The effect of participation in the neighborhood academic program on the autophotographic self-concepts of inner-city adolescents.

The current research investigates the effect of participation in the Neighborhood Academic Initiative Scholars Program (NAI) on the students' sense of self as viewed through autophotography. The NAI is designed to prepare inner-city middle school students, composed mainly of African-Americans and Hispanics, for entry into a four year university. Students who participated in the scholars' program were significantly more likely to view themselves as engaging in scholarly/responsible behaviors. Non-program participants viewed themselves as being involved in more playful activities, such as listening to music, dancing, and playing video games.

**********

A central concern of psychological research has been exacting a better understanding of adolescent self-concepts (Oyserman & Markus, 1990). Despite its centrality, "thousands of psychological studies of the self have left its secrets still intact" (McGuire, 1984, p. 73). These secrets remain hidden because traditional methodological approaches alone have not always allowed participants themselves to articulate their self-concepts. Researchers have tended to conceptualize and operationalize the salient dimensions of the self for their participants (Juhasz & Munshi, 1990; Ziller, 1990; Ziller & Rorer, 1985). The major obstacle confronting self research has been the almost exclusive reliance on "paper and pencil" measures (McGuire, 1984), a problem particularly acute for minorities who face culturally biased instruments (Spencer & Markstrom-Adams, 1990). Ideally, participants should be five to conjure up and communicate their own notions of themselves with minimal intrusion from researchers.

Juhasz and Munshi (1990) summarize the pressing need for new techniques for the study of the self: "Theory suggests that any attempt to obtain valid measures of children's self-esteem should actively involve the actual participants not only in the self-measurement but also in the content selection and test methodology" (p. 691). In sum, measures and instruments that facilitate participants' abilities to communicate their self concepts should enrich our understanding of adolescents and help us design curricula and build communication environments that foster and sustain academic achievement.

The current study attempts to move toward a better understanding of the self-concepts of at-risk adolescents from the inner-city, a place inundated with crime, drug abuse, alcoholism, low academic achievement, and ah overall sense of hopelessness and despair (Flemming, 1995). A relatively novel instrument for measuring self-concepts is complemented by a traditional measure of self-esteem. Following this protocol, this study gives participants a greater degree of autonomy in describing themselves. The overarching goal of this study is to investigate the differential effect participation in the Neighborhood Academic Initiative Program has on adolescents self--concepts as viewed through the autophotographic lens.

The Self

The search for the self traces its roots to symbolic interactionists (Cooley, 1902; Mead, 1934) who argue that communication plays a pivotal role in the emergence of the self: For symbolic interactionists, societal emergence is predicated on language: "In our statement of the development of intelligent we have already suggested that the language process is essential for the development of the self" (Mead, 1934, p. 135). Language enables us to develop and communicate and individualized self to others, and it allows us to initiate the process of interaction and understanding: "Interpersonal understanding begins with the process by the other to clarify and communicate the meaning of the self" (Ziller & Rorer, 1985, p. 629). At the core of symbolic interaction "the individual possesses a self only in relation to the selves of the other members of his social group" (Mead, 1934, p. 164) and "the mind arises through communication by a conversation of gestures in a social process" (p. 51). For symbolic interactionists, communication structures who we are, how we see ourselves, and how others see us; from these interactions, the self emerges.

The social comparison model (Festinger, 1954) extends symbolic interaction by identifying the critical interactions in the development of the self as being those which take place between individuals and their immediate social groups. Festinger's social comparison group resembles Mead's generalized other, which he defines as "the organized community or social group which gives to the individual his unity of self" (1934, p. 154). Each of these theoretical postulates proposes that individuals construct their sense of self by evaluating the feedback and the perceptions of others. The comparisons that individuals make to gain an understanding of the self are based on reflected appraisals (Felson, 1980), where individuals glean their self knowledge based on how they believe others perceive them. In this process, "The self-conscious human individual, then, takes or assumes the organized social attitudes of the given social group or community (or some one section thereof) to which he belongs" (Mead, 1934, p. 156). We should, then, expect the community in which people reside to exert a powerful influence on how individuals construct their sense of self.

Self-Perception theory (Bem, 1974) suggests that individuals develop their self-concepts by examining their actual behaviors. Accordingly, individuals ask themselves: "What must my attitude be if I am willing to behave in this fashion in this situation" (Bem, 1974,p. 28). If adolescents see themselves communicating in certain fashions, they might conclude that they favor certain communication styles over others. As a result, "individuals come to know their attitudes, emotions, and other infernal states partially by inferring them from observations of their own overt behavior" (p. 5).

Symbolic interaction and self-perception complement each other: one recommends that interactions serve as the basis of self concept, the other promulgates the view that individuals look at their manifest behaviors in forming their self concepts. Both perspectives should allow for the self to be defined as "the individual as known to the individual" (Murphy, cited in Ziller, 1990, (p. 28). According to Ziller, "in this definition of the self, the perceptions of the perceiver are raised to a new level of significance" (p. 29). This definition is also consistent with one of the major goals of the current research: to increase the role participants play in illustrating how they construct their sense of self. Joined in a multitheoretical framework, symbolic interaction and self-perception should enable us to better apprehend self development.

