Canadians assign more responsibility in the Senate controversy to suspended senator Mike Duffy and Prime Minister Stephen Harper than they do to Nigel Wright, a new survey has found.

The Nanos survey, released Friday, found that when assigning points out of 100 for the responsibility in the Senate controversy relating to the personal cheque, Canadians assign 40 points to Duffy. They assign 34 points to Harper and 18 points to Wright, Harper’s former chief of staff. The remaining points were not assigned and deemed “unsure.”

Wright resigned from the Prime Minister’s Office in May 2013, when it became public that he had written a $90,000 personal cheque to help Duffy repay ineligible expenses.

Nik Nanos, chairman of Nanos Research, said Canadians are primarily associating the controversy with Duffy and Harper.

"The prime focus of the controversy in the eyes of Canadians is on Duffy and Harper," he said in a statement. "Also, the performance of Harper on the economy is more likely to be important than the controversy at this time."

Duffy faces 31 charges of fraud, breach of trust and bribery in relation to his living and travel expenses. The suspended senator has pleaded not guilty to all of them.

His trial will resume on June 1, after a scheduled three-week break.

The survey also found the following:

  • While 55 per cent of Canadians believe that Harper's track record in promoting an environment to create jobs was most important to them in terms of their judgment of his performance, 31 per cent believed his management of the Senate expenses controversy was most important.
  • While 52 per cent of Canadians would prefer reforming the Senate as a possible way forward, 38 per cent would prefer abolishing the institution all together, and six per cent would prefer to leave it as it is.

A thousand Canadians were randomly selected to participate in the survey between April 24 and 27.

The survey was conducted through a mix of online and phone interviews. The sample included land lines and cell lines across Canada.

The margin of error for the survey is +/- 3.1 percentage points, 19 times out of 20.