This research was conducted in three phases: 1) formative research, 2) pilot study and 3) instrument development.

Formative Research Phase

This phase of the study involved 15 focus groups conducted with 120 participants who had not previously undergone genetic testing and had little knowledge of human genetics as assessed by screening questions during recruitment. Focus groups were used for their value in “learning how respondents talk about a phenomenon of interest” (p. 15), and collecting preliminary information about health phenomena at the aggregate level. The purpose was to find linguistically and culturally appropriate statements of the lay public’s perceptions related to genes, health and discrimination. Focus group questions stemmed from a literature review of relevant information, a prior genetic discrimination scale and discussions with three culturally diverse community advisory boards comprised of citizens and representatives of community organizations (including libraries, churches, schools and public health) to promote the participatory nature of the project. All focus groups lasted approximately two hours and included 60 persons self-identifying as blacks, 52 whites, seven Hispanic/Latinos, and one Tamare-an/Native American ranging in age from 18 to 51 years (M=32.6). Focus group discussions from the formative research phase were content-analyzed by trained coders who were listening for statements about genes and race, and genes and discrimination concerns. This resulted in the emergence of 48 unique statements about the nature of genetic discrimination and genetically based racism.

Pilot Study Phase

The second phase of the project consisted of pilot-testing the statements derived from the formative research to assess audience perceptions of clarity and meaning. The goal was to refine content for items to be used in the population-based survey. The content of the questionnaire was reviewed by the three multicultural community advisory boards that included men and women from varied backgrounds (as described in 36) as well as a population geneticist and genetic counselor who considered content validity. Readability was assessed using the Flesch-Kincaid index and found to be at the 8.6 grade level. The items were then pilot-tested using five-point Likert scales anchored by l=”strongly disagree” to 5 =”strongly agree” to assess perceptions of clarity, believability, comprehensibility and complexity as well as the emotions experienced as a result of exposure to the statements, including anxiety, fear or anger.
zyprexa medication

The pilot study involved 149 participants recruited from a large land grant university in the southeast, with 58.5% being female and 41.5% male. The majority of the sample self-identified as white (83.8%), with 6.3% self-identifying as African-American, 4.2% as Asian, 3.5% as bi-/multiracial, 1.4% as Hawaiian/Pacific Islander and 0.7% as other. These participants completed self-administered surveys and provided written confidential feedback in response to several open-ended questions about the perceived sensitivity associated with the questionnaire’s content.

In view of the reality that whites as the majority race in the United States are the race most often accused of racism and discrimination with regard to race, we were particularly interested in attaining their perceptions and reactions to the content of survey statements. Thus, our primary goal was to gather insights about perceptions to promote our ability to construct survey statements to which people would respond honestly, avoiding to the extent possible the termination of interviews in the population-based survey because of the sensitive nature of questions.

Pilot study participants’ ages ranged 19-40 years (M=21.66; SD=2.35). While participants perceived content to be sensitive, they did not react to it as being unduly so; content revisions primarily centered on feedback about the seeming redundancy associated with the survey items. Make your pharmacy dollar go further buy viagra uk

Instrument Development Phase

The third phase of the project centered on the development of measures of genetic discrimination and genetically based racism. A 103-item questionnaire was derived based on the formative research and pilot study results, and including the following scales added to validate the measurement model: 1) the Modern Racism Scale, alpha (oc)=0.78 (MRS); 2) the Racial Denial Scale ot=0.68 (RDS); and 3) the Motivation to Control Prejudice Scale, a =0.85 (MCPS). Four versions of the final survey instrument were created to control for possible effects associated with the order of responding to questions.

