Question 1

What are the advantages of computer-assisted telephone interviewing (CATI)?

Accepted Answer

Computer-assisted telephone interviewing (CATI) can be more efficient, have higher data quality, produce richer data, include a broader geography, and uniquely reach certain target segments compared to other data collection methodologies. On the other hand, it can be more disruptive to potential participants, has a more limited array of question types available, can be more difficult to contact potential participants or get responses to sensitive questions, and may also miss certain target segments compared to other methods. The degree of advantage or disadvantage can vary depending on which method is compared and the specific research situation. CATI shares some efficiencies with other computer-assisted methods, such as CAPI and CAWI, because both methods record data in the database as questions are answered. However, there are some ways in which CATI can be more efficient than CAPI or CAWI. For example, a live interviewer can keep the respondent engaged until they complete the interview, while a participant may be more easily distracted during a self-administered interview and finish it later or not at all. Also, if time is a factor, CATI interviewers can keep calling numbers until they reach the required number of completes while completion rates via other methods may be more difficult to accelerate if they lag requirements. CATI also shares some data quality advantages with other computer-assisted methods because all the questions must be answered, all answers must conform to the planned response categories, and there are no extra steps needed for data entry. However, CATI has unique data quality advantages because the interviewer can perceive and correct issues such as participant confusion, differences in vocabulary or language, and wandering attention. They also have more opportunity to determine if the participant is who they claim to be, and the use of telephone numbers for recruiting reduces the possibility of duplication or fraud. Some CATI systems make it easy for interviewers to record such observations. Also, CATI can provide richer data if open-ended questions are included and if interviewers are allowed to probe and record responses to any questions. Generally, it takes longer for people to type than to speak, and some freshness and detail is lost with self-administered open-ends which can be fully captured if the interviewer can type more quickly than a participant, knows shorthand, or is recording the call. Although computer-assisted web interviews (CAWI) and mail surveys share this advantage, CATI can cover a broader geography than central location methodologies, such as some self-administered or intercept interviews. More importantly, it can reach segments of the target market who cannot be contacted via email because they lack an email address (or you lack theirs) or because they do not have access to the internet. It can also be used to include people who are illiterate and therefore excluded from self-administered surveys, other than interactive voice response (IVR) surveys. There are also more subjective factors that are barriers to different methodologies, such as an individual’s inclination to respond to an email versus answer a phone call or their comfort level with talking with someone over the phone versus spending more time staring at a screen.

Question 2

What are the disadvantages of computer-assisted telephone interviewing (CATI)?

Accepted Answer

Computer-assisted telephone interviewing (CATI) have advantages over other data collection methodologies, but it can also be more disruptive to potential participants, have a more limited array of question types available, be more difficult to contact potential participants or get responses to sensitive questions, and may also miss certain target segments. The degree of advantage or disadvantage can vary depending on which method is compared and the specific research situation. Some people may find phone calls to be more disruptive or intrusive than emails, and perhaps most people will be irritated by them if they come at an inconvenient time. However, this disadvantage can be somewhat alleviated if the participants are panelists who have opted in and expect to be called, if the interview has been scheduled ahead of time, or if they have the option to call back at their convenience. Perhaps a more serious disadvantage for CATI is the stricter limitations on the types and numbers of questions. For example, if the topic relies on the participant’s ability to look at a visual, such as a product concept or choice-based exercise, CATI will not be an option. Generally speaking, CATI must have fewer questions than other methods because it takes longer to say them out loud than it does to read them to yourself, plus more time is taken up by building a rapport and providing clarifications in CATI interviews. Another disadvantage is that it can be more difficult to contact people by phone than by other means. For example, some people may be more oriented around a computer screen than their phone or they may be more willing to look at an unsolicited email from a familiar sender than to answer a phone call from a number that is not in their list of contacts. They might also be more willing to participate in an interview when visiting a brand’s website or via a banner ad than to answer a phone call regardless of whether the caller is known to them or not. Also, you may have access to more email addresses and web-based means of contacting people than you have phone numbers you can use. However, a provider of CATI interviewing may already have a database of phone numbers of willing participants who are pre-screened and verified. Willingness to answer sensitive questions may or may not be a disadvantage for CATI, depending upon which method you compare it to, the person asking the question, and the person answering the question. Compared to a face-to-face interviews, people may be more willing to answer a sensitive question over the phone because they aren’t distracted by seeing the person or how they are reacting, much like lying on the proverbial therapist’s couch. Early proponents of self-administered interviews argued that the computer removes interviewer bias from the process. However, self-administered methodologies also preclude a show of empathy or any other kind of rapport-building. Similar to the limitations of CAWI, CATI may be unable to reach certain segments who do not have phones or simply refuse to answer anonymous calls or participate in phone surveys. To ensure they are representing their entire target market, some researchers conduct multi-modal studies, and providers of CATI systems and CATI software may offer solutions that enable these different modes to be managed simultaneously and seamlessly.

Question 3

How is data quality managed in computer-assisted telephone interviews (CATI)?