In addition to the multitheoretical framework under which this research is grounded, autophography is used to better comprehend adolescents' self-concepts. Autophotography is a little used approach that promises to give researchers a better view into the actual workings of the self (Ziller, 1990; Ziller & Rorer, 1985). Autophotography stands in marked contrast to paper and pencil tests, where participants (normally in the classroom or laboratory) are asked to rate how they feel about particular statements generated by various researchers, for reasons which might or might not be tied to how individuals construct their sense of self. Autophotography improves upon traditional field studies because the participants are allowed to pick their field(s) of interest; researchers, then, see what participants see-from participants' vantage points.

Autophotography was pioneered by Worth and Adair (Ziller & Rorer, 1985) who gave movie cameras to a group of Navajo Indians and told them to film and edit images depicting who they were, how they saw themselves. Ziller and his colleagues (Combs & Ziller, 1977; Rorer & Ziller, 1982; Ziller, 1988, 1990; Ziller & Lewis, 1981, Ziller, Martel, & Morrison, 1977; Ziller & Okura, 1986; Ziller & Rorer, 1985) extended this technique to the use of instamatic cameras. Participants are told to take 12 pictures of items in their environment which best describe them. It is from pictures that the construction of the self can be best illuminated (Combs & Ziller, 1977; Ziller & Lewis, 1981). Autophotography allows participants freedom--freedom to use their actual surroundings, freedom to pick and choose the people (i.e.; their family, their friends, church, etc.) who are important to their self-concepts, and freedom to decide what issues and what artifacts are most salient to their construction of self. It is this immense freedom which the camera gives to participants which distinguishes it from traditional paper and pencil tests.

The Neighborhood Academic Initiative Program

The Neighborhood Academic Initiative (NAI) Program was established at the University of Southern California to help at-risk seventh and eight graders from the inner-city of South Central Los Angeles prepare for higher education. The ultimate goal of the NAI is to help these students qualify for regular admission into a four-year university. To prepare students for the rigors of a major university, the NAI focuses on enhancing the academic skills of its students.

Students selected for the program were required to maintain at least a C average in each of their classes, have no previous disciplinary problems, no drug or alcohol use, and no gang affiliation. Students were required to sign an agreement stating that they would attend at least 90% of their classes and make a conscientious effort to mm in all assignments on time.

Teachers in the program were selected from the Los Angeles Unified School District based on recommendations from their principals and students, their demonstrated commitment to improving the educational skills of inner-city youth, and their ability to foster in students ah increased sense of self-worth and self-esteem. The selection process was extremely rigorous because "teachers have a special opportunity to affect the self-esteem of their female students [as well as males].... by instilling confidence to shape their interests and aspirations (AAUW, 1990, p. 11).

One especially intriguing aspect of the program is the fact that it deliberately sets out to frame the students' sense of self-worth by labeling them scholars. According to the Director of the NAI, the program staff wants students to think of themselves as capable academics and they go far out of their way to instill that new sense of self. The NAI provides incentives to students to view themselves as scholars by encouraging and rewarding them for the appropriate scholarly behaviors (Fleming, 1995). When these scholarly expectations are effectively communicated, the prediction is that students will incorporate their sense of 'being scholars' into their self-perceptions; they will look at themselves, reflect and say "if I am acting like a scholar, communicating like a scholar, and being treated like a scholar, I must be a scholar."

Method

Participants

Ninety-one (50 girls, 42 boys) seventh and eight grade students from two middle schools located in South Central Los Angeles participated in the study. Forty students participated in the Neighborhood Academic Initiative Program (NAI).

Procedures-Phase 1

Participants' teachers were given a booklet containing the three dependent measures--described in detail in the next section--during one class period. The instructions on the cover sheet for the dependent measures assured the participants of confidentiality and anonymity. These instructions also explained that there were no right or wrong answers, that we were interested in their opinions, and that their teacher(s) would not evaluate or grade their responses. The teachers who administered the dependent measure read the instructions to the students. The booklets were collected by the teachers at the end of the class period.

Phase-2

In the following class period, participants were given disposal cameras with the these instructions.

   This activity involves taking two different
   kinds of pictures. For one kind of
   picture, we want you to describe yourself.
   To do this, we would like you to
   take 12 photographs that tell who you
   are. These photographs can be of anything,
   just as long as they tell something
   about who you are. these pictures
   should say: "This is Me!"

   For the other kind of picture, we want
   you to describe who you are not. To do
   this we would like you to take
   another 12 photographs. These photographs
   can be of anything as long as
   they say: "This is Not Me!"

   To help you keep track, we've put a
   label on top of the camera. When you
   take a picture of something that describes
   you, cross out a number under
   Me on the label. Similarly, when you
   take a picture that does not describe
   you, keep track by crossing out a number
   under Not Me!" When you are
   finished, you should have all the
   numbers under "Me" crossed out and all
   the numbers under "Not Me" crossed
   out. Remember, you are the only one
   who can take these pictures, so you
   cannot take any pictures of yourself.
   Please return the camera to your teacher
   in two weeks. We will be doing another
   project with the pictures you will be
   taking.

Students were given one week to use their entire 24 roll of film. After a week, the cameras were collected and the pictures were processed.

