This new poll says that people are opposed to cuts in Medicare and Social Security by vast margins, but also support raising the retirement age and reducing benefits to wealthier retired people. So apparently, some significant fraction of the population thinks something like this:

Pollster: Do you support cuts to Social Security.

Respondent:  No way! What are you crazy!

Pollster:  Well what about making most people work for a few years longer before they can collect Social Security and then paying some of them lower benefits.

Respondent:  Well, that's okay.

The thing is that the poll is making people's opinions sound more incoherent than they really are.  The respondent's answers make sense if you listen to how the pollster asked the questions (the link to the survey is in the story linked above.)  The question on cutting Social Security is phrased:

Let me you [sic] read you a number of programs that could be cut significantly as a way to reduce the current federal budget deficit.  For each one, please tell me if you think significantly cutting the funding for this program is totally acceptable, mostly acceptable, mostly unacceptable, or totally unacceptable as a way to reduce the federal deficit.

Social Security is then presented as a program that might be "cut significantly" to "reduce the current budget deficit."  A listener could reasonably assume that they were being asked to cut Social Security benefits right now for current retirees.  It is no surprise that only 22 percent of respondents answered unacceptable.  No major political figure is suggesting any such thing.  The question and answer options for means testing and raising the retirement age are phrased:

Let me you read you a number of other things that might be cut or eliminated as a way to reduce the current federal budget deficit.  For each one, please tell me if you think significantly cutting the funding for this program is totally acceptable, mostly acceptable, mostly unacceptable, or totally unacceptable as a way to reduce the federal deficit.

Reducing Medicare and Social Security benefits for wealthier retirees.

Gradually raising the Social Security retirement age to sixty-nine by the year 2075.

Those two options win 62 and 56 percent support respectively.  Unless you are trying to craft the Obama White House's reelection campaign, asking people about whether they want entitlement cuts is not useful  Most people are too unfamiliar with the policy options in dispute to give a meaningful answer in the absence of clear policy options. 

I remember watching C-SPAN a few days after the November 2010 election.  One of the conservative think tanks was putting on a post election discussion panel.  The panel included blogger Ed Morrissey, Byron York, and some leader of a Florida organization with "Tea Party" in its name.  The panel was asked about entitlement cuts.  The Florida person gave a confused answer about her elderly grandmother and how her elderly mother-in-law was willing to make some sacrifices, but only some sacrifices and how people who don't work should also have to make sacrifices too. She was a person who was active in politics but she seemed to have no clue about the various policy options for entitlement reform.  There is no reason to expect that the median American voter is any more familiar with those options.

This poll indicates that there is a potential majority constituency for entitlement reform, but conservatives will have to prepare the ground and choose their words very, very carefully.