Contents - Previous - Next


Ethnographic field study of Latinas in Los Angeles County, California began in 1987. The first study focused on the interrelationship of Latinas' knowledge and beliefs about other dangerous diseases and AIDS. The results of this exploratory study of 19 Latinas at a community based clinic indicated a lack of basic data on three aspects of Latinas and AIDS (author's unpublished master's degree research project, December 1987).

First, there was a lack of basic research describing the breadth and depth of attitudes, biomedical and folk knowledge about AIDS, and reported risk behaviours for different Latina sub populations in Los Angeles. Collecting very specific information regarding the range of sexual behaviour between and the variation within different Latina ethnic groups is essential to the study of AIDS risks for these populations.

Second, there was a need for the development of culturally appropriate AIDS educational programmes and materials for Latinas, programmes which could be adapted for use by the various Latina ethnic groups and/or sexual behaviour groups being served. It was also apparent that most of the available AIDS educational materials were not well suited to the majority of Latinas. Many of the materials had been translated from English to Spanish and so were not culturally sensitive. In addition, the materials often were at a literacy level exceeding the reading skills of the Latinas. Also, there were no materials which addressed the different needs of the Latinas who were lesbian or bisexual.

Third, there did not seem to be culturally appropriate methods of assessing AIDS attitudes and knowledge among Latinas. Most of the pre- and post tests which were being used consisted of a limited number of biomedically oriented AIDS knowledge questions, often in true/false format which may be unfamiliar to many Latinas. In order to provide a way to assess the range of beliefs, attitudes, and knowledge about AIDS, it seemed necessary to develop and test various educational and assessment tools within a culturally appropriate setting.

The Latinas and AIDS research project

In the fall of 1987, the Latinas and AIDS Research Project, developed by Laura Ramos and Loretta Samaniego, began with two main objectives: First, to collect basic data on biomedical and folk AIDS knowledge, perceptions of AIDS risks, and reported AIDS risk behaviours for Latinas of different ethnic and sexual behaviour groups. Second, to use the information to develop and test culturally appropriate AIDS educational materials and programmes for Latinas of these different subpopulations. (See Table 1 for study sample.)

For this study the following definitions were used: A Mexican Woman is defined as a woman born and raised in Mexico and came to the United States as an adult or born in Mexico and came to the United States after the age of 5 years. A Mexican American woman is defined as a woman born and raised in the United States or born in Mexico who moved to the United States before the age of 5 years. White women were defined as any Caucasian women.

As this is an ongoing study, we have completed about 130 interviews thus far. Complete detailed sociodemographic data is not currently available; however, the ranges of some of these variables are: Age: from 18 to 60 years old. Income: from no income to over $40,000 per year. Educational background: from no years of school completed through completed doctoral degrees. Women in each ethnic and sexual behaviour category have been interviewed. Of the women who have participated so far, Mexican women have been the least educated and the poorest. White women have been better educated and the most financially secure. They have come from every walk of life: a former teen prostitute, mothers, housewives, social workers, business executives, and custodians. All women had to be residing in Los Angeles County at the time of the interview. Some women have known a great deal about AIDS and others have known almost nothing (except that it can kill you). It seems that those whose lives have been directly affected by the HIV/AIDS epidemic know more (and with greater accuracy) than those who either do not work in the field of AIDS or who have not had any family member or friend affected. More precise analysis will be forthcoming when the collection of data is completed.

Due to the difficulty of reaching these specific subpopulations random sampling was not chosen. The actual numbers or proportions of lesbian and bisexual women within Latina populations in the United States are unknown. Equal quotas of ethnic and sexual behavioural groups are used to gain in-depth insights into the differences and similarities of perceptions, knowledge, and reported risks between and within these groups. Sexual and other reported AIDS risk data is collected for a period of the previous 10 years from the date of the interview. Convenience and snow-ball sampling methods were used. Mexican and Mexican American women are being studied because they are the most predominant Latino groups in Los Angeles County. The comparison group of White women is included to try to separate the effects of education and ethnicity/ acculturation on differences in AIDS perceptions, knowledge, and risk for AIDS.

