Volume 3, Spring 2010

8/9/2019 Volume 3, Spring 2010

1/58

Psychologya t B e r k e l e y

Highlights of Insight and Excellence in Undergraduate R

University of California, BerkeleyVolume 3 - Spring 2010


2/58


3/58

Psychologya t B e r k e l e y

Highlights of Insight and Excellence in Undergraduate R

University of California, BerkeleyVolume 3 - Spring 2010


4/58


5/58

E ditor -in -C hiEf Alexandra Carstensen

E xECutivE E ditorsGwen Bassett

Rachel GevertzMina Yadegar

A ssoCiAtE E ditorsCorey BrownLaurel Chun

Chardee GalanIrene Hong

Lucy Mendoza

C ovEr Artist

Gwen Bassett

L Ayout d EsignEr Polly Chen

W Eb d EsignEr Zain Syed

Journal Members


6/58


7/58

Table of Contents

P sychology A t B erkeley is an Associated Students of the University of California (ASUC) sponsoredpublication. The ideas and opinions contained in this publication solely re ect those of the authorsand not the University of California.

v Editors Note A LExAndrA C ArstEnsEn

vii Prefaced AChEr K ELtnEr

1 Sources of Media Exposure and their Effects on AltruismC orEy b roWn

6 Socioeconomic Status-Based Differences in Perceptionand Self-ExpressionC Linton C hEn

11 The Involvement Gap: A Comparison of School Involvement Between Chinese, Chinese American, and Caucasian Parents t iffAny W Ai M An L AM & Audun d AhL

Guest Writers

23 Culture and Gender Variations in EmotionWord Uset hiAgo C.s. M ArquEs

31 Munchausens Syndrome by Proxy: Flaws in Diagnostic Criteria J uLiA y Ang

37 Appendix 43 About the Authors


8/58

iv


9/58

cholarly pursuit is more than a career in the traditional sense; it is a creative process, a constant learning experience, and an unrelenting challenge. Like anygreat achievement, a research project is a labor of love above all else. It is thisunique combination of passion and enduring dedication that inspires and enables

the greatest of scholars to pursue their work on every level from that of scienti crevolutions to school science fair projects.

The same driving force is evident even early in the academic careers of manyaspiring researchers, and imbues the following articles. Corey Brown exposes theimportance of individual differences in empathy and the prosocial potential of media in uence on altruism. Clinton Chen demonstrates the pervasive in uenceof socioeconomic status on perception and expression of the self. Tiffany WaiMan Lam explores the effects of culture on parental involvement and educationalsupport. Thiago Marques examines differences in emotional expression acrossgender and culture. Lastly, Julia Yang argues for a more rigorous diagnosis cri-

teria for Munchausens Syndrome.This publication is dedicated to its mission to celebrate insight and promote

excellence in undergraduate research. These articles represent only a fraction of the high-quality research and theory of psychology undergraduates and attest to the innovation and dedication of these skilled student researchers. We are proud to introduce them to you in this third edition of Psychology at Berkeley.

Alexandra CarstensenE ditor -in -C hiEf

S

Alex Carstensen

Editors Note


10/58

vi


11/58

ere in a new area of psychological science with vigorous attention to culture and the brain and genetics which represent newfrontiers of empirical discovery. Very often the best discoveries innew areas of inquiry come from young people, young scientists, and

young minds. Berkeley has a long-standing commitment to rigorous thesis work in laboratories by undergraduate students and many of our best undergrads have gone on to exciting careers as professorsand research scientists. This journal is really the fruition of that com-mitment to new developments in psychological science by Berkeleyundergraduates. I hope that Psychology at Berkeleygrows, prospers,and ful lls its noble mission.

Dacher KeltnerP rofEssor , d EPArtMEnt of P syChoLogy

u nivErsity of C ALiforniA At b ErKELEy

W

Preface


12/58


13/58

Sources of Media Exposureand their Effects on AltruismCorey BrownUniversity of California, Berkeley

While there is a vast body of literature dealing with the effects of media on behavior, the extant literatureexamining its effects on prosocial behavior is lacking, as most research is done on the negative role that media plays. This study examines the relationship between self-reported levels of empathy (using Inter-

personal Reactivity Index scales) and type of media coverage (either in the form of a news article or staticimages of a hit-and-run accident), with regard to their effects on altruism (measured as donations to ahypothetical organization). Data from 236 college-aged participants revealed signi cant effects of mediatype on altruism, moderated by empathic concern. Furthermore, the empathic concern subscale served asa signi cantly stronger predictor of altruism ( = .24, p = .01) than the other subscales, supporting the ideathat empathy is not only a necessity, but a prerequisite, to altruism, as suggested by the empathy-altruism

hypothesis (Batson & Shaw, 1991).The effect of media type on altruism was also largely moderated byspirituality and a main effect of media type was demonstrated across all conditions in that participants inthe reading condition were signi cantly less willing to donate to the hypothetical non-pro t organizationthan those in the viewing condition ( = .22, p = .008). The results of this study build upon the extant litera-ture on media by revealing mediating effects of empathy and spirituality on altruism, which suggest that media can, in some contexts, have prosocial bene ts, in contrast to the damaging effects it is known for.

t can be a rare occurrence to come across a story in the newsmedia which tells of something good in the world. More often,those who watch the news on television or read news articlesare subjected to stories of violence, death, tragedy, and politicaland social tension. Prime-time television portrays many violentacts (Gerbner, Gross, Morgan, & Signorielli, 1986), yet some re-search points out that it commonly displays individual prosocialtendencies as well (Lee, 1988). One of the most notable sourcesof inspiration for the study of prosocial emotions, particularlyaltruism, is the case of Kitty Genovese, who was brutally rapedand murdered while none of approximately one dozen witnessestried to intervene or call for help (Greening, 2005; Stohl, 1987).An example such as this begs the question of why people behaveprosocially, and what it is that makes them refrain from doing so.

Studies of altruism, defined as the sacrificial behavior of oneself for the benefit of anothers condition (Bell, 2008; deWaal, 2008), have become widespread, delving into many relatedaspects of social psychology, including empathy. However, thequestion of whether or not empathy is needed for one to act inan altruistic manner is still debated (Batson & Shaw, 1991; deWaal, 2008). Altruism can be considered one of the most salientprosocial behaviors, but empathy has also been shown to play akey role. Defined as the bridge between egoism and altruism,between the self and others (Hodges & Klein, 2001; Hoffman,1981), empathy is often considered a necessity to behaving al-truistically (Batson & Shaw, 1991). As Batson and Shaws (1991)empathy-altruism hypothesis postulates, the initial decision of whether one should help another is determined primarily by thepresence of feelings of empathy for the other person. However,another factor raised by this hypothesis that merits consider-

ation is social exchange, which causes the individual to weigh

the potential social costs and rewards of his or her actiogiven social situation in order to maintain cohesion.

Some researchers argue that such evaluations forceacting in accordance with an altruistic ideal, in light ofexchange, to attempt to maximize the overall benefits whiimizing the overall costs of the situationboth to themand others (Deckop, Cirka, & Andersson, 2003; GouldnePeople behaving altruistically are assumed to be acting mas a result of their own egoism and pleasure (Batson &1991). This may sound rather egoistic to those who denexistence of true altruism (i.e., selfless concern for otherbeing) in favor of hedonistic explanations of prosocial bSuch critics argue that, despite anyones altruistic actionresulting feeling of pleasure for doing something goodently precludes acts of true altruism (ibid). However, BatShaw (1991) point out that there is often confusion betwedifferent forms of hedonism, which gives rise to such an ation. The strong form of altruistic hedonism maintains thachievement of pleasure is always the goal of an action; tform states that simply achieving a goal will bring pleasurploying the strong form of hedonism is inconsistent wselflessness of true altruism. In considering the weak formever, it becomes apparent that one can take pleasure meachieving a goal that increases anothers overall welfare.

Further research has identified additional factors whicplay an important role in moderating altruistic behavioras the finding that help is given more to those of in-groutus than of out-group status (Strmer, Snyder, & Omoto,and that religious affiliation predicts prosocial acts (SaPichon, Trompette, Verschueren, & Dernelle, 2005). As a

highly salient factor in our decision-making, the news me

s ourCEs of M EdiA E xPosurE And thEir E ffECts on ALtruisM | Corey Brown

I


14/58

2

2010PsychologyatBerkeleyhttp://www.ocf.berkeley.edu/~ujpb


serves more consideration as a potential moderating factor of prosocial behaviors such as altruism.

Much of the current research has largely focused on thedetrimental influences that media has on individuals, particu-larly emphasizing studies of its effects on children (e.g., mediaviolence and aggression; see Berkowitz, 1993; Boxer, Huesmann,Bushman, OBrien, & Moceri, 2009; Huesmann, Moise-Titus,Podolski, & Eron, 2003). As Barbara Wilson (2008) points out,

many studies have focused on the negative effects that media hason the behavior of youth and adults, yet there has been littleemphasis on the prosocial effects, which can be revealed throughsimilar research. The large majority of research done on the me-dia and prosocial behavior focuses on the impact of the media onchildrens prosocial tendencies and moral aptitude (Anderson &Bushman, 2001; Cosgrove & McIntyre, 1974). Undoubtedly, itwould be interesting and informative to explore the impressionthat television and other forms of media have on adults as well.

