Voters Consumption of General Elections in Transitioning Economies

1

Voters’ Consumption of General Elections in Transitioning Economies: Scale Development and Validation Introduction

According to Campbell et al (1960), “voters are categorized according to their levels of sophistication,” which includes ideological and non-ideological belief systems (Luskin, 1990; Converse, 1964), and depending on these levels, voting behaviors and patterns can be determined. In their study of political sophistication, Campbell et al (1960) observe that a group of respondents, Voters’referred to as category (A), were those whose understanding and evaluation of political actors had abstract connections that one would associate with ideologically. Respondents in category (B) were those who used “identity cues” and frames in their evaluation and understanding of political actors, such as “this candidate is friendly or hostile to people like us (our group)” This group thinks about whether a party/candidate is “for” their group (e.g. farmers, the working class, the poor).

For those in category (C), prevailing conditions were the most important stimuli for their voting behavior. This category largely makes references to public policies, though usually implicitly, by forming opinions based on whether times are good or bad. For example, a respondent in category (C) might feel that if conditions are “good” and their family is doing well, then the incumbent is doing well. Respondents in category (D) were those who failed to comment on anything related to policy differences and debate between the parties but rather worry about the candidates’ personal characteristics, such as “their popularity, their sincerity, their religious practices, or family background” (1960, p. 244). These categorizations of political sophistication identified by Campbell et al (1960) highlight attitudes that influence voter behavior and impact voter decision-making processes much in the same way that consumer purchasing decision-making processes influence commercial marketing.

From Campbell et al’s (1960) framework and their explications, one could infer that respondents in categories A and B are highly likely to be politically sophisticated, knowing what they seek in political products; thus, their voting decisions are more fully informed than those of respondents in categories D and C. As a result of this recognition of different levels of political sophistication, political parties and their operatives increasingly find means of continuous persuasion that influence voter decision-making. This is manifested in numerous ways, including having an inspirational leader, catchy campaign slogans, voter-centered campaign messages, use of celebrities in campaign advertisements, etc.

The Eurocentric attitudes that tend to dominate the literature on African economic and political dispensation (Meillassoux 1971) are now giving way to new and better-informed perspectives on Africa (Ninsin 2006; Mensah 2011; Blankson, Cowan, and Darley 2018). The dearth of research in this area relative to the importance of understanding political dispensation on the continent leaves a huge gap in our understanding of the African perspective. In the 25 years since Ghana returned to the path of democracy, the country—often referred to as a transitioning economy—has experienced tremendous changes in its political structure, electoral systems, party systems, media systems, voter sophistication, and freedom of speech and association. Having been embroiled in political instability as a result of military authoritarian rule after the overthrow of the first republican constitution, the country finally voted for constitutional rule once more in an April 1992 referendum. The country has experienced seven election cycles since then, alternating government between two dominant parties: the National Democratic Congress (NDC) and the New Patriotic Party (NPP). Each of these elections produced very close margins of victory for the

2

winning party, indicating a high level of competitiveness in the country’s electoral system. These two political parties have come to represent the ideological divisions in Ghana, where the NDC is of the left and the NPP is on the right of center. In addition to these two dominant parties are the Convention People’s Party (CPP) and the People’s National Convention (PNC). The CPP is the party of independence and the PNC emerged out of it. However, still other parties exist: at each electoral cycle of four years, the country registers a very competitive election of at least six political parties of varied strengths.

Like other transitioning economies in Africa, research on voting patterns in Ghana suggests a complex mix of factors underpinning political allegiance and voter choice (Anebo 1997; 2001), including family history, party traditions, personality, ethnicity, and economic basis. The foregoing intricacies in the literature underlie the motivation and contribution of this study. Literature review

Ninsin (2006) writes that most voters can make little sense of the abstract left-right ideological debate as because it has little practical relevance in their lives. These ideological/policy debates only prevail within the elite communities. As in most African countries, political parties instead play ethnic and religious sentiments implicitly to attract votes (Anebo 1997); a significant number of electorates also vote according to family attachment to political interests. The majority will go with the “electoral hype,” wherein parties appear most appealing during electioneering campaigns (Mensah 2011). This mix of factors underpinning voter choice in Ghana presents a behavioral sophistication (Luskin 1990), and thereby presents a challenge to campaign strategy; this situation is what informs our study. Luskin (1990) claims exposure to political information is one of three explanatory bases that describe whether people become politically sophisticated. The other two bases are voters’ ability to assimilate such information and their motivation to do so. Luskin therefore concludes that both internal and external stimuli impact voter choice and that the political cognitions of voters can be highly organized or constrained. Luskin also makes note of other findings in the study of political sophistication which suggest that interest and intelligence— representing motivation and ability—have major effects on voter attitudes.