Table 1. Sexual Behaviour Types

Women (N=180)

Sexual Behaviour Category


Mexican American


Sex only with men




Sex only with women




Sex with both men and women




Little is known about the differences in sexual behaviour within and between various Latina ethnic groups in the United States. Obtaining data on the differences in sexual practices between the three ethnic and three sexual behaviours groups will increase basic data on probable variation in HIV risk due to reported voluntary sexual behaviours. Other pertinent AIDS risks behaviours are included in the study as well: Intravenous drug use, use of injectable folk remedies, receipt of transfusions, rape and incest, artificial insemination, and the Latinas' sexual partners' risks. Beyond the effort to expand basic knowledge about the variation of AIDS risk behaviour and beliefs, the data are being collected with the purpose of developing culturally relevant small group workshops and materials to suit the specific AIDS prevention educational needs of Latinas of different ethnic and sexual behaviour groups.

The use of RAP methodology has been crucial in conducting this investigation. Observation, participant observation, informal interviewing, and a semi-structured ethnographic interview format have been used in this study. Experience gained with this population assisted in the development of the guides for the AIDS RAP written in 1989-90 [Scrimshaw, et al. in press]. Anthropologically based research methods have proven essential to locating and establishing rapport with women who belong to the bisexual and lesbian sexual behaviour groups within each ethnic sample. It is extremely difficult to study women on the basis of their reported sexual behaviours, as they do not gather together on the basis of their reported sexual behaviours, but rather on the basis of their sexual identity.

To reach these Latinas, contacts within gay and lesbian Latino organizations throughout Los Angeles County were established. Then snowball sampling was used to reach other Latina lesbians. A major difficulty was the fact that many Latina lesbians do not participate in most of the organized and visible part of the gay culture in Los Angeles. Fieldwork and recruitment were also conducted at social, music, religious, political and other significant events and places in order to contact the small networks of Mexican and Mexican American lesbians within the county.

Finding bisexual women was the most difficult, because many women who have had sexual contact with both men and women do not self identify as being bisexual. In the United States there is a tendency to define people either as exclusively homosexual or heterosexual, and people identified as bisexual are often considered unacceptable by both groups. Due to the stigma and complex nature of forming an alternative sexual identity, many of the lesbian-identified women actually report voluntary sexual behaviours with men and women within the last ten years.

Thus, it is important to note that one's identification with a particular sexual orientation does not in any way determine or define the limits of her possible sexual risk behaviours for AIDS. Obtaining information about a woman's reported risks is more valid than relying on a woman's reported sexual identity. The Latinas were placed in sexual behaviour categories based on their reported sexual risks over the previous ten years. Women who reported no voluntary sexual contact (with men nor women) were placed in the category which they identified as being a predictor of their future sexual behaviour (e.g. if a woman thought she would only ever choose to have sex with men, she was placed in the "heterosexual/women who have sex with men category"). Lesbians whose only sexual contact with men involved nonvoluntary contact (rape or incest) were categorized in the "women who have sex with women" category.

Decisions about how to categorize women's voluntary versus nonvoluntary sexual risk behaviours were made on the basis of this question: What types of AIDS prevention education do Latinas need? It is possible to try to teach women to change their voluntary sexual behaviours with partners. Nonvoluntary sexual contact (including forcible rape, pressured intercourse, incest, and any other type of sexual contact in which the women do not willingly participate) should be viewed as a form of violence perpetrated on these women and is not amenable to the same sorts of behavioural and attitudinal changes which one hopes to achieve within an AIDS prevention workshop. For example, having been raped or a victim of incest does not change a Latina lesbian's sexual identity, although it does change her actual sexual AIDS risk.

From observing Latinas in the community and learning from them, it was possible to avoid the pitfall of using sexual orientation labels as the basis of separating the risk groups. Two important points were brought out: First, many women do not understand sexual orientation labels. There is one section in the interview format which asks women to define the following words: sexual orientation, heterosexual, bisexual, homosexual, and lesbian. Most women understood what the terms homosexual or lesbian meant due to explicit cultural taboos which indicate that sexual contact with someone of the same gender is unacceptable. The most common incorrect definition for bisexual was "people who have sex with two people at one time". The term sexual orientation had two common incorrect definitions: "the first time you have sex" and "a workshop or a class on sex". Many Latinas and a few of the White women did not know the meaning of the word heterosexual. It was believed that women who had sex with men were "normal".