This study seeks to contribute to existing literature by ex-amining the medias influence on prosociality, with respect tothe effects of written or visual media on altruistic behavior. Itis predicted that subjects will be more likely to behave altruisti-cally after viewing pictures of a tragic event than after readinga news article describing the event. This prediction is based onthe argument that visual imagery provides a deeper and moresalient understanding of need, thereby provoking a strongerempathic reaction. Previous findings which show that viewingimages evokes a stronger affective response than reading wordshave provided a foundation for this theory (Maddock, Garrett, &Buonocore, 2003; Newhagen & Reeves, 1992). Because partici-pants in the reading condition will not be exposed to an explicit,salient image of the disturbing event depicted in the viewingcondition, these subjects are expected to exhibit less empathicconcern.

MethodsParticipants

The participants were two hundred and thirty-six collegestudents (92 male, 144 female) from the University of Califor-nia, Berkeley, who were randomly assigned to one of two condi-tions: the viewing condition, in which participants saw images of a tragic accident, or the reading condition, in which participantsread an article about the same incident without viewing the im-ages. Participants ages ranged from 18 to 60 (M = 22.9 years,SD = 6.75 years). Subjects participated voluntarily and did notreceive any class credit or monetary compensation for takingpart in the study.

Procedure The 49-question survey was posted online for participants

to complete at their leisure and subjects were contacted andrecruited through an e-mail list serve for psychology majors.Participants read about or viewed images from a hit-and-run ac-cident in Harford, Connecticut. The media (images or writtenstory) portrayed a 78-year-old man as he was tossed into theair and then fell onto the street after being hit by a car, whichsped away following the incident. No one offered to help the man,and while many looked from the sidewalks or passed by in cars,the victim continued to lie motionless. In one condition, subjectswere shown still images of the recent news story, while in the

other, participants read a news article describing the acwithout pictures. The article, written by Delfiner (2008adapted from a news piece featured in the New York Pohad been edited to omit mention of the release of a videoincident (from which the still images were created), allowreader to focus solely on the accident.

Each participant answered a series of questions thasured his or her scores on the Interpersonal Reactivity

scales (IRI; Davis, 1980, 1983). These scales include mof empathic concern, fantasy, perspective-taking, and pdistress. A short distracter task which followed the compof the index asked the subjects to rate the trustworthinea few faces. Next, participants were randomly divided igroups, one of which viewed the images while the other rarticle. Subjects then received a request for a hypotheticaetary donation to a fictional organization related to hit-anvictim support. Subjects provided the dollar amount thawould be willing to donate and were given the option of a letter of support to the organization. There was also tion in which the subjects filled in how much time theyhypothetically be willing to volunteer for the same organiThe final section of the survey was used to obtain demoinformation about the participants, including their level otuality, socioeconomic status, and political views.

Measures Levels of altruism were weighed according to su

willingness to write a supportive message to and voluntthe organization. The amount of money pledged as a doprovided a general indication of their support and empatthough participants were aware that their donations were shypothetical, this measurein accordance with past findis considered to be an accurate indicator of how mucheach participant would actually donate in a real-world si(Benz & Meier, 2008).

In addition to the measures for altruism obtained thsubjects willingness to donate, volunteer, or write a letsupport, participants completed the Interpersonal Reactivdex (see appendix). Responses on the IRI (Davis, 1980ranged from 1 to 5, with 1 corresponding to does not dme well, and 5 to describes me very well. The index cfour seven-item subscales, each examining a different feempathy. The empathic concern subscale measures particfeelings of warmth, compassion, and concern for otherson agreement with statements such as When I see someoing taken advantage of, I feel kind of protective towards

The items on the subscale for fantasy probe the participanity to identify with characters in movies, plays, novels, anfictional works (e.g., After seeing a play or movie, I havthough I were one of the characters). The perspective-scale measures respondents tendency to take on anotherof view (e.g., I sometimes try to understand my friendby imagining how things must look from their perspectivepersonal distress scale measures participants feelings of and discomfort when seeing anothers suffering (e.g., see someone who badly needs help in an emergency, I go es.). The overall IRI scale has proven to be a valid and instrument for assessing the individual dimensions of e(Davis, 1980, 1983).


15/58

ResultsThe data on monetary donations to the organization and

willingness to volunteer were analyzed using a multiple linearregression. With regards to the subscales on the IRI, the meanscore across both media conditions for the empathic concern sub-scale was 3.79 (SD = .61). The fantasy subscale produced a meanscore of 3.58 (SD = .68), while for perspective taking it was 3.53(SD = .58) and 2.72 (SD = .73) for personal distress. The overallIRI produced a mean score of 3.42 (SD = .38). Relative to themean scores of the other subscales, the lower average for thepersonal distress subscale indicates that respondents experiencedrelatively less discomfort when witnessing the negative experi-ences of others, compared to the strength of their tendenciesto take others perspectives, identify with fictional characters,and experience empathic concern. This is consistent with Davis(1980) observation that decreased personal distress is associatedwith increases in the other subscales. This can be explained bythe respondents levels of differentiation between themselves

and others, because seeing anothers distress will evoke feelingsof other-oriented concernrather than self-centered empathicdistressto a greater degree in those who differentiate more be-tween self and others.

Regarding the measured variables, the mean monetarypledge to the organization in support of hit-and-run victimswas significantly different between those in the reading (M =$15.82, SD = $27.07) and viewing (M = $87.38, SD = $280.80)conditions (t(234) = 2.65, p = .009) while controlling for income.Neither the mean amount of time per month participants werewilling to volunteer for the organization nor the offering of aletter of support differed significantly between the two mediaconditions.

In examining empathic concern, those in the viewing condi-tion were willing to donate more of their money to the cause,behaving with significantly more altruism than those in the read-ing condition ( = .16, p = .03). Within the viewing condition,those who were already highly empathic, as determined by theirIRI scores, were likely to donate more money than those in theviewing condition who exhibited low levels of empathic concern( = .24, p = .01; see Figure 1). This interaction between amountof empathic concern and media type demonstrates that individu-al scores on the empathic concern subscale are a stronger predic-tor of the amount of money participants are willing to donate.

There was no main effect of fantasy IRI scores on monetarydonation and no interaction between fantasy and media type; the

media condition was the only significant determiner of wor not participants were willing to donate their money ( = .20p = .01). This was also the case with the other two subscaperspective-taking and personal distress (respectively, = .18, = .01; = .19, p = .01).

Altruistic tendencies, as measured by the IRI and modby media condition, were not good indicators of volunteeoffered for the organization. Neither media condition nor the IRI subscales were significantly related to the number unteer hours offered by subjects. This result may be confby the busy schedules of college students, who might beto find substantial amounts of time to dedicate to assistiorganization.

The optional letter portion of the survey was used tovide a measurement of prosocial behavior, as indicatedchoice to write a letter of support. The prediction that ppants in the viewing condition would be more likely to conthe letter than those in the reading condition was support

the data; the chi-square goodness of fit test revealed thawas a higher frequency of individuals offering a letter oport in the viewing condition than in the reading conditi2= 83.97, p < .0001).

Further multiple regressions were conducted in ordanalyze the predictive value of several demographic chaistics on prosocial behavior. These factors included ageder, spirituality, political views, and socioeconomic statuwhich proved to be non-significant in determining the aof money pledged and the number of hours offered to voHowever, a significant interaction arose in that those whothemselves as highly spiritual were significantly more libe willing to donate money to the organization when in the

ing condition, as compared to reading ( = .22, p = .008). Suprisingly, those in the reading condition who were very spwere likely to donate less (M = $11.13 donated), than thowere not very spiritual (M = $25.51 donated; see Figure 2

DiscussionThis study examined the determinants of altruisti

havior as moderated by empathy and type of media exand sought to reveal which of the numerous factors relaempathy play the largest role in predicting prosocial behIt was hypothesized that participants exposed to imagesindividual in need would be more likely to behave in an tic manner than those who encountered the situation in w

Volume 3 | Spring 2010

Figure 1: Donations and Empathic Concern: The interaction between empathic concern and media type in predicting donations offered to a hypothetical organization to help victims of hit-and-run accidents. Those with high em-

pathic concern in the viewing condition were significantly more likely to donate larger sums of money than their counterparts in the reading condition ( = .24, p = .01), whereas those with low empathic concern donated similar amounts, regardless of media condition.

