Nine Ways to Rescue a Failing Phone Survey
Believe it or not, there are still plenty of phone surveys being done out there, especially for local needs assessments and evaluation surveys. Often these surveys are mandated by state or local governments to ensure that the right services are reaching people who need them.
But phone surveys are extremely difficult to field anymore, as evidenced by an SOS sent by a research colleague on the AAPOR discussion board last month:
We’re currently fielding a telephone survey … and our completes/hour rate has collapsed since the prior fielding in 2016. We’re currently completing surveys at something like 60% of our 2016 rate. This is a problem for our timeline—at our current productivity we’ll be collecting 2018 data until 2019….
Our cooperation rates and refusal rates remain about on track compared to our prior fielding. Unsurprisingly, it’s response rates that have taken the big hit—we’re just not getting people on the phone. We also don’t really have resources to solve the problem with labor-hours.
We were wondering if anyone had found promising solutions to getting people to pick up the phone, or had ideas for ways to make interviewer time more productive. We’re already rejiggering our callback rules to eliminate the least productive times and callbacks. What other strategies are out there? What have people tried?
Very quickly, a dozen researchers chimed in with ideas such as these:
- Make sure caller ID tags are accurate and not anonymous
- Leave voicemail and a call-back number, and designate interview staff to answer those calls
- Send pre-notification email or texts so that people are expecting (and will answer) a call
- Call during all times of day (e.g. late mornings) rather than exclusively evening
- When leaving voicemail, offer incentive payments
- Ensure that sample aligns with the topic as much as possible
- Pre-screen the sample to remove disconnected numbers
- When leaving voicemail make sure the survey sounds attractive (short, relevant, important)
- Use ABS (address-based sampling) and do a mail-push-to-web survey instead
Research projects hit serious speed (and cost) bumps all the time, and for beginners it is important to know that good brainstormed ideas from experienced colleagues can quickly turn around a failing project. Solving big problems is part of what makes research fun. And knowing that there is a large community of generous research colleagues to help is one thing I have always loved about AAPOR!