The role of contact and perceived attitudes of surroundings in children’s acceptance of “different” peers3

This study was conducted with Croatian fourth grade students (N=691) that were divided into two groups. The characters presented in vignettes had some type of stigmatizing feature (e.g., Roma nationality, wheelchair usage, or being overweight) in the experimental group, while these features were left out in the comparison group. The first goal was to examine the impact of each stigmatizing feature on the non-acceptance of the described characters. The second goal was to analyze the relationship between the non-acceptance of the stigmatized character and children’s established contact with people with similar features, as well as the relationship between the afore-mentioned and the perceived attitudes of parents and friends. Roma ethnicity had a significant influence on non-acceptance of the character, while contact was related to an increase in the level of non-acceptance of the Roma character. Negative perceived opinions of friends regarding individuals with stigmatizing features (regardless of the type) were related to a lower level of acceptance of that character.

Social attitudes that develop during childhood can persist during adolescence and adulthood (Nesdale, 2006).Studying different forms of children's prejudice and the factors that influence prejudice may help prevent its long-term consequences.Stereotypes and prejudice can result in social distance (Bogardus, 1925;Brown, 1995).Group membership is one of the bases for social rejection during childhood and the consequences of social rejection are often very serious, including depression, psychological maladjustment, poor academic achievement, violence, and school dropout (e.g., Boivin, Hymel, & Bukowski, 1995;Boulton, Don, & Boulton, 2011;Rutland et al., 2012).
Nevertheless, the phenomenon of social distance in children has been poorly studied, while the few existing studies indicate that even children of preschool age (Patterson & Bigler, 2006;Spears Brown & Bigler, 2005) and Corresponding author: jelena.maricic@hotmail.com(Maričić, Kamenov, & Horvat, 2012;Mihić & Mihić, 2003) perceive some groups as close and others as distant.
Previous research has shown that prejudice toward various groups in different ways changes with age and other factors (Fishbein, 2002).However, prejudice toward different groups can have many similarities, including the methods and tactics that are used to prevent their full integration (Connor & Ferry, 2006).
The lack of conclusions concerning the common fate of different groups arises from the fact that diverse target groups are rarely researched in the same study (Maričić et al., 2012;Powlishta, Serbin, Doyle, & White, 1994).This may prevent us from seeing the similarities and differences between how various groups are treated.For example, Roma children, overweight children and children with physical disabilities are three very different groups, but their members share a similar fate of being rejected by their peers.With the goal of finding their position and its determinants in the Croatian social context, we examined children's social distance toward these groups in light of contact with their members and perceived attitudes of children's closest surroundings toward these groups.
Roma people in Europe, and especially in some parts of Eastern Europe, (who are also known by the derogatory term Gypsies) represent a marginalized ethnic group that rarely has the chance to be integrated into wider society (Djurović, 2002).In Croatia, they are mostly perceived as dirty, beggars, uneducated, uncultured, thieves, poor and immoral (Grgić, 2006).Roma integration into the public school system is an example of their adjustment and adaptation to the majority society.Several empirical studies in Croatian schools described the existence of negative stereotypes toward Roma, as well as a great social distance toward them (Maričić et al., 2012;Previšić, Hrvatić, & Posavec, 2004).In neighboring countries, Roma are also in a worse position compared to other ethnic groups -for example, social distance toward Roma in Novi Sad or Vojvodina was higher than the distance toward five other national groups, and even Roma children expressed the greatest social distance toward their own ethnic group (Franceško, Mihić, & Kajon, 2005).When comparing social distance toward different ethnic groups, the situation is quite clear, however we wanted to go a step further and compare that ethnic with some non-ethnic groups.
Although there is a somewhat greater social sensitivity toward people with disabilities, this does not mean that prejudices toward them do not exist; they are only subtler.Of interest, is to see whether prejudice toward them is also expressed through social distance, and whether they will be explicitly rejected in some types of contact.Although Western European countries encourage the integration of children with disabilities into educational institutions, these children give poorer estimates of their social environment than do their peers without disabilities.They are also less involved in social and educational activities at school and they are not as socially favored as their "average" peers (Simeonsson, Carlson, Huntington, McMillen, & Brent, 2001;Wendelborg & Kvello, 2010).In Croatia, in spite of relevant legislation and national strategies, in practice many schools are inaccessible to children with physical disabilities, mostly because of an underdeveloped transport infrastructure and other environmental barriers.However, it is commendable that a project called Network of schools without architectural barriers, has partially improved the accessibility of education for children with disabilities in every county across Croatia (Ružić, 2010).
