Voter choice: a study of decision confidence and satisfaction

Understanding the dynamics of voter behaviour has been an area of interest to consumer researchers and marketers for sometime. This study focuses on specific issues related to voter behaviour via a study of voters in an Australia State election. Data were gathered from 159 voters using a survey administered using a drop off and pick-up. The results indicate that voter concern and involvement play a significant role in voter confidence and their desire to keep up to date during elections. Introduction to Voter Behaviour Theory Significant interest in voter choice can be found in research related to voter involvement and attitudes toward political advertising by researchers such as Burton and Netemeyer (1993), Faber, Tims, and Schmitt (1993) and Rothschild (1978), among others. However, little work has been undertaken to examine voter concern, involvement, confidence, and voters’ desire to keep up to date during elections and satisfaction. This study explores the dynamics of voter behaviour during a specific election by examining voter concern, voter involvement, decisionmaking confidence, desire to keep up to date during elections and satisfaction. Voter behaviour has been an area of interest to consumer researchers for many years (e.g., Faber, Tims and Schmitt, 1993; Hill, 1989; Newman and Sheth, 1985; Rothschild, 1978; Weaver Larscicy and Tinkham, 1999). Such research has often sought to understand the variance in voter behaviour that can be attributed to different voter characteristics, as well as voter decision criteria used to make choices (see Newman and Sheth, 1984) and the effect of different appeals in political advertising (e.g., Faber, Tims, and Schmitt, 1993; Hill, 1989; Weaver Larscicy and Tinkham, 1999). In this broad area of decision-making, consumer involvement has been recognized as an important influence on consumer decision-making (O’Cass 2000a). During the past decade or so, there have been many theoretical propositions regarding involvement and a wide variety of involvement types and effects (Mittal, 1989; Muehling, Laczniak and Andrews, 1993; Pinkleton, 1998; Rothschild and Houston, 1979). Political advertising researchers have recognized the value of understanding voter involvement in elections (e.g., Burton and Netemeyer, 1992; Faber, Tims and Schmitt, 1993; Rothschild and Houston 1979), because of its effect on search and advertising effectiveness. An important issue addressed periodically in the literature are the potential antecedents to voter involvement. In the broader literature, identified antecedents such as perceived risk, have been identified, however, an important issue raised in recent times has been that of consumer concern. For example, Bang, Ellinger, Hadjimarcou and Traichal (2000) highlighted the growing concern of consumers over environmental issues related to the earth’s resources. They examined issues related to consumer concern, consumer knowledge and the theory of reasoned action. However, in the context of voting a neglected area is the degree of consumer concern about political issues, including social and economic. Using a similar notion to Bang et al (2000) and that of Zaichkowsky (1985) it is argued that when issues in an election or broader economic and social concerns are raised in voters they will become more ANZMAC 2002 Conference Proceedings 1371 involved. As such we see voter concern directly influencing voter involvement and it is hypothesised that H1: Voter concern will significantly influence voter involvement. Related to involvement is search for information related to choice as a major element of consumer behaviour theory. There have been indicators that consumers, in general, seek little pre-purchase information (Beatty and Smith 1987). Information search, or seeking in the context of electoral choice, is defined as information seeking which voters engage in to facilitate decision-making in an election. In the context of commercial goods, it has been shown by Beatty and Smith (1987) and Bloch and Richins (1983), that involvement in a purchase is related strongly to external search by consumers. Overall, there appears to be some support for the view that consumers’ (or voters’) involvement will affect on the extent of information search. Although the empirical support is weak, the logical connection can be seen, in the work of Houston and Rothschild (1978), who fundamentally viewed intensity of search as simply one form of response involvement. Although not specifically defined in the literature, we may view extent of search as a desire to be informed or keep up to date. This is particularly relevant in the voter choice context if we consider the varied information sources available to voters to access information during campaigns and the significant volume of communications that occurs during an election. Considering the arguments of both Beatty and Smith (1987) and Bloch and Richins (1983), we expect that those more involved in politics will possess a greater desire to keep informed during an election, and as such, it is hypothesized that H2: Voter involvement will influence the desire to keep up to date during elections. Previous voter research has also shown different effects depending on level of involvement; for example, research by Burton and Netemeyer (1992) examined the effects of situational and enduring involvement on voters’ response