Party Identification
There is one important distinguishing feature of Michigan election studies. They attempt to understand electoral behavior and elections by conducting interviews on a nationally representative sample of individuals eligible to vote as close to the polling day as possible. Basically, elections are assessed through voters. Whatever influences or does not influence voters is found out by making a comparison on the attitudes and behavior of one particular group of voters with another group at the same election. The assumption is that, if what accounts for the disparities between voters at an election can be understood, the outcome and the major features of an entire election can be understood.
Practically, there are several actors in an election apart from the voters. The attitudes and behavior of actors is a product of these influences. If these other considerations are put aide, it is still important to question the strength of the Michigan approach. Central to any research enterprise is the desire to have something to say about causation. For instance, one may adopt a socio-psychological approach and ask the value of party identification in vote switching. Again, the rationality of the voters may be of interest especially with regard to whether individuals vote for the party the feel closest to or on the issues that they feel are important. This paper is an attempt to address these questions especially with regard to party identification. In Michigan style survey research, there are important issues regarding whether the relationship between party identification and volatility indicate the influence of party identification on the vote or whether it suggests that vote influences party identification. In this regard, party identification will be central to this assessment. The concept will be defined and its value in determining individual vote aggregate election outcome will be examined.
Party Identification Definition
Humans have a general tendency to attempt to understand where they are placed in the world, look for connections with other people and value individuals who are like them. It is a natural phenomenon for people to attribute labels to themselves and to establish group attachments. What they believe in, their attitudes, how they behave and their opinions are influenced by other people. Parties and politics are on the other hand far much removed from daily experiences. People hold their families, neighborhood, good opinions and social relations to be of great value to them. In this regard, it is easier for the majority of individuals to modify political preferences so as to correspond to these groups than selected memberships founded on the basis of political preferences. In this regard, identification comes before the establishment of political beliefs.
Party identification in this regard can be defined as persistent emotional attachment to a particular party (Bartle and Bellucci, 20084). A more direct definition is provided by Campbell et al who define the concept as the sense of personal attachment which one feels towards the party of his choice (Campbell et al., 1960321). As stated, the identity is assumed to be a consequence of passive socialization that is seldom changed once established. It was widely believed that the identification is ingrained in the political experiences of the forefathers that established loyalty to a party that was transmitted from one generation to another. In this regard, the tendency of people to view themselves as members belonging to a particular group while creating an emotional attachment to their membership, adjusting their behavior to correspond with the norms of the group and establish a general sense of identification can be termed as party identification. This sense of belonging to a particular group is fundamentally focused on basic social groups.
The concept of party identification may also be used to refer to an individuals psychological attachment to the party. This attachment is to some extent stable over time and is held to be independent of the actual vote. The persistent stability of party identification is its essential property.
Importance of party identity in explaining individual vote decision
Party identification is assumed to directly influence the choice of vote because there is likelihood of voters to express their identity with a particular party of with other party supporters (Sanders et al., 2002194). When all other factors are equal in weight, identity serves as a tie breaker. Identity also has the potential of exerting indirect influence on the opinions and attitudes of people. The view that ones sense of being either Democrat or Republican develops at an early age before the establishment of political preferences indirectly supports this proposition. According to Campbell et al., party identification has the potential of raising the perceptual screen through which a person seems to conceive what is favorable to his partisan inclination (Campbell et al., 1960195). However, the disparities in opinions and attitudes cannot be accounted for by identities even though the amount of influence that party identification has on perception and interpretation of political entities is so enormous that a person seldom establish a given attitude force that does not correspond with this allegiance (Campbell et al., 1960141).
The value of party identification is clearly seen when it is made central to comprehending the influence to all conceivable causes of the vote. Party identification is ingrained in the gradually changing group membership and is therefore a prime mover of variables situated later in the sequence. However, one may argue that if prior identities biases learning then conceiving how preferences can be formed by identifiers which do not correspond with their identities may be difficult. According to Bartel, there is an important role played by partisan bias in the way an individual perceives politics and in the perpetuation and reinforcement of sharp differences in opinion in major political parties (Bartel, 2002118). Party identity also has a major influence on the core values. As such, it plays a major role in explaining individual vote decisions.
Individuals often use their votes in expressing their party affiliation. As such, party identification is an important aspect in voting decisions. There is often a strong link between voting behavior and party identification. This is because ones evaluation of leaders and parties is affected by party identification and this in turn influences the choice of vote. Correspondence between voter preference and the activities of government lies in the critical response of the voters to the activities of those that they elect to represent them. There is a general tendency of individuals to be influenced by groups as much as it may be argued that a group is nothing more than the individuals who comprise it. Partisanship in this regard is a consequence of an enduring identity.
