Dynamics of Voter Satisfaction: Empirical Investigation of Young Indian Voters

The objective of the present study was to identify factors that sensitise voters towards the political system and stimulate them for active participation in the election process. The study also aimed to examine the role of these factors in increasing the voter satisfaction. Data was generated from a sample of 194 young voters. One contribution of the study was to validate a scale measuring political sensitivity of voters. Findings also revealed that the higher the economic and global consideration (EGC) among voters the lesser will be their satisfaction with the existing political system. The predictors of EGC are observed to be social considerations (SC) and perceived risk (PR). Findings of the study also suggest that higher perceived knowledge of voters will lead to increased involvement in the political system. The study is pioneering as it initiates efforts for comprehensive analyses of the Indian voters for better understanding of factors that affect their political sensitivity.


Introduction
India, being the largest democracy in the world and having around 850 million electorates, offer unique political scenario due to two factors. First, contrary to the international trends, Indian population is a young population. The proportion of 'youth' in total Indian population is 35% (Lu, Yiu, & Soman, 2016;Central Statistics Office, 2017). More specifically, around 50% of the Indian population is under the age of 25 years (Heslop, 2014). This trend is expected to remain same in future also. According to the estimates of the National Commission on Population (2006), the average age in India would be 31.39 years in 2026. In the 2019-General Elections in India, around 84 million first-time voters were eligible to exercise their right to vote (Deka, 2019;Jain, 2019).
Second, the recently held elections indicate that the young Indian population participates enthusiastically in elections (Deka, 2019). According to the estimates of a report prepared by Sahu (2014) and released by Research and Information Division of Lok Sabha Secretariat, around 66% of Indian voters exercised their right to vote in the 2014-General Election. An interesting observation in this election is the active participation of young and 'first-time' voters (Basu & Mishra, 2014). Thus, strategizing for elections in India is not possible without proper understanding of young voters (Vaishnay, 2015;Joshi & Kunduri, 2017).
With regard to election strategies, a new trend has been observed in the global context where politicians are using pure marketing concepts in reaching out to their voters (O'Shaughnessy 2001). Voter satisfaction is one such concept adapted from the concept of consumer satisfaction. Fundamentally, consumer satisfaction reflects the feeling that consumers show after consumption of products (Giese & Cote, 2000;O'Cass & Pecotich, 2005). It is considered as an outcome of various post purchase events (Gieese & Cote, 2000). In the context of voting, previous researchers have suggested that higher voter satisfaction results in more participation of voters during elections (Hadjar & Beck, 2010). Hence, it is imperative for politicians to have indepth understanding of the factors that lead to higher voter satisfaction and hence, contribute in active voter participation in the election process (Banerjee & Ray Chaudhuri, 2016;Ehrenberg 1997;Henn, Weinstein, & Forest, 2005;O'Cass & Pecotich, 2005;Winchester, Hall, & Binney, 2015).
However, there is limited literature available to understand factors crucial for voter satisfaction and their concerns with key political issues such as global reputation of the country, its economic condition, prevailing condition of underprivileged sections of the society, etc. In the Indian context, the researchers have paid little attention to understand young Indian voters (Joshi & Kunduri, 2017). The earlier studies have ordinarily focused on general issues such as party performance, election campaigns, media involvement and election results (Beg, 2017;Bhalotra, Clots-Figueras, Cassan, & Iyer, 2014;Harris, 2015;Heath, Verniers, Kumar, 2015;Kapur & Nangia, 2015;Vaishnav & Swanson, 2015). In majority of the earlier studies, the results are based on past election data (Bhalotra et al., 2014;Harris, 2015;Heath et al., 2015;Kapur & Nangia, 2015;Vaishnav & Swanson, 2015;Beg, 2017). Hence, there is a need for empirical studies based on direct responses from voters. Moreover, there is a need to identify the factors that results in voter satisfaction with the existing political system in the country. In light of this, objective of the present study was to identify the factors that sensitise voters towards the political system and stimulate them for active participation in the election process. The study also aimed to examine the role of these factors in increasing the voter satisfaction. This study is pioneering in the sense that it initiates efforts for comprehensive analyses of Indian voters.