Reading

Viewing


16/58

4



form. This prediction was based upon the assertion that seeingpictures of an event, rather than reading a news story, providesa better understanding of the sincerity of the situation (Mad-dock, Garrett, & Buonocore, 2003; Newhagen & Reeves, 1992).Previous research (ibid) suggests that images convey emotionalmeaning more directly than words, leading to the prediction thatreading would result in a weaker empathic response than view-ing images of an event, because it would produce a more distant

connection to the event. This effect of media type is attributed tothe idea that images are more explicit in their representation of what is depicted and consequently create more realistic internalexperiences and emotional reactions than written words, whichrequire extraction of symbolic meanings to create any emotionalresponse (Newhagen & Reeves, 1992).

Study results indicated that viewing images had a signifi-cant moderating effect on empathy as a predictor of participantswillingness to donate money to the organization; each of the in-dividual subscales for empathy on the IRI accurately predictedaltruism in subjects in the viewing condition. Of particular in-terest was the resulting linear regression for empathic concern,with = .24 and p = .011. This result indicates that those whoscored as highly empathetic would be more likely to offer a dona-tion after viewing the images compared to reading the article,whereas participants who scored low on empathic concern wereequally likely to donate in the viewing and reading conditions.This finding is perhaps best explained by the empathy-altruismhypothesis (Batson & Shaw, 1991), which proposes that what de-termines whether or not someone will behave altruistically to-ward another is based upon that individuals feelings of empathyfor the other. Thus, the theory suggests that without empathy,altruism would not be initiated, and thus could not be affected byvarious factors, such as media type.

Linear regression scores for subscales other than empathicconcern, especially those for fantasy ( = .20, p = .011), pro-vided further insight into the prosocial trends of participantsin the viewing condition. The fantasy subscale, as a measure of respondents tendencies to identify with fictional others, was ex-pected to predict participants propensities to transpose them-selves into the hit-and-run situation presented. Correspondingly,it was hypothesized that high fantasy participants would makecomparable donations in both the reading condition and viewingconditions (i.e., there would be no significant difference betweencontributions from high fantasy participants in reading vs. view-ing conditions), based on the idea that they would be especially

likely to readily identify with another in writing. Overalfantasy participants in the reading condition did indeedsubstantial contributions, however, the hypothesis was noported in that high fantasy participants contributed signifimore in the viewing condition, as opposed to reading (p =

While significant results were obtained for the perspetaking ( = .18, p = .018) and personal distress scales ( = .19p = .012), these scales were not as strong predictors of al

as the empathic concern and fantasy scales. This may berandom variability within the sample population, but can examined in relation to the empathy-altruism hypothesis (& Shaw, 1991). According to this hypothesis, empathic is the primary factor in determining altruistic behavior, wthe other factors of the IRI subscale are only secondary dicting whether someone will act altruistically.

There was no significant relationship between politifiliation and willingness to donate, which suggests that act altruistically regardless of their political values. Additthere was no interaction between media condition and pviews in determining altruistic tendencies, signifying theffects of differing types of media on altruism are not ated by political stance. The same pattern was found regsocioeconomic status, indicating that no one political viewcioeconomic level is prone to more prosocial behavior thother.

The results in this study concerning spirituality correswith what Saroglou, Pichon, Trompette, Verschueren, &nelle (2005) have found in their analysis of prosociality agion. Those that were more spiritual were more likely to bas altruistic and prosocial by their peers in the case of bquaintances and strangers. As the findings of the presendemonstrate, participants who rate themselves as beingspiritual are also more likely to act altruistically (as meby their hypothetical donations) after viewing images of matic event than after reading an article about the even =.22, p = .008). This finding parallels previous research shthat spiritual people are generally observed to be more cial, scoring higher on measures of altruism as well as e(Saraglou, 2005). Further research should aim to provide thorough explanation of why this is the case.

Limitations A few limitations to the study should be taken into c

eration. Because the surveys were administered online,

Figure 2: Donations and Spirituality: The interaction between spirituality and media type in

predicting donations offered to a hypothetical organization to help victims of hit-and-run accidents. Highly spiritual participants in the viewing condition contributed larger sums of money than their counterparts in the reading condition ( =.22, p = .008), while less spiritual individuals donated similarly regardless of media condition.

Reading

Viewing


17/58

Zillman (Eds.), Perspectives on media effects (pp. 17-40). Hillsdale, NJ:Lawrence Erlbaum Associates.

Gouldner, A. W. (1960). The norm of reciprocity: A preliminary statement. American Sociological Review, 25(2), 161-178.

Greening, T. (2005). Commentary. Journal of Humanistic Psychology, 45(2),140-143.

Hodges, S. D. & Klein, K. J. K (2001). Regulating the costs of empathy: Theprice of being human. Journal of Socio-Economics, 30(5), 437-452.

Hoffman, M. L. (1981). Is altruism part of human nature? Journal of Person-ality and Social Psychology, 40(1), 121-137.

Huesmann, L. R., Moise-Titus, J., Podolski, C. L., & Eron, L. D. (2003).Longitudinal relations between childrens exposure to TV violence andtheir aggressive and violent behavior in young adulthood: 1977-1992.Developmental Psychology, 39, 201-221.

Lee, B. (1988). Prosocial content on prime-time television. In S. Oskamp(Ed.), Television as a social issue (Applied social psychology annual,

Vol. 8, pp. 238-246). Thousand Oaks, CA: Sage Publications, Inc.Maddock, R. J., Garrett, A. S., & Buonocore, M. H. (2003). Posterior

cingulate cortex activation by emotional words: fMRI evidence from avalence decision task. Human Brain Mapping, 18(1), 30-41.

Newhagen, J. E. & Reeves, B. (1992). The evenings bad news: Effects of compelling negative television news images on memory. Journal of Communication, 42(2), 25-41.

Saroglou, V., Pichon, I., Trompette, L., Verschueren, M., & Dernelle, R.(2005). Prosocial behavior and religion: New evidence based on pro- jective measures and peer ratings. Journal for the Scienti c Study oReligion, 44(3), 323-348.

Stohl, M. (1987). Outside of a small circle of friends: States, genocide, masskilling and the role of bystanders. Journal of Peace Research, 24(2),151-166.

Strmer, S., Snyder, M., & Omoto, A. M. (2005). Prosocial emotions andhelping: The moderating role of group membership. Journal of Person-ality and Social Psychology, 88(3), 532-546.

Wilson, B. J. (2008). Media and childrens aggression, fear, and altruism. The Future of Children, 18(1), 87-118.

Supplementary material for this article is availableonline at http://www.ocf.berkeley.edu/~ujpb/past_editions.html


pants were able to complete them in many different settings andat any time of day. This limits the internal validity of the studyby creating extraneous environmental factors that could not becontrolled for or addressed. Furthermore, although participantswere randomly assigned to each condition, many of the subjectswere friends and acquaintances of the researcher, creating a se-lection bias, whereas a more representative sample would haveallowed for increased accuracy in generalizing the findings to the

general population.Software limitations prevented the use of video instead of static images of the event in the non-reading condition. Howev-er, research by Detenber, Simmons, and Bennett (1998) has sug-gested that stronger emotional responses can be obtained frommotion pictures than from still images, so utilization of a videocondition would likely enhance data acquisition in future studies.Lastly, the methods employed to measure empathy and altruismrelied entirely on self-reports, which provide only highly subjec-tive evaluations. Thus, the quality of the data could be improvedby using raters to score observable aspects of participants em-

References Anderson, C. A. & Bushman, B. J. (2001). Effects of violent video games on

aggressive behavior, aggressive cognition, aggressive affect, physi-ological arousal, and prosocial behavior: A meta-analytic review of thescienti c literature. Psychological Science, 12(5), 353-359.

Batson, C. D. & Shaw, L. L. (1991). Evidence for altruism: Toward a plural-ism of prosocial motives. Psychological Inquiry, 2(2), 107-122.

Bell, G. (2008). Selection: The mechanism of evolution. New York: OxfordUniversity Press.

Benz, M. & Meier, S. (2008). Do people behave in experiments as in theeld?Evidence from donations. Experimental Economics, 11(3), 268-

281.Berkowitz, L. (1993). Aggression. New York: McGraw-Hill.Boxer, P., Huesmann, L. R., Bushman, B. J., OBrien, M., & Moceri, D.

(2009). The role of violent media preference in cumulative developmentrisk for violence and general aggression. Journal of Youth and Adoles-cence, 38(3), 417-428.

Cosgrove, M. & McIntyre, C. W. (1974, March). The in uence of mister-ogers neighborhood on nursery school childrens prosocial behavior.Paper presented at the Biennial Southeastern Conference of theSociety for Research in Child Development, Chapel Hill, NC.

Davis, M. H. (1980). A multidimensional approach to individual differences inempathy. JSAS Catalog of Selected Documents in Psychology, 10, 85.

Davis, M. H. (1983). Measuring individual differences in empathy: Evidencefor a multidimensional approach. Journal of Personality and SocialPsychology, 44, 113-126.

de Waal, F. B. M. (2008). Putting the altruism back into altruism: The evolu-tion of empathy. Annual Reviews of Psychology, 59, 279-300.

