 1983) of the Test Anxiety Inventory (TAI, Spielberger, 1980) were assigned to a control condition or to either of two experimental conditions: attention directing or reassurance after receiving stress inducing instructions. Results indicated that attentiondirecting (task-focusing) prompts not only reduced self-preoccupying intrusive thinking, but also improved performance on an anagram task. Participants in the attention-directing condition performed better than those in the reassurance condition, who in turn performed better than those in the control group. Sud and Katoch (1994) investigated the cognitions of test-anxious children at different times during an actual examination. In line with the findings of Zatz and Chassin (1983), results of this study indicated that all participants, including highly test-

Bodas and Ollendick anxious children, tended to report more task facilitating than task debilitating cognitions. However, and unexpectedly, the task facilitating cognitions were not associated with better task performance. They also found that the level of anxiety and frequency of off-task (task-debilitating) cognitions was much higher when children were in the middle of the test as compared to the beginning or end of the test. Pursuing a different line of research, a few studies have explored the role of psychosocial factors on test anxiety (Kochgaway, 1993; Singh & Broota, 1992). For instance, Singh and Broota (1992) reported that parental pressure was related to higher levels of test anxiety. They also reported that parents’ educational and occupational status was positively correlated with test anxiety in children. Kochagaway (1993) reported positive correlations between test anxiety and various areas of adjustment including home, health, social, and emotional adjustment. More recently, a study by Sovani, Thatte, and Nadkarni (2000) assessed perceived sources of examination anxiety among high school and college students. Students, studying in 8th, 9th, and 10th grades (the latter representing the year of the School Board Examination) were assessed using the FRIEDBEN test anxiety scale (Friedman & Bendas-Jacob, 1997), a 23-item paper-pencil test in English and Marathi that was piloted before use in India. “Fear of Social Derogation” was found to be a principal factor in the year when the Boards were given, and the year prior to that. “Tenseness” was a stronger factor in the younger group (Standard 8th). Interestingly, neither group scored very high on the factor of cognitive obstruction, although most poor exam performances were attributed to this factor. Lastly, some studies have compared the nature of test anxiety among Indian youth with that among ` youth of other cultures vis-a-vis gender and educational level differences (Sharma, Parnian, et al., 1983; Sharma & Sud, 1990; Sud, 1990, 1991). For instance, Sharma, Parnian et al. (1983) compared the test anxiety levels of 160 Iranian and 160 Indian school children and undergraduates. Overall, the results indicated that the Iranian youth had higher levels of total test anxiety and worry than the Indian youth. No main effect of culture was evidenced for emotionality, however. Additionally, as seen in western and other cross-cultural research (Hembree, 1988; Seipp, 1991; Seipp & Schwarzer, 1996), females reported higher test anxiety than males across both cultural groups. With regard to the interactional effects of age, gender, and educational level, Indian

Test Anxiety: A Cross-Cultural Perspective school going females reported more test anxiety than those in college; whereas, Indian college males had higher test anxiety than their school going counterparts. In contrast, both male and female Iranian school children reported higher test anxiety than the college students. Sud (1990) also examined the nature of test anxiety among Indian and Italian high school and college students using the Italian normative data reported by Comunian (1985). In general, the overall level of test anxiety for both the cultures was similar and women in both the cultures had higher levels of overall test anxiety and worry than males. Differences between the samples were evident at the college level. Thus, Indian college females endorsed higher levels of anxiety (total, worry, and emotionality scores) than their Italian counterparts, and Indian males reported higher levels of worry and emotionality, but not total test anxiety as compared to their Italian counterparts. With regard to cross-cultural comparisons involving American children, Sharma, Parnian et al. (1983) compared the TAI scores of Indian and Iranian children to the normative TAI scores of American children. The results indicated differences among Indian, Iranian, and American youth at the college level only. Thus, Iranian college females and Indian college males endorsed higher levels of test anxiety than American college students. Among these three countries, females reported higher levels of test anxiety than males. Somewhat inconsistent with these findings, in a similar study comparing student samples in India and the United States, Sud and Sharma (1990) reported no significant differences in the overall test anxiety levels for the Indian and American college students. Moreover, no significant differences in overall test anxiety were reported for males from both the countries. Indian school females reported low levels of overall test anxiety as compared to their American counterparts. Moreover, Indian females had significantly lower emotionality and higher worry scores as compared to American females. However among males, only Indian college students reported greater worry than their American counterparts. Lastly, as reported in the previous studies (Hembree, 1988; Seipp, 1991; Seipp & Schwarzer, 1996), females in both countries had greater overall test anxiety than their male counterparts. Summarizing the studies discussed above, some studies have explored intra-individual variables such as the cognitions of test-anxious students while oth-

75 ers have investigated the role of social and interpersonal variables such as parental education and expectations. However, more extensive investigation needs to be conducted to substantiate these results. Finally, based on the three cross-cultural studies of test anxiety, one conclusion that can be reliably drawn at this time is that as reported in the western literature (Hembree, 1988) female students in India endorse higher levels of test anxiety than Indian males. These cross-cultural studies report a number of spe` cific comparisons vis-a-vis gender, educational level, and the components (worry and emotionality) of test anxiety between the youth in India and other countries; however, no clear findings have been obtained and no further generalizations can be made from these studies at this point in time.

Assessment Research on assessment has mainly involved the simple translation of the available western instruments into Indian languages. Initially, Hindi and Punjabi versions of the Test Anxiety Scale for Children (TASC, Sarason et al., 1960) were developed (Nijhawan, 1972). These measures were developed by simple translation and modification of some test items to make them culturally appropriate. Subsequently, reliability and validity of this scale was determined using a sample of Indian school children. This was followed by the translation and adaptation of the Test Anxiety Inventory (TAI, Spielberger, 1980) into Hindi (TAI-H, Sharma, Sud, et al., 1983). Sharma et al. prepared an experimental version of the TAI H by translating the original TAI into Hindi and having bilingual experts evaluate the translated version in terms of its adequacy and equivalence. Alternative wordings were then created for these items and a total set of 59 items was presented to psychometricians who rated the Hindi translation. Finally, based on these ratings, a set of 20 Hindi items, equivalent to the original TAI was selected and administered along with the original TAI, to a sample of 151 bilingual students in the tenth grade. Based on the obtained correlations between these measures, the items of two versions were considered equivalent. The internal consistency coefficients of the TAI-H scale also approximated the ones obtained for the original TAI using American participants. Furthermore, the concurrent and construct validity of the TAI H was also reported using the Test Anxiety Scale (TAS; Sarasaon, 1958) and measures

76 of intelligence and academic achievement, respectively (Sharma, Parnian et al. (1983). Using a similar methodology, Sud and Singh (1991) developed the Hindi version of Sarason’s Reactions to Tests scale (RTT, 1984). As in the Sharma et al. study, the equivalence of these measures was demonstrated using means, standard deviations, and correlations. Likewise, the internal consistency coefficients were calculated for the entire scale as well as the subscales and were reported to be comparable to the original RTT (Sarason, 1984). Lastly, in addition to these measures of test anxiety, efforts have been made to develop other measures used in research on test anxiety including performance measures such as parallel forms of difficult anagram tasks in Hindi (Sud & Prabha, 1993); measures to assess cognitions such as the Children’s Cognitive Assessment Questionnaire (CCAQ; Sud & Katoch, 1994); Cognitive Interference Questionnaire (Prabha, 1984); and academic stress (Verma & Gupta, 1990). In sum, preliminary efforts have initiated the process of assessing test anxiety and related constructs in India. However, more refined research may be required to further these developments and advance the field of test anxiety in India.