Literature Review and Conceptualization
Researchers across the globe have examined the voter concerns and preferences that can lead to voter satisfaction with the existing political system. In fact, the domain of voter concerns and preferences grabbed the interests of researchers for the first time in the 1952 USA elections (Rothschild, 1978;Winchester et al., 2015). Since then, the researchers have examined this domain of study employing different approaches (Cwalina, Falkowski, Newman, & Vercic, 2004;Newman & Sheth, 1985;O'Cass, 2003O'Cass & Pecotich, 2005. Researchers have opined that the concerns and preferences crucial for voter satisfaction depend upon the level of enduring and situational involvement of voters in the election process (Ben-Ur, 2007;Gad, 2014;Banerjee & Ray Chaudhuri, 2016).
Enduring involvement refers to the general interest of voters in the politics of their country whereas situational involvement refers to the interest generated among voters by the specific election campaigns (Ben-Ur, 2007). The higher enduring and situational involvement leads to better response from voters (Antil, 1984;Burton & Netemeyer, 1992;Ben-Ur, 2007;Faber, Tims, & Schmitt, 1993;Rothschild, 1978;Winchester et al., 2015). In fact, the high level of enduring and situational involvement of voters also enable them to use a combination of constructive and realistic approach for decision making during elections (Ehrenberg, 1991;Erasmus, Boshoff, & Rousseau, 2001;Peng & Hackley, 2009;Zaichkowsky, 1985). In constructive approach, voters make rational decisions by using specific information; while in realistic approach political decisions of voters depend on the prevailing issues in the country (Falkowski & Cwalina, 2012;Ki & Hon, 2012).
Broadly, concerns and preferences crucial for higher voter satisfaction can be explained by using the three models (Antunes, 2010): the sociological model (also identified as School of Columbia), the model of economic theory (also referred as School of Rochester) and the psychological model (referred as School of Michigan). The sociological model, explained by Lazarsfeld, Berelson, and Gaudet (1944), highlighted the influence of social factors (such as socio-economic status, religion and area of residence) on the voting decisions. On the other hand, the model of economic theory was based on the premises of rationality, self-interest, consistency and uncertainty (Downs, 1957). According to economic theory, decision making by voters about different options available is guided by rationality and self-interest with a certain level of consistency and uncertainty involved (Downs, 1957).
The psychological model, explained by Campbell, Converse, Miller, and Stokes (1960), suggested that the party partisanship is one of the important factors for affecting the voter behaviour. According to Campbell et al. (1960), voters prefer the ideology of one political party over that of other party. The party partisanship of voters was explained by Campbell et al. (1960) by linking long-term factors (such as race, ethnicity, gender, social class, occupation, values and partisanship) with shortterm factors (candidate evaluation and issue perceptions). According to Campbell et al. (1960), long-term factors create party identification among the voters that guide their short-term choices of candidate evaluation and issue perception. In the context of bi-party system, researchers have suggested that the psychological model explained by Campbell et al. (1960) is still very crucial in explaining political sensitivity of voters (Antunes, 2010;Dalton, 1984, 2000, Fiorina 1981Franklin, Mackie, & Valen, 1992;Kiewiet, 1983;Wattenberg, 1994).
In the context of countries where multiple parties are allowed to contest the elections, researchers have attempted to identify factors that are linked to higher voter involvement and their satisfaction with the existing political process (Beg, 2017;Cwalina et al, 2004;Cwalina, Falkowski, & Newman, 2010;Harris, 2015;Newman, 1999Newman, , 2002Newman & Sheth, 1985, 1987. These factors and the relationships among these factors have been discussed below

Voter Involvement (VI) and Voter Satisfaction (VS)
High involvement of voters assists them to make wise political decisions that result into the election of an efficient government (Richins & Bloch, 1991). Also, higher voter involvements enforce governments to improve their functioning (Geys, Heinemann, & Kalb, 2010). Thus, it is expected that high voter involvement will result into increase in voter satisfaction. In this regard, O'Cass (2002) comprehensively examined the role of higher voter involvement on increasing the voter satisfaction and suggested a positive linkage between the two variables. O'Cass and Natarajan (2003) emphasized that higher voter involvement infuse confidence about the existing political process that leads to higher satisfaction among voters. The above discussion helped in formulating the following hypothesis: H1: Voter Involvement (VI) positively affects Voter Satisfaction (VS)

Perceived Knowledge (PK) and Voter Involvement (VI)
One factor that is considered as crucial for influencing voter involvement in the election process is the perceived knowledge of voters. The construct perceived knowledge assist in capturing the awareness of voters about the issues related to national and international politics. An informed voter is expected to be more committed to democratic principles and hence, more sensitive to the political discourse in the country (Delli Carpini & Keeter, 1996). They are likely to be engaged in various forms of political participations (Banwart, 2007;O'Cass & Pecotich, 2005). Stockemer and Rocher (2017) had even suggested that increase in political knowledge can significantly increase turnout of young voters in the elections. Thus, the following hypothesis was formulated: H2: Perceived knowledge (PK) positively affects voter involvement (VI) in the election process