Concerning prejudice toward overweight and obese children, some studies have shown that these children experience prejudice which has become more negative over the last fifty years (Kraig & Keel, 2001;Latner & Stunkard, 2003;Penny & Haddock, 2007a).Obesity is a very serious and widespread problem among children in North America, where up to 50% of children experience obesity, while nearly 38% of children are obese in the European Union (Puhl & Latner, 2007;Wang & Lobstein, 2006).In the Croatian capital, Zagreb 27.7% of girls and 32.0% of boys were overweight between the ages of 7 and 15 in 2011, while 11.2% of boys and 7.2% of girls were obese at these ages (Croatian Institute for Public Health, 2012).Overweight individuals are often targets of discrimination, especially of teasing and derogatory comments (Neumark-Sztainer, Falkner, Story, Perry, & Hannan, 2002).It has been shown that there is a certain decrease in prejudice toward them between the age of 5-10, which increases just before adolescence (Klaczynski, Daniel, & Keller, 2008;Penny & Haddock, 2007b).
Based on Goffman's typology (1963), the three mentioned groups, Roma people, people with a physical disability and overweight people, are stigmatized related to different criteria -ethnicity, disability, or a "shameful".The term stigma is used to refer to an attribute that is deeply discrediting, but its discreditably is based on the relationship toward (un)usualness, and not on attribute itself (Goffman, 1963).
Although Goffman's Theory of Social Stigma (1963) is one of the most influential concerning the topic of stigmatization, it does not include a developmental perspective, therefore we based our study on Nesdale's Social Identity Development Theory (1999, 2004).Nesdale (1999Nesdale ( , 2004) ) emphasizes the dominance of social processes versus cognitive ones, in the development and reduction of prejudice, that are actually the foundation of his Social Identity Development Theory.Children often have negative attitudes which prevail in their surroundings, because they see themselves as members of the group that has a certain set of beliefs.Three factors lead to a possible change of in-group preferences into prejudice, around the age of seven: 1) the fact that people in the child's surroundings share and clearly express their biases, which become the norm; 2) the existence of competition, tension or conflict between members of dominant and minority groups; 3) and the conviction of members of their group that certain outgroups threaten their social status (Brown, 1995;Nesdale, 2004).Those influences can lead to different patterns of association between social distance and different personal or social phenomenon.Given that these groups vary considerably, and the prejudice against them were expressed in different ways, it is worthwhile to find out what is the relationship of social distance toward these three groups with external determinants, specifically in relation to contact with them and perceived attitudes of their surroundings toward them.
Contact is the first variable that will be considered in this study and is examined through a number of known persons with a certain feature.Contact is often seen as the most important component in prejudice reduction but has its limitations concerning diverse age groups and features (Allport, 1954;Oskamp, 2000;Pettigrew & Tropp, 2011).In adults, the negative correlation between inter-group contact and prejudice was confirmed in 94% of studies, with about half of those studies focusing on racial and ethnic groups (Pettigrew & Tropp, 2011).The positive effect of contact was also confirmed when considering other groups that face frequent discrimination, such as older persons, homosexuals, immigrants, obese people, and people with physical or mental disabilities (Fee & Nusbaumer, 2012;Oskamp, 2000;Pettigrew & Tropp, 2011).Sometimes, minimum contact is enough to reduce prejudices: for example, even brief exposure to the photographs of a few Asian faces increased later liking of different Asian faces, which suggests that even "knowing someone" through a photograph can reduce prejudice and contribute to the generalization effect towards the whole group (Zebrowitz, White, & Wieneke, 2008).Adolescents often assess contact as positive, allowing the positive conception of interracial contacts to increase with age (Killen, Kelly, Richardson, & Jampol, 2010).Quantity of contact with a certain group is shown as related to implicit attitudes, which are seen as being a consequence of associations within the environment and developed over a long period of time (Devine, 1989;Prestwich, Kenworthy, Wilson, & Kwan-Tat, 2008).In addition, quantity of contact, in the form of number of known members of a certain group, improves intergroup attitudes and is mediated by intergroup anxiety and empathy, which leads to long-term effects (Vezzali, Giovannini, & Capozza, 2010).In children, contact with some groups has proven to be related to a higher acceptance of peers from a different ethnic background or with a physical impairment (Nabors, Tipkemper, Lehmkuhl, Petre, & Nash, 2006;Stringer et al., 2010).