involvement, knowledge, confidence, and preference stability. Their findings indicate significant effects for involvement on voter knowledge, confidence, and preference stability. The degree of confidence a voter has regarding his or her views or decision making abilities to do with an election is important because it can affect the strength of the relationship between attitudes and behaviour. Confidence represents a consumer’s belief that his or her attitude, knowledge, or ability is sufficient or correct regarding a product (or object/stimuli). The degree of confidence could reflect certainty or uncertainty, as to which judgment is correct or the best, or even ambiguity, as to the meaning of an attitude object (O’Cass, 2000b) in the context of voting. As such, it is hypothesized that H3: Voter involvement will significantly influence confidence in making an electoral choice. H4: Voter desire to keep up to date during elections will influence confidence in decision-making. In the context of politics, satisfaction generally arises from a voter’s comparison of the perceived political performance with expectations for this performance. Satisfaction is indicated if performance meets or exceeds expectations. There is a growing body of consumer satisfaction research related to commercial goods and services; however, little if any has examined voter satisfaction. One impetus for examining the effects of voter involvement on satisfaction is Day’s (1977) observation that satisfaction decisions are contingent on the incidence of post purchase evaluation. For stimuli of low importance, evaluations might not be triggered and Barber and Venkatraman (1986) argue that satisfaction ANZMAC 2002 Conference Proceedings 1372 becomes an even more important issue for a high involvement product. This implies that, when a person is involved in the object, satisfaction is triggered and avoidance of dissatisfaction is important. This is supported by Richins and Bloch (1991) who argue that, because a product is an important part of a consumer’s life, highly involved consumers have stronger motivation to avoid post purchase dissatisfaction. They have more at stake in the purchase situation and a greater need to make a wise choice, and it is believed that voters who are highly involved in politics will also seek to avoid dissatisfaction. As such, it is hypothesized that H5: Voter involvement will significantly influence voter satisfaction. H6: Voter confidence will significantly influence voter satisfaction. H7: Desire to keep informed during an election will significantly influence voter satisfaction. Research Design The study was based on the design and administration of a self-completed questionnaire that was part of a larger study of voter behaviour, administered to a convenience sample of voters in a state election. The data were gathered immediately after the election, and the respondents were all registered to vote, and did cast a vote on polling day. A drop-off and pick-up approach was used, yielding 159 surveys, with respondent’s ages ranging between 18 and 79 years, with an average age of 32 years. The sample was evenly spilt between men and women and 20% of respondents had a junior high school education, 30% senior high school, 42% an undergraduate university degree, and 8% a postgraduate qualification. Data were gathered on voter involvement in politics, adapted from O’Cass (2000a). The seven-item scale measured the degree of what might be termed “product involvement” in a conventional sense. The scale is oriented toward measuring the position (centrality) that politics occupies in a person’ life (O’Cass, 2000a). Therefore, it measures voter involvement using items such as politics is a significant part of my life and politics means a lot to me. Voter confidence was measured via a 3-item scale adapted from O’Cass (1999), that measures the voter’s confidence in making his or her choice in the election with items such as I have confidence in my ability to make a good decision on who to vote for. Scales to measure voters’ desire to keep informed were developed for this study on the basis of work by Beatty and Smith (1987) and Mittal (1989). The items capture the extent that the individual is motivated or desires to keep up to date during the election via gathering information. Voter concern was measured using four items adapted from Bang et al (2000). Political satisfaction (general level with parties, politics etc) was measured via a four-item scale measuring satisfaction with political parties and politics in general and candidate satisfaction and party satisfaction. Responses to items were measured using seven-point scales with endpoints “strongly agree” to “strongly disagree”. The process adopted used a panel of expert judges (active consumer behaviour researchers) and a series of focus groups for assessment of content and face validity and survey refinement (O’Cass, 2000a; Zaichkowsky, 1985). Because of the nature of elections, this process was considered the most appropriate to test and validate the scales and keep the surveys as short as possible to aid completions. Also, because the scales had been used and validated in prior research, pilot testing the instrument on a larger sample was considered not critical. Because all scales were sourced from existing literature and validated in the pre-test, they were considered to possess content and face validity (O’Cass, 2001). ANZMAC 2002 Conference Proceedings 1373