Response rates to telephone public opinion surveys continue to decline, making them more expensive and less attractive than online panel research. We’ve been pointing to this trend for years. Now Pew Research confirms it.
The response rate on landline phones to survey calls in 1997 was 36 percent. In 2018, it fell to 6 percent. Potential telephone survey respondents have declined recently because of the surge in robocalls. Phones with caller ID also discourage answering unfamiliar rings and sometimes flag survey calls as spam.
This isn’t the end of public opinion research. Online panel research has represented an attractive and versatile alternative for some time. Participation rates tend to be higher, there is an ability to follow up with some or all of respondents and it appears participants are more candid online than on the phone. Participation is higher because respondents can answer survey questions when it is convenient for them, as opposed to when someone calls on the phone.
According to Pew Research, low response rates on telephone surveys, especially ones that include cell phones, don’t equate to lower accuracy of findings. The real impact is higher costs. “This reality,” Pew says, “often forces survey organizations to make trade-offs in their studies, such as reducing sample size, extending field periods or reducing quality in other ways in order to shift resources to the interviewing effort.” Those trade-offs can lessen confidence in results.
Lower participation rates on telephone surveys aren’t new. They have steadily declined since at least 1997. Rates stabilized around 9 percent in 2013, then started plunging again in 2016. Lower participation rates have persuaded Pew to conduct most of its US polling online using its American Trends Panel.
CFM has recommended online panel research to skeptical clients. To ease skepticism, we have benchmarked online results with results from telephone surveys, showing that results are comparable.
As telephone survey participation rates have declined and sample sizes have been trimmed, panel research offers an affordable opportunity for larger sample sizes, often larger than even healthy telephone survey samples.
Larger sample sizes can increase the confidence rate for panel research by ensuring the samples are representative of the audience being polled. Larger sample sizes have another practical value – they allow for greater segmentation of respondent results, which can be valuable in reading poll results. For example, in political polling, it is useful to have reliable results by congressional districts as well as statewide.
Maybe the greatest value of online research over telephone surveys is the ability to follow up with respondents. This can take the form of sharing findings, asking follow-up questions or seeking views on subsequent, related information.
Segmentation of panels allows segmented research. Follow-up questions can be directed at respondents based on their answers to questions. Online focus groups can be organized with respondents voicing a particular view. When we assisted Oregon officials in building a transportation funding proposal, we conducted online focus groups with respondents who expressed opposition to a gas tax increase, which produced useful information and an insightful dialogue among opponents that guided how the funding proposal was presented.
Telephone surveys have been a reliable research tool and still have utility. The ubiquity of cell phones, the surge of robocalls and the reluctance of people to interrupt dinner to answer survey questions are challenges that make telephone surveys a less effective option than before. The challenges are significant enough that panel research skeptics should put aside their doubts and talk to the firms that have spent time honing the use of online panel research.