Secondly, understanding how scientific or cultural labels are used within one's target population is very important in the development of AIDS prevention messages. For example, every time a television or radio announcer begins a message, "Heterosexual women are at risk for AIDS due to...", the women who identify as "normal" (whose behaviour is heterosexual) would have no idea that the message was designed to reach them. Also, many women who have sex with other women would never use the term "lesbian" to define their sexual identity. There is too much stigma attached to it. Thus, it is more important to find simple, yet clear ways to describe the behaviours which put women at risk than to assume that every one uses a sexual orientation label in the same manner.

AIDS prevention workshops for Latinas

In 1988, with Ms. MaryLou Gutierrez, an AIDS educational workshop for small groups of Latinas entitled, "De Mujer a Mujer", was developed. The groups are designed to be a safe place for women to gather together and discuss issues relevant to AIDS and AIDS prevention. It is important that the groups be led by women for women, as it is culturally acceptable for Latinas to discuss these topics with other women. If men were present, many of the women would be too uncomfortable to participate in any active way in the workshop. Many would not discuss sexual matters openly in front of men.

The workshop is based on an interactive discussion format, with the workshop facilitator attempting to actively involve all the participants in the discussions and group activities. To make the information more accessible to the women, the workshop begins with a poem. Statistics may give exact information, but they do not get the attention of a Latina's heart. Games and role playing activities were developed to raise intellectual and emotional awareness of AIDS. Inclusion of some joking behaviour within the workshop was necessary to diminish the tensions of discussing taboo topics. In order to evaluate the effectiveness of the workshop and to assess the basic AIDS knowledge, attitudes and beliefs of the Latinas, an illustrated AIDS fact and fiction card sorting test was created.

The illustrated fact and fiction cards were simple line drawings (like cartoons) depicting possible ways of transmitting AIDS. These illustrations included both biomedical and folk beliefs about AIDS transmission. Of the 38 card set, 30 cards were illustrated and eight cards were not illustrated. All the cards had simple text describing the behaviours in the illustrations. For eight cards, text only was used, as it was difficult to accomplish the following: make some drawings different than other ones (e.g. transfusions vs donating blood); create some illustrations within culturally acceptable mores (e.g. rape or incest); to show, in one simple picture, a complex type of risk behaviour (e.g. sex with a bisexual person or sex with an IV drug user). The text on the cards is in English, Spanish or both languages.

Each woman was asked to sort a set of cards into three piles: "You can get AIDS this way", "You can't get AIDS this way", or "I don't know if you can get AIDS this way or not". Collecting the data this way, as an interactive form of pretest seemed to benefit the women in two ways. First, the card game was seen as much more fun and interesting than the written test which is given to the women first. The women were actively involved in trying to figure out the answers to where the cards should be placed, more than when trying to read through the written pretest. The cards were then used as educational tools to generate discussion of AIDS transmission within the group. Second, private exposure to taboo topics through the card game seemed to lead to freer group discussion of these topics. The card game was played at the end of the workshop as part of the posttest.

Qualitative analysis, through directly observing the women's behaviour, indicates that the card game was a much more enjoyable and engaging task than the written pre- and posttests. Quantitatively analyzing the numbers of correct, incorrect, and don't know responses gave us a more definitive understanding of the changes in attitudes and knowledge from the workshop process. Pre- and posttest card data was collected from a total of 12 workshops for small groups of Latinas (primarily) and other women (a mix of Whites, Blacks, and other) in Los Angeles County from May 1988 through August 1990.