Importance of party identification in explaining aggregate electoral outcome in Britain and United States
The duration in which a country has experienced a stable political system is closely related to level of party identification. One way of explaining this aspect is that it is a function of the kind of experiences that individuals have with the party system and the transmission processes between generations. The degree of party identification or partisanship in a given country is suggestive of its political system.
There is often a danger of political instability in system in which party identification has been weakly established since there is high likelihood of flash parties garnering the majority of votes. Likewise, when an enduring attachment to a political party has been developed by the majority of individuals, there is less likelihood that flash parties will emerge since party identification serves as a barrier against such kind of unprecedented changes in the party system.
United States is an example of a country that has a strong party identification. American voters are usually defined by their strong and stable affective relationships with political parties, relatively low levels of interest and involvement, and a bigger strength of long term over short term forces (Sacchetti, 20098). In this regard, United States provides a stable party identification context which is critical in shaping the preferences of voters. The concept of party identification is also applicable to Britain even though the circumstances vary.
For the concept of party identification to be meaningful, it must be equally applicable in different countries. In other words, it must have the same properties and meaning across different systems. Attempts to validate the concept outside the United States have been minimal. It is only in the then Federal Republic of Germany that a serious attempt to validate the concept took place. However, the main question that I wish to address in this section concerns the value of party identification in explaining aggregate election outcome. Various studies carried out on party identification often suggest an almost perfect match between party identification and vote preference (Kaase, 1976123). Within the setting of the United States, party identification corresponds with a model in which it precedes vote preference. Since it has a two party system, the two parties are compelled to be at the center of the political spectrum in order to increase their vote. The consequence is that the political differences that exist between them reduce. Political platforms, therefore, fail to offer the voters sufficient clues for making a decision on which party to vote for as time goes.
Under such circumstances, voters learn to identify themselves with a specific party so as to avoid the daunting task of collecting enough political information to deliberate. Party identification is particularly valuable in the United States as an analytical concept since it provides the opportunity to differentiate short term factors from long term influences (Thomassen, 1996416). It especially introduces the possibility of determining the role of each candidate and the main issues in an election.
A differentiation between short term forces is only possible when vote preference and party identification are held to be different concepts. A perfect match between vote preference and party identification may imply that in a given election, short term factors are not in operation and hence voting occurs according to party identification. In a system like that of the United States, the way vote preference and party identification relate varies with how strong party identification is. A large majority of individuals vote for the parties that they identify with. This is an indication that party identification may predict the outcomes of an election.
Since party identification is an enduring psychological attachment to a particular party, it follows that a change in party identification should not immediately precede changes in vote preference. However, a change of party identification is unlikely to take place unless there is a change in vote preference. This level of independence between vote preference and party identification is prominent in the United States. According to Butler and Stokes, Britain has more dynamic party identification that he United States (Butler Stokes, 1969234). There are always a small number of voters who normally change their party identification. However, the rate in Britain is double that of the United States. There is a higher percentage of individuals changing their vote preference and retaining their party identification than those who change their party identity and keep their vote preference (Thomassen, 1996427). The difference is more defined in Britain than in the United States.
The value of party identification in these countries is reflected by these finding which suggest that party identity is a reflection of the vote. It may suggest the outcome of an election since there is a high likelihood of a party that many people identify with garnering votes in an election. As much as party identification varies between United States and Britain, it is a major factor in explaining the vote. This does not imply that the role of short term influences is overlooked but rather, party identification is a powerful indicator of electoral behavior.
Measuring party identification
The importance of party identification in studying electoral behavior cannot be denied. However, there is no universally accepted measurement of party identification. Party identification being the persistence attachment that an individual feels towards a given political party, the attachment has been translated in various ways. The attachment is seen to be of an effective or emotional kind derived from past social experiences by the proponents of social-psychological approach while the rational choice theorists see it as a kind of cognitive shortcut representing a running tally of retrospective analysis of party performance (Fiorina, 1981123).
In both instances, the most important point is that an individual is disposed by partisan identity to vote for a particular party in various elections and to assess emerging political information is ways that are likely to be sympathetic to the interests of the party and policy stands. The phenomenon of close disparity between major parties during general elections can be explained by the relative stability of partisanship over time. The majority of voters seem to vote for the same party in the various elections. Most frequently, it is the independents that often change their position thereby resulting in one party winning the election rather than another.
It has often been suggested that the view that one votes for a given party because heshe identifies with it does not contribute much in explaining vote (Bartle Ballucci, 200895). The criticism would be fair is there is not much disparity between party identity and voting for that particular party. For those who advocate for party identification theory, there exists an important difference. Partisanship is lasting predisposition. Voting on the other hand is a particular act in which the prevailing circumstances may override partisanship resulting in the partisans voting for parties other than the ones they identify with. From this, it then follows that the measurement for party identification must be independent of the prevailing voting preference. If this is not so then any model that claims to explain the prevailing preference by making reference to partisanship would be a mere rhetoric.
