The link between immigrant backgrounds and parental achievement values

For many immigrant parents, the importance assigned to academic achievement is particularly high (1). Upward social mobility is a powerful factor that drives immigrants to settle in new countries, especially among those who have moved from developing to industrialized countries. Immigrant parents tend to perceive education as a primary avenue through which social mobility can be achieved by their children in the new country (2). It has been proposed that immigrant parents are optimistic about their offspring’s educational and socioeconomic prospects (3,4). This may prompt their children to try harder to achieve academically. It is unclear, however, how immigrant parents’ optimism leads their children to adopt behaviors that influence their emotional health. Given that social mobility is usually an integral motivation for immigrants, higher levels of parental achievement values (i.e., attitudes that emphasize academic achievement) are expected among immigrant parents as compared with non-immigrant parents. Several studies have supported this finding (5-7), particularly among Asian immigrant groups in the United States, and this has given rise to the Asian “model minority” stereotype (8-10). This stereotype permeates American society and refers to the general tendency to regard Asian immigrants as high-achieving students and employees.

The link between parental achievement values and emotional problems

Although parental achievement values have been linked to positive outcomes among adolescents (11-14), they have also been linked to emotional problems in groups such as the Chinese (15) and Korean (16) immigrants as well as among affluent American suburban adolescents (17). The latter suggests that the link between strong parental achievement values and emotional problems may not necessarily be limited to immigrant groups. Although research in these areas has been in demand (17-19), not much is known about the relationship between parental achievement values and emotional problems among adolescents, particularly immigrant adolescents. Why might strong parental achievement values be related to emotional problems? Strong feelings of pressure to achieve in accordance with parental (or the child’s own) values may be part of the explanation for this.

It has been found that achievement pressure, whether parental or personal, appears to increase emotional problems (15,17;20-25). According to this line of thought, the link between parental achievement values and emotional problems may depend on the amount of perceived achievement pressure. As compared with parental achievement values (which measure how strongly parents value their child’s achievement in terms of attitudes), the relationship between achievement pressure (which measures parental behavior with the aim of influencing children to achieve better) and emotional problems is more thoroughly studied and supported in the research literature (15,17,20-25). Although parental achievement values and pressure are usually linked, it is possible to hold strong achievement values without pressuring or actively coercing the child to excel academically. Parents can exert academic pressure in many ways, such as by offering conditional rewards (e.g., going to the cinema with friends) or punishments (e.g., not being allowed to participate in desired leisure activities) depending on the child’s academic results. Another way of exerting academic pressure is through comparison; this involves parents openly comparing their child’s achievements with those of others (e.g., siblings, peers) with the aim of coercing the child to exceed academically. Comparison is a type of psychological pressure or control that has been found to be more frequently used among groups with predominantly collectivistic values (16,26). Thus, comparison as an expression of pressure may mediate the relationship between parental achievement values and filial emotional problems. As such, strong comparison could explain why strong parental achievement values are sometimes related to high levels of filial emotional problems.

The unique effect of parental achievement values on emotional problems

To properly assess the unique effect of parental achievement values on emotional problems, it is important to adjust for a series of factors related to the child’s academic situation. First and foremost, it is important to distinguish between the effects of parental achievement values from those of parental achievement pressure, because the latter could confound the relationship between the former and emotional problems. Actually, these terms have been used interchangeably in the research literature, thus making it difficult to tease their effects apart from each other (17,27). In studies based solely on children’s own self-reporting, there is also a need to adjust for factors that could influence the relationship between child-reported parental achievement values and emotional problems. For example, academic competence, which is generally inversely related to emotional problems (22), can potentially confound the relationship between parental achievement values and emotional problems because all variables, including parental ones, are based on the child’s perception. Perfectionistic students with high levels of perceived academic competence could report high levels of both perceived parental achievement values and personal emotional problems (21), because all of these are perceived by the same person at the same point in time. Other examples of potentially confounding variables are academic factors that have been associated with both emotional problems and parental achievement values, such as school hassles, lack of parental help with homework, and domestic chore burden, which is included as an academic factor as a result of its negative association with academic competence (28). More school hassles have been associated with emotional problems (29). Parental help with homework has been associated with positive academic outcomes (30), although general findings are mixed (31). Despite higher levels of parental achievement values, immigrant youth in Norway have not reported equivalently higher levels of parental help with homework as compared with their ethnic majority peers (28). Domestic chore burden has been suggested as a reason why some groups of immigrant children, especially girls, appear to engage less than their ethnic majority peers in leisure activities that seem to protect them from emotional problems (32). These perceived variables are likely to influence reported levels of either emotional problems, parental achievement values, or both in studies comparing immigrants with non-immigrants.