In relation to contact with the groups that are the focus of this study, contact between the Roma and non-Roma population in Croatia is often characterized as weak because Roma often live in segregated settlements and have low employment rates in the formal labor market.For this reason contact from an early age through school is very important for Roma children and their families, as a way of breaking social exclusion patterns.Unfortunately, as prejudices are prevalent and there is a widespread lack of mutual trust, Roma children often do not feel accepted at school (Ombudsperson for Children, 2010).Regarding contact with persons in wheelchairs, chances of contact are higher where there is better adaptation of public spaces, while contact with overweight persons is expected to be the most frequent, since there is a higher prevalence of overweight people in the population.
Another important element in reducing prejudice could be in-group emotions, in situations when a person does not have direct contact with a member of an out-group, but indirect contact, through (real or imagined) friendship between a member of their group and a member of an out-group (Cernat, 2011).Namely, positive emotions toward members of an in-group could spread positive emotions toward an out-group.The role of the surroundings can be important in both directions: for example, the Social Identity Development Theory emphasizes that if people in the child's environment share and clearly express their prejudice, they become norms and it is expected that even children will accept them (Nesdale, 1999(Nesdale, , 2004)).In other words, people usually consider parents to be the main source of children's prejudice, but it is possible that parental influence on children's attitudes varies greatly with respect to the discriminated group (Fishbein, 2002).It is also possible that a source of prejudicial attitude modeling exists in the peer environment (Fishbein, 2002;Harris, 1995).Children that express significant prejudice between the ages of nine and ten reduce their prejudice after a discussion about racial issues with their non-prejudiced peers (Aboud & Doyle, 1996).
Besides the influence of parental and peers' real attitudes, their perceived attitudes are a very significant type of influence, whose relevance has been proven in different age groups (Castelli, Zogmaister, Tomelleri, & Amari, 2007;Goel, Mason, & Watts, 2010).
The goal of our study was to understand the role of social variables, including the number of known people from a particular group and perceived attitudes of people close to the child in children's attitudes toward peers who are different from them.These attitudes are evident by analyzing children's social distance toward their peers who exhibit differences, such as disability or being overweight.With this goal, we studied social distance toward a character described in vignettes as a typical 10-year old who likes some activities which are currently preferred among Croatian children of the same age.Thereby, the character in the comparison group does not have any atypicality, while he or she has some stigmatizing feature in the experimental one.Since children choose friends primarily on the basis of perceived similarities and differences, we consider it to be the foundation for studying and developing inter-group contact.Generally, the most important factors considered for children's choice of friend include age, gender, ethnicity, common interests, and preferences for certain activities (Fishbein, 2002;Nabors et al., 2006).The situation that consists of differences in some aspects, including ethnicity, and similarities in other aspects, including preference for similar activities, is the foundation for studying and developing inter-group contact.
On the one hand, we examine the role of direct contact with stigmatized persons, while on the other, we question the role of perceived attitudes in their surroundings that children absorb, trying to fully understand their impact on children's acceptance of "different" peers.
Based on previous research, we can propose the following hypothesis: 1) The presence of a stigmatizing feature (Roma ethnicity, wheelchair usage, being overweight) influenced children's expression of social distance in a negative sense.2) Lower children's social distance toward their peers with some stigmatizing characteristics can be predicted through the number of known people from a particular stigmatized group (contact) and through perceived positive attitudes of parents and friends toward people from these particular groups.
In essence, prejudices are part of the complex value system of an individual.Therefore, this study tries to examine the role of possible children's prejudice determinants toward a few specific groups, as well as to comprehend biases which can become norms, as the Social Identity Development Theory emphasizes.