A summary of preliminary results from these workshops is shown in Annex A and Annex B. For every group, an increase in knowledge is shown from pre- to posttest, for an average increase in correct responses of 11% overall. However, the two most promising results of this type of culturally appropriate illustrated test are:

1. "Don't Know" responses show a sizeable decrease from pre- to posttest. In the pretests "Don't Know" responses are about 10% of the total, while in the posttests "Don't Know" responses virtually disappear (0-2%). This indicates that the women are becoming much more sure about what they do know. Including the option of a "Don't Know" response, rather than the more limited true/false or yes/no type of answer format, gives a more accurate picture of the women's knowledge. There is a reduction in the amount of guessing that goes on when informants are allowed to indicate that they simply do not know an answer.

2. The success of the workshop is also seen in how the women change their "Don't Know" and incorrect pretest responses. Overall the women change their minds about 7 behaviours in the card sets. Analysis shows that 5 (71%) of those 7 changes are in the correct direction (i.e., from an incorrect or "Don't Know" response to the correct response).

The amount of change in knowledge and attitudes is approximately the same for the overall group and the total subset of Latinas. However, when the subset of Latinas is divided into those who have 10 or more years of education and those who have 9 years or less of education, there seems to be a difference in the effects of using this method for testing. The Latinas with less education knew less at the pretest (65% compared to 81% overall). And they had the highest percentage of "Don't Know" responses (20% compared to 10% overall). Their pastiest scores seem to indicate that the workshop and the illustrated test helped them more than any other group. They had almost double the increase in correct responses (20%) compared to the group overall (11%). And even though the pretest percentage of "Don't Know" responses was twice that of the group overall, the low literate Latinas also decreased this to approximately the same low (between 0-2%) posttest response rate.

In Annex B. the Latina subset responses are analyzed by each behavioural concept depicted on the cards. The percentages indicate the proportion of Latinas who answered correctly for each card type. For example, 44% of the Latinas answered the card about tattoos correctly in the pretest, but 78% answered correctly in the posttest. By analyzing the individual card data, one can see where the workshop needs to be improved, what kinds of information or AIDS beliefs are the most amenable to change in this group of Latinas, and which beliefs are the most difficult to change. At the bottom of the page, one can see the value of grouping the AIDS knowledge and belief cards by types of beliefs. The women did well overall on biomedical knowledge, but still need to improve on folk beliefs. Just addressing biomedical knowledge will not be enough in designing AIDS educational materials or workshops for Latinas nor for many other groups.

There are other ways in which this information can be categorized or organized to improve the utility of the information for policy making, research design, and health education planning. Taxonomic analysis and cognitive mapping can be used to examine how Latinas' knowledge, attitudes, and beliefs about AIDS affect their perceptions of risk and risk behaviours. As shown in the card game results, Latinas concurrently hold biomedical and folk beliefs about AIDS, even if these beliefs contradict one another. The beliefs are categorized by different types of AIDS transmission, the names and descriptions of the behaviours they associate with each transmission category, and who they perceive is at risk by each specific transmission category. Understanding the ways cultural beliefs and practices affect Latinas' application of AIDS biomedical and folk knowledge to their AIDS prevention behaviours leads to developing more effective AIDS health education intervention techniques. This type of analysis is also useful in determining the knowledge and beliefs of health workers. Comparing the health worker analysis and the target groups' taxonomic analysis can give real insights on the gaps in knowledge and attitudes which must be bridged before useful communication and education may take place in a particular community.

Implications for future research and application

While the illustrated fact and fiction cards show promise, there is much work to be done in improving the test design and scientifically testing the format and literacy levels of the test for Latinas. (This type of testing would need to be done for each target group for which the test and materials are used; it is not just transferable from one culture or subculture to another.)

As a result of the efforts of the Latinas and AIDS Research Project, funding was obtained by Dr. Erin Quinn and Laura Ramos through the University-wide AIDS Research Program, University of California, to found the AIDS Literacy Project for Low Literate and Non Literate Latinas in Los Angeles County in the University of Southern California, Department of Family Medicine. This project is continuing the efforts to design and test effective, low technology (black and white line drawings), culturally appropriate AIDS educational materials and assessment tools for Latinas.

The illustrated fact and fiction cards are going through extensive revision through the following process: New drawings are being made of the behaviours which may or may not transmit AIDS. These drawings are being done by two Latino artists (one male and one female). Then all the pictures are taken to three local community-based clinics in Los Angeles serving primarily Latino populations. Individual Latinas are interviewed and asked to describe (verbally) what is going on in the illustrations. The interviewers write down verbatim whatever the women say is happening in the pictures.