However, education and media exposure do not necessarily provide the impetus for sophistication (Luskin 1990). This means that voters, just like consumers, are not equal in their understanding of—and interest in—political products and how they impact their motivation to vote, abstain from voting, and/or decide how to vote. Voters can be categorized according to their level of sophistication (Campbell, Converse, Miller, and Stokes 1960), which includes ideological and non-ideological belief systems (Luskin 1990; Converse 1964); depending on these levels, voting behaviors and patterns could be determined.

In the case of Ghana, after several years of military rule interruptions, these ideological leanings gradually translated into other subliminal identifications, such as tribal, religious, and other divisions as the nation departed further from the early years of independence. Research on the early part of the fourth republic elections (see Ninsin 2006; Youde 2005; Gyimah-Boadi 2001; Nugent 2001) suggest an overwhelming reference to regionalization and ethnocentric patterns of voter perceptions and choice-making. These researches reveal a significant proportion of voters making reference to regional and ethnocentric associations for partisan identification. In a study on voter choice and electoral decisions in Ghana’s election in 2000, for example, Anebo (2001) notes that citizens had little incentive to evaluate parties and candidates on issues and competency. Anebo attributed the lack of evaluation to the low level of political knowledge in the majority of the Ghanaian electorates coupled with other primordial characteristics of the Ghanaian social

3

structure (Anebo 2001, p. 85). Recently, Tweneboah-Koduah et al (2010) have made similar claims of a confused state of voter knowledge and perceptions of political “brands.” Their study attempted to find out whether voters in Ghana could link certain constructs to the two political brands—NDC and NPP—and if such constructs could influence their political choice (Tweneboah-Koduah et al. 2010, p. 83). The authors’ findings, however, revealed that voters were unable to recall any relevant functional component of a political brand when the names of the two leading political parties were mentioned.

Regarding brand as customer reflection and voter self-image constructs, which seek to measure the likelihood of a voter choosing a political brand based on their ethnicity to the political brand in question, Tweneboah-Koduah et al (2010) reported that less than half of the focus group respondents in their study made reference to ethnicity-party associations. The authors’ findings indicate a gradual shift from previously held views, offering insight into the possible impact of modern political systems, media systems, and voter dynamics in Ghana. The authors note an emergence of voter-party relationship sophistication resembling the product-consumer relationship akin to the factors that underpin consumer decision-making (Engel, Blackwell, and Miniard 1986). Interestingly, marketing scholars overlook studies geared toward voters’ consumption of general elections, especially in transitioning economies such as Ghana.

Studies have shown that some voters are influenced by cognitive shortcuts in political decision making (see Needham 2005; Lupia and McCubbins 1998). When any of the political elements—such as the party, candidate, or policy—gains visibility unequal to the rest and is used as a representation of the rest in the party’s political structure, that element is most likely to influence voter perception; therefore, voters will have a “remarkable” knowledge of the element which is most promoted. In this direction, Needham (2005) observes that voters’ impressionistic perceptions about candidates could influence their electoral choice-making. In Ghana, findings of electoral campaign studies (Mensah 2009; Gyimah-Boadi 2001; Ayee 2000) show that the political candidate, for example, is more highly promoted in recent Ghanaian elections than the party and the policy, leading to the increasing influence of candidate brands on voter choice (Mensah 2011). It is unfortunate to realize that in spite of the growing interest in Ghana’s peaceful political and economic transition (see Whitfield 2005), to our best knowledge, not a single documented study has examined voters’ (i.e., the general public) consumption of general elections despite the call for such research by Lupia and McCubbins (1998), Ninsin (2006), and Mensah (2011). To fill this gap in the literature, the aim of this study is to untangle the factors underpinning Ghanaian voters’ voting habits in general, thereby identifying voters’ consumption of general elections. More specifically, this paper develops and then validates a scale delineating the key factors explaining why people in transitioning economies such as Ghana vote in general elections the way they do. Pursuant to this study’s aim, the main objective of the study is to develop and validate a scale designed to identify key factors reflecting voting habits/patterns in general elections. We use Ghana as an illustration of a transitioning economy (see also Whitfield 2005).