Method Participants
The participants (N=691) in this research were fourth-grade students between the ages of 10 and 11 years old.The research was conducted in 17 elementary schools in two Croatian cities, Zagreb and Split.The sample of convenience was divided into an experimental group (N=338; N boys =185, N girls =153) and a comparison group (N=353; N boys =171, N girls =182).Sampling was done arbitrarily, selecting one school for the experimental and another one for the comparison group, so the number of both schools was even.However, the geographical closeness of experimental and comparison schools was taken into consideration -schools were distanced from each other, as well as schools from the center and from the suburbs were included.The aim was to cover a wide geographic range of schools within these two cities.In addition, an appropriate rotation of questionnaires was carried out due to the process of sampling.In addition, a pilot research study was conducted with a sample of fourth-grade children from two elementary schools in Split (N=44).
Since young children participated in this study, only those children whose parents signed consent for their participation were included in the study.Approximately 1000 consent letters were prepared to account for the number of children in all the schools in both cities.We did not receive a response from approximately 30% of the parents.A 30% non-response rate may be attributed to parents who did not wish to give their consent or children who forgot to bring the signed consent form back to school.

Measures and Procedure
The Modified Social Distance Scale Related to the Spending of Leisure Time.This instrument was based on the Scale of Social Distance Related to the Spending of Leisure Time for Children (Maričić et al., 2012).It was constructed after the researcher's previous study of children in which children's ways of spending leisure time with their peers was explored.The researcher attempted to find out from which popular peer activities children can be excluded.
In order to ensure that the included activities were generally acceptable to children from both Zagreb and Split, a replication study was conducted in two fourth-grade classrooms in two separate elementary schools in Split.Based on the responses of the children in Split, some items of the scale were modified, excluded, and amended, to include the activities that are generally acceptable to children in both Split and Zagreb.The degree of closeness was finally expressed on a 13-item social distance scale, which was arranged from the smallest ("I would agree to go to the same out-of-school activity with him/her") to the largest degree of closeness ("I would agree to share my secret with him/her").Factor analyses (principal component, one main component, without rotation) of Scales were conducted, to insure the acceptable value of social distance expression.Since the typical root (λ) of the first factor was significantly higher than other factors, it was reasonable to keep the one-factor structure of Scales.Factor analyses of Scales are given in detail in the table below.In the main study, the scales measured social distance toward the Roma child, the child in a wheelchair, and the overweight child.Characters were described in vignettes with the purpose of personalization.Each vignette included a short description of a child engaged in some activities that are attractive to 10-year-old Croatian children today, such as playing computer games or playing cards (see Appendix A).It was expected that participants would share some common interests with the characters, and that characters would be attractive to the participants.Moreover, the character was the same sex as the participant and the stigmatizing feature was mentioned in the vignette's first sentence of the experimental group.While the characters in the experimental group were members of one of the three mentioned stigmatized groups, those features were not mentioned in the comparison group, so the presence vs. nonpresence of a certain feature was an independent variable.
Participants were asked to mark either a "+" or "-" next to the type of contact that they would agree or not agree to have with a character from each vignette.Any rejection of the common activity scored one point and a higher score on the scale showed greater social distance toward the character with a possible range from 0-13.

Number of known members of discriminated groups (contact).
Questions relating to the number of known members of discriminated groups contained three items, one item for each discriminated group.The participants needed to numerically express how many such individuals he or she knows.An example of an item on the questionnaire was: "How many people do you know who are Roma?Write a number."

Perceived attitude of parents and friends about members of discriminated groups.
The respondents were asked to respond to items using a three-point scale where a higher score indicates a more positive attitude toward members of a specific group.An example item was: "What opinion do your friends have about people who are in a wheelchair?1: bad opinion; 2: neither good nor bad opinion; 3: good opinion."These six items were not added together, but each item was observed individually instead.They were presented separately for each discriminated group and separately for the experimental and comparison group.

