CATI (Computer Assisted Telephone Interviewing) is a time-tested approach to conducting surveys by phone.
Live phone polling consistently delivers high completion rates on surveys—even complex ones that demand a lot from the participant. It also gives more accurate data as interviewers can assist participants and explain things to them they might otherwise misunderstand.
Unlike online polling, it enables your study to access older and less tech-savvy demographics whose opinions you would otherwise miss. Finally, it allows you to uncover richer data through deeper conversations that coded responses just miss.
According to Pew Research, the decline in the use of live phone polling that began about ten years ago was precipitated by rising costs and falling response rates. Let’s examine how these issues can be addressed.
Reducing Costs
- Targeted Sampling: Strategic focus on populations most likely to provide detailed phone responses (older demographics, niche interest groups) minimizes wasted calls and keeps interview length short, boosting efficiency while potentially mitigating response bias.
- Automated Interviewer Tools: Optimizing CATI systems for rapid, intuitive data entry saves time per call, lowering the cost per respondent. Integrating these tools with predictive dialers reduces wasted interviewer downtime through intelligent scheduling.
- Collaboration & Outsourcing: Smaller firms may lack infrastructure for large-scale CATI. Partnering on studies where there is a shared interest can maximize interviewer utilization, while contracting out phone survey work to a specialized provider can further reduce costs.
- Offshore Interviewers: In a globalized world, individuals are more than used to speaking on the phone with overseas contact center staff. A skilled interviewer armed with a CATI system can be located anywhere as long as they can be understood by the target audience and connect with them on a cultural level. This can reduce the cost of conducting live phone polling by around 50%.
- Hybrid Approaches: Online or SMS surveys can be used as an initial screening step to gather standard demographic data. More costly interviewer time can be reserved for more detailed questions where live interviewer input yields a maximum return.
Boosting Response Rates
- Advanced Call Management: Tools from adaptive dialers (pacing outreach based on response data) to algorithms suggesting optimal contact times improve chances of connection, reducing abandoned calls compared to manual attempts.
- Personalized Outreach: Addressing people by name in introductions and using survey data to customize scripts demonstrates attention, reducing the perception of an impersonal robocall.
- Mixed-Mode Options: Allowing an online follow-up for time-pressed respondents may retain partial data. Although not ideal, this will reduce the non-response rate without introducing additional bias or reducing the sample’s representativeness.
- Transparency & Engagement: Briefly outlining the study’s purpose and potential benefit can reduce instant hang-ups. Emphasizing the value of gathering feedback from a specific population sector to which a respondent might belong is particularly useful in reaching under-sampled groups.
Addressing Methodological Concerns
- Interviewer Training & Bias: Thorough training must go beyond the survey itself to develop interviewers’ sensitivity to cues in tone of voice, recognizing signs of discomfort, and navigating refusal in a neutral way. Ongoing review and quality control remain crucial.
- Social Desirability Bias: Combining direct questions with indirect queries that seek to piece together the respondent’s real viewpoint—for example, asking how many peers hold a certain opinion as a proxy for understanding the respondent’s own—can potentially reduce inaccuracy.
- Mobile Phone Inclusion: Using samples that include both landline and mobile phones is essential, as some demographics are exclusively cell phone users while others are easier to reach by landline.
- Non-response Bias: While no strategy offers a flawless solution, careful use of re-weighting and statistical imputation guided by relevant external data can help counter biases in raw phone data compared to the whole population.
For more information about TKW Research’s Live Phone Polling solutions – and to find out how we’ve made it cost-competitive again – visit this page.