Deckop, J. R., Cirka, C. C., & Andersson, L. M. (2003). Doing unto others: The reciprocity of helping behavior in organizations. Journal of Busi-ness Ethics, 47(2), 101-113.

Del ner, R. (2008, June 6). Hit-&-run-&-watch: Heartless Hartford crowdsignore car-plow victim. New York Post. Retrieved from http://www.nypost.com/seven/06062008/news

/regionalnews/hit__run__watch_114208.htmDetenber, B. H., Simons, R. F., & Bennett, G. G. (1998). Roll em!: The ef-

fects of picture motion on emotional responses. Journal of Broadcast-ing and Electronic Media, 42(1), 113-127.

Gerbner, G., Gross, L., Morgan, M., & Signorielli, N. (1986). Living withtelevision: The dynamics of the cultivation process. In J. Bryant & D.

pathy and altruism or by gathering information from obstion of real behavior, which could then be studied and anaenabling more objective measurements.

Future Directions Building on the findings of this study, further rese

should be conducted to investigate the importance of spirity in determining altruism and to further examine why t

rated as highly spiritual donated less money after readinnews article than after viewing the images. It would alvaluable to study how the amount of time that people arposed to media affects their tendencies to act prosociallywould produce information from a perspective that is notiunderrepresented in the vast body of research on the medlarge majority of which has been devoted to characterizinthe negative effects it has on society. As the results of thisrobustly convey, media has numerous potentially prosocfluences that are too often ignored.


18/58

6


sEs-b AsEd d iffErEnCEs in P ErCEPtion And s ELf -E xPrEssion | Clinton Chen

Socioeconomic Status-Based Differencesin Perception and Self-Expression

Clinton ChenUniversity of California, Berkeley

Socioeconomic status (SES) is often linked to health and well-being, but its effects on perception and self-expression are just beginning to be explored. Recent theoretical work on social status (Keltner, Grunefeld,& Anderson, 2003) suggests that upper SES people tend to be more dispositionally-minded and lower SES

people tend to be more contextually-minded. Disposition refers to ones characteristic attitude or mental outlook, while context refers to ones background and situation. In this study, the pronouns that partici-

pants used during an interaction with a stranger were coded for their reference to disposition or context.Disposition was operationalized as pronoun usage for references to the participant themselves (e.g., I, me, my) and context was operationalized as pronoun usage for references to the background or situa-tion (e.g., you, they, their, we). In support of the hypothesis, lower SES participants were found tobe more likely to use contextual pronouns when describing themselves. Implications for future researchon SES, as well as the consequences of this SES-based pattern of perception, are discussed.

ocioeconomic status (SES) is defined by ones material re-sources and access to social institutions such as education andhealthcare (Lachman & Weaver, 1998; Oakes & Rossi, 2003).People of low SES have material constraints that render themvulnerable to signs of poor health and well-being such as de-pression, lower self-esteem, higher anxiety, greater incidence of

mental illness, and overall lower motivation (Adler, et al., 1994;Link, Lennon, & Dohrenwend, 1993). Since individuals of highSES have high access to material resources, they are more like-ly to enjoy indicators of positive health and well-being such ashigher self-esteem, lower anxiety, lower incidence of mental ill-ness, and heightened motivation. These studies imply that SESimpacts individuals health and well-being, which in turn affectstheir perceptions of themselves and the outside world (as isthe case with depression, anxiety, self esteem, and motivation).However, few studies to date have examined how SES impactsthe ways in which people perceive and relate to the social world.

Differences in SES should lead to different patterns of so-cial perception. The hypothesis is that upper SES individuals willdescribe themselves and perceive the social world in ways thatreflect their overall focus on peoples dispositions (i.e., an indi-viduals mental outlook or characteristic attitude). In contrast,individuals of lower SES will describe themselves and perceivethe social world in ways that reflect their overall focus on theircontexts.

The Impact of Culture on Psychological Processes At one time, psychologists were most interested in univer-

sal explanations of behavior. One example is the fundamentalattribution error (FAE), (Ross & Sicoly, 1979), which sought toexplain the tendency for a person to consistently attribute causesof the behavior of others to internal, rather than external or

situational factors. At first, this phenomenon was thougha universally-held truth of social perception, and much data supported this view (e.g., Malle, 1999; Ross & Sicol

However, over the past two decades, cultural differenpsychological processes previously considered universbeen uncovered. The converging results found in the e

studies of FAE used participants exclusively from Westetures. More recent work, with samples from other culturerevealed cultural differences in attribution style (e.g., MPeng, 1994; Miller, 1984). For example, Western and Easparticipants were found to have differences in attention jects, explanations of events, and self-descriptions (MaNisbett, 2001; Choi, Dalal & Kim-Prieto, 2001; TriandisSuch findings called claims of universality into questiopushed psychologists to take the cultural context more serIn fact, as understanding of cultural and situational factcreases, claims of universality in psychology seem less plausible.

SES as a Cultural Context As in previous research comparing social environ

this-study argues that culture has a profound impact on hindividual thinks about and perceives the social world.that a persons SES affects the environments they interacit is proposed that SES greatly influences the ways in whdividuals think about themselves and perceive the socialThe hypothesis of this study is that SES-based differencimpact self and social perception; upper SES individutend to focus on their own dispositions, whereas lower SEviduals will tend to focus on their context.

Individuals of upper SES have life circumstances thdramatically from those of lower SES. Higher SES indi

S


19/58

are characterized by better access to healthcare and education,superior health, and increases in positive affect, social standing,sense of control, and resources. Individuals of lower SES arecharacterized by a lack of access to healthcare and education,lower health, and decreases in positive affect, social status, senseof control, and resources (Oakes & Rossi, 2003; Adler et al.,1994; Lachman & Weaver, 1998; Snibbe & Markus, 2005; Keltner,Gruenfeld, & Anderson, 2003; Domhoff, 1998).

SES, Status, and Dispositionalism vs. Contextualism Parallel research on social power, a measure of ones ability

to control the surrounding environment, anticipates the abovehypotheses. A review of research on social power, including SES-based indicators of power, suggests that high-power individuals(who are usually individuals of upper SES) tend to focus more ontheir own dispositions, goals, and emotions, whereas low-pow-er individuals (usually individuals of lower SES) tend to focusmore on their own contexts, others goals, and others emotions(Keltner, Gruenfeld, & Anderson, 2003). High-powered individu-als tend to report elevated mood (Watson & Clark, 1997), whilelow-powered individuals report higher levels of negative mood(Link, Lennon, & Dohrenwend, 1993). Since previous researchsuggests that positive mood facilitates goal-oriented behavior(Davidson, 1992; Higgins, 1997), it follows that high-poweredpeople would be more likely to actively pursue their own goalsand thus be more likely to focus on their disposition. High poweris also associated with prosocial behavior, as high-powered in-dividuals tend to engage in higher levels of touching (Major &

Heslin, 1982). These findings suggest that high power is ated with behaviors related to ones own thoughts and wpossibly irrespective of the wishes of others. Conversepower is associated with inhibition, for example, inhibited(Hosman, 1989; Holtgraves & Lasky, 1999) and a tendperceive threat. For example, children of low sociometriare more likely than children of high sociometric status ceive threat in ambiguous situations (Schwartz, Dodge,

1993). In addition, adults of lower SES tend to report motrust in others (Mirowsky & Ross, 1983) as well as highries about crime (Riger, Lebailly & Gordon, 1981) than ahigher SES. These studies suggest that people of lower Smore attention to the context because they are more suscto dangers from it.

Further research supporting the hypothesis come fromies on SES and sense of control. Sense of control is dean individuals ability to change his or her own outcomesoutcomes of others (Lachman, 1986). Upper SES has besistently shown to be associated with a higher sense of pcontrol (Johnson & Krueger, 2006; Lachman & WeaverFor example, high-school-educated participants were foprefer music emphasizing a lack of control over a harshnal environment, while college-educated participants prcultural products that emphasized control, uniqueness, anencing others (Snibbe & Markus, 2005). Given that indiwith low SES tend to lack a sense of control, it is reasonexpect that those individuals will focus more on their callowing them to prepare for potential dangers. Conversecause people with high SES feel more in control, they pected to focus on their own dispositions, worrying lessthe context and concentrating instead on themselves anpossible rewards they may obtain.

Taken together, the above evidence suggests that SES-differences in status, sense of control, and available reshape the ways in which an individual perceives and inwith the social environment. It is predicted that individuupper SES, because of their greater influence and contrtheir circumstances, will focus on what drives their socicomes: their own dispositions and abilities. In contrast, inals of lower SES, because of their relative lack of conresources, will tend to focus on what most often drives thcial outcomes, namely, the surrounding context.

