On the preferred prime minister measure, Conservative Leader Stephen Harper is in the lead, followed by Liberal Leader Justin Trudeau, and with NDP Leader Thomas Mulcair in third, according to the latest nightly tracking by Nanos Research for CTV News and The Globe and Mail.

Survey respondents were asked: "Of the current federal political party leaders, could you please rank your top two current local preferences for prime minister?"

The latest numbers, which were released at 2 p.m. on Wednesday, Sept. 30, show:

  • 32.0 per cent said they preferred Harper
  • 28.0 per cent preferred Trudeau
  • 23.5 per cent preferred Mulcair
  • 4.4 per cent preferred Green Party Leader Elizabeth May

Compared to a month ago, the latest results show Trudeau has gained nearly six percentage points on this measure, Harper has gained nearly three percentage points, and Mulcair has lost about four percentage points.

When asked to pick their second choice for preferred prime minister:

  • 39 per cent of those who picked Harper first said they had no second choice
  • 50 per cent of those who picked Trudeau first said Mulcair would be their second choice
  • 59 per cent of those who picked Mulcair first said Trudeau would be their second choice

Nanos Party Power Index

In the latest Nanos tracking, the Liberals scored 54.6 out of a possible 100 points on the Nanos Party Power Index, the NDP scored 52.5, and the Conservatives scored 51.0.

The Party Power Index uses a series of questions on ballot preferences and leadership impressions to give parties a composite score out of 100.

The results are based on the views of 1,200 respondents, compiled into a diffusion brand index for each party.

A "0" means that the party has no brand strength, while a "100" means that it has maximum brand strength.

Survey methodology

A national dual-frame (land and cell) random telephone survey, is conducted nightly by Nanos Research throughout the campaign using live agents. Each evening a new group of 400 eligible voters are interviewed. The daily tracking figures are based on a three-day rolling sample composed of 1,200 interviews. To update the tracking a new day of interviewing is added and the oldest day dropped. The margin of error for a survey of 1,200 respondents is ±2.8 percentage points, 19 times out of 20. The margin of error will vary for subpopulations such as supporters for the respective parties.

Full survey at Nanos Research

Follow @niknanos on Twitter