The Norwegian context and hypotheses

The contrast between non-Western immigrants and non-immigrant ethnic Norwegians is particularly interesting in terms of parental achievement values. This is both because Norway’s non-Western immigrants are more traditionalistic and collectivistic as a group and because there is a relatively low emphasis on academic achievement among the non-immigrant population (again, as a group). Norway is one of the most socially egalitarian countries in the world in which additional higher education pays the least (33). It is possible that, within this social context, Norwegian parents may not have the same eagerness and urgency with regard to pushing their children toward academic achievement. By contrast, immigrant parents are willing to pay the price of a loss of social network and cultural competence through immigration to achieve social mobility for themselves or their children in the new country (4). The fact that first-generation immigrants sometimes have to postpone social mobility for the next generation may explain why second-generation immigrants report more parental achievement values (5,28,34). As for collectivistic parenting strategies, they may promote interdependent bonds between generations, and they may strengthen the association between parental and child behavior (35). This suggests that the link between parental comparison and emotional problems could be stronger among children of immigrants as compared with non-immigrants.

This article attempts to bridge together two research domains: one attending to the link between achievement values in immigrants and the other addressing the link between pressure and emotional problems. This is a novel framework, especially for adding to studies of the mental health of children of immigrants. On the basis of the previously mentioned existing research, we first hypothesize that immigrant background will be associated with higher levels of both parental achievement values and comparison. Thus, we attempt to differentiate between the effects of parental achievement values and achievement pressure, which are operationalized by comparison in this study. Second, we suggest that the relationship between parental achievement values and emotional problems will be mediated through comparison and moderated by immigrant background in such way that the relationship will be stronger for those with immigrant backgrounds.

Data collection

Data was provided by the Youth Culture and Competence (YCC) study undertaken by the Norwegian Institute of Public Health and approved by the Regional Committee for Medical Research Ethics (REC). During 2006 and 2007, the YCC invited preadolescents between the ages of 10 and 12 years from fourteen schools in Oslo and Bergen to participate in the first wave of this study. Although the proportion of immigrants in Bergen is equal to the 11%, which is the national average, 27% of the population of Oslo is foreign born, or Norwegian born with two foreign-born parents. Immigrants are from more than 200 different countries, most of which are non-Western (33). Although there are few ethnic enclaves in Norway, immigrants tend to settle together in areas where housing prices are reasonable. Schools in these areas were targeted during the initial stage of recruitment. In addition, children of Tamil and Turkish origin from other schools were recruited through collaboration with immigrant non-governmental organizations.

Parents were informed about the YCC study through the preadolescents’ school or the immigrant non-governmental organization and asked to provide written informed consent if their preadolescents were allowed to participate in the study, in accordance with REC guidelines.

The YCC questionnaire was completed by the preadolescents in their respective classrooms during two class hours. For those recruited through immigrant non-governmental organizations, questionnaire completion was performed after school hours. Research assistants were available if needed during data collection. Of the children invited, 1042 preadolescents in the fifth, sixth, and seventh grades participated in the YCC, which resulted a participation rate of 65%.

Identifying immigrants and non-immigrants

For ethical reasons, we were not allowed by the REC to directly ask participants about their ethnic backgrounds in the questionnaire. Thus, we relied on parental and grandparental place of birth to categorize participants according to parental immigrant background. During the first phase of categorizing the participants, we grouped participants into three broad categories according to parental immigrant background: 1) immigrant parents (n = 473; parents born abroad and at least three grandparents born abroad); 2) non-immigrant parents (n = 476; parents born in Norway and at least three grandparents born in Norway); and 3) dual parents (n = 91; one parent born in Norway and one parent born abroad). Participants in the third category were excluded from further analyses, because their complex mix of immigrant and non-immigrant backgrounds required special attention that exceeded the scope of this study.