4

SCALE DEVELOPMENT AND VALIDATION STAGE 1: GENERATION OF VOTING PATTERNS STATEMENTS

Methodology We followed Mentzer, Flint, and Hult’s (2001) and Grohmann’s (2009) multi-item and

multi-study scale development and validation process involving a literature review, focus group interviews, face-to-face interviews, and large-scale surveys using members of the general public. Our efforts encompass systematic methodical adherence and attention to the key steps of the scale development process (Churchill 1979; DeVellis 2017). Our approach also follows that of Homburg and Pflesser (2000), who employed qualitative research and subsequently a survey in their development and measurement of relationships among different components of market-oriented culture.

Thus, we developed our constructs from original scales. Specifically, we conducted field studies aimed at capturing voters’/consumers’ perceptions and their natural responses to rationales for engaging in general elections. In spite of its merits, we did not embark upon experimental study in that an experimental study does not reflect more general population voting situations, and experimental studies have often solely used students as samples, which limits the study’s external validity (Walsh and Wayne-Mitchell 2005). Instead, we surveyed members of the public.

Ten months prior to Ghana’s general elections in November 2016, we used snowball and foot-in-the-door techniques to conduct four focus group interviews among a convenience sample of the public reflecting rural, urban, and city residents in Ghana. Sixty participants attended four interview sessions and were compensated with snacks, beverages, and GHC10.00 (USD3.00) cash each. The participants ranged from age 18 to 74. Discussion involved their perceptions, views, likes, and dislikes about Ghana’s general elections, political parties’ campaigns, politicians, and political party communication via radio, television, newspaper, and billboards in the country. (Ghana is an Anglophone country and the official language is English; also worthy of note is its advertisements are in English). The interviewees also discussed their involvement in party politics and their reasons for involvement in a certain political party. The discussions lasted between 45 and 90 minutes after which participants were asked to identify and then list “top-of-mind” statements and words describing their impressions of general elections and political campaigns in Ghana.

Next, in line with suggestions from experts, we followed the focus group interviews with 30 in-depth, face-to-face interviews with senior party executives, members of parliament, academics with research expertise in political marketing, opinion leaders from churches and mosques, and traditional chiefs. Inspired by Hult et al. (2017), these discussions centered on party executives’ and opinion leaders’ positions on general elections and political dispensation in Ghana, as well as their views on Ghanaians’ perceptions of the general elections and politicians (see also Mensah 2011). We used inductive reasoning to identify a list of statements generated from the focus group sessions and interviews with the executives and opinion leaders, making sure that duplicate and overlapping statements were deleted. A list of statements resulting from the focus group and the face-to-face interviews was given to four academic experts on scale development for their comments (Parasuraman, Zeithaml, and Berry 1988). The latter was an attempt to clarify any ambiguity in the statements prior to the first stage of data collection. Furthermore, we pre- tested the refined list of statements among a convenience sample of 30 MBA and EMBA students and 30 undergraduate students in a business school of a major university in Ghana. A pool of 75 statements (items) resulted following the focus group sessions and interviews.

5

Finally, to establish face validity, two academic colleagues and two doctoral students screened the items by examining their meaning in consonance with the aim of the study. The 75 items were reduced to 58 because of similarities in wording and meaning. Thus, the final list contained 58 items. We next examined the face validity of each item and further eliminated those items that were judged to overlap between the focus group interviews and the face-to-face interviews. We retained statements that were closely related to the divergent thinking process as well (Smith et al. 2007). To further reduce conceptual redundancy and achieve parsimony, we refined the list of items with the help of an academic expert and a master of philosophy student. This resulted in 45 items. The 45 items formed the final basis of the questionnaire for our survey in stage 2 of the scale development.