Phase-3

In the next class period, participants were given a manila envelope with their name on it. Participants were then given the following instructions:

   We would like you to open the big
   envelope with your name on it. Inside
   you should find the pictures you took
   and two smaller white envelopes--one
   labeled "Me" and one labeled "Not Me."
   Place the white envelopes on the table
   in front of you. You can take out your
   pictures now and look at them. You will
   notice that each picture has a number on
   the back of it. Try to keep the pictures
   in order.

   You will see that there is a form in front
   of you that has your name on it. This
   form asks you to explain why each
   picture represents you or does not. Take
   the first picture in your pile, and check
   the back to be certain that it is picture
   #1. Go to the first question on your
   form. If the picture is one that describes
   who you are, please circle "Me" on
   question I and write a short sentence
   explaining why the pictures describes
   you. If the picture describes who you
   are not, circle "Not Me" and write a
   short sentence explaining why the picture
   does not describe you. When you
   have finished explaining the first picture,
   place it on the white envelope in
   front of you labeled "Me" if the picture
   describes you, of "Not Me" if the picture
   does not describe you. Follow this
   same procedures for all of your pictures.

   After you have described all of your
   pictures, put all of your me pictures in a
   pile on the "Me" envelope and all of
   your "Not Me" pictures in a pile on the
   "not me" envelope.

   Now turn to the last page of the form.
   Please choose the three pictures that
   describe best who you are, and the three
   pictures that describe best who you are
   not. When you have decided which three
   pictures describe you best, write down
   the numbers of the pictures in the spaces
   underneath "Me." Follow the same
   procedure for the "Not Me" photos.
   Place of the "Me" pictures in the "Me"
   envelope, and all of the "Not Me" pictures
   in the "Not Me" envelope.

   Now there is one final question. Please
   indicate which was easier, taking the
   "Me" or taking the "Not Me" pictures in
   the space provided.

Participants were then given a copy of the pictures they had taken to keep and tanked for their participation.

Dependent Measure

The Rosenberg' Self-Esteem Scale (1965), a 10 item, 4-point (1=strongly agree, 4=strongly disagree) measured global self-esteem. The Rosenberg scale is widely used and is a reliable estimator of global self-esteem, containing only one underlying factor (Marsh, 1996).

Coding

To organize the photos into analyzable units, the researchers examined the photos and their accompanying cutlines and attempted, based on the previous autophotographic schemes employed by Ziller and his colleagues, to propose enough categories so that each photo could be assigned to one category. To aid the coders, they were provided with the "cutlines," the one sentence descriptors that the participants wrote about each photo, which were described in the procedures sections. The cutlines were invaluable because "language is a tool for representing experience" (Belensky, Clinchy, Goldberger, & Tarule, 1996,p. 25). In our situation, we invited the participants to use language to complete their pictorial narrative. Consequently, if the coders had disputes over the proper categorization of a photo, they would refer back to the cutline, which constituted the participants' actual description of their photographs.

Thirty-three categories were generated from this exercise; most of our categories have been employed in other autophotographic studies. For example, Dollinger and Clancy (1993) and Dollinger et al. (1996) also used achievement, pets, sports, religion, television, family, possessions, and interests, as categories. Ziller and Rorer (1985) used 21 categories including tire that were the same as our categories: school, drugs, working around the house, and art.

Three research associates, who were unaware of the purposes of the project, were used to code the photos. They coded each picture based on what the picture represented. They were asked "What does this picture refer to" and they were given the thirty-three categories from which to choose. Coders were told to select the category which best represented what the picture referred to. The coders also coded the photo according to whether or not it referred to the subject or it did not refer to the subject, the "Me" or "Not Me" dimension. This dimension was actually generated by the participants who had previously coded their pictures according to whether or not they represented them or did not represent them.

Analysis

The Rosenberg Self-Esteem scale was analyzed using Oneway ANOVAs. Summary scores, where higher scores indicated higher levels of global self-esteem, were used.

To obtain a usable statistic from the photo categories, a summary score indicated the percentage of times that a particular photo category was mentioned. To calculate that statistic, the number of mentions of any particular category by any individual subject was divided by the total number of mentions for all categories, which yielded the summary score for each category, totaling 100%.

Summary scores flora the Rosenberg Self-Esteem scale were then correlated with the percentage of mentions for each photo category. Oneway ANOVAs were used to test if differences existed among the various groups and to correct for experiment wide error because of the huge number of comparisons that were computed.

Results

Rosenberg Self-Esteem

A 2 (race) X 2 (gender) X 2 (program/ non program) revealed significant main effects for gender F(1 87) = 6.61. p<.01 and race F(1, 87) = 7.776, p<. 01. Follow up t-tests pinpointed the following differences: the program students reported higher self-esteem than the non program students t(92) = 1.95, p<.06; Hispanic girls from the program were marginally higher than Hispanic girls from the non program group t(39) = 1.95, p<.06.

Correlations--Self-esteem and photo mentions.

Correlations between scores from the Rosenberg Self-Esteem scale and the percentage of photo mentions revealed the following significant correlations. For the entire sample, self-esteem was positively correlated with achievement (r=.32, p<.001) and worry (r=.27,p<.01). For African American females, self-esteem was negatively correlated with scholarship (r=-.75, p<.01). For African American males from the program, self-esteem was positively correlated with videogames (r=.87, p<.05), gender (r=.81, p<.05) and sports (r=.83, p<.05). For Hispanic females from the program group, self-esteem was significantly correlated with empathy and nurturing (r=.55, p<.05) and worry (r=.71,p<.003). For non-program African American females, self-esteem was negatively correlated with self-presentation (r=-.88, p<.05). For non program Hispanic females, self-esteem was positively correlated with nature (r=.50, p<.01), and negatively correlated with gender (r=-.45, p<.05). No other significant correlations were discovered.