During the second phase of categorizing the participants, we grouped those with immigrant parents into national groups according to maternal place of birth. (There were a few participants whose parents were born in two different countries outside of Norway.) The result was a broad but scattered representation of a total of 49 national backgrounds. The only countries that made up more than 5% of the sample were Norway (n = 485), Pakistan (n = 126), Turkey (n = 45), and Sri Lanka (n = 43).

During the third phase of categorizing the participants, we divided the immigrant group into two main groups in accordance with a distinction used by Statistics Norway: 1) individuals originating in the European Union or the European Economic Area, the United States, Canada, Australia, and New Zealand; and 2) individuals originating in European countries outside of the European Union, Asia, Africa, Latin America, and countries in Oceania other than Australia and New Zealand (36). Because there were only sixteen participants in group 1, they were excluded from analysis. In other words, the parents of the preadolescents included in this study have non-Western backgrounds.

Sample

The sample consisted of 902 preadolescents who were evenly divided across age and gender. Forty-seven percent of the sample had immigrant backgrounds, which means that minorities were overrepresented in line with the recruitment strategy of the YCC. The largest minority groups were from Pakistan (14%), Turkey (5%), and Sri Lanka (5%). Of those recruited through cultural centers, eighteen preadolescents with Turkish backgrounds (40% of the Turkish sample) and four with Tamil backgrounds (9% of the Tamil sample) completed the questionnaire. It was not possible to compare participants and non-participants in terms of background characteristics, because this was a convenience sample and information about non-participants was limited.

Outcome variables

Emotional problems were measured by the self-report version of the emotional problems subscale of the Strengths and Difficulties Questionnaire, Norwegian version (see www.sdqinfo.com) (34). This questionnaire has adequate psychometric properties that have been replicated in Norway (37-39). The emotional problems subscale consists of five items that indicate symptoms of somatization, anxiety, and depression. The items are as follows: “I get a lot of headaches, stomachaches, or sickness,” “I worry a lot,” “I am often unhappy, depressed, or tearful,” “I am nervous in new situations. I easily lose confidence,” and “I have many fears. I am easily scared” (α = 0.68). The items were rated 0 (“Not true”) to 2 (“Certainly true”), and a sum score of between 0 and 10 was computed.

Predictor, moderator, and mediator variables

The predictor variable was parental achievement values, which was a measure of how preadolescents perceived parental attitudes regarding achievement and which was extracted from the Parental Pressure Questionnaire validated by Kim (35,40). Four items were rated on a Likert scale that ranged from 1 (“Never”) to 5 (“Always”) (α = 0.70), and a mean score was computed. The items were as follows: “My parents carefully monitor what I do in school,” “My parents think that doing well in school is very important,” “My parents say that I should work hard at everything I do,” and “My parents often ask me about schoolwork.” The moderator variable was immigrant background, which was dichotomized as 0 for those with non-immigrant backgrounds and 1 for those with immigrant backgrounds.

The mediator variable was comparison, which measured how parents actively compared the child’s achievement with that of others as a form of parental pressure or control (26). This is done as a parental strategy to coerce the child to desire to outperform his competitors, thereby conforming to parental expectations. These items were developed and validated by Sumer and Kagitsibaci (26) and added to the comparison subscale of the Egna Minnen Beträffande Uppfostran scale for children. This is a validated scale, and its name, translated from Swedish, means “My Memories of Upbringing” (41). These items were added as a result of their relevance for measuring parental rearing styles in collectivistic societies (26). Three of the items added measured achievement comparison and were included in the present study due to their relevance for exploring the construct of parental achievement pressure. The achievement comparison items were as follows: “My parents are more concerned about how I do relative to other children than what I actually accomplish,” “My parents compare my school performance with that of my friends,” and “My parents compare my school performance with that of my siblings or relatives” (α = 0.63). Each question was rated on a scale that ranged from 1 (“Never”) to 5 (“Always”), and a mean score of the scale was computed.