STAGE 2: DATA COLLECTION, SCALE DEVELOPMENT AND PURIFICATION, FACTOR

DIMENSIONS, AND VALIDATION In line with good practice, prior to the survey, we pre-tested the questionnaire for

readability, relevance, and face validity among 40 respondents: 30 members of the public and 10 university secretaries and university computer technicians. This helped to synthesize and then reduce the items obtained in stage 1 into generic constructs that will inform the basis of a typology of general elections voting patterns. To that end, respondents were asked to indicate the degree to which each statement in the questionnaire fit with (or was in agreement with) their perceptions of voting in the forthcoming Ghana general elections. We measured the statements on a 7-point Likert scale, where 1 indicated I do not agree at all with the statement and 7 indicated I very much agree with the statement. We conducted the data collection in two successive stages (initial and validation) approximately one month apart with members of the general public.

We hired six masters in philosophy and 20 undergraduate students who dropped off and collected a package of envelopes containing a cover letter on university letterhead formally introducing the survey and a three-page questionnaire about consumers’ perceptions about the general elections and political parties and politicians. In all, a simple random survey distribution of 2,000 questionnaires via the drop-off-and-collect technique (see Ibeh, Brock, and Zhou 2004) yielded 800 responses: a 40 percent response rate in a one-wave survey. Following the elimination of incomplete responses, 702 questionnaires were kept for analysis. The survey was conducted in January and February 2016 in Accra, Tema, and Kasoa: three major metropolitan areas in Ghana by a total of 26 students. The students were compensated with GHC200 (USD50.00) cash each, excluding transportation and feeding expenses.

In order to achieve validation of the data, in July and August 2016 (five months after the first survey and three months prior to the 2016 voting day), similar to the first survey, we embarked upon our second survey among the general public residing in Accra, Takoradi, and Kumasi. In the absence of a reliable sample frame, we relied on the drop-off-and-collect technique. Thus, similar to survey 1, we measured the statements on the questionnaire on a 7-point Likert scale, where 1 indicated I do not agree at all with the statement and 7 indicated I very much agree with the statement. Again, the sample in both surveys are members of the general public over the age of 18. We received 600 completed questionnaires out of the 3,000 questionnaires distributed to respondents: a 20 percent response rate in a one-wave survey. Following the deletion of incomplete responses and cleaning, we kept 584 for analysis. Although a one-wave method of data collection characterized the two stages of the survey, in view of the size of the completed questionnaires and the similarity of the sample with the general population in the study context, we believe non- response bias is minimal. Furthermore, a simple non-response bias test in the form of follow-up

6

visits to a sample of the population in the study setting two weeks after the data collection revealed one main reason for non-completion of questionnaires: did not have time.

Scale Development and Purification

We explored the underlying structure of the first survey data by carrying out principal component analysis (PCA). Initial visual assessment of the correlation matrices showed a considerable degree of inter-statement correlation. On average, strong positive correlations in the range of r = .29, p < 0.001 or above and higher means were identified for the first survey, indicating the importance of the items for the model (see Appendix 1).

In addition, the Bartlett test of sphericity (p < 0.000; Approx. Chi-Square = 8377.681; df = 990) and the Kaiser-Meyer-Olkin (KMO) test (index = 0.829) for survey 1 confirmed the appropriateness of the data for exploratory factor analysis. We explored the PCA, and using an eigenvalue greater than 1 and a cumulative percentage of variance explained greater than 50, 13 factors were extracted, explaining 58% of the variance in the data for survey one. The results revealed that an acceptable number of factors accounted for the maximum portion of the variance of the original items (Steenkamp and van Trijp 1991). Moreover, the results showed high commonalities for the data set, indicating a strong variance between the variables.

Identification of Main Factor Dimensions

We employed a varimax rotation to the data to obtain a clear picture of the structure and thus identify representative reflective indicators. Internal reliability with a Cronbach α value cut- off of 0.70 was established. Further examination of the item-total correlation using a cut-off of 0.30 revealed 12 items providing evidence of the robustness of the structure with high factor loadings, high Cronbach α, and high means across the data set (Appendix 2). All variables measured on multi-item scales as reflective indicators and under three constructs/factors, namely, (a) personal gratification ethos, (b) democratic ethos, and (c) civil and patriotic ethos (Appendix 2).