Autophotographic Differences

A oneway ANOVA revealed that the program group was significantly higher than the non program group on scholarship F(1,91) = 8.84, p<. 01, housework and chores F(1,91) = 11.63, p<.001 , and general F(1,91) = 4.97, p<.05, marginally higher on sports F(1,91) = 3.65, p<.06. The non program group was significantly higher on music and dance F(1,91) = 4.49, p<.05, marginally higher on physical attributes F(1,91) = p<.06. Table 1 presents all of the results for each of the autophotographic items.

Girls flora the program group were significantly higher than girls from the non program group in the percentage of photos that mentioned USC scholarship F(1, 49) = 7.28, p<.01, material possessions F(1,49) = 3.93, p<.05), housework and chores F(1, 49) = 7.50, p <.01, marginally higher on sports and school F(1, 49) = 3.69, p<.06.

Boys from the program group were significantly higher than boys from the non program group in the percentage of photos that mentioned housework and chores F(1, 40) = 4.10, p<.05, music and dance F(1, 40) = 4.01, p<.05, marginally higher on food F(1,40) = 3.18, p<.08, and USC scholarship F(1, 40) = 3.37, p<.05. The non program boys were higher on reading F(1,40) = 4.72, p<.03.

African Americans in the program group were significantly higher than African Americans in the non program group on material possessions F(1,17) = 9.04, p<.01.

African American girls in the program group were significantly higher than African American girls in the non program group in the percentage of photos that mentioned hobbies F(1, 8) = 8.08, p< .02, material possessions F(1, 8) = 10.09, p<.01, and television and movies F(1, 8) = 6.76, p<. 03.

African American boys in the program group were significantly higher than African American boys from the non program group in the percentage of photos that mentioned empathy and nurturing F(1,7) = 7.72, p< .03.

Hispanics in the program group were significantly higher than Hispanics in the non program group in the percentage of photos that mentioned USC scholarship F(1, 70) = 8.86, p<.004, housework and chores F(1, 70) = 13.64, p<.001. Non program Hispanics were significantly higher on music and dance F(1, 70) = 4.09, p<.05. Hispanic girls from the program were significantly higher than Hispanic girls from the non program group in the percentage of photos that mentioned housework and chores F(1,38) = 6.14,p<.02, marginally higher on sports F(1, 38) = 3.57, p<.07, and cares and concerns F(1, 38) = 3.87, p<.06.

Discussion

Several methodological limitations of this study need to be detailed before we can launch into a profitable discussion about the findings. First, groups were not randomly assigned, violating a basic tenet of experimental design (Stanley & Campbell, 1963). Absent randomization, the possibility of self-selection biases looms large. Students from the NAI were, in part, a self-selected group. They chose to apply for and participate in the program. Given this self-selection process, it is entirely possible that the program group could have started with higher levels of self-esteem than their counterparts in the non-program group. They might have been more motivated and more confident of their academic abilities. The possibility of a self-selection bias is important to examine because we hypothesize that participation in the program might have a differential effect; we need to be alert to fallacious causality.

On the other hand, other factors might have minimized the effects of any self-selection bias. The entire sample of students was selected from the same pool. Many students who applied for the program, who would have otherwise been eligible, were not selected (Fleming, 1995). Of the approximately 300 students who were eligible for admission to the NAI, only 60 were selected, which was later pared down to 42 who remained in the program. This suggests that, while the NAI's selection process was rigorous, the selection criteria were not; some of the students who were not selected still remained in the control group, which indicates that many of the students from the control group shared similar characteristics to members from the experimental group. Additionally, we detected no significant differences in the relative academic skills of the two groups. The NAI students were not required to be exceptional in order to gain admission into the program. They merely needed to maintain a C average in their classes.

The marginally significant difference on the Rosenberg self-esteem between the experimental and control group also tends to dispel the self-selection bias. The marginal difference indicates that, at the very least, there were no significant self-esteem differences between the groups. Notwithstanding these potentially mitigating factors, caution is the benchmark when interpreting the findings.

Second, the use of autophotography is Not a panacea (Ziller, 1990). This instrument has shortcomings. We found that not every student used his/her entire role of film, some students ended up taking more pictures than others, though most of them used their entire roll. It is also difficult for autophotography to be used to answer direct self-esteem questions.

Program vs. Non Program

The overall trends that emerge from these findings lend some empirical support to the hypothesis that participation in the NAI helps frame students' self-concepts. This finding is consistent with Marsh, Barnes, and Richards (1986) who found that the "Outward Bound Program [was] ah effective intervention for changing multiple dimensions of self-concepts" (p. 202). The Outward Bound Program is similar to the NAI except that it focuses on students whose ages range from 17-27 (Marsh, Barnes, & Richards, 1986). Both programs were able to successfully flame students' self-concepts, the ultimate aim of intervention programs.