Treatment Treatment outcome research has mainly involved examining the efficacy of a variety of cognitive behavioral interventions for test-anxious individuals in India, based on research conducted in the West. A series of studies conducted by Sud and colleagues have reported the efficacy of attentional skills training in the treatment of test anxiety. Attentional skills training is a short-term intervention designed to reduce the worry component of test anxiety. Based on Wine’s (1971) cognitive attentional theory of test anxiety, it involves providing instructions to direct the participant’s attention to the task and to reduce self-awareness. It also involves training participants to counter off-task thoughts, and to engage in rewarding self-statements for task-oriented behavior. In one study (Sud & Sharma, 1990), the efficacy of attentional skills training was compared with that of the Systematic Rational Restructuring approach (Golfried, Linehan, & Smith, 1978). The latter approach involved training participants to realistically reexamine imaginally a presented hierarchy of stressful test-taking situations. In contrast to the findings obtained by Goldfried, Linehan, and Smith

Bodas and Ollendick (1978), Sud and Sharma (1990) found that neither test anxiety reduction nor performance improvement was evidenced for persons receiving the Systematic Cognitive Restructuring. Attentional Skills training, on the other hand, was effective in reducing test anxiety but not in improving task performance. The Attentional Skills training approach was also compared to cognitive modeling in a subsequent study involving test anxious girls (Sud, 1994). The cognitive modeling approach, which involved a live model who demonstrated the use of a task-orienting, self-instructional approach during task performance, was found to be less effective in reducing test anxiety as compared to attentional skills training. Finally, the attentional skills training approach was found to be more successful in reducing test anxiety in school girls than relaxation training (Sud & Prabha, 1995). However, neither approach resulted in improved task performance (Sud & Prabha, 1996, 2000). Additionally, Sharma, Kumaraiah, and Mishra, (1996) have successfully used stress inoculation training (Meichenbaum, 1985) to reduce test anxiety in a sample of high school girls. Lastly, Sanghvi (1995), based on the deficits model of test anxiety (Culler & Holahan, 1980), provided support for the efficacy of study skills training not only in reducing test anxiety but also in improving academic performance in test-anxious high school boys and girls. Overall, a considerable number of studies appear to have addressed, to some extent, the issues of conceptualization, assessment, and treatment of test anxiety in India. However, as is true of most crosscultural studies on test anxiety, Indian research is significantly westernized. Thus, most of the studies have unfortunately adopted an imposed etic approach and can be classified as psychological differences studies or generalizability studies. An approach focusing on individual differences, though not entirely inappropriate in the context of an individualistic western society, may not be sufficient for understanding test anxiety or any other phenomenon in a collectivist society such as India.

STATE OF THE ART/SCIENCE What We Know or Don’t Know: A Critique of the Existing Literature Western research on test anxiety has flourished since the pioneering work of Sarason and Mandler in 1952. It has made important advances in all aspects of

Test Anxiety: A Cross-Cultural Perspective investigation including theory and conceptualization, assessment, and treatment. Specifically, from being mechanistically conceived as a unidimensional construct, the concept of test anxiety has evolved into a multifaceted construct comprising cognitive, physiological, and behavioral aspects. Furthermore, important strides have been made in terms of assessment of this complex construct. Thus, in response to evolving conceptualizations of test anxiety, more sophisticated and psychometrically advanced measures to assess the different aspects of test anxiety have been developed. Finally, a variety of cognitive behavioral interventions have been examined for their effectiveness in reducing test anxiety and improving task performance. With regard to cross-cultural research on test anxiety, it is interesting to note that the construct of test anxiety has been investigated in so many different countries and cultures. The impetus to this research came primarily from the translation and adaptation of the TAI into many different languages of the world. This tool provided norms and data that could be compared across cultures. However, it is also important to consider some of the shortcomings of this research.

Decontextualized and Atheoretical Approach A significant limitation of the research conducted in the field of test anxiety is that it lacks a context-based approach. Consistent with the individualistic perspective, much of the western research has focused on individual differences in test anxiety with respect to conceptualization, assessment, and treatment considerations. In other words, the contextual factors including interpersonal and ecological factors have not been measured and systematically analyzed to determine their impact on test anxiety. This focus on individual differences appears to be based on the assumption that test anxiety lies within the individual and that contextual variables are common to all children and contribute less significantly to the nature or expression of test anxiety. Much of the crosscultural research in general and Indian research in particular also lacks a context-based approach. Having its roots in western research, much of the crosscultural research in test anxiety is individualistic in its orientation. The individualistic approach in the western literatures is evident in our review of test anxiety. For example, in a metanalytic review by Hembree (1988), 562 studies on test anxiety were examined. Among

77 all of these studies, socioeconomic status was the only contextual variable that was examined and then only in 8 (1.43%) of the studies. Likewise, all of the treatment outcome studies (N = 192) focused on individual-oriented treatments. In the same way, a meta-analytic review by Seipp (1991) that included 126 studies examining the impact of general and test anxiety on academic performance, did not take into consideration any contextual variables. Lastly, in a review of test anxiety literature by King et al. (1991) that examined the nature and treatment of test anxiety in 87 studies; all but 4 studies investigated intraindividual variables. Furthermore, all the assessment measures and treatments considered in these studies were individual-oriented. This pattern of research, centering on intra-individual characteristics for conceptualization, assessment, and treatment is mirrored in the cross-cultural and Indian literature as well. In a review of cross-cultural studies on test anxiety by Seipp and Schwarzer (1996), all the reported studies from 13 countries other than the United States, were primarily concerned with scale adaptations, comparison of levels of anxiety between national groups, and factor analysis or scale validity issues. Cultural and contextual hypotheses related to the role of high stake examinations and values related to academic achievement were not examined in any study and were only proposed as post hoc explanations for obtained differences in the review. Lastly, with regard to Indian research on test anxiety, out of approximately 35 studies on test anxiety examined for this review, only 3 studies investigated the impact of contextual variables such as parenting and the nature of examinations. Another problem associated with cross-cultural research on test anxiety, but not with the western research, is a lack of a clear theoretical framework. As aptly noted by Zeidner (1998) cross-cultural research on test anxiety lacks specific hypotheses based on an understanding of a culture, its norms, values, socialization practices, and other variables relative to the construct of test anxiety. Furthermore, although the researchers have demonstrated some cross-cultural differences in the levels of test anxiety, they have not examined the causal mechanisms through which cultural variables systematically influence test anxiety (Zeidner, 1998). As is true for cross-cultural research on test anxiety in general, Indian research too lacks a theory-driven, context-based approach. Consequently, the research has not generated a cohesive body of knowledge on the nature of test anxiety in Indian children.

78 One approach to classifying cross-cultural studies has been provided by van de Vijver and Leung (2000). According to van de Vijver and Leung’s taxonomy of cross-cultural studies, two core dimensions may be used to classify research. The first dimension, which is important for any research, differentiates between studies that are exploratory (atheoretical) and those that involve hypothesis testing (theory based). The second dimension, which is particularly relevant to cross-cultural research, distinguishes between studies based on whether or not they emphasize the role of contextual variables such as family relations and socioeconomic variables. This taxonomy provides a useful framework for evaluating the cross-cultural studies in the field of test anxiety. For instance, a significant amount of research has focused on translating the TAI (Spielberger, 1980) and validating its factor structure on a sample from a different country to examine the universality of the construct of test anxiety. Based on van de Vijver and Leung’s (2000) taxonomy of crosscultural studies, these studies may be considered as “generalizability studies” where the research is based on a theoretical foundation but does not examine the contextual factors. Numerous studies have compared the concept of test anxiety in two or more different ` cultures vis-a-vis their scores on the respective versions of TAI (Spielberger, 1980). These studies have merely attempted to investigate the differences in the magnitude of test anxiety across cultures and may be labeled as “psychological differences studies” (Van de Vijver & Leung, 2000) Thus, the “generalizability studies” and “psychological differences studies” mentioned above both disregard the contextual dimension which is considered to be the crux of a cross-cultural approach. In this regard, Bronfenbrenner’s contextualist philosophy provides a broad theoretical framework for examining the sociocultural variables impacting behavior.