Confirmatory Factor Analysis

In order to test and validate the three initial constructs, a second survey was undertaken involving 3,000 members of the public as noted earlier. Following Anderson and Gerbing (1988), we subjected the second survey data to confirmatory factor analysis (CFA) using AMOS. Internal consistency reliability was assessed using the composite reliability (CR) index. One value exceeded the generally recommended level of 0.70 while two constructs were marginal (0.65 and 0.68) (see Appendix 3). The psychometric properties of the measures were assessed in terms of convergent and discriminant validity. Two primary measures were used to evaluate the convergent validity (Hair et al. 2012): (1) factor loadings of the measures must be greater than 0.60 and statistically significant and (2) the average variance extracted (AVE) estimates of the constructs should be greater than 0.50. All three factor loadings ranged from 0.59 to 0.94 and were also statistically significant.

In addition, the AVE value for “democratic ethos” and “civil and patriotic ethos” constructs met the 0.50 threshold (i.e., 0.52 and 0.50, respectively) while the remaining construct (personal gratification ethos) was lower (0.35). Notwithstanding the latter weakness, overall, the results exhibit adequate to marginal convergent validity. Furthermore, discriminant validity is evident if the square root of the AVE of a construct is greater than its correlations with other factors in the model (Hair et al. 2012). From CFA, the AVE (diagonal values) are greater than the correlations

7

between factors (Appendix 3). Moreover, the fit indices for the three-factor measurement model are moderate in terms of the cut-off ranges (GFI = .91, AGFI = .84, NFI = .81). We infer that the analysis and results provide support—albeit in a few cases, marginal—for convergent and discriminant validity. In view of the foregoing results, we conclude that voters’ consumption of general elections in transitioning countries reflects three factors: “personal gratification ethos,” “democratic ethos,” and “civil and patriotic ethos.” Each dimension is grounded in practice and the three constructs are generic in nature. Thus, on the basis of this research’s theoretical underpinnings and empirical development, the constructs are important and will be appreciated by marketing theorists and policymakers and can be generalized within the context of investigating voters’ consumption of general elections in transitioning economies (see Winer 1999).

References

Anebo, F. K. G. (1997). “Voting Pattern and Electoral Alliances in Ghana’s 1996 Elections.” African Association of Political Science, 2(2), 38–52. Anebo, F. K. G. (2001). “The Ghana 2000 Elections: Voter Choice and Electoral Decisions.”

African Journal of Political Science/Revue Africaine de Science Politique, 6(1), 69–88. Anderson, J. C. & Gerbing, D. W. (1988). “Structural Equation Modeling in Practice: A Review

and Recommended Two-Step Approach.” Psychological Bulletin, 103(3), 411–423. Ayee, J. R. A. (1997). “The December 1996 General Elections in Ghana.” Electoral Studies, 16, 416–427. Ayee, J. R. A. (2000). “Deepening Democracy in Ghana: Politics of the 2000 Elections.” Freedom

Publication, 1–11. Blankson, C., Cowan, K., & Darley, W. K. (2018). “Marketing Practices of Rural Micro and Small

Businesses in Ghana: The Role of Public Policy.” Journal of Macromarketing, 38(1), 29– 56.

Campbell, A., Converse, P., Miller, W., & Stokes, D. (1960). The American Voter. Chicago: University of Chicago Press.

Center for Democratic Development (CDD) Ghana (2005). “Financing Political Parties in Ghana: Policy Guidelines.” Accra: CDD.

Converse, P. E. (1964). “The Nature of Belief Systems in Mass Publics.” In David E. Apter (Ed.), Ideology and Discontent. Free Press.

Churchill Jr, G. A. (1979). “A paradigm for Developing Better Measures of Marketing Constructs.” Journal of Marketing Research, 16(1), 64–73.

DeVellis, R. F. (2017). Scale Development: Theory and Applications (Fourth Edition), SAGE, Los Angeles, CA.