Religious and Caste Considerations (RCC) and Voter Involvement (VI)
Previous researchers have opined that the voters are more attractive to candidates who are similar to them (Blalock, 1967;Chandra, 2009;Cook, 1994;Dolan, 1998;Heath et al., 2015;Huddy & Terkildsen, 1993;Tate, 1993). More specifically, voters are more inclined to elect candidates belonging to caste and religion similar to their caste and religion. This is also supported in the observation by Cwalina et al. (2010) that emotions (such as those related to religion and caste) can be crucial in stimulating voters towards the political process. In this regard, the construct Religious and Caste Considerations (RCC) measures the voter concerns regarding the issues related to their religion and caste. This construct is particularly important in the Indian context where the religion and caste affiliations always remain at the focal point in the voter decision making (Beg, 2017;Bhalotra et al., 2014;Harris, 2015;Heath et al., 2015;Huber & Suryanarayan, 2016;Pande, 2003). Heath et al. (2015) suggested that religious affiliations affect election-related decisions of Indian voters. The other researchers such as Bhalotra et al. (2014), Harris (2015) and Beg (2017) also used religious identity as a factor in their study on political marketing in India and asserted that religious concerns can increase voter involvement. Similarly, researchers such as Banerjee Green, McManus, and Pande (2014) and Huber and Suryanarayan (2016) opined that the caste concerns of Indian voters also influence their involvement in the election process. Recently, Vaishnav (2015) also emphasized the importance of religious and caste identity of politicians in Indian elections. Based on the above discussion, the following hypothesis was formulated: H3: Religious and caste considerations (RCC) positively affect voter involvement (VI) in the election process

Economic and Global Considerations (EGC) and Voter Satisfaction (VS)
The construct EGC is intended to measure concern of voters related to prevailing economic policies and impact of these policies on global reputation of the country (Acs & Szerb, 2009;Bengoa & Sanchez-Robles, 2003;Hazelkorn, 2014). The economic considerations of voters are considered to be crucial for influencing election results in the global context (Anderson, 2007;Lewis-Beck & Stegmaier, 2000). In the Indian context, Vaishnav and Swanson (2015) have suggested that voters make election decisions after careful evaluation of economic policies of incumbent governments. This suggests that the perception of voters about economic policies of successive governments could be a predictor of their satisfaction with the existing political system in the country. Moreover, researchers have also suggested that the sustained economic growth of a nation improves its global rankings (Acs & Szerb, 2009;Bengoa & Sanchez-Robles, 2003;Hazelkorn, 2014). In fact, the economic policies and global reputation are considered to be interrelated and hence, it is expected that the construct EGC is likely to be a key predictor of voter satisfaction (Ezrow & Xezonakis, 2016). Thus, the following hypothesis was formulated:

Perceived Risk (PR) and Economic and Global Considerations (EGC)
The concept of risk perceived by voters covers uncertainty, anxiety and risk to future generations (Slovic, 2015). Voters perceive election process as a high-risk process because they believe that the selection of an incompetent government by them may disturb the economic stability of a country and may also damage its global reputation (Bauer, 1960;Dowling & Satelin, 1994;O'Cass, 2002O'Cass, , 2003O'Cass & Pecotich, 2005). As a result, the construct perceived risk is expected to mobilize voters for higher participation in political process by making them believe that their election decisions has a bearing on the economy and global rankings of their country (Bauer, 1960;Dowling & Satelin, 1994). Thus, the following hypothesis was formulated: H5: Perceived risk (PR) positively affects economic and global considerations (EGC) of voters.

Social Considerations (SC) and Economic and Global Considerations (EGC)
The construct of social considerations measures the concerns of voters regarding the underprivileged classes in their country. Farmers and women are the two underprivileged sections of the Indian society which capture a major part of the political debate in the country. Issues such as low prices of agricultural products, crop failure, high debt and poor income have led to around 300,000 farmer suicides in the last two decades (Basu, 2016;Kalokhe et al., 2017;Desai & Mehta, 2017;Kaushal, 2018;Menon & Allen, 2018;Sunder, 2018). With reference to women, issues such as domestic violence, women security, workplace harassment, etc. have been observed in the recent past (Kalokhe et al., 2017;Menon & Allen, 2018). Previous researchers have indicated that voters perceive that the prevailing conditions of underprivileged sections of the society is a reflection of economic status of the country and affects reputation of the country at the global stage (Kapur & Nangia, 2015;Vaishnav & Swanson, 2015). Hence, the following hypothesis was formulated: H6: Social considerations (SC) positively affect economic and global concern (EGC) of the voters.
Based on the above hypotheses, the conceptual model presented in Figure 1 was proposed to be validated in the present study:

Item Integration and Validation of the Scale
The scale items were integrated from two sources. First, some items (Table 1) were adapted from the study by O'Cass and Pecotich (2005). These items (S13-S22; S29-S31) were added with a purpose to refine the scales proposed by O'Cass and Pecotich (2005) based on the Indian context. Second, items from S1 to S12 and from S23 to S28 were added by the researchers in the present study based on prevalent issues in the country such as economic conditions of India, global reputation of India, conditions of farmers, women, backward classes, minorities and issues specific to different caste and religion (Beg, 2017;Desai & Mehta, 2017;Press Trust of India, 2017;Vaishnav, 2015;Vasudeva, 2017;Kaushal, 2018;Menon & Allen, 2018;Singh, 2018;Sunder, 2018). These questionnaire items were based on a 5-point Likert scale (1 = strongly disagree; 5 = strongly agree).
Before going for a final data collection, the questionnaire was distributed to a selected group of subject experts to check for proper wording, ease of understanding and to remove other inconsistencies. The experts were selected from a reputed university located in the northern region of the country and funded by the federal government of India. Based on the feedback from the experts, some of the items were rephrased to make them more relevant to the Indian context.

Sample
Internationally, the United Nations has defined 'youth' as an individual within the age bracket of 15-24 years (Ministry of Youth Affairs and Sports, 2014). Based on the Indian context, the 'youth or young population' was defined by Ministry of Youth Affairs and Sports (2003) as the population that falls in the age bracket of 13-35 years. This definition was revised by Ministry of Youth Affairs and Sports (2014) that considered population in the age bracket of 15-29 years as the young population.
For the purpose of sample selection, purposive sampling technique was employed. Several factors favour the use of purposive sampling technique in the present study. First, due to absence of a reliable and perfect sample frame, the researchers had to rely on their judgement to generate a sample that can represent the population (Khan & Kirmani, 2018). Second, the purposive sampling technique ensures advantages such as generation of data from a large sample within a short duration of time and at a low cost (Gravetter & Forzano, 2003;Malhotra & Dash, 2011).
Recent studies have highlighted that the global youth is the frequent users of the internet and likes to remain online always (Abeler, Bekker, & Falk, 2014;Ahmad & Khan, 2017;Guth, Schmidt, & Sutter, 2007;Jones 2002;Yoo & Donthu 2001;). In India, 74% of internet users are under the age of 35 years (Statista.com, 2016). The politicians have also identified online platforms as an effective medium for election campaigns, especially for the global and Indian youth (Chadha & Guha, 2016). In light of above stated facts, it was believed that the online platforms can be employed to generate quality feedback from young population in India. Hence, it was decided to generate data for the present study from young internet users. In order to comply with the definition of 'youth' by Ministry of Youth Affairs and Sports (2014), care was taken to generate data only from the internet users who fall in the age bracket of 15-29 years.
The data was generated from a sample comprising 56% male and 46% female respondents. Around 59% of the respondents were below 25 years of age and other respondents (41%) had already attained the age of 25 years. The respondents were a heterogeneous mix of different religions and caste. Majority of respondents (69%) were affiliated to Hinduism followed by respondents affiliated to Islam (19%), Christianity (4%) and other religions (8%). For caste affiliations, around 65% of respondents belonged to general category, 23% to Other Backward Classes (OBC) and 12% to Scheduled Castes (SC) and Scheduled Tribes (ST). Sample comprised respondents belonging to northern (33%), central (29%), western (19%), southern (15%) and eastern (4%) parts of the country.

Data Collection
The methodology for data collection was adapted from the study by Bashir and Madhavaiah (2015). The questionnaire was created in Google-Docs and posted on various online pages, forum, pages and circles of students across the country. The survey ran for one month and stopped when the response rate dropped significantly. The survey resulted in generation of data from 263 respondents. Of these, 69 questionnaires were incomplete or unfit and hence, they were eliminated. Thus, only 194 questionnaires were declared fit for further analysis. Therefore, the final sample size for the present study was 194. This sample size is adequate for the studies employing statistical tools such as exploratory factor analysis and structural equation modeling (Comrey & Lee, 1992;Gorsuch, 1983;MacCallum, Widaman, Zhang, & Hong, 1999). This sample size also complies with the subject to variable ratio of 5:1, suggested and used by a significant number of previous researchers (Ahmad & Khan, 2017;Hatcher, 1994).