Results
The aim of this article is to analyze the role of (non)existence of some stigmatizing feature in social distance toward children and to also observe some determinants on children's social distance toward their peers with a stigmatizing feature.T-tests between groups were conducted, for each vignette separately.Higher social distance toward the Roma child was confirmed; average result on the social distance scale toward the Roma character was M=4.49 (σ=3.75), while the average result toward identical non-Roma character was M=2.67 (σ=2.54) (t=7.468, p<0.001, df=589,689).Differences between the experimental and comparison group concerning vignettes about the character with disability (t=-1.259, p>0.05, df=677,689) or the overweight character (t=0.341,p>0.05, df=680,689) were not found, so we can conclude that the social distance toward children with those two features was not increased in comparison with the social distance toward majority children.
The following analysis refers to the correlation between social distance and the number of known members of target groups.Before running the results of this analysis, the number of known people with specific stigmatizing features by children in the experimental group is presented.Most of the children in the experimental group did not know any Roma or anyone from the wheelchair user group.The percentage generally decreases with the increased number of people they know as illustrated in Table 2.This observation is not applicable to overweight people as almost one third (30,8%) of children knew four or more overweight people and quite a small percentage of them (12.7%) did not know any individuals with that feature.Since we wanted to determine the amount of variance in the rejection of characters, and to determine the role of independent significant predictors in the amount of variance in the rejection of characters, we conducted a series of regression analyses, as shown in Table 3.
Among the number of known members of discriminated group variables, only knowing of Roma people is shown as a significant predictor; children who know more Roma people tend to reject the Roma character to a greater extent.
Furthermore, as significant negative predictors, perceived parental attitudes about Roma and friends' attitudes about all three groups were confirmed.
All regression models showed significance at least to the 5% level.The highest amount of explained variance is reached in the case of Roma character rejection (9.5%).