Methodological Issues A number of methodological concerns arise out of this largely western approach. Specifically, and as noted by Van de Vijver and Leung (2000), these studies have a “western bias” in their choice of a research topic, the methodology used to investigate the topic, the conceptualization of a construct under study, and “blind importation” of assessment tools. Such an approach has been criticized on the basis that it fails to capture the essence of the new culture that is being

Bodas and Ollendick studied in a cross-cultural paradigm. In other words, using a western framework for investigating a phenomenon will yield “western results” that may be inapplicable or uninterpretable for any one culture. Another methodological problem with the current cross-cultural research on test anxiety is related to the interpretation of the cross-cultural differences observed. By and large, most of the cross-cultural studies on test anxiety have reported mean differences in the levels of test anxiety between samples of participants from two or more countries. The difficulty of interpreting such large and simple crosscultural differences has been called the “interpretation paradox” of cross-cultural differences (van de Vijver & Leung, 2000). It is a paradox because, although these differences are apparently simple and straightforward, there is a problem of identifying the real cause. The two cultures are likely to differ in so many respects such as socialization practices, values, and the like, that many alternative explanations can be proposed to explain the differences in test anxiety. In other words, as the specific cultural variables have not been delineated, measured, and analyzed, it is unclear as to what cultural factors are implicated in these explaining these mean differences. Finally, most of these studies have assumed cross-cultural equivalence. Construct or conceptual equivalence refers to whether a construct has the same meaning in the two cultures; operational equivalence refers to whether the factor structure is equivalent for the two cultures; item equivalence refers to whether the construct is measured by identical or equivalent items of an instrument; and scalar equivalence involves similarity between units of measurement (Berry, Poortinga, Segall, & Dasen, 2002; Hui & Triandis 1985). The problem of equivalence arises when instruments are translated into another language. Some of the translated measures, including the Indian versions have indeed attempted to establish operational, item, and scalar equivalence. However, construct equivalence has been presupposed and not examined in most cases. The concept of construct validity is also pertinent to the issue of translation and adaptation of instruments for use in a different culture. To what extent do the various test anxiety instruments indeed measure test anxiety among Indian children? Anastasi (1976) has recommended a variety of techniques to examine the construct validity of an instrument such as computing correlations with other tests, internal consistency coefficients, and conducting factor analysis. A multitrait multimethod matrix (Campbell

Test Anxiety: A Cross-Cultural Perspective & Fiske, 1959) may also provide a useful method for evaluating the construct validity of the test anxiety measures (Anastasi, 1976). Thus, consistent with the contextualist approach of Bronfenbrenner, discussed earlier, it is essential that the current research findings as well as future endeavors be understood in the light of two additional factors important for cross-cultural work: namely, interpersonal and ecological variables. These contextual variables are important in creating a theoretical framework for developing culture based hypotheses regarding test anxiety. Understanding Specific Cultural/Contextual Factors Interpersonal Relationships The problem of test anxiety is much more significant in Indian and other Asian children as compared to non-Asian children (Chambers, Yeragani, & Keshavan, 1986). In particular, feelings of shame and disgrace to oneself and one’s parents and family may be associated with examination failure (Vishwanathan, Shah, & Ahad, 1997; Wang & Ollendick, 2001). Achievement motivation in the Indian society goes beyond individual achievement and includes loyalty to the in-group as well. For instance, in a study by Singhal and Misra (1989; 1994), the expectations of significant others such as parents, teachers, and peers were reported to be the most important determinants of achievement goals among Indian youth. Kagitcibasi (1996) aptly labels this tendency, which is also seen among youth from other collectivistic cultures such as Japan (DeVos, 1968), as “social achievement motivation.” Consistent with the collectivist perspective, the traditional Indian family values interdependence, hierarchical relationships, and reverence toward elders (Verma & Saraswathi, 2002). Also, a high level of parental involvement and control, particularly in middle class families, is typical. Like other Asian cultures, Indian parents are active participants in their child’s education. They emphasize academic achievement, and often enroll their children in after-school educational activities such as “tuitions” to advance their academic achievements even though their children are not “struggling” in school (Cho, 1995; Verma, Sharma, & Larson, 2002). They also adjust their leisure and social activities to supervise their children’s homework (cited in Verma et al., 2002). Finally, in Asian cultures, higher education tends to be associated with higher income, social sta-

79 tus, and better career prospects (Cho, 1995; Dhesi, 2001). American families, in contrast, encourage independence and tend to be less involved with their children’s school work especially as they progress to higher grades (Steinberg, 1996).

Schooling The Asian education system has been criticized from a western perspective for a number of reasons such as large class sizes and a competitive and stressful examination system requiring excessive rote memorization (Cho, 1995). A number of these concerns are now being examined and openly expressed even within these cultures. Education System in India. A National Advisory Committee was constituted in March 1992 by the Ministry of Human Resource Development under the Chairmanship of Prof. Yash Pal, former Chairman of the University Grants Commission (UGC), to suggest ways to reduce the academic burden on school students. The Committee stated that the learning in classrooms in India was “joyless” and submitted its recommendations report “Learning without burden” to the Ministry in July 1993. More recently, a National Curriculum Framework for School Education was developed by the National Council of Educational Research and Training (2000) which proposes to eliminate the “pass-fail” system followed by the school boards, and plans for structural changes in the evaluation practices. However, India still continues to struggle with these issues (Verma et al., 2002). Nature of Examinations. In 1959, the University Education Commission headed by Dr. Sarvapalli Radhakrishnan commented: “If we have to suggest one single reform in university education, it would be that of examination.” The current education system that stresses the acquisition of knowledge at all costs to the students in India, initially introduced during the British rule, is considered lacking by many. The traditional education system involving learning of life skills in addition to knowledge acquisition has been considered as a better way of educating children (Kumar, 1991). The current education system involves primary, middle, and secondary school levels which span over a period of 10 years and 2 years of higher secondary education (Association of Indian Universities, 1997). In the current system, children are required to take about three to four exams per academic year before they are promoted

80 to the next grade. At the end of the 10- and 12-year period, children typically appear for a school board exam. These exams are conducted either statewide or nationwide, depending on the board that the school is affiliated with, are centrally prepared, graded, and administered. According to the report of the Ministry of Human Resource Development (1993), the education system in general and the examination system in particular have contributed to “joyless learning.” Particularly, the public education system at the end of the 10th and 12th grades has had a dramatic impact on the lives of children and their parents. Performance in these exams is crucial for seeking admission for higher education at universities and colleges and to gain access to prestigious careers. Such practices put considerable pressure on the students. This is evident in the kind of preparation that is undertaken for these examinations such as attending “tuition” or coaching classes, using study guides and the like, to the extent that classroom teaching has become a “chore.” Moreover, much of this preparation begins much before the children enter these grades and during vacation as well (National Advisory Committee, 1993). School Stress. In a recent study by Verma et al. (2002), schoolwork was correlated with level of stress in Indian adolescents. Specifically, adolescents in India were found to spend about one-third of their waking time in school-related activities. Moreover, they found that schoolwork generated negative subjective states, most frequently experienced during homework. In particular, those who spent more time doing homework experienced more internalizing problems. The “do or die” competitive educational system was thought to induce high levels of stress in students as well as parents. In contrast, adolescents in the United States report spending nearly half of their waking time in leisure activities (which exclude work and maintenance activities), although some variability was noted in relation to age and other factors (Larson & Seepersad, 2003). Different Pathways and Different Expression: Some Hypotheses In examination of these contextual factors, it is apparent that the causes of test anxiety among Indian children may be very different as compared to the causes of test anxiety in western children. Having identified some of the contextual factors that may be related to the construct of test anxiety, what also ensues is that the conceptualization of test anxiety

Bodas and Ollendick in India may need to be different from that in the western countries. Particularly, given the emphasis on academic achievement, the time invested in academic activities, parental and social pressure to perform, the stressful nature of examinations, and the high stakes on performance in examinations, the preoccupation with and relatively intense preparation for exams on the part of Indian children may not be surprising and may be perceived as “a way of life.” It is reasonable to imagine that “normal” examrelated behavior of Indian youth may be construed as “test anxious behavior” which is similar to that observed from the western perspective. Furthermore, it is acknowledged that such an intense focus on academic achievement may even be eventually adaptive; especially if it leads to higher income, status, and better careers (Cho, 1995; Dhesi, 2001). Still, what is known at this time is that academic stress tends to be associated with negative emotional states (Verma et al., 2002). Whether such a stressful environment has long-term negative consequences has not been investigated. On the other hand, despite a stressful “way of life” academically, individual differences within such a population will undoubtedly be seen and hence cannot be ignored. In sum, intraindividual, inter-individual, and ecological variables are all implicated in the expression of test anxiety in Indian children. Another issue that is relevant in this context is related to the differential manifestation of test anxiety in Indian children, not unlike other psychological syndromes. A number of studies investigating the nature of anxiety and depression in Asian cultures in general, and in India in particular, have reported that Indian people tend to express psychological conflicts through somatic symptoms. With regard to depression, at least five studies including two crosscultural studies using adults from India indicate that depressed individuals tend to present with somatic problems more commonly (Raguram, Weiss, Keval, & Channabasavanna, 2001; Sethi, Nathawat, & Gupta, 1973; Srinivasa, Subramaniam, & Chatterji, 1990) than either American (Derasari & Shah, 1988) or British participants (Teja, Narang, & Aggarwal, 1971). Such studies involving samples of children from India are scant. However, a recent study by Verma and Gupta (1990) reported that primary symptoms of academic stress in school children included headaches, stomachaches, fever, and tenseness. Specifically with regard to test anxiety, one study (Sovani et al., 2000) reported that children endorsed physiological symptoms more

Test Anxiety: A Cross-Cultural Perspective than cognitive symptoms. In general, the studies reviewed above suggest that contrary to the trend observed in the western literature, it is likely that somatization or the emotionality factor of test anxiety may be more significant than the worry component in Indian children.

