As silver linings go, it’s pretty thin. With millions of Canadians unemployed, working from home, staying inside or just looking for something to do, pollsters are finding it a lot easier these days to convince people to take their surveys.
Response rates have been dropping for years as people have moved from landlines to mobile phones, and as they’ve become more reluctant to answer calls from numbers they don’t recognize.
The Pew Research Center reports the response rate to its telephone polls in 2018 was just six per cent. That means 94 per cent of households contacted either didn’t answer the phone or refused to participate in the survey.
In 1997, the reported response rate was 36 per cent. The decline in response rates has been steady ever since, although it was still over 20 per cent until the mid-2000s.
But something has changed over the past few weeks. Those rates are going up.
Call me, call me any, anytime
Margaret Brigley, CEO of Narrative Research, said that “in general, when contacting research respondents over the past approximately 45 days, we have seen significantly better cooperation and response rates in our projects.”
Brigley said that the number of people who refuse to participate in their telephone surveys has decreased by somewhere between 10 and 30 per cent. She’s also seeing better participation in Narrative’s online surveys as well.
Christian Bourque, executive vice-president at Léger, said he has seen an increase of roughly 10 per cent in the response rate for their online surveys, and an increase of more than 10 per cent to the response rate for phone polls.
And more of the people who do agree to participate in online surveys are sticking it out until the end. David Coletto, CEO of Abacus Data, said that he has “seen that completion rates are higher, and we can field longer surveys than usual.”
Normally, Coletto said, the number of people who start an online survey without finishing it can range between 10 and 25 per cent of the total, depending on the length of the questionnaire and the topic it covers. (More interesting surveys have better completion rates.)
This doesn’t necessarily change much when it comes to the costs of running a poll; Brigley said that the cost of a poll depends on things like the number of completed surveys a client wants. But it does mean, he said, that “we’ve been able to complete studies within a slightly shorter timeframe, in most instances.”
For a pollster, it’s better to be in the field for the shortest possible amount of time — especially in a fast-changing opinion environment like the pandemic. Polls are supposed to serve as snapshots of public opinion; the longer they take, the likelier it is that their observations will be out of date.
On the face of it, the reason behind this spike in both response and completion rates seems obvious. With many businesses still closed across the country and a significant chunk of the workforce stuck at home throughout the day, more people have free time on their hands. They might also be looking to have a conversation with someone outside their social distancing cohort.
That doesn’t hold across the board, however. Greg Lyle, president of the Innovative Research Group, said that while response rates have gone up for the general public, “it appears it is more difficult to reach people in a business that is closed but easier to reach people in open businesses.”
Someone to talk to
This trend in response rates is not limited to Canada. Pollsters in the United States have found the same thing. But boredom isn’t the only thing driving this phenomenon; fear is playing a role as well.
Last month, the New York Times reported that while more people are willing to participate in telephone polling these days, many are also grateful just to have someone to talk to about the anxiety they’re feeling. One American pollster said that response rates also climbed after the terrorist attacks of September 11, 2001, for the same reason.
Interviewers normally have to follow their assigned scripts carefully in order to avoid biasing the interviewee, ensuring that every interview is done the same way. That can be a challenge when the person on the other end of the line is baring their soul to a stranger. (That isn’t a factor in online polling, of course.)
“Canadians have more time and are easier to reach, so that likely has increased response and completion rates,” said Coletto.
“But if the survey is about COVID-19, as almost all of ours have been, I think there’s a genuine desire to share experiences and views on something that has impacted and captivated almost all of us.”