Engel, J. F., Blackwell, R. D. & Miniard, P. W. (1986). Consumer Behavior. 5th edition., Hinsdale, IL: Dryden.

Gyimah-Boadi, E. (2001). “A Peaceful Turnover in Ghana.” Journal of Democracy, 12(2), 103– 117 Hair, J. F., Sarstedt, M., Ringle, C. M., & Mena, J. A. (2012). “An Assessment of the Use of Partial

Least Squares Structural Equation Modeling in Marketing Research.” Journal of the Academy of Marketing Science, 40(3), 414–433.

8

Homburg, C. & Pflesser, C. (2000). “A Multiple-Layer Model of Market-Oriented Organizational Culture: Measurement Issues and Performance Outcomes.” Journal of Marketing Research, XXXVII (November), 449–462.

Hult, G. T. M., Morgeson III, F. V., Morgan, N. A., Sunil Mithas, S., & Fornell, C. (2017). “Do Managers Know What Their Customers Think and Why?” Journal of the Academy of Marketing Science, 45, 37–54.

Ibeh, K., Brock, J. K., & Zhou, Y. J. (2004). “The Drop and Collect Survey Among Industrial Populations: Theory and Empirical Evidence.” Industrial Marketing Management, 33(2), 155–165.

Lupia, A. & McCubbins, M.D. (1998). The Democratic Dilemma: Can Citizens Learn What They Need to Know? Cambridge: Cambridge University Press.

Luskin, R. (1990). “Explaining Political Sophistication.” Political Behaviour, 12(4), 331–361. Meillassoux, C. (1971) (Ed.). The Development of Indigenous Trade and Markets in West Africa.

New York, NY, and London, UK: Oxford University Press. Mensah, A. K. (2011). Political Brand Management: Forms and Strategies in Modern Party

Politics. Ph.D. Thesis. The University of Sheffield. Mentzer, J. T., Flint, D. J., & Hult, G. T. (2001). “Logistics Service Quality as a Segment-

Customized Process.” Journal of Marketing, 65(October), 82–104. Needham, C. (2005). “Brand Leaders: Clinton, Blair and the Limitations of the Permanent

Campaign.” Political Studies, 53, 343 Mensah, A. K. (2011). Political Brand Management: Forms and Strategies in Modern Party Politics 361

Ninsin, K. A. (2006) “Political Parties and Political Participation in Ghana.” Konrad Adenaur Stiftung, Accra, Ghana.

Nugent, P. (2001) “Ethnicity and Recent Democratic Experiments in Africa.” African Issues, 29(1/2), 2–7.

Parasuraman, A., Zeithaml, V. A., & Berry, L. L. (1988). “SERVQUAL: A Multiple-Item Scale for Measuring Consumer Perceptions of Service Quality.” Journal of Retailing, 64(1), 12– 40.

Smith, R. E., MacKenzie, S. B., Yang, X., Bucholz, L. M., & Darley, W. K. (2007). “Modelling the Determinants and effects of Creativity in Advertising.” Marketing Science, 26(6), 819– 833.

Steenkamp, J. E. M. & van Trijp, H. C. M. (1991). “The Use of LISREL in Validating Marketing Constructs.” International Journal of Research in Marketing, 8, 283–299.

Tweneboah-Koduah, Y. E., Akotia, M., Akotia, S. C., & Hinson, R. (2010). “Political Party Brand and Consumer Choice in Ghana.” Journal of Management Policy and Practice 11(5), 79–88.

Walsh, G. and Mitchell, V. (2005). “Consumer Vulnerability to Perceived Product Similarity Problems: Scale Development and Identification.” Journal of Macromarketing, 25(2) (December), 140–152.

Whitfield, L. (2005). “Trustees of Development from Conditionality to Governance: Poverty Reduction Strategy Papers in Ghana.” Journal of Modern African Studies, 43, 641–664.

Winer, R. S. (1999). “Experimentation in the 21st Century, The Importance of External Validity.” Journal of the Academy of Marketing Science, 27(3), 349–358.

Youde, J. (2005). “Economics and Government Popularity in Ghana.” Electoral Studies, 24(2005), 1–16.