A Culturally Sensitive Approach to Test Anxiety in India A culturally sensitive approach needs to consider the role of individual, interpersonal, and ecological factors in conceptualization, assessment, and treatment of test anxiety. It is apparent that the individual as well as contextual factors vary in different cultures. Indeed, for a better understanding of the ` culture-behaviors relationship vis-a-vis test anxiety, it is pertinent to take these factors into account in developing an approach to test anxiety. (See Fig. 1.)

81 ilar cultural milieu, what distinguishes children who become test-anxious from those who don’t? A research agenda, adopting a derived-etic approach to address these questions, needs to take into consideration a variety of issues including translation and adaptation of available instruments for measurement of individual factors, developing or adapting assessment tools for measuring contextual variables, and theory testing. Based on the literature reviewed above these issues fall in two main categories that need to be considered when planning future crosscultural research in test anxiety: namely, theory development and incorporation of a contextual approach. These two issues are consistent with van de Vijver and Leung’s (2000) dimensions of classifying cross-cultural studies based on whether or not they involve hypothesis testing (exploratory versus theory-driven studies) and whether or not they measure contextual variables.

Theory Development A RESEARCH AGENDA Based on the literature reviewed above, a number of questions regarding the nature of test anxiety can be raised. For example, how should test anxiety be conceptualized in the Indian culture? Is the test-related behavior of Indian children “anxious” or “adaptive,” or both? How can test anxiety be assessed in a context-sensitive manner? Is the pattern of anxiety (with respect to worry-emotionality factors) the same in Indian children? How do they cope with the educational demands placed on them? Finally, given that all Indian children experience a sim-

As noted earlier, cross-cultural research in test anxiety has been primarily atheoretical. Specifically, much of the research has imported western theory and applied it to a cross-cultural setting. Thus, using the integrative or derived etic approach, it is important to explore the applicability of the existing theories in the Indian or any other cross-cultural setting as well as conduct exploratory research to determine the culture-specific aspects of test anxiety. An important issue related to examining the appropriateness of existing theories to a different culture involves translation and adaptation of existing

Fig. 1. A culturally sensitive approach to test anxiety in India.

82 measures of test anxiety. An international committee of professionals in the field of psychology, affiliated to various international organizations involved in instrument translations, was formed in 1993 by the International Test Commission (Van de Vijver & Hambelton, 1996). The committee developed a set of guidelines describing recommended approaches to test translations. Four primary areas covered by the guidelines included: context, development, test administration, and interpretation (see Van de Vijver & Hambleton, 1996 for review). These guidelines are comprehensive and provide a framework for undertaking test translation. Likewise, Van Widenfelt, Treffers, de Beurs, Siebelink, and Koudijs (in press) have provided an extensive review of issues pertaining to translation and adaptation of measures for conducting research in different cultures. The authors have highlighted common errors made in assuming the equivalence of translated measures as well as in assuming the adequacy of the simple translation-back translation process in developing instruments for research in different cultures. They suggest that a culturally sensitive approach be adopted in this process. Specifically, van Widenfelt et al. (in press) propose that certain test items may need to be modified to be culturally appropriate. Likewise, some concepts in clinical psychology may be more or less pathological depending upon their relevance in the cultural context and hence need to be examined for cultural significance. Finally, these authors recommend a complex translation process involving a combination of a variety of techniques. Examples include using a team approach, involving the original author, involving a native bilingual translator, and piloting or pre-testing the measure. Thus, careful attention needs to be given to the process of translating and adapting available test anxiety instruments into different languages to avoid the drawbacks of the existing adaptations. Keeping in mind the derived-etic approach, it is important to explore the concept within the new culture using the “emic” approach. Thus, it may be important to ask more open-ended questions regarding the concept and issues relevant to it. An example of such an approach in the Indian setting is seen in the study by Verma and Gupta (1990). They developed an interview schedule checklist for exploring various aspects of academic stress based on a combination of open-ended interview questions and 3-point rating scale checklist. Development of such tools for measuring different constructs related to test anxiety such as cognitions and study habits will provide

Bodas and Ollendick rich qualitative information regarding the nature of test anxiety in the Indian setting. Thus, existing measures of test anxiety such as TAI-H (Sharma, Parnian et al. (1983)) that have been developed primarily using simple translation-back translation methods should be reexamined for cultural appropriateness and construct and operational equivalence. It is important that these measures incorporate some open-ended questions, at least in the pilot trials, to explore aspects of test anxiety specific to the Indian culture. New items generated from this procedure can then be pooled with the old one and the combined set can then be factor analyzed to examine the factor structure. A somewhat similar procedure was used by Friedman and Bendas-Jacob (1997) when they developed the FRIEDBEN test anxiety scale (FTA). The authors presented high school students with four open-ended questions and factoranalyzed the pool of items generated from their responses. Indeed, they discovered another factor from this procedure, which was called “social derogation.” Since the authors were developing rather than adapting a test, they did not include items of an already existing instrument. However, for adapting an instrument and exploring for culture specific aspects at the same time, pooling of items of an existing measure with those generated from openended question may be a suitable “derived-etic” approach. This approach needs to be extended to other related measures including those examining the cognitions and study habits of test-anxious children. Emotional aspects of test anxiety measured using physiological variables such as heart rate measures in evaluative situations (Beidel, 1988) is another avenue for future research. Such measures combined with the use of adapted instruments will result in a step forward in the direction of a derived etic understanding of test anxiety.

Incorporating a Contextual Approach A contextual approach has been advocated throughout this paper. An earlier section made a case for incorporating interpersonal and ecological level variables in research along with the traditional individual level analysis. An approach for including contextual variables into the research design and analyzing their effects has been proposed by Poortinga and Van de Vijver (1987). According to these researchers, the contextual variables in a cross-cultural

Test Anxiety: A Cross-Cultural Perspective study should ideally account for all the differences between two cultures on a dependent variable such as test anxiety. They describe a methodologicalstatistical framework for examining these context variables. When two cultures differ significantly on the outcome variable, specific contextual variables that have been operationalized and selected apriori are systematically introduced into the analysis to determine how much of this initial variance can be explained by them. Examples of specific contextual variables include psychological as well as ecological variables such as parenting styles and socioeconomic status. According to Poortinga and Van de Vijver (1987), in an ideal cross-cultural study, the variance due to culture is increasingly explained by bringing in more specific contextual variables, such that eventually there is no variance remaining to be explained. This process thus involves “unwrapping of the packaged variable culture.” Poortinga and van de Vijver (1987) further emphasize that the choice of contextual variables should be theory-driven and should be done prior to data collection. Indeed, Berry et al. (2002) recommend that researchers collect adequate amount of information on different contextual variables to be able to fully account for cultural differences in the construct of interest. Contextual variables can be incorporated into the research design in a variety of ways. Based on the literature reviewed above, contextual variables particularly important for examining test anxiety in India appear to be related to family and social relationships and those related to the education system. Thus, variables such as parental expectations, parental education, social pressures, perceived level of stress due to examination system, as well as achievement-related values held by school, family, and the individual need to be examined. Instruments for measuring these variables need to be adapted or developed. Other culture-specific contextual variables that may be discovered during emic exploration of the construct can also be incorporated into the research design. One approach for obtaining data involves having the participants report their thoughts and feelings on the different contextual factors such as family and peer relationships, examination, and school stress. A good example of this approach is the interviewchecklist developed by Verma and Gupta (1990) that examined aspects of school stress among test-anxious children. On the other hand, information about these factors can be obtained from other sources such as teachers and parents. Data obtained from multiple

83 sources is likely to be more comprehensive and informative as compared to that obtained through selfreports only (Achenbach, 1998; Ollendick & Hersen, 1993). Conceptually, using the statistical-methodological approach of Poortinga and van de Vijver (1987) appears to be promising because the primary purpose of this analysis is to break up a broad contextual variable such as culture into more specific components that can then be subjected to analyses to determine their effects on the outcome variable. This helps us overcome one of the common problems of cross-cultural research; the interpretation paradox (van de Vijver and Leung, 2000). A significant shortcoming of cross-cultural studies has indeed been cultural explanations of results based on insufficient information of contextual variables (Berry et al., 2002). Incorporating such an approach to conduct ` cross-cultural research on test anxiety, vis-a-vis Indian and American cultures would first involve examining cross-cultural differences in the testanxiety scores. If these two cultures differ significantly on the test-anxiety measures, then the contextual variables such as parental expectations and school stress can be systematically introduced into the analyses to determine their role in explaining these cross-cultural differences. In sum, such a theory-driven contextualist approach thus produces more meaningful and wellfounded results by overcoming some of the major limitations that have questioned the validity of much of cross-cultural research in test anxiety.

