Disclaimer. Don't rely on these old notes in lieu of reading the literature, but they can jog your memory. As a grad student long ago, my peers and I collaborated to write and exchange summaries of political science research. I posted them to a wiki-style website. "Wikisum" is now dead but archived here. I cannot vouch for these notes' accuracy, nor can I say who wrote them.
Dalton and Wattenberg. 2000. Parties without partisans. Oxford: Oxford University Press.
The authors argue that parties matter primarily to the extent that they organize voters. Thus, although other authors view parties as having three primary roles (party in government, party as organization, party in electorate; see Aldrich 1995), Dalton and Wattenberg look only at the party in the eletorate.
The authors see a decline in partisanship ("dealignment") in all the advanced democracies. They attribute this decline to increasing voter self-sufficiency (as a result of increased education, more candidate-centered elections, and so on).
The "decline of parties" hypothesis. Argues that researchers who focus only on what parties do miss the mark; since the goal of parties is to get votes (at least for these authors), then we should be greatly concerned with levels of partisanship/partisan identification among voters. Parties matter because they lower voters' information costs and provide an easier way of mobilizing activists, so individual partisanship matters. Dealignment has happened primarily as people have become politically self-sufficient over time (due to higher education, more availability of information about politics, and an increased valuation of individualism).
Dealignment has been called a myth. Seeking to prove the reality of dealignment, the authors examine several observable implications of dealignments besides survey results. (1) Electoral behavior shows evidence of dealignment, as seen by an increase in the effective number of parties, more split ticket voting, and more variability across elections. (2) Campaigns have become less partisan and more candidate-centered. This is driven largely by opened primaries and TV campaigning. This effect is strongest in presidential systems. (3) Volunteering and campaign activity are falling, even though interest in politics appears to be rising. (Conclusion) This could be good or bad. Good if it means that we come closer to the "model of rational and deliberate choice enshrined in democratic theory," but more likely bad, if narrow topics are able to overshadow "serious debate about the future of the nation" (60-61). The wedge issues of recent elections (gay marriage, abortion) might be something like this.
Research on similar subjects