SES-based Differences in Self Expression The present study examines SES-based differences i

expression. The case has been made that SES has an ef

how individuals view the world as a result of determiningaspects one must attend to. It is predicted that people of SES will attend more to their dispositions and less to theground or context. Additionally, people of lower SES contextually-minded (i.e., more aware of their backgroucontext).

Based on previous work in culture and disposition vtextualism (Kitayama, Duffy, Kawamura & Larsen, 2003dis, 1989), this study is designed to test SES-based diffin perception and self-expression. It examines self-descduring a getting acquainted interaction between two pIt is predicted that participants with upper SES will bedispositional, describing themselves using pronouns that


Table 1: Frequency of measures of family income, parental education,ethnicity, and gender


20/58

8



their focus on internal thoughts and wishes. In contrast, partici-pants with lower SES will be more contextual, describing them-selves with pronouns that reflect their focus on involvement withgroups or others.

The current literature on SES primarily focuses on its ef-fects in the realms of health and mental well-being. The affectof SES on how people think about and view the world has beenlargely unexamined. This study adds to the current literature

by investigating how SES influences the way people perceive theworld and describe themselves. While other studies have inves-tigated the relationship between SES and self-definition (e.g.,Hart & Edelstein, 1992), the current study is the first to examineparticipants self-definitions when becoming acquainted with astranger.

MethodThis study looked at the relationship between participants

SES and their self-descriptions. Participants in the study met inpairs for the first time for a five-minute getting acquainted in-

teraction, in which each participant was asked to describe them-selves to their partner. The initial self-descriptions were tran-scribed, then coded for the use of first-person singular (e.g., I,me, my) and third-person pronouns (e.g., them, they, he/she, him/her). Pronoun usage is used to characterize contex-tual versus dispositional tendencies, as it indicates how often a

person refers to him or herself as opposed to others. It is ex-pected that high SES individuals will use more first-person sin-gular pronouns, referring to their own dispositions, whereas lowSES individuals will use more third-person pronouns, referringto others in the surrounding context, during the interaction.

Participants Participants were 106 undergraduate students enrolled in

a large public university. Students participated in the study inexchange for course credit. Gender and SES varied within eachpair. The age group of the sample was approximately 19-21 yearsold. The largest ethnic group in the sample was Asian Americanat 44.3%, followed by European American at 42.5%. The remain-ing sample was made up of African American, Native American,Latino and other ethnicities at 13.2%.

Procedure Each session of the experiment involved two participants

meeting for the first time. They were brought into a 10 by 14foot room with two chairs positioned a few feet away from andangled toward each other. A camera in the back of the roomfilmed the interaction.

In the first part of the experiment, a getting acquaintedinteraction, participants were instructed to take turns describ-ing themselves to each other. The participants were told thatgetting acquainted before a job interview has been shown to

improve the results of the job interview itself, and wereto describe themselves and then talk casually until the ementer returned. If they ran out of things to talk aboutwere instructed to ask each other questions from a list owhich included items such as What extra-curricular actclubs, or jobs do you participate in?, How would you yourself?, and What is your current major at Cal, and wyou like most about it?.

After giving the instructions, the experimenter left thefor five minutes before returning. Participants then tooin a mock job interview and allocation task as part of sexperiments, before completing a demographic questiowhich included questions about socioeconomic status. Thanalyzes the pronoun usage data collected during the gacquainted portion of the experiment, because participascribed themselves freely to each other at that time, wherexperimenter asked specific questions during the intervietion.

Measuring Pronoun Usage The initial self-descriptions that occurred in the dya

teraction were coded for pronoun usage. The interaction wcorded by a camera and subsequently, the getting acqupart of the interaction was transcribed into text. Then, ablind to participants SES counted all occurrences of first

singular (I, me, my), and third-person pronouns they, he/she, him/her).

ResultsOne participant was excluded because of substantia

ing data. The remaining sample included 105 participants

Objective SES and Dispositional vs. Contextual Pronoun Usage To measure objective SES, the annual income was ass

numerical value of 1-7 (M = 5.48, SD = 1.78), with highbers indicating higher income. In measuring parental edutwo numbers were assigned: one for the mother (M = 2.530.62) and one for the father (M = 2.64, SD = 0.62), with va1-3, in which a higher number indicates higher educationthese three measures, a sum composite measure (averagetal education plus annual income) of objective SES was(out of a possible 10 points, M = 5.97, SD = 1.99). PleaseTable 1 for complete demographic information.

To investigate whether higher or lower SES was corrwith dispositional pronoun usage (I/me/my), (M = 9.2= 5.92), the correlation coefficient was computed betwnumber of first-person singular pronouns used per partiand their objective SES. There was no significant correlattween objective SES and number of dispositional pronouper participant However, there was a correlation (r = -0.10.048) between the number of contextual pronoun words

Table 2: Correlations between Objective SES and Pronoun Usages

Contextual Pronouns Dispositional Pronouns

-.194* -.002Correlation Coef cient

Note: *p < .05


21/58


Table 3: Regression of Contextual Pronoun Usage on Total Words and Objective SES

Note: *p


22/58

10



a prevalent element of their lives. This study focused on provid-ing evidence for this claim, showing that SES-based differencesdo indeed impact the way people perceive the world and describethemselves. Understanding SES as a cultural variable is crucialbecause of its impact on health, well-being, and as shown, onperception and self-definition. This study, among numerous oth-

References Adler, N. E., Boyce, T., Chesney, M. A., Cohen, S., Folkman, S., Kahn, R.

L., & Syme, S. L. (1994). Socioeconomic status and health: The chal-lenge of the gradient. American Psychologist, 49, 1524.

Anderson, C., & Galinsky, A.D. (2006). Power, optimism, and risk-taking.European Journal of Social Psychology, 36, 511-536.

Agnew, C.R., Van Lange, P.A.M., Rusbult, C.E., & Langston, C.A. (1998).Cognitive interdependence: Commitment and the mental representa-tion of close relationships. Journal of Personality and Social Psychol-ogy, 74, 939-954.

Atkinson, J. W. (1964). An Introduction to Motivation. Princeton, NJ: VanNostrand.Brewer, M.B., & Gardner, W. (1996). Who is this We?Levels of collective identity

and self representations. Journal of Personality and Social Psychology, 71,83-93. Choi, I., Dalal, R., Kim-Prieto, C., & Park, H. (2001). Cultureand judgment of causal relevance. Journal of Personality and SocialPsychology, 84, 46-59.

Davidson, R. J. (1992). Emotion and affective style: Hemispheric substrates.Psychological Science, 3, 3943.

Domhoff, G.W. (1998). Who Rules America. Mountain View, CA: May eldPublishing.

Exner, J.E., (1973). The self focus sentence completion: A study of egocen-tricity. Journal of Personality Assessment, 37, 437-455.

Glass, D.C., & Singer, J.E. (1973). Experimental studies of uncontrollableand unpredictable noise. Representative Research in Social Psychol-ogy, 4, 65-183.

Hart, D., & Edelstein, W. (1992). The relationship of self-understanding inchildhood to social class, community type, and teacher-rated intel-lectual and social competence. Journal of Cross-Cultural Psychology,23, 353-365.

Higgins, E. T. (1997). Beyond pleasure and pain. American Psychologist,

52, 12801300.Holtgraves, T., & Lasky, B. (1999). Linguistic power and persuasion. Journal

of Language and Social Psychology, 18, 196205.Hosman, L. A. (1989). The evaluative consequences of hedges, hesitations,

and intensi ers: Powerful and powerless speech styles. Human Com-munication Research, 15, 383406.

Johnson, W., Krueger, R.F. (2006). How money buys happiness: geneticand environmental processes linking nances and life satisfaction.Journal of Personality and Social Psychology, 4, 680-691.

Keltner, D., Gruenfeld, D.H., & Anderson, C. (2003). Power, approach andinhibition. Psychological Review, 110, 265-284.

Kitayama, S., Duffy, S., Kawamura, T., & Larsen, J.T. (2003). Perceiving anobject and its context in different cultures: A cultural look at New Look.Psychological Science, 14, 201-206.

Kulik, C.T., Oldham. G.R. & Hackman, J.R. (1987). Work design as an ap-

proach to person-environment t. Journal of Vocational Behavior, 31,278-296.Lachman, M.E. (1986). Locus of control in aging research: A case for multi-

dimensional and domain-speci c assessment. Psychology and Aging,1, 34-40.

Lachman, M. E., & Weaver, S. L. (1998). The sense of control as a modera-

tor of social class differences in health and well-being. Journal of Personality and Social Psychology, 74, 763-773.

Link, B. G., Lennon, M. C., & Dohrenwend, B. P. (1993). Sociometric statusand depression: The role of occupations involving direction, control,and planning. American Journal of Sociology, 98, 1351 1387.

Major, B., & Heslin, R. (1982). Perceptions of cross-sex and same-sex non-reciprocal touch: It is better to give than receive. Journal of NonverbalBehavior, 6, 148162.

Malle, B. (1999). How people explain behavior: A new theoretical frame-work.