Discussion
The first goal of this study was to explore social distance toward a Roma child, a child in a wheelchair and an overweight child in comparison with social distance toward identical characters without those features.As expected, social distance toward the Roma child is significantly higher than social distance toward an identical character in the comparison group, which has been shown in the previously described literature (Franceško et al., 2005;Maričić et al., 2012;Mihić & Mihić, 2003;Previšić et al., 2004).The expectations concerning social distance toward an overweight child and a child in a wheelchair were not confirmed -which can be explained by the subtler prejudice toward them in comparison to prejudice toward Roma, or partially by social desirability, especially in the case of the child with a disability (Leutar & Štambuk, 2006;Antonak & Livneh, 2000).
In view of our second objective, we analyzed a few possible determinants of children's social distance toward stigmatized peers.More specifically, these determinants included the number of known members of certain groups and the perceived attitude of others in their immediate surroundings.
We can conclude that this study did not confirm the positive effect of contact (number of known people), and in the case of the Roma population, the study even found a negative effect, which is contrary to the hypothesis.We could clarify our findings by drawing on Petigrew's and Tropp's (2008) conclusion that anxiety reduction and empathy, which are products of increased intergroup contact, can take a major role in prejudice reduction.In the case of Roma, since they are mostly seen as thieves or beggars (Grgić, 2006), contact with them probably provokes anxiety.The research of Franceško et al., (2005) found that both children who have day to day contact with Roma as well as children that only have sporadic contact express extremely high social distance toward them.Various studies consistently show that Roma are represented as very poor, chronically unemployed, uneducated, members of large families and individuals without permanent residence (Ringold, Orenstein, & Wilkens, 2005).Additionally, there are further challenges for Roma who educate themselves and who try to come out from this circle of poverty, as they find it difficult to get a job, despite affirmative actions (Novak, Feldman, & Tomljenović, 2007).As a consequence, educated Roma people often try to conceal their ethnic identity, which results in people's unawareness of a "non-stereotypical" Roma (Novak et al., 2007).It is a great loss, because that awareness is a very important element in prejudice reduction (Low Stanic, 2014).
Concerning the other two "non-ethnic" features (knowing persons in a wheelchair and overweight persons), they do not have a significant role in the tendency to reject characters with those same features, but we have to consider that these features did not have an impact on the social distance itself.However, other researchers have found that the presence of negative stereotypes regarding being overweight generally increases during the transition from childhood to adolescence (Klaczynski et al., 2008), but it is probably more present through other forms of prejudice expression (e.g.teasing) than through social distance.
However, we must emphasize that we did not take into account frequency, intensity and amenity of contact, but only the number of known people from certain groups.Additionally, our study did not focus on the fulfillment of preconditions of contact, such as equal status and interdependence (Allport, 1954), but the aim was to determine whether knowing a person(s) with a certain feature has a positive influence on the acceptance of peers with the same feature.
Regarding the role of perceived attitudes of children's closest surroundings in relation to the tendency to reject members of certain groups (Table 3), results indicate that perceived friends' negative attitudes about particular groups were followed by a higher tendency to reject characters from that group, in all three cases, which is in accordance with our hypothesis.
On the other hand, the role of perceived parental attitudes was not significant, except for rejection of the Roma character.Based on the results, we can propose several conclusions: that the role of perceived friends' attitudes is stronger than the role of perceived parental attitudes regarding rejection of members of specific groups, which corresponds to Harris (1995); the perceived friends' attitudes, at age 10 and 11, have a greater influence regarding a wider range of groups in comparison with perceived parental attitudes and that surroundings have significant role in the rejection of Roma, in cases of both actors.We can therefore assume that prejudices about Roma are largely based on social norms, which confirms the Social Identity Development Theory (Nesdale, 1999(Nesdale, , 2004)).
Regarding the similarities and differences in the determinants of social distance toward different groups, results confirmed that all of the investigated determinants were not the same.Some of them could be regarded as more general, while the effects of some depend on a concrete group.Namely, the positive effect of perceived positive attitudes of friends on acceptance was confirmed for all investigated stigmatized groups.However, the effect of contact was found only with regards to social distance toward the Roma, which was contrary to our expectations regarding contact, but suggests the importance of norms.This result might be explained with the basics of the Social Identity Development Theory (Nesdale, 1999(Nesdale, , 2004)).
One of the limitations of this study is that the sample was not representative, but we used a nonrandomized sample of convenience instead.In addition, the research was only conducted in two of the largest Croatian cities. Consent for children's participation in the study may have been given by parents who are more likely to accept minorities, which may have led to a sample that was more accepting than the general population of children belonging to that age group.In this study, it should also be pointed out that children assessed the attitudes of their parents and friends, which may not have given us correct information about the actual attitudes of those actors.On the other hand, perceived attitudes probably have more impact when compared with the real attitudes of parents and friends.
Furthermore, in the questionnaire, we often only used one item to measure the desired concept, which reduces reliability.In addition, social distances toward different groups and the numbers of known members of the different groups are, on average, quite low.This implies that the distribution is asymmetric which requires greater caution when drawing conclusions.
In future research, it certainly would be useful to focus on other measures of contact, such as frequency, intensity, and amenity.As for the attitudes of the closest surroundings, it would be interesting to focus on the real attitudes (e.g. to examine the real attitudes of parents and a few friends) to gain insight into the role of the real versus perceived attitudes.Moreover, based on the fact that distribution of the results related to the social distance is asymmetrical, construction of the scale that is more focused on subtler forms of rejection is planned, since differences between the experimental and comparison group were mostly found on those items.It would be meaningful to examine the tendency to avoid children from the mentioned groups in other activities, such as joint projects at school, to identify prejudices associated with cognitive abilities of those children.
Based on our results we can give some recommendations that would lead to a better understanding and utility of contact, parental and peers' perceived attitudes.
Ignoring or avoiding contact with a person is seen as a relatively mild form of discrimination, when compared to direct aggression with physical consequences (Fishbein, 2002), which will probably be seen as more problematic, and stopped.However, it is necessary to bear in mind that the so-called milder forms of discrimination can lead to severe emotional infringement, and that a society which ignores these forms of discrimination creates "fertile soil" for the more severe forms.Given that most children grow up in a specific system (family and educational), which mainly emphasize similarities with others, there is a necessity to integrate a diversity consideration through various activities, workshops and programs, especially in elementary schools.In addition, we must emphasize that our results show that a much greater expertise and effort needs to be implemented in the field of combating ethnic prejudice.Specifically, in the Croatian context, Roma minority experiences are characterized by a relatively high degree of prejudice starting in childhood.Similar conclusions can be drawn in the wider European context, as multicultural and multiracial dimensions exist to a greater extent.This study also shows how peers' perceived attitudes have a great impact on child's creation and vision of somewhat different persons in their daily life, so education which include peer groups and positive use of peer influence could be good way reducing prejudice toward various stigmatized groups.
Currently, this study could be a contribution primarily to the creation of prevention programs for combating children's ethnic prejudices in this region of Europe, but also as an assessment of the necessary changes that need to be implemented concerning general prejudicial norms in a particular society.

Figure 1 .
Figure 1.Social distance toward characters in the experimental and comparison group

Table 1
Reliability and some factorial aspects of The Modified Social Distance Scale Related to Spending Leisure Time

Table 3
Regression analysis examining the role of contact and perceived parents' or friends' attitudes about various members of discriminated groups in children's tendency to reject a character from that group a a Beta standardized coefficients