Home :: Research Focus Boxes :: Research Issues :: Internal and External Validity

Internal and External Validity: Cholas and Gang Girls

By Michael Wapner

In Los Angeles County alone there are more than 600 gangs with approximately 100,000 members. Gangs, once restricted to low-income inner city neighborhoods, now stretch into surrounding suburban neighborhoods and schools. Mexican American adolescents make up an increasing number of gang members in the suburbs surrounding Los Angeles, with females composing a significant subgroup in all Mexican American gangs.

Mary Harris (1994) of Bloomsburg University sought to understand the world of gang members, specifically that of Mexican American gang girls, or cholas, from their perspective. Accordingly, Harris interviewed 21 current and former gang girls in suburbs of Los Angeles, sometimes individually and at other times in groups, asking what motivated them to join a gang and how they felt about their gang activities. She conducted these interviews in different settings—in parks or street corners, neighborhood centers, the barrio, and their homes.

As a science, psychology has developed many powerful and subtle research techniques to ensure, as much as possible, the validity of its conclusions. Some of these techniques have been presented in the Research Focuses, like this one, distributed throughout the book. However, sometimes the most sophisticated techniques cannot deliver as dramatic and moving a look at living human experience as can a straightforward approach, such as simply approaching individuals in the daily contexts of their lives and asking them to recount their experiences.

Harris found a strong cohesion and an unwavering loyalty to the gang. When asked their reasons for joining, for instance, these girls spoke of their sense of belonging, as well as their need for group support.

Benita: All it really is that you want to be a Chola because you see the other girls that want to be a Chola and it looks as if they have fun and everything. You want to put the make-up on like them too. (Harris, 1994, p. 293)

Reselda: You can belong as long as you can back up yours—and don’t rat on your own homegirls or back away. If you don’t back them up and you run we’ll jump that girl out because she ain’t going to take care of nothing. All she wants is our backup and our support but she ain’t going to give us none of hers, so what’s the use of her being around. She has to be able to hold up the hood. (pp. 292–293)

These girls also spoke of their gang as a family; for many, loyalty to the gang came first. Most gang members derived their status, self-esteem, identity, and sense of belonging from the gang, which substituted for often weak support from family and the absence of any real connections to school.

Reselda: I used to hate my dad because of what he did to my mom. I grew up with this hatred and anger. . . . A lot of them do come from families that are messed up. A lot of girls, like they ain’t got backup in their families. If they get into a gang they got more backup. They’ve got more girls to really hang around with. . . . They ain’t got too much love in the family. So they don’t care what’s going on. If their family don’t care, she don’t care. Nothing’s going right in her house so what should she care about what’s going on out there. (p. 294)

As powerful and as clearly worthwhile as research such as this is, there can be problems. Harris conducted her interviews in the girls’ homes and neighborhood settings, even at times having several members of a gang present for a single interview. An advantage to conducting interviews in this way is that the girls were most likely to be comfortable, and hence candid, when in familiar settings or with their friends. A problem, however, is that one cannot be sure that what they said about their experiences was not influenced by the presence of other members of the gang or even by their knowledge that, though alone with Harris, their remarks might be overheard. It is possible, in other words, that they may have been less willing to admit to conflicts about belonging to a gang or to other feelings under such conditions. In short, there was more than one possible explanation for some of the findings. When research does not provide an unambiguous answer to the question it was designed to address, we say that it lacks internal validity.

A second type of validity is external validity. Does the answer we get apply only to the people we have observed, or can we generalize the findings to others? Just how representative are the findings? Can we assume that we have an accurate picture of the life of female gang members all over Los Angeles? California? The United States? This is the problem of external validity. The impression we get of gang life can be totally valid for these girls (a question of internal validity) and yet be largely invalid for other gangs (a question of external validity). External validity can also be affected by the very conditions that are necessary to achieve internal validity. Were Harris to have attempted to experimentally control for some of the factors that might affect the way gang girls responded in these interviews—such as by conducting individual interviews on a university campus—she might have obtained data that were unrepresentative of the way these girls would normally have responded, or might not even have been able to interview any gang girls at all!



M. G. Harris. (1994). Cholas, Mexican-American girls, and gangs. Sex Roles, 30: 289–301.