More About Computer Assisted Telephone Interviewing
Computer Assisted Telephone Interviewing (CATI) utilizes state-of-the-art computing to reduce the time, effort, and money required to conduct large surveys. The CATI/Ci3 survey laboratory is designed to implement telephone surveys efficiently. CATI and Ci3 are two separate software packages that were developed to work independently or together. These packages allow the research team to tailor their questionnaires, randomly select a sample, contact respondents, enter the responses, and conduct rudimentary statistical analyses.
The Ci3 software was developed to assist in designing questionnaires. The software allows researchers to tailor their interview schedules to meet their needs. The questionnaire is displayed on the screen while interviews are conducted. One question appears per screen thereby reducing the likelihood of interviewer mistakes. Ci3 also controls contingency questions so respondents are asked only relevant questions.
The Ci3 software also can generate lists based on the respondent's previous answers. For example, a respondent may be asked which airlines they had used in the last year. Based on their answers, Ci3 will build a list consisting of only those airlines. This function permits interviewers to ask additional questions concerning the selected airlines while skipping those questions about airlines with which the respondent had no contact.
In addition to Ci3's question formatting capabilities, the program is designed to allow direct data entry. Interviewers can type the respondent's answers as they are given. The Ci3 software saves these data and can be used to convert the raw data to files that can be analyzed using advanced statistical packages such as SAS or SPSS. These functions eliminate several steps in the survey process and therefore reduce the likelihood of coding and data entry errors.
The CATI software enhances the researcher's control over the entire survey process. The software can be used to set up pre-interview specifications that will manage the survey. First, the software can be used to assign interviewers to specific studies. Our laboratory, for example, has 9 workstations from which interviews can be conducted. CATI can assign five of these to one survey, two to a second survey, and two to a third study. Thus, several surveys can be conducted at one time. Moreover, CATI can be programmed to dial numbers at specific times across several time zones.
Next, CATI can assist in the sample selection process. The program can generate random numbers. The software also can be used to monitor sample quotas when selecting a stratified random sample. For example, if the sample needs to be comprised of 113 females over the age of 42, CATI will monitor how many such respondents have been interviewed. The program will then notify the supervisor when the quota is met. This function helps assure the representativeness of the sample.
The CATI program can automatically dial randomly generated telephone numbers or randomly select numbers from a list that has been entered into the computer. The program automatically re-dials a number that was busy when first called. The time between re-dials can be specified by the researcher. Moreover, the CATI program allows interviewers to schedule call-backs for later times or dates.
These functions, and the many others of CATI/Ci3, enhance our ability to conduct time and cost effective surveys. Data are directly entered into the computer, and results can be obtained instantaneously. CATI/Ci3 puts the social world at your fingertips.