FULL PACKET_2008-04-07 - Laserfiche WebLink

Home Browse Search

Language Capabilities We provide multicultural research strategies as appropriate. We find the literal translation of English instrumentation to be insufficient, and we emphasize the development of linguistically and culturally appropriate research instruments and methods. We regularly conduct research in the Latino and Vietnamese communities, and have conducted surveys in Korean, Cambodian, Armenian; Tagalog, and Mandarin. Our university setting allows us to access the linguistic a11d cultural expertise of members of many raciaUethnic acid national groups. The Cal State Fullerton campus affords access to abroad range of faculty expertise and student perspectives, representing many etlnlic backgrounds and heritages. Translated survey instruments are programmed into Ci3 scripting software that interfaces with the CATI system. This allows interviewers conducting English and non-English language sur<~eys to read the questions on the screen and type verbatim responses into the CATI system. Procedures for handling these open-ended text responses are discussed in Appendix B. Sample Design A confidence interval for survey data refers to the level of precision required for estimates of population parameters based upon sample data. A generally accepted statldard for policy-relevant research is a confidence interval of plus or minus five percent. When this standard is met, we can be 95% confident that the true population parameter lies within an interval extending five percent above and below any proportion derived from survey (sample) data. A population paranieter is the result one would obtain if all universal access customers were interviewed. Sampling error, as indexed by the confidence interval around reported proportions, varies in relation to sample size and to the variability of survey responses (among other factors). In general, as the sample grows larger; the confidence interval grows narrower. That is, inferences about population parameters based upon larger samples are more precise (are. associated with narrower confidence intervals) than inferences based upon smaller samples. Also, as the proportion of soiree attribute in the sample (e.g. overall satisfaction with the Santa Ana WORK Center) approaches afifty/ fifty split, (say, 50% are dissatisfied and 50% are satisfied) sampling en-or increases, resulting in a wider confidence interval. Conversely, sampling error decreases as the proportion of a given attribute approaches a five/ ninety-five split, (5% are dissatisfied and 95% are satisfied) resulting in a more precise estimate and a narrower confidence interval. Assuming a population of unique Universal Access customers of roughly 5,000 in the program year under study, the table on the following page depicts the required sample sizes for three confidence intervals: 1) Calculated leniently, that is, based upon the 6 25N-32