Procedures and participants. Data were collected via telephone through the use of random digit dialing (RDD) procedures with oversampling of area codes associated with high proportions of black residents designed to obtain nearly equal representation of both blacks and whites. The phone survey was administered by a professional survey research center and utilized trained interviewers and a computer-assisted telephone interview (CATI) system. After introducing her/himself, the interviewer introduced the survey in the following fashion: “We are assisting a professor here at the university in conducting an opinion survey about genes and health, and you have been randomly selected to participate. All answers that you give us will be completely confidential. The survey will take about 10 minutes of your time. In order for the results of the survey to be representative of the state’s population, I need to speak with the youngest male 18 years of age or older who lives in the household.” If the potential participant responded that he met this criterion or sought the participation of someone in the household who met the criteria, the survey continued. If the potential participant responded that there was no one in the household who met the requirements, the interviewer asked, “May I speak to the oldest female 18 years of age or older?” By this introduction, efforts to obtain male participants and older female participants were enhanced. If an eligible person indicated that they did not have time at the moment, the interviewer sought to set up a callback time. Participants were told that: 1) five individuals would be randomly selected to receive $50 for their participation, 2) all information would be kept strictly confidential, 3) participation was completely voluntary, and 4) they did not have to respond to any question that made them uncomfortable. levitra plus

Participants were also told, “My supervisor may also listen to part of the interview for quality-control purposes. The questions that we are asking you do not reflect our opinions on the subject, and some of them may not reflect your attitudes either. There are no right or wrong answers. We ask such a wide range of questions in order to be sure that we capture the full range of attitudes. We hope you will feel free to give us your own honest opinion.” The overall response rate for the completed survey (N=644) was 35.1%, using the completed interviews as a percentage of the contacted, including eligible refusals and partial interviews. This level of response is a reality observed in other survey research and is attributed to the increased number of solicitations by phone and the response gaps associated with software packages that work with CATI to match a live interviewer with a live respondent but too often leave someone on the phone listening to silence. The nature of a study also contributes to the completion rate, with the sensitive nature of the questions asked in this project, including items about racism and measures for validity purposes, which likely contributed to a lower response rate, a finding consistent with previous survey research. In this research, findings may actually be conservative in nature, as respondents may have avoided more extreme responses by refusing to participate, failing to complete participation or tailoring responses in a socially desirable direction. Surveys were completed by 644 randomly selected adults from the state of Georgia, with 224 of these individuals having provided contact information at the end of the survey to allow their names to be included in a random drawing for one of five $50 cash incentives. There was nearly equal participation by whites (48%) and blacks (43%) obtained by targeting area codes known to have a high percentage of black American residents. The remaining participants self-identified as biracial (3%), Hispanic (1%), Asian (0.5%), Native American (0.3%), and other (2%) or refused (2.5%). Among participants, 65% were female, with ages ranging from 18 (3.4%) to 87 (.2%) years of age and a mean of 43.2 years (SD=16.82); the average age of black participants was 41 years, and 44.5 years for whites. Income and education information is included in Table 1, including the baseline characteristics of black and white participants. viagra plus

Table 1. Characteristics of Participants (N=644) for Development of the Genetic Discrimination Instrument (GDI) and the Genetically Based Racism Instrument (GBRI)

Percentage (%)
Characteristic Overall Blacks (n=276) Whites (n=309)
Less than high school




Completed high school




Had some post-high school




College graduate








Income (Annual)




















$100,000 or more




Don’t know








Data analysis. The Statistical Package for the Social Sciences (SPSS) was used to conduct exploratory factor analysis, an appropriate method for data reduction when a grounded theory approach has been used as was the case in this research. Criteria for factor and item retention were: 1) eigenvalues >1.0 for retained factors, 2) primary factor loadings of >0.50 with the exception of one item that had theoretically meaningful content, 3) use of the Scree test to assess relative increase in variance accounted for, and/or 4) interpretability of the resultant factor structure. Prior to conducting the analyses, items were reviewed for nonnormality and eliminated if they demonstrated high levels of skew associated with ceiling or floor effects, or kurtosis greater than or less than -2-+2. An exception was made for two items that related to previous genetic discrimination research that had high face validity. These were: 1) insurance companies should not be permitted to discriminate against those who have genetic flaws; and 2) employers should have the option to not hire someone who is more likely than average to get a genetic disease. As a result of the previous step, 10 items were removed from further analyses (e.g., “Schools should not have access to a child’s genetic tests without permission from a parent”). In general, all participants strongly agreed with this statement (M=4.51; SD=1.09). Principal axis factoring with varimax rotation was selected toward the aim of creating a parsimonious representation of the data and factors to be extracted from statements about genetic discrimination. buy abilify

Category: Diseases / Tags: genetic discrimination, health communication, instrument development, racism, racism and genetics

Leave a Reply

Your email address will not be published. Required fields are marked *