However, this does not mean that partisanship is essentially a rhetorical phenomenon. The current voting preference for many voters can be comprehended as emanating from their persistent party identification. A major question that is left to be asked is whether a given way of party identification measurement differentiates partisanship from the prevailing voting preference. Skepticism concerning the validity of BES question sequence has risen among many analysts concerning its validity as a measuring instrument based on this context (Sanders et al., 2002345). The perspective that there is significant overestimation of the number of voters classified as identifiers has been expressed by these analysts.
Party identification of respondents have been assessed by BES based on whether one sees himselfherself as Conservative, Liberal-Democrat, Labor or any other party. Respondents who single out a particular party are further asked how strong they their party affiliation based on three categories very strong, fairly strong, and not very strong. Those who answer none or dont know to the first question are asked if they feel closer to any party and if they do, they are which party it is.
The responses got from successive BES survey with regard to these questions are indicative of three major conclusions. First, it suggests a progressive fall in the proportion of very strong identifiers. Secondly, the percentage of voters identifying themselves with one of the major parties has generally maintained an average of 85 percent of the voters. Finally, the overall stable proportion of identifiers is indicative of the fact that the drop in very strong identifiers has been to some extent matched by the increase in not very strong identifiers (Sanders et al., 2002347). However, it may be asked if the identifiers that have been measured can be correctly seen as the genuine partisans. There are numerous arguments that indicate that they cannot.
According to Brynin and Sanders (1997), there is close correspondence in the way that partisanship is measured by aggregate Gallup time-series. Their argument is that the correlation between the times series suggest that the standard BES partisanship questions do not measure voters but instead measures the current party preference. In order to test this hypothesis, they used the data from the 1991-1992 British Household Panel Study which indicate that party identification and voting choice measures correlate in an identical manner with numerous other theoretical variables. They attribute this to the fact that both the measures are a reflection of the current political preferences and not a distinct phenomenon.
Respondents may also not possess partisan identification even though the standard partisanship battery has the potential of making respondents to give in to partisan identification. The initial question fails to make provision for respondents to say that they have no party affiliation. If this were to be included, then the estimates are likely to increase for non-partisans.
BHPS data has also been used to show the level of instability in responses among individuals with regard to party identification questions. Johnston and Pattie discovered that when the same people are asked at varying times if they support a given party, a small percentage of respondents name the same party consistently than would have been expected by convention al measurements (Johnston et al., 1996295-305). They arrived at a conclusion that partisanship only persists for a small percentage of electorate. The BHPS result pattern is in sharp contrast with traditional BES pattern. The BES partisanship question measures hardly arrest the persistence of voters attachment to the main political parties. In fact, data from BHPS clearly demonstrate that partisan attachments expressed by the majority of British voters are not persistent.
What these BHPS findings indicate is that the only way that actual partisanship can be effectively measured is through panel data. The cost of carrying out such panels implies that electoral research based on surveys will continue to require approximating partisanship founded on single shot cross section surveys (Sanders et al., 2002348). Under such conditions, there is still need to devise single time-point approximations of the status of partisanship. The approach that this paper seeks to advance is derived from Blais et al., (2001) in their use of split-sample design and a filter to find out the resonance of the notion of usually supporting a given party with the respondents.
Within this approach, the control half of the sample is asked the initial traditional question about their affiliation. The test half on the other hand asks about party politics within the particular country of origin with emphasis being laid on what other people think with regard to their support for a particular political party rather than another. It further asks the respondents to set aside hisher current biases concerning the government and the opposition whether hehe think of himselfherself as supporter of one specific party or not. If answered in the affirmative, the respondent is further asked which party it is. If on the negative, heshe is further asked why this is so. Within this question, there are four options for the respondent heshe may answer on the negative because heshe does not think that heshe supports one party rather than another heshe does not take part in politics heshe does not know or heshe has simply refused to say.
The test question and the control question differ in two important ways. First, it makes provision for respondents to say that they normally do not support a party. Secondly, the respondents are not artificially cued to suggest their identification with one of the main parties because it excludes mentioning any particular party labels. It is highly improbable that a respondent genuinely identifies with a party unless heshe is specific about that particular party without prompting.
The questions have been applied to a random sample of British voters and the results indicate relative accuracy of the measurement. In this regard, party identification can be best measured by making clear provisions for the respondents that allow them to indicate their lack of long term partisan attachment. In this regard, it can be best measured using Split Sample Survey.
Conclusion
Party identification has been an important and contentious variable in the contribution to the explanation of voting and voting behavior. It is important because there are many electorates who consider themselves long-term party supporters and hence vote for the same party in different elections. At the same time, it is contentious in the sense that there exists a general skepticism concerning how it should be measured. In this paper, the conceptualization of party identification, its value in explaining individual vote and aggregate election outcome has been assessed.
Categories:
0 comments:
Post a Comment