SUMMARY AND CONCLUSIONS School and exams are an inevitable aspect of most children’s lives in our modern world. Academic stress and test anxiety are thus ubiquitous problems in today’s world. Because of the detrimental effects of test anxiety such as poor academic performance and achievement, this construct will continue to be researched by psychologists, educationists, and other professionals who work with children and youth. The primary objective of this review was to pursue a cross-cultural analysis of the construct of test anxiety. An attempt has been made to understand the concept of test anxiety within a broad framework of a contextualist approach, specifically, that provided by Bronfenbrenner’s ecological system’s approach. A contextualist approach is fundamental to

84 cross-cultural psychology because of the assumption that cultural contexts shape the behaviors of individuals. The culture behavior interaction can be studied from within (emic) or across-culture (etic) perspectives. However, the derived-etic approach that appropriately combines features of both the emic and etic viewpoints is considered a better alternative. This paper has adopted a derived-etic position in reviewing and evaluating cross-cultural research on test anxiety. Cross-cultural studies have often used broad categories as names of countries or dimensions such as individualism-collectivism to examine cultural hypotheses. However, such comparisons pose the problem of multiple alternative explanations for observed differences. Essentially, use of such vague categories yields uninterpretable results because it is difficult to pinpoint the exact cause. Thus splitting up the broad contextual variable into more specific contextual categories such as intra-individual factors, interpersonal factors, and ecological variables provides a better framework for cross-cultural comparison between two nations. The concept of test anxiety has been extensively investigated in the western countries, especially the United States. Important strides have been made in theory construction, assessment, and treatment of test anxiety. However, western research has given insufficient attention to contextual variables such as socialization practices and value systems and their impact on the development and maintenance of test anxiety in children. Cross-cultural as well as Indian research has been significantly influenced by western research. In fact, the methodology and instruments used in the latter are significantly westernized. In general, research has not attempted to explore culture-specific aspects of test anxiety and thus culture-based hypotheses related to test anxiety have not been proposed. Overall, cross-cultural and Indian research has been primarily atheoretical and decontextualized in nature. Contextual factors falling in the two broad categories of interpersonal and ecological variables, especially relevant to test anxiety are family relationships and school-related variables. Given the interdependent and hierarchical nature of family relationships in India, academic achievement goals are socially oriented. Moreover, in developing countries like India, higher education is often associated with better careers and higher socioeconomic status. Academic stress experienced by Indian children is often high because of parental expectations as well as competitive, high-stake examinations.

Bodas and Ollendick In the context of existing psychosociocultural conditions and in the light of existing cross-cultural research it is hypothesized that Indian children are more likely to express test anxiety as physiological symptoms. It is not known whether this intense concentration on academic goals is adaptive or harmful in the long run. A research agenda adopting a derived etic approach to address some of the questions and concerns that have been raised in this review has been proposed. It is hoped that such a research approach designed to examine the construct of test anxiety more holistically will advance our understanding of the construct and related concepts and help develop better interventions to alleviate the stresses associated with evaluative situations. REFERENCES Achenbach, T. M. (1998). Diagnosis, assessment, taxonomy, and case formulations. In T. H. Ollendick and M. Hersen (Eds), Handbook of child psychopathology (3rd ed., pp. 63–87), New York, NY: Plenum Press. Ahlawat, K. (1989). Psychometric properties of the Yarmouk Test Anxiety Inventory (Y-TAI). In R. Schwarzer, H. M. van der Ploeg, & C. D. Spielberger (Eds.), Advances in test anxiety research (Vol. 6, pp. 263–278). Lisse, The Netherlands: Swets & Zeitlinger. Alpert, R., & Haber, R. (1960). Anxiety in academic achievement situations. Journal of Abnormal and Social Psychology, 61, 207–215. Anastasi, A. (1976). Psychological testing (4th ed.). New York: Macmillan. Araki, N. (1992). Test Anxiety in elementary school and junior high school students in Japan. Anxiety, Stress, and Coping, 5, 205–215. Araki, N., Iwawaki, S., & Spielberger, C. D. (1992). Construction and validation of a Japanese adaptation of the Test Anxiety Inventory. Anxiety, Stress, and Coping, 5, 217–224. Association of Indian Universities. (1997). Data for academic year: 1998–1999. New Delhi, India: Author. Aysan, F., Thompson, D., & Hamarat, E. (2001). Test anxiety, coping strategies, and perceived health in a group of high school students: A Turkish sample. Journal of Genetic Psychology, 162, 402–411. Bauermeister, J. J., Collazo, J. A., & Spielberger, C. D. (1983). The construction and validation of the Spanish Form of the Test Anxiety Inventory: Inventario de Auto-Evaluacion Sobre Examenes (IDASE). Stress and Anxiety, 2, 67–85. Bedell, J. R. (1976). Systematic desensitization, relaxation-training and suggestion in the treatment of test anxiety. Behaviour Research and Therapy, 14, 309–311. Beidel, D. C. (1988). Psychophysiological assessment of anxious emotional states in children. Journal of Abnormal Psychology, 97, 80–82. Beidel, D. C., & Turner, S. M. (1988). Comorbidity of test anxiety and other anxiety disorders in children. Journal of Abnormal Child Psychology, 16, 275–287. Benjamin, M., McKeachie, W. J., Lin, Y. G., & Holinger, D. P. (1981). Test anxiety: Deficits in information processing. Journal of Educational Psychology, 73, 816–824. Berry, J. W. (1969). On cross-cultural comparability. International Journal of Psychology, 4, 119–128.

Test Anxiety: A Cross-Cultural Perspective Berry, J. W. (1976). Sex differences in behaviour and cultural complexity. Indian Journal of Psychology, 51, 89–97. Berry, J. W. (1980). Ecological analyses for cross-cultural psychology. In N. Warren (Ed.), Studies in cross-cultural psychology (Vol. 2, pp. 157–189). London: Academic. Berry, J. W. (1989). Imposed etics-emics-derived etics: The operationalization of a compelling idea. International Journal of Psychology, 24, 721–735. Berry, J. W. (1999). Emics and etics: A symbiotic conception. Culture and Psychology, 5, 165–171. Berry, J. W. (2001). A psychology of immigration. Journal of Social Issues, 57, 615–631. Berry, J. W., Poortinga, Y. H., Segall, M. H., & Dasen, P. R. (1992). Cross-cultural psychology: Research and applications. Cambridge, UK: Cambridge University Press. Berry, J. W., Poortinga, Y. H., Segall, M. H., & Dasen, P. R. (2002). Cross-cultural psychology: Research and applications (2nd ed.). New York: Cambridge University Press. Bronfenbrenner, U. (1993). The ecology of cognitive development: Research models and fugitive findings. In R. H. Wozniak & K. W. Fischer (Eds.), Development in context: Acting and thinking in specific environments. The Jean Piaget symposium series (pp. 3–44). Bronfenbrenner, U., & Morris, P. A. (1998). The ecology of developmental processes. In W. Damon (Series Ed.) & L. M. Lerner (Vol. Ed.), Handbook of child psychology: Vol. 1. Theoretical models of human development (5th ed.). New York: Wiley. Bruch, M. A. (1978). Type of cognitive modeling, imitation of modeled tactics, and modification of test anxiety. Cognitive Therapy and Research, 2, 147–164. Bruner, J. (1993). Do we “acquire” culture or vice versa? Behavioral and Brain Sciences, 16, 515–516. Campbell, D. T., & Fiske, D. W. (1959). Convergent and discriminant validation by the multitrait-multimethod matrix, Psychological Bulletin, 56, 81–105. Cassady, J. C., & Johnson, R. E. (2002). Cognitive test anxiety and academic performance. Contemporary Educational Psychology, 27, 270–295. Chambers, J., Yeragani, V. K., & Keshavan, M. S. (1986). Phobias in India and the United Kingdom: A trans-cultural study. Acta Psychiatrica Scandinavica, 74, 388–391. Cho, H. (1995). Children in the examination war in South Korea: A cultural analysis. In S. Stephens (Ed.), Children and the politics of culture (pp. 141–168). Princeton, NJ: Princeton University Press. Comunian, A. L. (1985). The development and validation of the Italian form of Test Anxiety Inventory. In H. M. van der Ploeg, R. Schwarzer, & C. D. Spielberger (Eds.), Advances in test anxiety research (Vol. 4, pp 215–220). Lisse, The Netherlands: Swets & Zeitlinger. Comunian, A. L. (1989). Some characteristics of relations among depression, anxiety, and self-efficacy. Perceptual and Motor Skills, 69, 755–764. Cornish, R. D., & Dilley, J. S. (1973). Comparison of three methods of reducing test anxiety: Systematic desensitization, implosive therapy, and study counseling. Journal of Counseling Psychology, 20, 499–503. Culler, R. E., & Hollahan, C. J. (1980). Test anxiety and academic performance: The effects of study-related behaviors. Journal of Educational Psychology, 72, 16–20. Davidson, A. R., Jaccard, J. J., Triandis, H. C., Morales, M. L., & Diaz-Guerrero, R. (1976). Cross-cultural model testing: Toward a solution of the etic-emic dilemma. International Journal of Psychology, 11, 1–13. Derasari, S., & Shah, V. D. (1988). Comparison of symptomatology of depression between India and U.S.A. Indian Journal of Psychiatry, 30, 129–134.