Personality and Social Psychology Review, 3, 23-48.Masuda, T., & Nisbett, R.E. (2001). Attending holistically vs. analytically:

Comparing the context sensitivity of Japanese and Americans. Journalof Personality and Social Psychology, 81, 922934.

Miller, J. (1984). Culture and the development of everyday social explana-tion. Journal of Personality and Social Psychology, 46, 961-978.

Mirowsky, J., & Ross, C. E. (1983). The multidimensionality of psycho-pathology in a community sample. American Journal of CommunityPsychology, 11, 573591.

Morris, M.W., & Peng, K. (1994). Culture and cause: American and Chineseattributions for social and physical events. Journal of Personality andSocial Psychology, 67, 949971.

Nygard, R. (1969). Motive to approach success and motive to avoid failurein boys from different social groups. Scandinavian Journal of Educa-tional Research, 4, 222232.

Oakes, J.M., & Rossi, R.H. (2003). The measurement of SES in healthresearch: Current practices and steps toward a new approach. SocialScience and Medicine, 56, 769-784.

Riger, S., Lebailly, R. K., & Gordon, M. T. (1981). Community ties and urban-ites fear of crimes: An ecological investigation. American Journal of

Community Psychology, 9, 653665.Ross, M. & Sicoly, F. (1979). Egocentric biases in availability and attribution.

Journal of Personality and Social Psychology, 37, 322-336.Schneider, B. (1978). Person-situation selection: A review of some ability-

situation interaction research. Personnel Psychology, 31, 281-297.Schwartz, D., Dodge, K. A., & Coie, J. D. (1993). The emergence of chronic

peer victimization in boys play groups. Child Development, 64,17551772.

Snibbe, A.C., & Markus, H.R., (2005). You cant always get what you want:Educational attainment, agency, and choice. Journal of Personalityand Social Psychology, 88, 703-720.

Triandis, H.C. (1989). The self and social behavior in differing cultural con-texts. Psychological Review, 96), 506-520.

University of California, Santa Cruz. (2003). Results of the Fall 2002 CIRPFreshmen Survey. Institutional Research and Policy Studies.

Watson, D., & Clark, L. A. (1997). Extraversion and its positive emotionalcore. In R. Hogan & J. A. Johnson (Eds.), Handbook of personalitypsychology (pp. 767793). San Diego, CA: Academic Press.

Zander, A., & Forward, J. (1968). Position in group, achievement motivation,and group aspirations. Journal of Personality and Social Psychology,8, 282288.

er convergent findings, illustrates that SES is a profound minant of cognition, confirming that future research on Sthought is surely warranted. In order to explicate the diffeamong people, it is necessary to understand the effects of addition to those of culture and gender.


23/58


The Involvement Gap:

Tiffany Wai Man Lam and Audun DahlUniversity of California, Berkeley

Parental involvement in schools has been encouraged because of the demonstrated bene ts for schools, parents, and students. Although parental participation is on the rise, certain cultures, such as the Chinese,are relatively less involved in schools. This study examines the potential cultural factors that affect Chi-nese parental participation at a San Francisco Bay Area elementary school. Survey responses from 132

parents (77 Chinese immigrants, 24 Chinese Americans, and 31 Caucasians) are analyzed, and it is found that Chinese immigrant parents are signi cantly less involved in at-school volunteering, special events,and parent-teacher associations. Different obstacles to participation were also found for each cultural

group. Implications and suggestions for culturally sensitive ways of increasing parental involvement inschools are discussed.

ducation is generally viewed as a bidirectional, interactiveprocess between the student and teacher, in which each compo-nent of the link contributes and benefits from the other. Thisbidirectional relationship has recently transformed into a triadicone, with parents comprising the third party. This new three-way relationship is especially relevant in the early years of ed-ucation, when parents are often most involved in the learningexperiences of the child. Parental involvement is increasinglyemphasized, as the insights and participation contributed by par-ents are recognized as valuable benefits to students, parents, andeducators (Epstein, 1986).

Grolnick and Slowiaczek (1994) defined parental involve-ment as a dedication of resources by the parent to the childwithin a given domain (p. 238). In this study, the domain of interest is education, and the dedication of resources spreadsacross several dimensions. Some researchers see parental in-volvement as a one-dimensional construct (Stevenson & Bak-er, 1987), while others see it as multi-dimensional (Epstein &Dauber, 1991; Grolnick & Slowiaczek, 1994). For the purposeof this study, the Epstein and Dauber (1991) parental involve-ment model will be used to study parental participation acrosscultures. This model was chosen because it incorporates par-ents participation in many different realms, including the home,school, and the processes involved in decision-making. Epsteinand Daubers (1991) model also acknowledges the collaborativerole that schools play in parental participation.

This multidimensional model enables a critical evaluation of the cultural factors that associate with parental involvement. Inthis study, the following five dimensions of involvement from theEpstein and Dauber (1991) model were assessed:

1. Basic Obligations at Home: necessities that prepare studentsfor school and help maintain a supportive learning environ-ment.

2. School-Home Communication: communication from theschool (phone calls, parent-teacher conferences, etc.) that in-

forms parents of the childs academic progress.

3. Volunteering or Being an Audience at School: volunvolvement in assisting teachers, staff, or students in clschool, or on field trips, including parental attendancecial events.

4. Involvement in Home Learning Activities: helping, ming, and discussing homework or schoolwork with stu

5. Decision Making, Governance, and Advocacy Roleinvolvement in the school or community to create chanrepresent the parental perspective by contributing idethe school through participation in parent-teacher astions (PTAs), parent-teacher organizations (PTOs), orsory councils.

Epstein and Dauber (1991) also introduced a sixth caregarding community collaboration, which addresses the bution of the community network, including businessescies, and other relevant organizations. However, the comcategory is not addressed in this study, as the focus is mathe nature of the triadic relationship between parents, teaand students.

Benefits of Parental Involvement Parental involvement has a positive impact on childre

ents, and the school (Comer & Hayes, 1991; Epstein, 19Neal, 1999). Parents can offer more individualized acadesistance to students in the classroom and help them buildskills by interacting with adults from diverse backgrounaddition, school staff can teach parents to be positive roleand to help their children learn more efficiently. Thus, painvolvement at schools is sometimes described as a typecial capital, in which the social relationships between tand family help students reach their goals (McNeal, 1999Lin, 2005). Social capital helps to create better learning ements, largely because parents and teachers goals are inPrevious research shows that when parents are involved, pteacher cooperation is maximized and antagonism min

moreover, teachers are perceived as more professional an

E

A Comparison of School Involvement BetweenChinese, Chinese American, and Caucasian Parents


24/58

12


t hE i nvoLvEMEnt g AP | Tiffany Wai Man Lam & Audun Dahl

fied (ibid). Participation gives parents a sense of pride and sat-isfaction, and children show approval and appreciation for theirparents involvement. These studies have also demonstrated thatwhen parents develop a strong positive relationship with theschool, students are more likely to create the same attachment.Although parental involvement is seen as a positive influence andis often encouraged, many factors determine the frequency orwillingness of participation. One main factor is culture.

Impact of Culture Culture plays a role in moderating parental involvement in

schools. Parents and children from different cultures hold di-verse educational values and aspirations, which are likely to in-fluence educational practices. Most Chinese immigrants values,for example, are deeply influenced by Confucian philosophy andteachings, which in turn affect personal study habits and paren-tal monitoring of work (Schneider & Lee, 1990; Zhang & Car-rasquillo, 1995; Chao, 1996, Yan & Lin, 2005). Common Confu-cian teachings applicable to education include an emphasis oneffort, discipline, delay of gratification, respect for elders, a senseof family obligations, and reverence of learning. These valuespromote cultural respect for learning, parental monitoring, andclose family structures, thus encouraging diligence from stu-dents. A strong parallel exist between the values emphasized athome and at school, such that teachers support and encouragethe behavior socialized at home (Siu, 1994).

Confucian values often contrast with the Socratic westerneducational model (Tweed & Lehman, 2002), which is reflectedby the United States school system. The Socratic model valuesquestioning, evaluating, and generating knowledge through di-rect student-teacher interaction. The difference in educationalvalues between Confucian and Socratic models has substantialimplications on educational practices. Chao (1996) found thatEuropean American mothers emphasized social skills in order tobuild up self-esteem, thereby encouraging interest in learning.They allowed children to learn at their own pace and providedresources to trigger their creativity. On the other hand, Chineseparents emphasized academic performance, believing that theirchildrens performance was central to child rearing, so failurein school performance was equated with failure in child rearing.Based on this view, Chinese parents felt they only needed to workon activities that could be completed and monitored at home, aplace central to child rearing.

Future socioeconomic implications also factor into Chineseparents assessments of educational importance. Because manyChinese parents see education as the path to social mobility

(Pearce & Lin, 2007; Sy, 2006), they hold high academic expecta-tions for their children to fulfill. Such expectations are especiallyemphasized in immigrant families that have moved to Americain hopes of a better life. These high academic expectations en-courage certain parental educational habits, such as tutoring ormonitoring homework, and contribute to childrens academicperformance in schools (Schneider & Lee, 1990).