The major finding indicates that participation in the NAI evinced a positive self-concept of being labeled a "scholar". In a sense, the NAI students had been inducted into their own community of scholars; they joined a new social group, a new generalized other, a symbolic outcome that Mead seems to have anticipated; it is these interactions that appear to have succeeded in shifting the NAI students' self-perceptions from "just plain old student" to "scholar," a special type of student; a student who behaves and communicates in a scholarly manner.

Theoretically, labeling is a two-edge sword, it can have both positive and negative effects. On one edge of the sword, researchers have generally found that educational labeling, where students are placed in classes with students of their same academic skill, has a deleterious effect on those students who end up in the lowest level classes (Dembo, 1988). Other labeling studies found that the same negative effect of labeling holds true for delinquent youths (Ziller & Lewis, 1981; Matsueda, 1992), females who participate in traditionally male sports and are labeled lesbians (Blinde & Taub, 1992), and students who are labeled nerds (Kinney, 1993). What generally arouses the ire of labeling theorists is the selectivity of labeling. Matsueda (1992) summarizes this problem: "A hallmark of labeling theory is the proposition deviant labels are not randomly distributed across the social structure, but are instead more likely to apply to the powerless, the disadvantaged, and the poor" (p. 1588).

Matueda's observation is important for this study because we are primarily concerned with students from the inner-city who suffer the brunt of labeling (Reed, 1988); these students are poor and come from low-income families who lack strong educational backgrounds and political muscle. However, while the basis for the condemnation of tracking and labeling remain, a new perspective on labeling emerges through the NAI. The NAI has moved labeling onto more positive footing. The NAI's labeling protocol suggests to labeling theorists that labeling can enhance self-esteem. It also tells educators that using communication to affix positive labels onto students can motivate some of them to engage in scholarly behavior, a significant issue for all scholars.

Being labeled a scholar might be a mixed blessing for adolescents. From the view-point of academic professionals, being labeled a scholar is one of the greatests compliment one can receive. For adolescents, however, being labeled a scholar can be ah unwanted designation, an insult, or a put down. According to Kinney, one of the reasons some adolescents are called Nerds, an unsavory term, stems from their academic prowess. Fordham & Ogbu (1986) reported that many African-American adolescent and high school students avoid behaviors that suggested "brainiac" or "acting white." Nerds, brainiacs, dweebs, geeks, acting white are unpopular labels generally assigned to individuals who do well academically, are generally shy, lack the requisite adolescent social skills, are unattractive, and are non-athletic (Kinney, 1993). Nevertheless, for students in the NAI, being a scholar had other advantages; they were able to take classes at a major university; they were taken on a number of field trips and provided free lunches from time to time. These benefits do not discount the importance of being labeled a nerd of a brainiac, labels which might affect scholarly behavior (Fordham & Ogbu, 1986). An example best illustrates this point. African-Americans will sometimes do poorly in school to avoid being accused of acting white (Fordham & Ogbu, 1986). Acting white includes but is not limited to some of the following "(1) speaking standard English; (2) listening to white music or white radio stations; (3) going to the opera of ballet; (4) spending a lot of time in the library studying; (5) working hard to get good grades in school; (6) getting good grades in school those who get good grades are labeled "brainiacs" (p. 186). This finding is buttressed by Rotherman-Borus (1990) who found that African-American adolescents were strongly ethnically identified; that is, they were more likely to choose "the patterns of[behavior from] their ethnic group, whose norms often deviate from those of the dominant group" (p. 1075). The point here is that it is unclear whether being labeled a scholar is good or bad from adolescents' perceptions, though our evidence continually points us toward concluding that such a label is positive. We don't know if these students had learned effective coping strategies to overcome or compensate for some of the negative labels that might have been assigned to them by their peers who viewed scholars as nerds. In the main, we do not know if participation in the NAI changed language or cultural patterns in the students, an area ripe for future investigations.

The extensive mention of housework and chores by the program group highlights the second major difference. It appears as though the program group was more oriented toward responsible behaviors; they were apparently more disciplined. The non program group seemed more oriented toward focusing on pleasurable activities as evidenced by their significant mention of music and dance. Adolescents in the program group are obviously expected to do their fair share of the family's chores. These students are required to be disciplined, to attend to their responsibilities.

One of the responsibilities that students in the NAI had to attend to was education, which demands self-discipline. Discipline is not only critical for educational success, it lies at the core of adolescents' transition to maturity: "The adolescent's key to a successful transition to adulthood is learning to attend to and focus on the activities that are necessary for adult life rather than on the immediately pleasurable activities of childhood" (Oyserman & Markus, 1990, p. 112). Wayson and Lasley (1984) define discipline as adherence to some socially acceptable pattern of behavior. They suggest that discipline is learned during the process of growing up and requires a positive self concept and a sense of belonging. Students from the NAI might have gleaned positive self-concepts from the publicity the program garnered. The Los Angeles Times wrote about the program twice. Dan Rather devoted a special segment on the Nightly News to the NAI. Several publications at the University of Southern California prominently featured the program and its scholars. Many of the students were repeatedly interviewed and quoted. In sum, they received a great deal of positive attention, the type of attention that could account for the self-concept differences detected in our study. Consequently, we raise the caution flag because it is still unclear what role the program played in this difference. It is difficult to determine, with any reasonable degree of certainty, whether students were highly disciplined before they entered the program or whether they learned discipline during the program; the evidence is tentative and mainly circumstantial. Another option, one that provides a better explanation, remains. Even if these students did not learn responsible behaviors from their participation in the NAI, the program appears to have exerted some influence. For students who were already somewhat disciplined, the program provided the impetus to stay disciplined. Conversely, lf students were undisciplined, the program staff certainly encouraged them to attend to their responsibilities (Fleming, 1995).