85 De Vos, G. A. (1968). Suicide in cross-cultural perspective. In H. L. P. Resnik (Ed.), Suicidal Behaviors: Diagnosis and Management (pp. 125–146), Boston: Little Brown. Dhesi, A. S. (2001). Expectations and post-school choice: Some data from India. Education and Training, 43, 14–24. El-Safty, M. M. (1995). Test anxiety and its relation to achievement motivation: A cross-cultural study between high school students in Egypt and Arab Emirates. Derasat Nafseyah, 5, 71–106. Ergene, T. (2003). Effective interventions on test anxiety reduction: A meta-analysis. School Psychology International, 24, 313–328. Friedman, I., & Bendas-Jacob, O. (1997). Measuring perceived test anxiety in adolescents. Educational and Psychological Measurement, 57, 1035–1046. Galassi, J. P., Frierson, H. T., & Sharer, R. (1981). Behavior of high, moderate, and low test anxious students during an actual test situation. Journal of Consulting and Clinical Psychology, 49, 51–62. Georgas, J., & Berry, J. W. (1995). An ecocultural taxonomy for cross-cultural psychology. Cross-Cultural Research: The Journal of Comparative Social Science, 29, 121–157. Georgas, J., van de Vijver, F. J. R., & Berry, J. W. (2004). The ecocultural framework, ecosocial indices, and psychological variables in cross-cultural research. Journal of Cross-cultural Psychology, 35, 74–96. Goldfried, M. R., Linehan, M. M., & Smith, J. L. (1978). Reduction of test anxiety through cognitive restructuring. Journal of Consulting and Clinical Psychology, 46, 32–39. Gonzalez, H. P. (1995). Systematic desensitization, study skills counseling, and anxiety-coping training in the treatment of test anxiety. In C. D. Spielberger and P. R. Vagg (Eds.), Test anxiety: Theory, assessment, and treatment (pp. 117–132). Washington, DC: Taylor and Francis. Greenfield, P. M. (2000). Three approaches to the psychology of culture: Where do they come from? Where can they go? Asian Journal of Social Psychology, 3, 223–240. Hagtvet, K. A. (1984). A Norwegian adaptation of the Test Anxiety Inventory: A first tryout. International Review of Applied Psychology, 33, 257–265. Hembree, R. (1988). Correlates, causes, effects, and treatment of test anxiety. Review of Educational Research, 58, 47–77. Hocevar, D., & El Zahhar, N. E. (1985). Test Anxiety in the USA and Egypt: A paradigm for investigating psychometric characteristics across cultures. In H. M. van der Ploeg, R. Schwarzer, & C. D. Spielberger (Eds.), Advances in test anxiety research (Vol. 4, pp. 203–214). Lisse, The Netherlands: Swets & Zeitlinger. Hodapp, V., & Benson, J. (1997). The multidimensionality of test anxiety: A test of different models. Anxiety, Stress, and Coping, 10, 219–244. Hodges, K., McKnew, D., Cytryn, L., Stern, L., & Kline, J. (1982). The Child Assessment Schedule (CAS) diagnostic interview: A report on reliability and validity. Journal of American Academy of Child Psychiatry, 26, 654–661. Hong, E. (1999). Test anxiety, perceived test difficulty, and test performance: Temporal Patterns of their effects. Learning and Individual Differences, 11, 431–447. Hui, C. H., & Triandis, H. C. (1985). The instability of response sets. Public Opinion Quarterly, 49, 253–260. Hwang, K. R. (1997). Comparison of three training methods of reducing test anxiety: Behavioral method, cognitive method, combination of cognitive and behavioral methods. Korean Journal of Counseling and Psychotherapy, 9, 57– 80. Jahoda, G. (1977). “In pursuit of the emic-etic distinction: Can we ever capture it?” In Y. H. Poortinga (Ed.), Basic problems in cross-cultural psychology. Lisse, The Netherlands: Swets and Zeitlinger.

86 Jahoda, G. (1992). Foreword. In J. W. Berry, Y. H. Poortinga, M. H. Segall, & P. R. Dasen (Eds.), Cross-cultural psychology: Research and applications (pp. x–xii). Cambridge, UK: Cambridge University Press. Kagitcibasi, C. (1996). Family and human development across cultures: A view from the other side. Hillsdale, NJ, England: Lawrence Erlbaum Associates, Inc. King, N. J., Mietz, A., Tinney, L., & Ollendick, T. H. (1995). Psychopathology and cognition in adolescents experiencing severe test anxiety. Journal of Clinical Child Psychology, 24, 49–54. King, N. J., Ollendick, T. H., & Gullone, E. (1991). Test anxiety in children and adolescents. Australian Psychologist, 26, 25– 32. King, N. J., Ollendick, T. H., & Prins, P. J. M. (2000). Test-anxious children and adolescents: Psychopathology, cognition, and psychophysiological reactivity. Behaviour Change, 17, 134– 142. Kochagaway, V. (1993). A study of relationship between academic anxiety and adjustment among high school students. Indian Journal of Behavior, 17, 16–21. Kovacs, M. (1985). The Interview Schedule for Children (ISC). Psychopharmacology Bulletin, 21, 991–994. Kreitler, S., & Kreitler, H. (1987). Modifying anxiety by cognitive means. In R. Schwarzer, H. M. Van der Ploeg, & C. D. Spielberger (Eds.), Advances in test anxiety Research (Vol. 5, pp. 195–206). Lisse, The Netherlands: Swets and Zeitlinger. Kumar, K. (1991). Political agenda of education: A study of colonialist and nationalist ideas. New Delhi, India: Sage. Lang, P. J. (1968). Fear reduction and fear behavior: Problems in treating a construct. In J. M. Shlien (Ed.), Research in psychotherapy. Washington, DC: American Psychological Association. Larson, R., & Seepersad, S. (2003). Adolescents’ leisure time in the United States: Partying, sports, and the American experiment. New Directions for Child and Adolescent Development, 99, 53–64. Liebert, R. M., & Morris, L. W. (1967). Cognitive and emotional components of test anxiety: A distinction and some initial data. Psychological Reports, 20, 975–978. Lonner, W. J., & Adamopoulos, J. (1997). Culture as antecedent to behavior. In J. W. Berry, Y. H. Poortinga, & J. Pandey (Eds.), Handbook of Cross-Cultural Psychology: Theory and Method (Vol. 1, pp. 43–83). Boston: Allyn & Bacon. Man, F., Budejovice, C., & Hosek, V. (1989). The development and validation of the Czech form of the Test Anxiety Inventory. In R. Schwarzer, H. M. van der Ploeg, & C. D. Spielberger (Eds.), Advances in Test Anxiety Research (Vol. 6, pp 233–243). Lisse, The Netherlands: Swets & Zeitlinger. Mandler, G., & Sarason, S. B. (1952). A study of anxiety and learning. Journal of Abnormal and Social Psychology, 47, 166–173. Markus, H. R., & Kitayama, S. (1991). Culture and the self: Implications for cognition, emotion, and motivation. Psychological Review, 98, 224–253. McIlroy, D., & Bunting, B. (2002). Personality, behavior, academic achievement: Principles for educators to inculcate and students to model. Contemporary Educational Psychology, 27, 326–337. Meichenbaum, D. (1985). Stress inoculation training. New York: Pergamon. Miller, P. H. (2000). Theories of developmental psychology (4th ed.). New York: Worth. Ministry of Human Resource Development. (1993). National Advisory Committee Report. New Delhi, India: Author. Morris, L. W., Davis, M. A., & Hutchings, C. H., (1981). Cognitive and emotional components of anxiety: Literature review and a revised worry-emotionality scale. Journal of Educational Psychology, 73, 541–555.

