Video Computer-Assisted Self-Interviewing (VCASI)

Video computer-assisted self-interviewing (VCASI), sometimes referred to as audio-visual CASI (AVCASI), involves administration of an electronic self-administered questionnaire, which includes prerecorded video clips of an interviewer asking survey questions. This mode of data collection is an extension of audio computer-assisted self-interviewing (ACASI). Responses are recorded either by keyboard, [Page 946]mouse, or touch screen data entry. This technique is typically used in two settings. First is in conjunction with a face-to-face computer-assisted interview, where part of the interview is conducted by the interviewer and part is left to the respondent using the computer in a self-administered mode. Second is in a completely self-administered setting where the CASI questionnaire runs on a stand-alone terminal or kiosk. In an interviewer-administered setting, the respondent uses headphones for complete privacy. In a private kiosk setting the questions might be delivered to the respondent using built-in computer speakers.

VCASI was developed to collect information on sensitive topics, such as illegal drug use, sexual preference, criminal activity, in as valid and reliable a manner as possible. When responding to sensitive questions in the presence of a live interviewer or with other persons present, respondents are sometimes reluctant to answer certain questions completely and honestly. Instead, they tend to provide partial responses, socially desirable responses, or refuse to answer altogether. To overcome such resistance, interviewers equip respondents with headphones, which prevent others from overhearing the questions. The respondent then enters his or her response directly into the computer, thereby ensuring greater confidentiality. This approach is assumed to be less prone to social desirability bias and other response errors related to the presence of an interviewer or third parties. Unlike ACASI, in which the respondent typically hears only the questions being asked and sees text or a visual representation of the question on the computer screen, with VCASI the video of the interviewer is thought to mimic some of the benefits associated with face-to-face interviewing, such as personal interaction.

In addition, VCASI helps address the issue of illiteracy among some respondents, which can be a problem with some self-administered survey modes (such as mail surveys). Because the questions are asked in a video format, respondents do not need to be able to read the questions. They do, however, need to be able to enter the correct response category, which typically is an easier task requiring a lower level of reading competency.

VCASI also provides standardization of interviewer behaviors. Because the interviewer is prerecorded, there is no interviewer variation per se, such as interviewer mood or voice inflection, which might affect how a person responds. Conversely, interviewers cannot react to personal characteristics of the respondent, the interview setting, or to the responses provided and thereby bias the data. In addition, because skip instructions and branching, as well as the sequence of the questions, are predefined and programmed into the VCASI system, no accidental or intentional omission of questions can occur, and all questions are administered exactly as worded. VCASI is not suitable, however, for complicated concepts and questions that require probing or clarification. Also, lengthy questionnaires that may require motivational efforts from an interviewer to maintain the respondent's interest in the survey are not good candidates for this approach.

...