Each woman is only asked to review seven to 10 pictures so that the interview does not get to be overwhelming (there are currently over 120 pictures being tested). When enough complete sets of pictures are reviewed in the clinics, the data collected will be used to determine which pictures are adequately visually literate to the Latinas. The accumulated comments allow for appropriate revision and retesting of the other pictures until adequate visual literacy is obtained for all - meaning that the Latina's descriptions about what they see in the illustrations match what the researchers/health educators think is represented in the pictures. This process makes the concepts in the illustrations and the test available to be used with Latinas who have no reading skills.

Then, once the pictures are found to be visually literate for our population, we can begin testing various simple text with the pictures to make sure that the text is also culturally literate and readable for the Latinas. Appropriate text can reinforce the concepts portrayed in the illustrations and clarify any confusion for those Latinas who can read.

This application of visual literacy is different from how it is generally used. Most of the time educational materials are made for target populations with a standard level of literacy established by the researchers or educators which is selected to serve the target population; yet many members of the target population may not be literate enough to use the materials. This manner of visual literacy and testing of materials allows for them to be tailor-made to the target population's current literacy level and allows them to be accessible to those who do not read at all. The target population does not have to attain a new level of educational achievement nor literacy in order to be reached by the health prevention message.

In addition to the behavioural cards, we are developing similar kinds of illustrations for sexual and reproductive anatomy and possibly for depicting the common sexual behaviours mentioned in many AIDS brochures. While this part of the testing has not yet been conducted, previous research has shown that this topic is even more culturally sensitive than other types of AIDS related behaviours. The illustrations which have been completed are a cross between very artistic and medical drawings. The tone of the pictures is warm and realistic, nonmedical and yet not erotic. This will be a challenging part of the project, yet a very important one.

It is far too often assumed that adults know the sexual and reproductive anatomy, and furthermore, AIDS prevention literature often misses the target populations, because the words are either totally unknown (medical) or offensive (vulgar slang). This process will allow us to find out what parts of the body (for male and female anatomy) Latinas recognize and what words they use to describe the different bodily parts. Then the materials can be adapted to the educational/sexual literacy level of the Latina target population.

When the entire process is whole, the complete package of illustrations with simple text will become a functional AIDS literacy test. With this test we will be able to determine the functional literacy that the Latinas have in the areas of AIDS. Then we will be able to refine our AIDS educational materials, brochures and workshops to include the most needed topics and to reinforce what they already know. Functional literacy refers to the ability of a person to read/comprehend visually enough language and other cultural symbols in a particular subject or practical area to function at an adequate level. Functional AIDS literacy would then mean the ability of Latinas to recognize behaviours which will transmit AIDS, behaviours which will not transmit AIDS, sexual and reproductive anatomy that is essential to understanding sexual AIDS transmission and in applying preventive sexual measures, and the ability to effectively apply their knowledge in their everyday AIDS prevention practices.

It is particularly important to test the level of sexual literacy among all health care workers who will be assisting in the implementation of the AIDS health education, services, or planning programmes. Many times even well educated individuals are unable to draw male and/or female internal and external reproductive or sexual anatomy, and are unable to accurately label the drawings. A small sample of health professionals (72), who participated in conference workshops given by the author, was unable to do the drawings and was also unable to accurately label medically oriented drawings which had fill-in responses. It is not possible to conduct an adequate AIDS programme where even the staff does not have adequate sexual literacy levels.

The concept of drawings is also an important tool, because cross-cultural ideas about how the body works and how sex or reproduction occurs vary from society to society. Drawings may be one way to communicate with low literate populations about these topics. Then the AIDS programmes could be adapted as necessary to the particular anatomical beliefs held in that culture.