However, despite this cultural emphasis on education, thereis a paradox in that the Chinese parental role appears to be lim-ited to the domestic realm. Research has found that Chinese im-migrant and Chinese American parents, compared to Caucasian,Hispanic, and African American parents, tend to be less involvedin school activities (Pearce & Lin, 2007; Schneider & Lee, 1990;

Yan & Lin, 2005), but more hands-on in home activitieCaucasian parents (Chen, 2001). The home is a setting inparents can take charge and stress their values, and thuChinese parents at-home emphasis suggests that Chinesents use private, family time to play an active role in theireducation (Chao, 1996). However, other studies have fouChinese parents were not significantly more involved atthan Caucasian, Hispanic, and African American parents

2006; Peng & Wright, 1994), but rather more likely to strictive and limit leisure time during at-home activities (2006; Sy, 2006).

Barriers to Parental Involvement Because Chinese parents have been found to be high

cerned with their childrens academic performance, oneexpect them to be more involved in school activities, in ato academic work in the home. However, it has been showthis is not the case (Pearce & Lin, 2007; Schneider & LeYan & Lin, 2005). One explanation is that immigrants arwith several barriers to at-school involvement (Siu, 1996;2001). A lack of English proficiency and the resulting ito communicate with teachers and school personnel canparents, while financial problems may keep parents conat work, taking time away from attending conferences anactivities hosted by the school. Limited participation alsofrom a lack of communication from the school and theto establish specific matters to discuss during meetings (D2001). These reports indicate inadequate school-home comcation, which may contribute to a lack of goal unity andinvolvement. Consequently, parents may miss importantmation, fail to advocate for their childrens needs, and appthetic to their childrens studies.

At-school involvement can be negatively affectedChinese, and more broadly Asian, cultural values clash wAmerican educational model. According to Turney an(2009), Asians tend to face more barriers in participatioother cultural groups in the American education systemConfucian value of respect for authority, for example, mcourage Asian parents from giving suggestions to the schvolunteering in the classroom, because they may find itspectful to the teachers and administrators who are the auity figures in the school setting (Sy, 2006; Li, 2006). Thuparents who have stronger connections to Confucian educvalues may be less likely to participate in school activit2006). English proficiency and time spent in America apositively associated with involvement: the longer the du

of their time in America, the more familiarity parents havthe school system, and the more they participate in it (T& Kao, 2009). Additionally, better language ability creadiscomfort for parents when speaking with school officia

As children become older, parental involvement in gregardless of culture, seems to taper off, partly because pbelieve they do not possess sufficient educational skills ttheir children (Epstein & Dauber, 1991; Sy, 2006). Thispecially true for immigrant parents, who may not have aa level of education as non-immigrant parents, and werlikely instructed in a different language, making it difficthem to provide academic help to their children. Moreovrental involvement with at-home activities is harder to or


25/58


as academics become more rigorous and time-consuming, and isconsequently implemented less frequently. These beliefs and bar-riers regarding education make parental involvement difficult toencourage and enable in immigrant families, and these concernsneed to be addressed.

Responsibility of School Parental involvement is not the sole responsibility of par-

ents. The opportunities and encouragement provided by theschools also affect the likelihood of parent participation in schoolactivities. Unfortunately, many teachers are limited in theirmethods of contacting parents, largely due to parental inacces-sibility (Bempechat, 1992). These teachers also often hold expec-tations that parents have no time to become involved, despite theestablished benefits of parental participation. Thus, the school isa frequently underutilized, but key source for encouraging andincreasing parental involvement, and its responsibility cannot beoverlooked.

The Purpose of this Study Although Chinese students, as a whole, are academically

high-achieving, they can still benefit from communication be-tween their parents and schools. Potential benefits from school-based involvement include consensus of goals, appreciation fromstudents, parental knowledge of school functioning, and a maxi-mization of available educational resources (McNeal, 1999; Yan& Lin, 2005). These elements have been demonstrated to lead tobetter academic and social responses (ibid). Such benefits are aprimary reason that parental involvement is continuously beingpromoted in school settings. However, there are many factorsthat limit participation, such as timing inconveniences, financialstress, and cultural barriers.

These issues are characteristic of the obstacles that preventimmigrants from participating. As the United States becomes in-creasingly populated with immigrants and their families, Ameri-can schools must adapt to meet their cultural needs and expec-tations. This is especially true in the San Francisco Bay Area,where the Chinese population is as high as 20% (U.S. CensusBureau, 2005). However, very little research has been conductedon parental involvement in San Francisco Bay Area schools, andthere is none available examining cultural differences and the is-sue of meeting multicultural needs. The present study focuseson cultural differences in parental involvement between Chineseimmigrant, Chinese American, and Caucasian parents in a SanFrancisco Bay Area elementary school and takes a step forwardin examining the types of obstacles these cultural groups face

relative to each other and the types of obstacles associating withparental engagement at school.We hypothesize that Chinese immigrant parents will be less

involved in at-school contexts, in comparison to Chinese Ameri-can and Caucasian parents, because of translation issues, timinginconveniences, and the cultural tendency to defer academic au-thority in public settings to school officials. Chinese immigrantparents are also expected to exhibit lower participation in thedecision-making, advocacy, and governance domain because of their high respect for the authority of school officials. By exam-ining these rates of parental involvement and barriers to partici-pation, we hope to understand how school involvement amongChinese immigrant, Chinese American, and Caucasian parents

can be improved.

MethodThe School

The elementary school1 in this study is located in an acomprised of approximately 39.5% Chinese, 39.7% Ca1.1% African American, 4.8% Hispanic, 0.6% Amerdian, and 7.0% other Asian (1.9% Japanese, 1.5% Kore

Filipino) households (U.S. Census Bureau, 2005). Accoschools 2008-2009 records, the school itself is composedstudents, 41.6% Chinese, 23.1 % Caucasian, 14.2% othspecified-Whites (a total of 37.3% Whites), 1.4% Africancan, 3.7% Hispanic, 0.9% American Indian, and 7.7 % oth(2.3% Japanese, 2.3% Korean, 3.1% Filipino). Students reported ethnicities2 make up 7.4% of the population. Thusschool is fairly representative of the neighborhood in whsituated. School records also state that 44% of the studesocioeconomically disadvantaged and 32% are English le

Opportunities for parental involvement: The elementary school offers a number of opportunit

parents to become involved. These include monthly PTAings; periodic meetings; school, classroom, and field tripteering; event organization, participation, set-up, and cleand monetary or material donation and fundraising. Parenhelp fundraise and decide how PTA funds are used, and chelp monitor the schools budget, academic plans, and prthrough participation in Council Meetings. In order to amodate working parents, these opportunities take place nights and weekends. The opportunities are regularly poleast a month ahead of time on the schools PTA website. are also announced in school verbally and through distriof fliers. Involvement at home with children is also encoParents are advised to help children with schoolwork andabout school regularly.

Participants The participants of this study were parents of eleme

school students. Three hundred and fifty copies of the were issued and 221 were returned, which is a responof 63%. Of these 221 surveys, 136 were from immigranese (N=80), Chinese American (N=25), and Caucasian families. The demographic question regarding ethnicity a fill-in format, and relevant responses were categorized tthe three groups (Chinese immigrant, Chinese AmericaCaucasian) in this study. Chinese immigrant participant

identified in three ways: 1) those who identified as Chithe survey and said they were born outside of America, 2who identified as Asian in ethnicity and said they were bChina or Taiwan, and 3) those who completed the transurvey in Chinese and said they were born in China or TChinese Americans were individuals who identified as be

1 The name of the specific San Francisco Bay Area elementarhas been left anonymous as requested by the school, and for ttection of the schools students, parents, and staff.

2 The term ethnicity, as used in this study, refers to the specificlassification identified by the participant on the survey. Thculture, is used to reflect social influences acting upon paethnic groups.


26/58

14



nese and were born in America. Caucasian participants are thosewho identified as being Caucasian, White, or having Europeandescent. Eighty-five surveys were omitted because participantseither did not fit under the specified ethnic categories or did notprovide enough information to be included in one of the threestudy groups. Of the 136 surveys, several were omitted fromsectional analysis because of completion errors.

There were a total of 55 men (47% Chinese, 9% Chinese

American, 44% Caucasian), 78 women (67% Chinese, 24% Chi-nese American, 9% Caucasian) and 3 unidentified gender (2 Chi-nese, 1 Chinese American) participants who completed the 136surveys. The education level in each group was similar; Chineseimmigrants (69%) and Chinese Americans (86%) tended to behigh school and college graduates and Caucasian (90%) partici-pants tended to have college degrees or higher education. Nodistinction was made between parents or caregivers who com-pleted the survey, as the distinction is not particularly relevantto the studys objective.