Exploratory Factor Analysis
Exploratory factor analysis (EFA) was performed using SPSS 20.0 (see Table 1). Techniques such as the principal component analysis were employed for factor extraction and varimax rotation was used for the purpose of factor rotation. The items with factor loadings less than 0.4 were eliminated (Kline 1994;Hinkin, 1995 The Cronbach's alpha values for all seven variables were found to be in the acceptable range of 0.771 to 0.916. The first variable (EGC), represented by three items, had an alpha value of 0.771. The second and third variables (RCC and SC), represented by six-item and three-item scales, had alpha values of 0.916 and 0.836, respectively. The fourth variable (VI), comprising three items, had an alpha value of 0.877. The fifth (PR), sixth (VS) and seventh (PK) variables, each comprising three, four and seven item scales, had alpha values of 0.790, 0.831 and 0.902 respectively. Thus, the scale was found to be reliable with Cronbach's alpha values of more than 0.6 for all seven variables (Hair, Anderson, Tatham, & Black 1998;Kerlinger & Lee 2000;Khan & Adil 2013).

Measurement Model
Previous researchers have advocated that confirmatory factor analysis (CFA) should be followed by EFA as EFA fails to describe the cross loadings of items and correlation among the variables (Ahmad & Khan, 2017;Ahire, Golhar, & Waller, 1996;Khan & Adil, 2013;Lee, 2008). Hence, CFA was performed on the seven factors using AMOS 20.0 (see Figure 2). The majority of the items was loaded on the intended variable with loadings greater than 0.5 (Ryu, Han, & Jang, 2010). The loadings for items S6, S9, S22, S26 and S27 were poor and hence, these items were eliminated for further analysis. The result of model fit indices is presented in Table  2. The model fit indices were within the acceptable range (Gerbing & Anderson, 1988;Hu & Bentler, 1999;Hair, Black, Babin & Anderson, 2010).  Malhotra and Dash (2011) For the purpose of reliability and validity, techniques such as average variance extracted (AVE) and composite reliability (CR) were employed (Fornell & Larcker 1981;Hair et al., 2010). AVE values for all seven variables were above 0.5, indicating adequate convergent validity (Fornell & Larcker 1981;O'Leary-Kelly & Vokurka 1998;Hair et al., 2010;Khan & Adil 2013). The square root of AVE (diagonal values in Table 3) is greater that the 'inter-construct' correlation, confirming acceptable discriminant validity (Fornell & Larcker, 1981;O'Leary-Kelly & Vokurka, 1998;Hair et al., 2010;Khan & Adil, 2013). The values for CR were also within acceptable range (>0.7) indicating adequate composite reliability (Fornell & Larcker, 1981;Hair et al., 2010;Malhotra &Dash, 2011).  The study has validated the model presented in Figure 3. The model fit indices for structural modeling were also within acceptable range (CMIN/df=2.030; GFI=0.881; AGFI=0.849; CFI=0.930; RMSEA=0.060). The study findings revealed that the construct voter involvement is not significantly linked to voter satisfaction (p>0.05) suggesting that the hypothesis H1 was not supported. However, the construct EGC was observed to be significantly (p<0.05) but negatively (β=-0.222) linked to the construct VS. Thus, the hypothesis H4 was partially supported. The study findings also suggested that perceived knowledge (PK) is significantly (p<0.05) and positively (β=0.522) linked with voter involvement. But, the relationship between the religious and caste considerations (RCC) was not observed to be significant (p>0.05) with voter involvement. Thus, the hypothesis H2 was supported but the hypothesis H3 was not supported. Both predictors (perceived risk and social considerations) of Economic and Global Consideration (EGC) were observed to be significantly (p<0.05) and positively ((β=0.305 and 0.212 respectively) related with EGC. Hence, the hypotheses H5 and H6 were supported.