Correlations

Scores on self-esteem for the entire sample were significantly correlated with the percentage of photos that mentioned achievement. This finding implies a fairly strong convergent relationship between two disparate measures of the self; it is also consistent with other research linking self-esteem to academic achievement (see Kohr et al., 1988). However, Kohr ct al. warned that although a number of research efforts have linked self-esteem to academic achievement, the association is not particularly robust. His warning is bolstered by our findings that did not consistently link self-esteem with academic achievement for the majority of groups. For African-American girls self-esteem was negatively correlated with scholarship, a finding in the opposite direction. It is also a finding that is different from all other groups whose self-esteem scores were positively, if not significantly, correlated with scholarship. Several explanations for this anomalous finding are possible. It might simply indicate that African-American girls' self-esteem (in this sample) is not tied to academic scholarship and that scholarship may not be an intrinsic component of their self-concept; or it may suggest that African-American girls do not view scholarship as ah important element in their lives. To resolve theses anomalies, further theoretical and methodological refinements are required.

The most unusual finding is the positive correlation between self-esteem and worry, which implies that the more people worry about issues, the higher their self-esteem. It might suggest that individuals with high self-esteem worry because they are constantly trying to achieve their goals, and the goals that they set for themselves are very high, which, in turn, causes them to worry.

Differences between African-Americans

African-Americans in the NAI mentioned material possessions more than African-Americans in the control group. Gibbs (1988) describes African-American males as being at a "endangered species" (p. 1), which she defines as "a class of individuals having common attributes and designated by a common name ... [and] in danger or peril or probably harm or loss" (p.1). If African American males are endangered, low on the totem pole, the expectation would be that they would have low levels of self-esteem. However, previous research indicates that African-Americans have equal levels to other groups, including Asians, Hispanics and Whites (Rotheram-Borus, 1990). For example, if a person is endangered, low on the hierarchical totem pole, then that person might possibly lust for the material possessions which have eluded him or her thus far. This result might also suggest that African-Americans want to envision a scenario where they can acquire the material possessions that have escaped their grasp. It might also indicate that the program gives students a sense that they may be able to acquire those material possessions. They know if they can qualify for regular admission to USC, they will have a free education. They also know that with an education, they can be competitive for well-paying jobs (Fleming, 1995).

The current research is of special interest because it uses a multimethodological design, combining two disparate approaches to studying the self: autophotography and the Rosenberg Self-Esteem scale. It illuminates their convergence and allows a view of adolescents that was previously unattainable in some studies. It also supports the usefulness of the NAI.

Future Research

Future researchers may further refine the autophotographic approach and focus on designing measures and instruments which are more subject-friendly. As Juhasz & Munshi (1990) argue: "We need new techniques with the power to identify components which rail outside those measured by instruments currently used" (p. 692).

Multimethodological designs, where more than one instrument of measure is used, can significantly improve our ability to investigate self-concepts. Given the multifaceted, dynamic nature of the self, the instruments should be complementary, each adding a unique contribution to understanding the self. Combining, for example, Kuhn & McPartland (1954) "Who am I" measure, where participants are asked to write down who they are, with the autophotographic depictions of the self might lead researchers to a greater understanding of the self. The multimethod approach also allows for tests of discriminant and convergent validity between and among various measures.

Comparisons between inner-city students from low-socio-economic backgrounds and students from middle to high socioeconomic backgrounds should be conducted. It would be interesting to sec if and what role socio-economics play in the acquisition of self-concepts of adolescents. For example, self-concepts of students from the inner-city of Los Angeles should be compared to students from Beverly Hills, and other areas of Los Angeles characterized by rich and upper middle-class residents.

Future research may concentrate on one of the most significant questions concerning self-esteem: to whom do students say they compare themselves? The theoretical formulations of the symbolic interactionists suggests the importance of the generalized other, the group from which individuals draw many of their perceptions of self. The fundamental assumption is that adolescents compare themselves to their immediate social group. However, researchers rarely ask participants whom do they compare themselves against. So while we can infer that the immediate social group plays a central role, it would be far more instructive to simply ask students about their base of comparison.

The categories that participants mention need to be more consistently codified. While certain categories that were generated by the current study overlapped with previous research efforts, further refinement of the categories is needed in order to make possible meta-analytic reviews.

Thus far, little has been said about the min-narratives, though they are critical components in apprehending adolescents' self-concepts. Our project employed a slightly different technique than Ziller and his colleagues and Clancy and his associates. Instead of asking participants to write a grand narrative where they explain the photos as a whole, we asked our participants to write a short sentence about each photo. Rather than debate the merits of each approach, it would be better to combine these approaches. Future researchers should use our cutlines, which function as mini-narratives, and the grand narratives employed by other autophotographic researchers. Following this protocol, researchers would have available richer textual offerings to complement the photos themselves.