For the purposes of this study, Chinese immigrants will bereferred to as Chinese, Chinese Americans as Chinese Ameri-cans, and White or Caucasian participants as Caucasians.

Design and Materials A survey was used to examine parental involvement in the

elementary school. The complete survey was formed by compil-ing three separate surveys: one designed by the principal, one bythe PTA president, and one compiled by the primary researcher.There were a total of 84 questions, 7 of which addressed basicdemographics.

The questions selected in the researcher-compiled surveyexemplified the five categories in Epstein and Daubers (1991)model that measure parental participation. There were five mainconstructs with a different number of questions per construct,based on programs available at the elementary school, as well asthe necessity of questions to address each category. The follow-ing is the breakdown of questions per category, with the numberof questions included in parentheses:

1. Basic Obligations at Home (4)2. School-Home Communication (13)3. Volunteering or Being an Audience at School (6)4. Involvement in Home Learning Activities (5)5. Decision Making, Governance, and Advocacy Roles (3)

The obstacles and barriers section contained ten categories.Eight of these were based on the prior research on participation

(Siu, 1996; Dyson, 2001). The two new categories were OtherReasons, which encompassed all other explanations, and NoDifficulty, which indicated that there were no barriers to par-ticipation. For this section, all ten options were provided for eachquestion regarding activity participation. The specific categorieswere as follows:

1. Not Enough Information or Notice2. Inconvenient Time3. No Childcare Available4. No Translation Provided5. Transportation Issues6. Not Interested

7. No Reason to Go8. Do Not Feel Comfortable9. Other Reasons10. No Difficulty

The final survey, integrating the researcher, principa

PTA president questions, was composed of ten sections (of questions per section is in parentheses):

1. Demographics (7)2. Academic/Programs (17)3. Importance of Academic Programs (7)4. School Culture & Environment (8)5. School-Home Communication (9)6. Parental involvement at School (10)7. At-Home Activities & Involvement (6)8. Activity Participation (6)9. Obstacles and Barriers (6)10. General PTA Participation & Opinion (8)

The survey sections did not adopt the same titles as thstructs of Epstein and Dauber (1991), but instead were rein accordance to general description of all questions particular category. Most survey sections consisted of mchoice questions with the options of Yes, No, and Know, although some sections provided more varied refrom which to select and a few open-ended questions. nal draft was constructed after careful and repeated revithe principal of the elementary school and the presidentschools PTA. The final translated draft of the survey wviewed by three native Chinese speakers who spent the mof their lives in China.

A total of 37 questions (included in the appendix wdemographic questions) were analyzed in this study. Allanalyzed questions adopted the Yes/No/I Dont Knowmat, except for questions regarding participation specat school, which had a Yes/No format, and questionsObstacles and Barriers section, where participants checobstacles to their participation. Other survey sections, incacademic program and PTA funding, are not discussed study, but were present in the survey for the interest of thecipal and PTA president.

Procedure Three hundred and fifty English and Chinese survey

given to teachers for student distribution. Each child was

a survey to take home to his/her parents, and per instructiothe survey, each family only had to fill out one survey.Parents were given a school week to return surveys; th

veys were first distributed on a Monday, and all were coon the Friday of that week. There were no rules addressinto fill out the anonymous survey, with the exception of itions on completing specific sections (e.g., Please place() under your answer). All applicable surveys collecFriday after survey distribution were used in this study.

ResultsResults were examined by categorical construct (E

and Dauber, 1991). Questions from four survey sections


27/58


sic Obligations at Home, (2) School-Home Communication, (3)At-Home Involvement & Activities (Epstein and Dauber (1991)category: Involvement in Home Learning Activities), and (4)Decision Making, Governance, and Advocacywere numeri-cally coded and a composite score was created to represent eachparticular construct. Yes answers were coded as +1, No an-swers as -1, and I Dont Know as 0. Because the scores werenegatively skewed, the use of tests assuming a normal distri-

bution was not appropriate. Instead, Kruskal Wallis and Mann-Whitney tests were performed to compare the three culturalgroups on each of the four composite measures (see Table 2ain the supplementary materials). Due to small sample sizes andcategorical responses, Fisher Exact tests were used for each of the six questions of the Volunteering or Being an Audience atSchool construct (see Table 3a in supplementary materials). Aidk Bonferroni error correction was made for post-hoc testscomparing each cultural group to the two others, using a family-wise alpha level of FW = .10. Since three comparisons werenecessary for each measure, the error correction gave a compar-ison-wise significance level of CW=.03.

Cronbachs alphas were obtained for every construct, andwere typically quite low. However, because of ceiling effects,which limited variation in individual scores, we cannot concludethat the low alpha values imply that the questions included in thecomposite measure are in fact tapping into different constructs.

Finally, a set of loglinear models was fitted to examine therelationship between ethnicity, activity, and participation obsta-cles3 (see Table 1 in supplementary materials). For each com-bination of ethnicity group and activity, response frequencies

were obtained for each obstacle. Because the overall freqof certain obstacle categories were very low, some obstaegories were collapsed into meaningful superordinate cate

Results are outlined in accordance to Epstein and D(1991) categories:

Basic Obligations at Home The Cronbachs alpha for survey questions in this s

was c = 0.29. The omnibus test of group difference wsignificant, so the completion of parental obligations as rto involvement could not be said to differ between the thrtural groups.

School-Home Communication Answers to three questions regarding translation wer

cluded from the calculation of overall parental perceptschool-home communication, because translation servinot be relevant to participants whose native language islish, and thus it was difficult to interpret their answers regatranslation.

Cronbachs alpha for survey questions in this sectioc = 0.60. The result of the omnibus test of group diffewas significant ( = 7.52, d = 2, p = .023). Mann-Whitney phoc tests showed that Chinese Americans were more likeChinese (U = 471.50, p = .026) to be satisfied with schoocommunication (as illustrated in Figure 1). The differesatisfaction between Chinese and Caucasian parents for Home Communication almost approached significance879.0, p = .06; see Table 2b in the supplementary materimore detailed School-Home Communication results).

To examine whether translation services had any relaship with perception of School-Home Communication inal groups for which such service may be relevant, Kendbwas calculated separately and as a group for Chinese and CAmericans as a test of ordinal association. The associatefficients did not differ significantly from zero, suggestithere was no relationship between translation and School

Figure 1. Group Comparison of School-home Communication

3 Although most reported either zero (considered No Difficulty) orone obstacle for each activity, a few participants reported two obsta-cles to their participation in a given activity. Including both answerswould violate the assumption of independent observations underly-

ing the tests of fit of the models. Hence, for each such pair of data,one answer was randomly selected, and the loglinear models wererefitted. This procedure was repeated 10 times, and each iterationyielded conclusions identical to those of the models reported here.


28/58

16



Communication.

Volunteering or Being an Audience at School Responses to each of the six questions here were analyzed

with Fishers Exact test because Chi Square approximation wasinappropriate due to low cell frequencies (< 2). There was a sig-nificant relationship between cultural groups and attendance of parent-teacher associations (p = .049), special events (p = .029),and volunteer opportunities (p = .001), such that Chinese im-migrant parents were less likely than Caucasians to be involvedin all of these occasions (p = .025, p = .02, and p = .001, respec-tively). Chinese immigrants were also less likely than ChineseAmericans to volunteer (p = .01; see Table 3a-d in supplemen-tary materials for detailed results).

There were no significant differences between Chinese im-migrants, Chinese Americans, and Caucasians in the reportedparticipation in parent-teacher conferences, workshops, andSchool Site Council advocacy meetings. (Fisher Exact test re-sults presented in Table 3a in supplementary materials; see Fig-ure 2 for a comparison of cultural groups with regard to theat-school involvement construct.)

Involvement in Home Learning Activities Cronbachs alpha for survey questions in this section wasc

= 0.64. There were no significant differences between the homeinvolvement of Chinese immigrants, Chinese Americans, andCaucasian parents (see Table 2a in supplementary materials fordetailed results).

Decision Making, Governance, & Advocacy Roles Cronbachs alpha for survey questions in this section wasc

= 0.56. The three cultural groups did not differ significantly onthe decision making, governance, or advocacy involvement mea-surements (see Table 2a in supplementary materials for detailedresults).

Obstacle and Barriers to Participation

This section investigated obstacles preventing parents from

participating in school contexts. Participants were direccheck all responses that apply, however, most only providresponse per category, and a number chose not to respond tain activities. Ten alternative response categories were prfor each of the six activities (parent-teacher conferences, Teacher Association meetings, volunteering, advocacy meworkshops, and special events). However, because the numresponses for some of the activity categories was low, soegories were collapsed into meaningful superordinate catin order to compare the likelihood-ratio statistics to a chi-distribution4.

There were a total of five collapsed categories. One original obstacle categories remained while the other ninregrouped into the following four new collapsed categori

1. Other Reasons encompassed Not Enough InformaNotice, No Childcare Available, Transportatio