Bodas and Ollendick Morris, L. W., & Liebert, R. M. (1970). Relationship of cognitive and emotional component of test anxiety to physiological arousal and academic performance. Journal of Consulting and Clinical Psychology, 35, 332–337. Morris, L. W., Smith, L. R., Andrews, E. S., & Morris, N. C. (1975). The relationship of emotionality and worry components of anxiety to motor skills performance. Journal of Motor Behavior, 7, 121–130. Mwamwenda, T. S. (1994). Test anxiety and academic achievement among South African university students. Psychological Reports, 75, 1593–1594. National Council for Educational Research and Training. (2000). National Curriculum Framework for School Education, New Delhi, India. Naveh-Benjamin, M. (1991). A comparison of training programs for intended different types of test anxious students: Further support for information processing model. Journal of Educational Psychology, 83, 134–139. Naveh-Benjamin, M., McKeachie, W. J., & Lin, Y. G. (1987). Two types of test anxious students: Support for an information processing model. Journal of Educational Psychology, 79, 131– 136. Nijhawan, H. K. (1972). Anxiety in school children. New Delhi, India: Wiley-Eastern. Oener, N., & Kaymak, D. A. (1987). The transliteral equivalence and the reliability of the Turkish TAI. In R. Schwarzer, H. M. van der Ploeg, & C. D. Spielberger (Eds.), Advances in test anxiety research (Vol. 5, pp. 227–239). Lisse, The Netherlands: Swets & Zeitlinger. Ollendick, T. H., & Hersen, M. (1993). (Eds.). Handbook of child and adolescent assessment. Boston: Allyn & Bacon. O’Neil, H. F., & Fukumura, T. (1992). Relationship of worry and emotionality to test performance in a Juku environment. Stress and Coping, 5, 241–251. Padua, U. (1993). Anxiety, cognitive interference, and school performance of Italian Children. Psychological Reports, 73, 747– 754. Pang, V. O. (1991). The relationship of test anxiety and math achievement to parental values in Asian-American and European-American middle school students. Journal of Research and Development in Education, 24, 1– 10. Peleg-Popko, O., & Klingman, A. (2002) Family environment, discrepancies between perceived actual and desirable environment, and children’s test and trait anxiety. British Journal of Guidance and Counselling, 30, 451–466. Pike, K. L. (1967). Language in relation to a unified theory of the structure of human behavior. The Hague, The Netherlands: Mouton. Poortinga, Y. H., & Van de Vijver, F. J. (1987). Explaining crosscultural differences: Bias analysis and beyond. Journal of Cross-Cultural Psychology, 18, 259–282. Prabha, I. (1984). Effect of social support on the performance of high school subjects as a function of stress and test anxiety. Unpublished M. Phil dissertation, Himachal Pradesh University, Shimla, India. Prins, P. J. M., & Hanewald, G. J. F. P. (1997). Self-statements of test-anxious children: Thought-listing and questionnaire approaches. Journal of Consulting and Clinical Psychology, 65, 440–447. Raguram, R., Weiss, M. G., Keval, H., & Channabasavanna, S. M. (2001). Cultural dimensions of clinical depression in Bangalore, India. Anthropology & Medicine, 8, 31–46. Richmond, B. O., Rodrigo, G., & Lusiardo, M. (1989). Measuring test anxiety among children in Uruguay. In R. Schwarzer, H. M. van der Ploeg, & C. D. Spielberger (Eds.), Advances in test anxiety research (Vol. 6, pp. 215–222). Lisse, The Netherlands: Swets & Zeitlinger.

Test Anxiety: A Cross-Cultural Perspective Rocklin, T., & Ren-Min, Y. (1989). Development and adaptation of the Chinese Test Anxiety Inventory: A Research Note. In R. Schwarzer, H. M. van der Ploeg, & C. D. Spielberger (Eds.), Advances in test anxiety research (Vol. 6, pp. 245–251). Lisse, The Netherlands: Swets & Zeitlinger. Sanghvi, C. (1995). Efficacy of study skills training in managing study habits and test anxiety of high test anxious students. Journal of the Indian Academy of Applied Psychology, 21, 71– 75. Sarason, I. G. (1958). Interrelationships among individual differences variables, behavior in psychotherapy, and verbal conditioning. Journal of Abnormal and Social Psychology, 59, 272– 275. Sarason, I. G. (1984). Stress, anxiety and cognitive interference: Reactions to tests. Journal of Personality and Social Psychology, 46, 929–938. Sarason, S., & Mandler, G. (1952). Some correlates of test anxiety. Journal of Abnormal and Social Psychology, 47, 810–817. Sarason, S. B., Davidson, K. S., Lighthall, F. F., Waite, R. R., & Ruebush, B. K. (1960). Anxiety in elementary school children. New York: Wiley. Schwarzer, C., & Kim, M. J. (1984). Adaptation of the Korean Form of the Test Anxiety Inventory: A research note. In H. M. van der Ploeg, R. Schwarzer, & C. D. Spielberger (Eds.), Advances in test anxiety research (Vol. 3, pp. 277–285). Lisse, The Netherlands: Swets & Zeitlinger. Segall, M. H., Dasen, P. R., Berry, J. W., & Poortinga, Y. H. (1990). Human behavior in global perspective: An introduction to cross-cultural psychology. New York: Pergamon. Seipp, B. (1991). Anxiety and academic performance: A metaanalysis of findings. Anxiety Research, 4, 27–41. Seipp, B., & Schwarzer, C. (1996). Cross-cultural anxiety research: A review. In C. Schwarzer & M. Zeidner (Eds.), Stress, anxiety, and coping in academic settings (pp. 13–68). Tubingen, Germany: Francke-Verlag. Sethi, B. B., Nathawat, S. S., & Gupta, S. C. (1973). Depression in India. Journal of Social Psychology, 91, 3–13. Sharma, S., Kumaraiah, V., & Mishra, H. (1996). Behavioral intervention in test anxiety. NIMHANS Journal, 14, 57–60. Sharma, S., Parnian, S., & Speilberger, C. D. (1983). A crosscultural study of test anxiety levels in Iranian and Indian students. Personality and Individual Differences, 4, 117–120. Sharma, S., & Sud, A. (1990). Examination stress and test anxiety: A cross-cultural perspective. Psychology and Developing Societies, 2, 183–201. Sharma, S., Sud, A., & Spielberger, C. D. (1983). Development of the Hindi form of Test Anxiety Inventory. In H. M. van der Ploeg, R. Schwarzer, & C. D. Spielberger (Eds.), Advances in test anxiety research (Vol. 2, pp. 183–190). Lisse, The Netherlands: Swets & Zeitlinger. Shweder, R. A., & Bourne, E. J. (1984). Does the concept of the person vary cross-culturally? In R. A. Shweder & R. A. LeVine (Eds.), Culture theory: Essays on mind, self and emotion (pp. 158–199). Cambridge, UK: Cambridge University Press. Singh, A., & Broota, A. (1992). Socio-personal variables and examination anxiety. Journal of the Indian Academy of Applied Psychology, 18, 73–78. Singhal, R., & Misra, G. (1989). Variations in achievement cognitions: The role of ecology, age, and gender. International Journal of Intercultural Relations, 13, 93–107. Singhal, R., & Misra, G. (1994). Achievement goals: A situationalcontextual analysis. International Journal of Intercultural Relations, 18, 239–258. Sinha, D. (1997). Indigenizing psychology. In J. W. Berry, Y. H. Poortinga, & J. Pandey (Eds.), Handbook of cross-cultural psychology: Vol. 1. Theory and method (2nd ed., pp. 129–169). Needham Heights, MA: Allyn & Bacon.