RAP methodology has proven essential in studying heterosexual, bisexual and lesbian subpopulations of White, Mexican, and Mexican American women in Los Angeles County. The active process of seeking out of these special target populations and the interactive approach of learning from the women within the county greatly helped in the development of a very sensitive, culturally appropriate and useful semi-structured interview format with these women. It would not have been possible to study these special groups of women without utilizing ethnographic street research, observation, participant observation, and informal interviewing to reach these women in the community.

Information gathered from ethnographic fieldwork and through face to face interviews has resulted in the development of a culturally appropriate AIDS educational workshop and specific AIDS educational materials. It is leading to the development of a AIDS assessment tool for low or non literate Latinas. If the new research project's efforts prove fruitful, they will dead to the development of a type of functional AIDS literacy test. Such a test would make it easier to assess the literacy level of Latinas on the specific subject areas involved in AIDS education and prevention. This could lead in turn to more tailored programmes for the needs of specific target groups.

More research needs to be done in developing, testing and applying the concepts of visual literacy and culturally appropriate illustrations within AIDS education and prevention programmes and materials. While the Latinas and AIDS Research Project has been working on the development of such materials since 1988, these concepts of providing useful illustrated materials for use in occupational and health educational programmes have been present throughout the AIDS epidemic and for many decades. A recent programme with interesting results is Younger and Friel's "How to develop culturally appropriate condom instructions". This project, sponsored by the World Health Organization's Global Programme on AIDS, reviewed many types of condom related illustrations with varying success in four different countries. The success reported from their project and the preliminary success reported here strongly suggest that much more effort should be placed in the development and testing of the efficacy of illustrated AIDS educational materials. Furthermore, since many small clinics have tried to put together their own versions of illustrated tests, because written ones did not work with their clinic populations, a need to have some more coordinated international effort is apparent.

Culturally appropriate AIDS assessment tools and illustrations can assist not only in educational programmes, but in other aspects of research and programme planning as well. If one has tested the illustrations for use with a particular target population, there is no reason that these illustrations could not be used for the following:

• In brochures, posters, or other visual materials.

• In knowledge and attitude assessment tools.

• In interviews with members of the target populations. For example, in the form of small illustrations which would allow shy or timid members of the target population to point to rather than speak directly about taboo topics.

• In better planning the AIDS educational and service needs of the target group. These types of illustrations will add a valuable tool to the RAP inventory.


Arguelles L., Rivero A., HIV Infection/AIDS and Latinas in Los Angeles County: Considerations for Prevention, Treatment, and Research Practice. California Sociologist 1988; 11(1-2).

Centers for Disease Control, HIV/AIDS Surveillance Report, US AIDS cases reported through May 1991. Atlanta, GA: CDC, June 1991.

Chu SY, Buehler JW, Berkelman RL, Impact of the Human Immunodeficiency Virus Epidemic on Mortality in Women of Reproductive Age, United States. J Am Med Assoc 1990; 264(2): 225-229.

Chu SY, Buehler JW, Fleming PL, Berkelman RL, Epidemiology of Reported Cases of AIDS in Lesbians, United States 1980-89. Am J Publ Health 1990; 80(11): 1380-81.

Scrimshaw SCM, Carballo M, Carael M, Ramos LJ, and Parker R. AIDS Rapid Anthropological Assessment Procedures. Boston, MA: International Nutrition Foundation for Developing Countries (INFDC), in press.

Scrimshaw SCM, Carballo M, Ramos LJ, Blair BA, The AIDS Anthropological Assessment Procedures: A Tool for Health Education Planning and Evaluation. Health Educ Quart 1991; 18(1): 111-123.

Worth D. Women at High Risk of HIV Infection: Behavioral, Prevention, and Intervention Aspects. In: Ostrow DG, ed. Behavioral Aspects of AIDS. New York: Plenum Medical Book Co. 1990.

Worth D. Minority Women and AIDS: Culture, Race, and Gender. In: Feldman DA, ed. Culture and AIDS. New York: Praeger Publishers. New York. 1990.

Additional reading

The ACT UP/NY Women and AIDS Book Group, Women, AIDS, and Activism. Boston, MA: South End Press, 1990.

SIECUS Report. Special Issue on Women and AIDS. New York: Sex Information and Education Council of the United States, December 1990-January 1991.

Contents - Previous - Next