Discussion and Conclusion
The present study is based on the premise that designing strong campaigns is crucial for winning elections in India and hence, political parties and leaders need to be aware of the complexities involved in decision making process of voters in India. In this regard, the study has two important contributions. First, the study has validated a scale for measuring the factors crucial for active participation of voters in the political process. In fact, the 23-item scale will improve the understanding of academicians and practitioners about the issues vital for determining voter participation in the political process. Second, the study also proposed and validated a model examining the linkages among the variables covering the 23-item scale. This model will greatly help academicians and researchers for deeper understanding of the dynamics involved in satisfaction of voters and their eagerness to participate in the political process. The improved understanding of voters will also assist politicians in developing appropriate strategies. In fact, politicians may prioritize voter concerns by focusing on findings of the present study.
More specifically, the variable Economic and Global Considerations (EGC) was observed to be crucial for predicting voter satisfaction. Study findings revealed that more the voters are concerned about the economic status and global reputation of their country, difficult it would be for politicians to satisfy them with their policies. The incumbent government is required to emphasize on finding solutions for the economic problems and building a strong reputation for the country across the globe. This will create a positive perception among voters about the incumbent government and will contribute in increasing the voter satisfaction. This assertion find support in the recently held General Elections (2019) in India where it was noticed that the positive perceptions about the economic and foreign policies led to return of the incumbent government with a clear majority (Devlin, 2019).
The two variables that are observed to be crucial for predicting EGC are social considerations (SC) and Perceived Risk (PR). The present study has affirmed that the Indian politicians ought to design appropriate policies for the underprivileged sections of the society including farmers, women and other backward classes to create a positive perception among the voters with regard to economic status and global reputation of the country (Kalokhe et al., 2015;Kapur & Nangia, 2015;Vaishnav & Swanson, 2015;Basu, 2016;Menon & Allen, 2018). Similarly, the present study also supports the previous findings that higher the risk perceived by voters in their election decisions, more concerned they will be for the economic and global reputation of their country (Bauer, 1960;Dowling & Satelin, 1994;O'Cass, 2002;O'Cass & Pecotich, 2005). In order to reduce concerns of Indian voters regarding economic status and global reputation of the country, politicians are required to propose strong economic and foreign policies in their election manifesto.
It was also observed in the present study that the perceived knowledge of voter results in their greater involvement in the election process. This finding supports the similar observation of the previous researchers in different cultural contexts (Delli Carpini & Keeter, 1996;O'Cass & Pecotich, 2005;Banwart, 2007;Stockemer & Rocher, 2017). This is crucial in the Indian context where it has been observed that voter participation remain low in elections. This suggests that the key to increase involvement of voters in the political system is to increase their perceived knowledge i.e. making them believe that they possess sufficient knowledge about the political issues. However, findings of this study revealed that the higher voter involvement does not necessarily ensure the satisfaction of voters with the existing political system. This finding contradicts the observations of the studies in other countries and suggest that the Indian politicians ought to embrace different approach for increasing voter involvement and for increasing voter satisfaction (Richins & Bloch, 1991;O'Cass, 2003;Geys et al., 2010).
In a nutshell, the present study has broken the myth that the Indian voters are guided by their concerns related to their religious and caste affiliations (Pande, 2003;Bhalotra et al., 2014;Harris, 2015;Heath et al., 2015;Huber & Suryanarayan, 2016;Beg, 2017;Vaishnav, 2015). In fact, findings of the study suggest that the Indian voters adopt a realistic approach in their voting preferences and hence, Indian politicians must also employ a practical approach by focusing on issues such as economy, global reputation, progress of underprivileged sections of society and increasing voter trust.

Limitations and Future Research Directions
The scale and model validated in the present study can help politicians to design appropriate strategies for the election campaigns. However, there is a scope of improvement in some areas. First, the focus of the present study was on the urban voters. However, in India, politicians are required to modify their election strategies depending upon the specific needs of the voters living in villages. In fact, the fate of leaders and political parties in India is decided by the rural population, which is around 70% of the total population in India (Ministry of Home Affairs, 2011; Lu, Yiu, & Soman, 2016). Thus, the future researchers must undertake fresh studies for the rural voters. In fact, a comparative study on urban and rural voters will also provide useful insights.
Second, the study was based on responses from young voters only. The results may vary for other demographic categories such as voters belonging to different age groups, women voters, etc. Especially, the segment of women voters has attracted the attention of Indian politicians recently. Hence, the model presented in this study can also be revalidated for the women voters.
Third, the items to measure EGC, RCC and SC were added keeping current issues in mind. However, the issues may change with time and hence, future researchers may have to adapt items depending upon the issues prevailing at the time of their study. Finally, the present study is an example of cross sectional design that fails to capture the changes in voter preferences over a period of time. Thus, the future researchers may prefer a longitudinal study design to capture the changes over a period of time.