Background variables

In addition to gender (dichotomized as 0 for boys and 1 for girls) and city background (0 for Bergen and 1 for Oslo), grade level (1 for fifth grade, 2 for sixth grade, and 3 for seventh grade) and economic hardship were included as background variables. The two latter variables were used as proxies for age and socioeconomic status, respectively. Age was computed by transforming grade level values from the fifth, sixth, and seventh grades into 10, 11, and 12 years of age, respectively. Economic hardship, which served as a proxy for socioeconomic status, was measured with four items from the scale Adolescent Perceptions of Family Hardship (42). The items were as follows: “How often do your parents argue with each other about not having enough money?”, “How often do you argue with your parents about not having enough money?”, “How much of a problem does your family have because your parents do not have enough money to buy things your family needs or wants?”, and “How upset or worried are your parents because they do not have enough money?” (α = .65). The first two items were rated on a scale that ranged from 1 (“Never”) to 5 (“Always”), whereas the latter two items were rated on a scale that ranged from 1 (“Never”) to 4 (“Very much”). All four economic hardship items were standardized, and a mean score was computed. The unstandardized scale, which consisted of the first two items, was used when reporting mean differences (α = .48); otherwise, the standardized scale was used.

Academic factors

School hassles (i.e., everyday concerns related to schoolwork) were measured by five items from the General Everyday Hassles scale (43). The preadolescents listed how often during the last year they had experienced the following: “I am afraid of not doing well enough at school,” “Big problems in understanding the teacher when he/she is teaching,” “Strong pressure from those around me to succeed and do well at school,” “Problems with one or more teachers,” and “Arguments or problems with other(s) in class” (α = .57). Each question was rated on a scale that ranged from 0 (“No, never”) to 3 (“Yes, very often”), and summed scale scores were computed.

Parental help with homework was measured by the mean score of two questions from the Lamborn Parenting Scale (44): one concerning maternal help and the other concerning paternal help. These were worded as follows: “My mother/father helps me with school assignments and home work if there is something I do not understand” (α = .54). Adolescent domestic chore burden was measured with one question from the General Everyday Hassles scale (43): “Too much responsibility at home (for taking care of siblings or doing household chores).” Participants could rate this on a scale that went from 1 (“Not at all”) to 5 (“Always”). This measure is considered an academic factor because it has been associated with lower achievement and stronger emotional problems and because it is particularly relevant in specific immigrant communities in which girls are expected to contribute more than boys to household chores (32). Self-rated academic competence was measured by two questions concerning how competent preadolescents perceived themselves to be in math and Norwegian, respectively. Each subject was rated from 1 (“Not at all”) to 5 (“Very competent”), and the mean competence score was computed (α = .53).

Missing analysis

We used a maximization-expectation algorithm to impute missing responses for each scale included in the analyses with more than one item. These were emotional problems, parental achievement values, comparison, school hassles, domestic chore burden, academic achievement, and parental help with school work. Missing values were imputed if at least 50% of the items had been completed by each participant. The scale with the highest percentage of missing values was comparison (7.8% for non-imputed items and 3.2% with imputed items). The number of cases involving emotional problems was unchanged before and after imputation (n = 889). After imputation, the number of valid cases increased from 742 to 849 (94.1%). The imputed variables were approximately normally distributed for emotional problems (skewness, 1.17; kurtosis, 2.08), parental achievement values (skewness, –0.73; kurtosis, –0.08), comparison (skewness, 0.54; kurtosis, –0.44), and academic achievement (skewness, –0.49; kurtosis, 0.08), whereas values for school hassles (skewness, 0.61; kurtosis, –0.28) and parental help with school work (skewness, 1.74; kurtosis, 2.70) fell within acceptable values.

Analytic strategy

The level of significance was set at .05. We used PASW Statistics 18 software (45) to conduct correlations, to perform t-tests to investigate whether preadolescents with immigrant backgrounds reported higher levels of parental achievement values and comparison, and to determine the mediating role of comparison on the relationship between parental achievement values and emotional problems among minority and majority preadolescents with the macro ‘Process’ developed by Hayes (46). More specifically, we tested Model 59 in ‘Process,’ which tested moderated mediation; this looked at whether the effect of parental achievement values (X) on emotional problems (Y) was significantly mediated by comparison (Mj) across immigrant background (the moderator, W) (46). Background (i.e., gender, school grade, economic hardship, and city) and academic factors (i.e., school hassles, age, parental help with homework, domestic chore burden, and academic competence) were adjusted for in the model, because these variables could confound the relationship between X and Y given that all measures in the study stem from child self-reports at one point in time. Because sampling distributions of indirect effects are seldom normally distributed (47), bootstrapping was performed to test the significance of the indirect effect among minority and majority preadolescents and to address the difference between the magnitudes of the indirect effect in these two groups. To estimate 95% confidence intervals, 5000 bootstrap samples were drawn from the original sample, all of which had the same N as the original sample.