Accepted Answer

Computer-assisted telephone interviewing (CATI) has some data quality advantages intrinsic to it due to the direct entry of data, the role the interviewer can take, and the more personal nature of phone numbers. Providers of CATI also follow best practices in training interviewers and monitoring field work. Intrinsic to CATI, direct entry of data into the computer eliminates a separate data entry step and, therefore, the potential for data entry errors. The interviewer can ensure data quality by helping the participant through parts of the interview they may find difficult, being attuned to how engaged the participant is, and managing or flagging suspicious participants. Also intrinsic to the process, phone numbers can be more personal than email, so there is greater assurance that when you reach someone, it is the person you intended to reach, and the interviewer can confirm this. The potential for fraud is reduced relative to other methods, and the potential for duplicate interviews is extremely low. CATI managers follow several best practices to ensure high quality data, and some CATI systems and CATI software offer functionality to facilitate this. Typically, interviewers are well-trained before they attempt live interview work, and part of the training would include listening in on other interviews. CATI managers establish a certain percentage of interviews to monitor across all interviewers in each project, and may increase the percentage for less experienced ones or those who have been flagged for one difficulty or another. They may also have the ability to advise the interviewer without interrupting the interview or even to take over the interview if it gets off track. CATI managers also take care to assemble a team of interviewers who can not only meet each project’s productivity expectations, but who are best suited for the target market. For example, an interviewer who is fluent in health care issues can be more helpful during the survey, can ask more insightful probes, and gain more credibility with doctors and other hospital staff than one who is only fluent in computer technology. Certain interviewers may be more compatible with potential participants based on region, gender, education level, and other factors. CATI projects typically include a soft launch or pre-test to ensure that the interview and the process are up to expectations. Clients may be able to listen live or to recordings, on-site or remotely, and give their feedback. CATI systems or CATI software may also help managers to schedule interviews and monitor progress, ensuring that sample and quota sizes are met and not exceeded. Especially across time zones, calling and interviewing schedules must be optimized for local time and expectations around the availability of a potential participant. This can impact data quality because surly participants are not always meaningful participants, and calling at time when people are typically unavailable may make it more difficult to complete surveys, and this takes away from your flexibility to purge lower quality interviews. Perhaps not strictly a data quality issue, monitoring and managing sample and quota needs has been particularly challenging for CATI, but especially important if time and budget are also challenged. Because interviews are distributed across teams of people, each person needs to be aware of which quotas have been filled in real time. Otherwise, more interviews may be completed than are necessary, sending the project over its budget. CATI system and CATI software can help with real-time coordination of progress across interviewers, and this can be especially important when members of the interviewing team are not co-located. Also not strictly a data quality issue, CATI interviewers have been challenged not only by the widespread use of cellular phones versus landline, but also by recent regulations in different countries and regions that dictate how they may be contacted. For example, manual dialing may be allowed while auto-dialing or robo-calling may be illegal. CATI interviewers are aware of these laws and have innovated ways to continue their work without violating them. Some CATI systems and CATI software have added functionality that specifically addresses these restrictions. Some data quality advantages are “baked in” to the CATI methodology, and experts in CATI interviewing have best practices they follow that further ensure high quality.

Question 4

What kinds of questions can be asked in computer-assisted telephone interviews (CATI)?

Accepted Answer

Unlike interviews that are self-administered via computer or paper and pencil, CATI has stricter limitations on the types and numbers of questions it can ask due to visualization requirements, question complexity, and overall number of questions. If the interview topic relies on the participant’s ability to look at a visual, such as a product concept, choice-based exercise, or even a mock up, CATI will not be an option, or at least it will be an extremely challenging one. Sometimes, this challenge can be addressed by mailing, emailing, or faxing an image, or mailing a product or mock up to the participant, then scheduling a call to discuss it. However, simply scheduling a call and having the participant show up for it can be challenging, and that challenge is exacerbated if the person has to have the sent material available at the time of the call. If it is impossible or otherwise impractical to conduct the research any other way, sending the material to the participant may be a necessary evil. CATI questions themselves need to be kept to minimum verbiage and complexity so that the participant can follow them. Even seemingly simple ranking questions or constant sums in which the participant’s answers have to sum to a certain amount, such as 100%, are difficult to visualize and answer with only oral discussion. Plus, you risk frustrating the participant by making them try to answer a question that requires multiple interdependent responses, all of which they need to keep in mind simultaneously. They can provide answers, but will they be motivated more by accuracy or by trying to please the interviewer and moving on to the next question? Also, if there are long sentences, a lot of words to process, or many answer choices, it can be difficult for the participant to retain the flow and understand the question properly. This is a challenge for public opinion polls conducted by telephone that require exposition in order to communicate assumptions, level expectations, or establish a scenario. You can be sure that the ability to follow such exposition varies greatly across the target population, unless that population is somehow psychologically homogeneous. Generally speaking, CATI must have fewer questions than other methods. It takes longer to say questions and answer choices out loud than it does to read them to yourself. Plus, the average person may not be willing or able to spend as much time on a phone interview as one via computer or paper and pencil as interviews in these latter modes can, rightly or wrongly, be broken up across multiple sessions. Plus, some of the precious time they are willing to give you over the phone must be allocated to building a rapport, providing clarifications, and potential disruptions or delays. One advantage of CATI, however, is the potential for richer verbatim responses to open end questions. The participant can concentrate on their thoughts and speak freely instead of typing or writing, and a trained interviewer may be able to take the response down faster than a participant could type or write it. They may be able to record the conversation and provide a transcript. The interviewer can also probe for more detail or clarifications, something that cannot be done in a self-administered interview. It’s best to stick to simple questions with a few closed end answer choices plus some open end topics and to avoid the need for visuals and props, exposition, or complex questions and response choices. CATI providers generally run a soft launch or pretest to identify any consistent issues due to the survey itself, including typical length. They may also have the ability to let you listen live or to recordings so you can be certain that the interviews meet your standards. CATI providers can also draw on their deep experience and advise you on best practices before you test the interview.

Top CATI - Computer-Aided Telephone Interviewing Companies

What are CATI - Computer-Aided Telephone Interviewing Companies?

Featured Experts in Top CATI - Computer-Aided Telephone Interviewing Companies

Compare CATI Interviewing

Bellomy Research - Data Collection Center

Bellomy Research - Data Collection Center

NewtonX

NewtonX

Precision Opinion, Inc.

Precision Opinion, Inc.

The Logit Group

The Logit Group