87 Sipos, K., Sipos, M., & Spielberger, C. D. (1985). The development and validation of the Hungarian form of Test Anxiety Inventory. In H. M. van der Ploeg, R. Schwarzer, & C. D. Spielberger (Eds.), Advances in test anxiety research (Vol. 4, pp. 221–228). Lisse, The Netherlands: Swets & Zeitlinger. Smith, R. J., Arnkoff, D. B., & Wright, T. L. (1990). Test anxiety and academic competence: A comparison of alternative models. Journal of Counseling Psychology, 37, 313– 321. Sovani, A., Thatte, S., & Nadkarni, A. (2000). Perceived sources of examination anxiety among school and college students. Paper presented at fourth annual conference of the Indian Association of Mental Health, Hyderabad, India. Sowa, C. J. & LaFleur, N. K. (1986). Gender differences within test anxiety. Journal of Instructional Psychology, 13, 75–80. Spangler, G. (1997). Psychological and physiological responses during an exam and their relation to personality characteristics. Psychoneuroendocrinology, 22, 423–441. Spielberger, C. D. (1980). Preliminary professional manual for the Test Anxiety Inventory (“Test Attitude Inventory”) TAI. Palo Alto, CA: Consulting Psychologists. Spielberger, C. D., & Vagg, P. R. (1987). The treatment of test anxiety: A transactional process model. In R. Schwarzer, H. M. van der Ploeg, & C. D. Spielberger (Eds.), Advances in test anxiety research (Vol. 5, pp. 179–186). Lisse, The Netherlands: Swets & Zeitlinger. Srinivasa Murthy, R., Subramaniam, K. V., & Chatterji, S. (1990). Psychiatric research in India—annual review in 1988. NIMHANS Journal, 8, 13–21. Steinberg, L. (1996). Beyond the classroom: Why school reform has failed and what parents need to do. New York: Simon & Schuster. Sud, A. (1990). A comparative study of test anxiety across Indian and Italian cultures. Psychological Studies, 35, 69–75. Sud, A. (1991). Educational level differences on test anxiety within Indian and US cultures. Journal of Personality and Clinical Studies, 7, 131–136. Sud, A. (1993). Efficacy of two short term cognitive therapies for test anxiety. Journal of Personality and Clinical Studies, 9, 39– 46. Sud, A. (1994). Attentional skills training/cognitive modeling: Short term therapeutic cognitive interventions for test anxiety. Psychological Studies, 39, 1–7. Sud, A. (2001). Test anxiety research in India: Twentieth century in retrospect. Psychology and Developing Societies, 13, 51–69. Sud, A., & Katoch, S. (1994). Cognitions of test anxious students during an actual test situation. Journal of the Indian Academy of Applied Psychology, 20, 23–30. Sud, A., & Prabha, I. (1995). Test anxiety and academic performance: Efficacy of cognitive/relaxation therapies. Psychological Studies, 40, 179–186. Sud, A., & Prabha, I. (1996). Efficacy of cognitive/relaxation therapy for test anxiety. Journal of Personality and Clinical Studies, 12, 37–47. Sud, A., & Sharma, H. (1989). Test anxiety, intrusive thoughts and attentional process. Journal of Personality and Clinical Studies, 5, 139–145. Sud, A., & Sharma, S. (1990). Two short-term, cognitive interventions for the reduction of test anxiety. Anxiety Research, 3, 131–147. Sud, A., & Singh, J. (1991). Development of Hindi version of the Reactions to Tests Questionnaire. Psychological Studies, 36, 137–143. Suzuki, L. K., & Greenfield, P. M. (2002).The construction of everyday sacrifice in Asian Americans and European Americans: The roles of ethnicity and acculturation. Cross-Cultural Research: The Journal of Comparative Social Science, 36, 200– 228.

88 Szapocznik, J., & Kurtines, W. M. (1993) Family psychology and cultural diversity: Opportunities for theory, research, and application. American Psychologist, 48, 400–407. Teja, J. S., Narang, R. L., & Aggarwal, A. K. (1971). Depression across cultures. British Journal of Psychiatry, 119, 253–260. Tobias, S. (1985). Test anxiety: Interference, defective skills, and cognitive capacity. Educational Psychologist, 20, 135–142. Triandis, H. C. (1995). Individualism and collectivism. Boulder, CO: Westview. Triandis, H. C. (1999). Cross-cultural psychology. Asian Journal of Social Psychology, 2, 127–143. Tyron, G. (1980). The measurement and treatment of test anxiety. Review of Educational Research, 50, 353–372. Vagg, P. R., & Papsdorf, J. D. (1995). Cognitive therapy, study skills training, and biofeedback the treatment of test anxiety. In C. D. Spielberger and P. R. Vagg (Eds.), Test anxiety: Theory, assessment, and treatment. Series in clinical and community psychology (pp. 183–194). Washington, DC: Taylor & Francis. Van der Ploeg, H. M. (1982). The relationship of worry and emotionality to performance in Dutch school children. In R. Schwarzer, H. M. van der Ploeg, & C. D. Spielberger (Eds.), Advances in Test Anxiety Research (vol. 1, pp. 55–56), Lisse, Netherlands: Sweets & Zeitlinger Publishers. Van der Ploeg, H. M. (1983). The validation of the Dutch form of Test Anxiety Inventory. In H. M. van der Ploeg, R. Schwarzer, & C. D. Spielberger (Eds.), Advances in test anxiety research (Vol. 2, pp. 191–202). Lisse, The Netherlands: Swets & Zeitlinger. Van de Vijver, F., & Hambleton, R. K. (1996). Translating tests: Some practical guidelines. European Psychologist, 1, 89–99. Van de Vijver, F. J. R., & Leung, K. (2000). Methodological issues in psychological research on culture. Journal of CrossCultural Psychology, 31, 33–51. Van Widenfelt, B. M., Treffers, P. D. A., de Beurs, E., Siebelink, B. M., & Koudijs, E. (in press). Translation and cross-cultural adaptation of assessment instruments used in psychological research with children and families. Clinical Child and Family Psychology Review. Verma, S., & Gupta, J. (1990). Some aspects of high academic stress and symptoms. Journal of Personality & Clinical Studies, 6, 7–12. Verma, J., & Triandis, H. C. (1999). The measurement of collectivism in India. In W. J. Lonner, D. L. Dinner, D. K.

Bodas and Ollendick Forgays, & S. A. Hayes (Eds.), Merging past, present, and future in cross-cultural psychology: Selected papers from the fourteenth international congress of the International Association of Cross-Cultural Psychology (pp. 256–265). Lisse, The Netherlands: Swets & Zeitlinger. Verma, S., & Saraswathi, T. S. (2002). Adolescence in India: Street urchins or Silicon Valley Millionaires. In B. B. Brown, R. W. Larson, & T. S. Saraswathi (Eds.), The world’s youth: Adolescence in eight regions of the globe (pp. 105–140). Cambridge, UK: Cambridge University Press. Verma, S., Sharma, D., & Larson, R. W. (2002). School stress in India: Effects on time and daily emotions. International Journal of Behavioral Development, 26, 500– 508. Wang, Y., & Ollendick, T. H. (2001). A cross-cultural and developmental analysis of self-esteem in Chinese and Western children. Clinical Child and Family Psychology Review, 4, 253– 271. Warren, M. K., Ollendick, T. H., & King, N. J. (1996). Test anxiety in girls and boys: A clinical-developmental analysis. Behaviour Change, 13, 157–170. Wine, J. D. (1971). Test anxiety and evaluation threat: Children’s behavior in the classroom. Journal of Abnormal Child Psychology, 7, 45–49. Wine, J. D. (1979). Test anxiety and evaluation threat: Children’s behavior in the classroom. Journal of Abnormal Child Psychology, 7, 45–59. Wisocki, P. A. (1973). A covert reinforcement program for the treatment of test anxiety: Brief report. Behavior Therapy, 4, 264–266. Yerkes, R. M., & Dodson, J. D. (1908). The relation of strength of stimulus to rapidity of habit-formation. Journal of Comparative Neurology and Psychology, 18, 459–482. Yue, X. (1994). A comparative study of test anxiety among Chinese and American high school students. Hong Kong Psychological Society Bulletin, 32–33, 47–59. Zatz, S., & Chassin, L. (1983) Cognitions in test anxious children. Journal of Consulting and Clinical Psychology, 51, 526–534. Zeidner, M. (1998). Test anxiety: The state of the art. New York: Plenum. Zeidner, M., & Nevo, B. (1993). Test Anxiety Inventory/Hebrew adaptation (TAI/HB): Scale development, psychometric properties, and some demographic and cognitive correlates. Megamot, 35, 293–306.