Table 1

Mean Responses on each Auto Photographic Variable

Photo                Scholars    Control      F       P

achievement            1.91        1.59       .23    .63
age                     .00         .19      1.63    .20
scholarship            2.29         .10      8.84    .00
school                 5.51        4.92       .24    .63
chores                 3.84        1.12     11.63    .00
empathy                 .67         .78       .57    .81
sex identity            .75         .36       .73    .39
family                 4.46        2.91      1.37    .25
food                   4.38        5.54       .32    .57
friends                1.11        1.34       .14    .70
gender                  .85         .75       .05    .82
miscellaneous          7.32       12.15      4.97    .03
hobbies               11.91       11.64       .02    .88
illegal acts           6.20        6.02       .01    .89
intellectual            .45         .38       .04    .83
communication          1.36        2.09      1.25    .26
possessions            9.93        7.49      1.38    .24
music/dance            2.32        4.43      4.49    .04
name                    .00        1.15      1.91    .17
nature                 3.83        3.35       .14    .70
occupation             2.12        3.37      1.30    .26
pets                   3.66        3.15       .25    .06
physical                .99        4.45      3.61    .06
sports                 8.10        5.15      3.65    .06
poverty                 .83         .39      1.25    .26
race                    .43         .75       .65    .42
reading                1.89        2.66      1.49    .22
church                  .84         .58       .42    .51
self-presentation      3.22        3.54       .06    .79
personality            1.65        1.48       .06    .80
tv/movies              3.22        2.99       .08    .77
videogames             1.29        1.45       .08    .77
cares/concerns         2.34        1.49      1.55    .21

References

American Association of University Women. (1991). Shortchanging girls, shortchanging America. (Washington, D.C.).

Bern, D.S. (1974). Self-perception theory. In L. Berkowitz (Ed.). Advances in Experimental Social Psychology (pp. 321-345). New York: Academic Press.

Blinde, E. M., & Taub, D. E. (1992). Homophobia in sport. In p. L. Adler & P. Adler (Eds.). Constructions of Deviance: Social power, context, and interaction. Belmont, CA: Wadsworth.

Byrne, B. M. (1983). Investigating measures of self-concept. Measurement and Evaluation in Guidance, 16, 115-126.

Campbell, D.T., & Stanley, J.C. (1963). Experimental and Quasi-experimental design for research on teaching. In N. L. Gage (Ed.). Handbook of research on teaching (pp. 171-246). Chicago: Rand McNally.

Clinchy, B. M., Goldberger, N. IL, & Tarule, J.M. (1996). Women's ways of knowing: The development of the self, voice, und mind. US: Basic Books.

Combs, J. M., & Ziller, 1L C. (1977). Photographic self-concepts of counselees. Journal of Counseling Psychology, 24, 452-455.

Clancy, S. M., & Dollinger, S. J. (1993). Photographic depictions of the self. Gender and age differences in social connectiveness. Sex Roles, 19, 477-495.

Cooley, C. H. (1902). Human nature and the social order. New York: Charles Scribner's Sons.

Dembo, R. (1988). Delinquency among Black male youth. In J. T. Gibbs (E.d.). Young, Black, and male in American. An endangered species (pp. 129165). New York: Auburn House.

Demo, D. D., & Parker, K. D. (1986). Academic achievement and self-esteem among black and white college students. The Journal of Social Psychology, 127, 344-355.

Dollinger, S. J., & Clancy, S. M. (1993). Identity, self and personality: Glimpses through the autophotographic eye. Journal of Personality and Social Psychology, 64, 1064-1071.

Dollinger, S. J., Preston, L. A., O'Brien, S. P., & DiLalla, (1996). Individuality and relatedness of serf: An auto photographic study. Journal of Personality and Social Psychology, 71, 1268-1278.

Fairbanks, C. M. (1993). Labels, literacy, and enabling learning: Glenn's story. Harvard Educational Review, 62, 475-493.

Felson, R. B. (1980). Communication barriers and the reflected appraisal process. Social Psycology Quarterly, 43, 223-232.

Fastinger, L. (1954). A theory of social comparison processes. Human Relations, 7, 117-140.

Fleming, J. (1995). Personal communication.

Fordham: S., & Ogbu, J. U. (1986). Black students' school success coping with the burden of acting white. Urban Review, 18,176-206.

Gibbs, J. T. (1988). Young Black males in America. Endangered. embittered and embattled. In J. T. Gibbs (Ed.). Young, Black, and male in America. An endangered species (pp. 1-36). New York: Auburn House.

Gordon, I. J. (1966). Studying the child in the School. New York: John & Wiley & Sons.

Housley, K., Martin, S., McCoy, H., Greenhouse, P., & Stigger, F. (1987). Self-esteem of adolescent females as related to race, economic status, and area of residence. Perceptual and Motor Skills, 64, 559-566.

Hughes, M. & Demo, D. H. (1989). Self-perceptions of Black Americans: Self-esteem and personal efficacy. American Journal of Sociology, 95, 132-59.

Jensen, G. F., White, C. S., & Galliber, J. M. (1982). Ethnic status and adolescent self-evaluations: An extension of research on minority self-esteem. Social Problems, 30, 225-239.

Juhasz, A. M, & Munshi, A.N. (1990). "Faces and feelings": A projective technique for measuring young children's self-esteem. Perceptual and Motor Skills, 70, 691-702.

