To state the matter more generally, most people aren't really sure what their opinions are on most political matters, including such completely personal matters are their level of interest in politics. They're not sure because their are few occasions, outside of a standard interview situation, in which they are called upon to formulate and express political opinions. So, when confronted by rapid fire questions in a public opinion survey, they make up attitude reports as best they can as they go along. Because they are hurrying, they are heavily influenced by whatever ideas happen to be at the top of their mind.... This sort of dependence on immediately salient ideas is ubiquitous in public opinion polling....
Zaller proceeds to argue that 'salience effects' are crucial in influencing responses including question order effects and the priming effects of TV news. A key influence of course is the influence of the words in the question.
One perception that has dominated the Liberal leadership race has been the idea that the popularity of Justin Trudeau would lead to large Liberal gains in an election. One reason for this has been several polls, including a series of surveys by Forum Research, that have asked the question: ‘If a federal election were held today, which party would you be most likely to vote for if Justin Trudeau were the Liberal leader?’ It leads Forum to describe the result in the headline in the particular poll just cited: Trudeau Liberals to Run Table.
This question does not mention the names of the other leaders that Trudeau would be competing against. Might this influence how some respondents reply to the question? Another influence is the enduring popularity of Justin's father Pierre Trudeau who still tops polls such as this Angus Reid survey where respondents were asked to name the best PM since 1968.
TC's view on this, which I have blogged about before citing the example from 1993 of Kim Campbell, is that polls of this nature can be extremely misleading. A more recent Forum poll, released on March 8, does report an improvement in Liberal support in their standard party preference question where no leaders are mentioned. One can make a reasonable argument that Justin Trudeau is having a modest positive impact on Liberal support but the headlines generated by the Forum poll such as this one in the National Post: "Liberals tied with Tories… but would win commanding victory with Justin Trudeau as leader: poll" are unreasonably inflating the hopes of Liberals.Such polls capture momentary perceptions rooted in Justin Trudeau's reputation, which is influenced by memories of this father. What will matter is not how Mr. Trudeau performs against the other relatively weak Liberal candidates in debates (his most formidable potential opponent, Bob Rae, did not run), but how he carries out the job as leader once chosen as clearly he will be. When it comes to the next election due in October 2015 what will matter will be two and a half years of doing the job of party leader.To quote from a comment I made about the NDP leadership race, what matters in the end are a list of the "important aspects of leadership: the role of the party chief as motivator, cheerleader, reconciler of the inevitable party factions, lead party organizer, most important fund raiser, and chief executive officer."
The polls published during a leadership race should be viewed as of limited importance: the skills, experience and character of the person selected are what matters. Justin Trudeau has a legendary name but limited experience and a party weakened by decades of decline (see my series from 2010 on this). He faces a highly uncertain future.