Kinney, D. A. (1993). From nerds to normals: The recovery of identity among adolescents from middle school to high school. Sociology of Education, 66, 21-37.

Kohr, R. L., Coldiron, J. IL, Skiffington, E. W., Masters, J. R., & Blust, R. S. (1988). The influence of race, class, and gender on self-esteem for fifth, eight, and eleventh grade students in Pennsylvania schools. Journal of Negro Education, 57, 467-481.

Kuhn, M. H., & Mcpartland, T. S. (1954). An empirical investigation of self-attitudes. American Sociological Review, 19, 68-76.

Marsh, H.W. (1996). Positive and negative global self-esteem: A substantively meaningful distinction or artifacts? Journal of Personality and Social Psychology, 70, 810-819.

Marsh, H. W., Barnes, J., & Richards, G. E. (1986). Multidimensional self-concepts: The effect of participation in an outward bound program. Journal of Personality and Social Psychology, 50, 195-204.

Matsueda, IL L. (1992). Reflected appraisals, parental labeling, and delinquency: Specifying a symbolic interactions theory. American Journal of Sociology, 97, 1577-1611.

Martinez, R., & Dukes, R. L. (1987). Race, gender and self-esteem among youth. Hispanic Journal of Behavioral Sciences, 9, 427-443.

McGuire, W. J. (1984). Search for the self: Going beyond self-esteem and the reactive self. In R Z. Zucker, J. Aronoff, & A. I. Rabin (Eds.). Personality and the Prediction of Behavior (pp. 73-120). New York: Academic Press.

Mead, G. H. (1934). Mind, self and society. Chicago: University of Chicago Press.

O'Brien, E. J. (1991). Sex differences in components of self-esteem. Psychological reports, 68, 241-242.

Osborne, W. L.,& LeGette, H. IL (1982). Sex, race, grade level, and social class differences in self-concept. Measurement and Evaluation in Guidance, 14, 195-201.

Oyserman, D., & Markus, J. R. (1990). Possible selves end delinquency. Journal of Personality and Social Psychology, 59, 112-125.

Purkey, W. W. (1970). Self-concept and school achievement. New Jersey: Prentice-Hall. Reed, R. M. (1988). Education and achievement of young Black males. In J. T. Gibbs (Ed.). Young, Black, and male in America. An endangered species (pp. 37-96). New York: Auburn House.

Richman, C. L, Clark, M. L., & Brown, K. P. (1985). General end specific self-esteem in late adolescent students. Race x gender x SES effects. Adolescence, 20, 555-566.

Rosenberg, M. (1965). Society and adolescence self-image. Princeton, New Jersey: Princeton University Press.

Rosenberg, M., Schooler, C., & Schoenbeck, C. (1989). Self-esteem and adolescences problems: Modelling reciprocal effects. American Sociological Review, 54, 1004-1018.

Rorer, B.A.,& Ziller, R C. (1982). The iconic communication of values. Journal of Cross-Cultural Psychology, 13, 352-366.

Rotheram-Borus, M. J. (1990). Adolescents' reference group choices, self-esteem, and adjustment. Journal of Personality and Social Psychology, 59, 1075-1081.

Spencer, M. B., & Markstrom-Adams, C. (1990). Identity processes among racial and ethnic minority children in America. Child Development, 61, 290-310.

Thorton, B. & Ryckmam, IL M. (1991). Relationship between physical attractiveness, physical effectiveness, and self-esteem: A cross-sectional analysis among adolescents. Journal of Adolescence, 14, 85-98.

Wayson, W. W., & Lasley, T. J. (1984). Climates for excellence: Schools that foster self-discipline. Phi Delta Kappan, 65, 419-421.

Wylie, IL (1980). The self-concept: Vol. 2. Theory and research on selected topics (rev. ed.). Lincoln: University of Nebraska Press.

Ziller, IL C. (1990). Photographing the self. Methods for observing personal orientations. Newbury Park: Sage.

Ziller, IL C. (1988). Orientations: The cognitive link in person-situation interactions. Journal of Social Behavior and Personality, 1, 1-18.

Ziller, R. C., & Lewis, D. (1981). Orientations: Self, social and environmental percepts through auto-photography. Personality and Social Psychology

Bulletin, 7, 338-343.

Ziller, R. C., Martel, IL T., & Morrison, R. H. (1977). Social insulation, self-complexity, and social attraction. A theory chain. Journal of Research in Personality, 11, 398-415.

Ziller IL C. & Okura, Y. (1986). The psychological niche of older Japanese and Americans through auto-photography. Children's Environments Quarterly, 6, 186-196.

Ziller, IL C. & Roar, B.A. (1985). Shyness-environment interaction. A view from the shy side through auto-photography. Journal of Personality, 53, 626-639.

Ziller, IL C., Vern, H., & de Santoya, C. C. (1989). The psychological niche of children of poverty or affluence through autophotography. Children's Environments Quarterly, 6, 186-196.

Mark J. Jones, Ph.D, Department of Speech and Communication Studies, San Francisco State University.

Correspondence concerning this article should be addressed to Mark J. Jones, Department of Speech and Communication Studies, San Francisco State University, 1600 Holloway Avenue, San Francisco, California 94132; Email: [e-mail]

Ah earlier version on this paper was presented at the 1995 Annual Meeting of the Speech Communication Association.