9

Appendix 1: Survey 1 - Descriptive statistics of items (N=702)

Survey 1 Items Mean Std. Deviation Party communication influences my vote for a particular party 4.03 2.02

Media influences my vote for a particular party 3.19 1.91

General elections in Ghana have changed over the years 4.96 1.78 I am in favor of freedom of speech 5.87 1.47

I am in favor of an economically strong Ghana 5.77 1.75 The Electoral Commission has the capacity to conduct credible elections 4.80 1.79

I am in favor of Ghana succeeding as a whole 6.26 1.22

Political parties have good intentions for the people 3.68 1.68 I am afraid of chaos during the General Elections 4.19 2.04

I am happy about the General Elections 4.70 1.71

I enjoy party political campaigns 4.07 1.88 Party manifestos are important 5.34 1.76

Party political campaigns are good for the country 4.87 1.71

My party has a clear political message 5.03 1.60 I enjoy discussing politics with my friends 4.09 1.98

I enjoy discussing politics with my family 4.02 1.98

I like political leaders when they show different leadership styles 5.07 1.69 Resource mobilization for political parties is important 4.79 1.79

I pay dues (money) to my party 2.50 1.78

It is important to pay dues (money) to political parties 3.55 2.05

It is important for companies to fund political parties 3.05 1.84

Social groups influence voter intentions 4.67 1.80 I am fond of engaging with political parties via social media 3.24 1.94

Social media is important in political campaigns 5.01 1.78

I am in favor of multiparty democracy 5.33 1.85 Democracy is good for Ghana 5.75 1.73

Democracy is good for every country 5.47 1.91

I make my own decisions about which party to support 6.01 1.53

My friends influence my perceptions about which party to support 2.38 1.87

My family influences my perceptions about which party to support 2.69 1.87 I can be persuaded by a political party with money 2.14 1.69

I can be persuaded by a political party with their campaign promises 2.96 1.90

I enjoy voting 4.48 2.03 I am aware of parties’ messages 4.95 1.77 I enjoy discussing politics 3.76 2.04 I have concerns about the way Ghana’s politics are conducted 5.05 1.75

I have concerns about politicians 4.79 1.90

My perception about the general elections is that it helps the image of Ghana 5.30 2.03 My perception about the general elections is that it opens doors for future opportunities 5.11 2.06

My perception about the general elections is that it will encourage foreign investment 4.99 2.10 Compared to my friends, I spend a lot of time listening to politicians 2.89 2.16 I have contributed resources to my party 2.11 1.84 Political campaigns affect my decision to vote for a particular political party 3.12 2.22 The flagbearer (leader) of a political party influences my decision to vote for a particular political party

3.92 2.42

A party’s manifesto influences my decision to vote for a particular political party 4.36 2.34

10

Appendix 2: Survey 1 - Descriptive statistics, Varimax - rotated principal component loadings, Cronbach alpha values (N= 702) (final refinement and re-specification)

Variable Mean SD Factor 1: Personal gratification ethos

Factor 2: Democratic ethos

Factor 3: Civil & patriotic ethos

Corrected Item-Total Correlation

I enjoy discussing politics with my friends

4.09 1.98 0.86 0.51

I enjoy discussing politics among my family

4.03 1.98 0.84 0.50

I enjoy discussing politics 3.77 2.04 0.70 0.43 I enjoy party political campaigns

4.07 1.88 0.57 0.40

Party political campaigns are good for the country

4.87 1.71 0.55 0.39

Party Manifestos (i.e., stated policies and objectives) are important

5.34 1.76 0.55 0.39

Democracy is good for Ghana

5.76 1.73 0.86 0.44

Democracy is good for every country

5.48 1.91 0.82 0.38

I am in favor of multiparty democracy

5.34 1.85 0.57 0.37

My perception about the general elections is that it opens doors for future opportunities

5.11 2.06 0.77 0.42

My perception about the general elections is that it will encourage foreign investment

4.99 2.10 0.74 0.41

My perception about the general elections is that it helps the image of Ghana

5.30 2.03 0.73 0.44

Cronbach alpha 0.75 0.74 0.70 Extraction Method: Principal Component Analysis; Rotation Method: Varimax with Kaiser Normalization; Cronbach’s Alpha = 0.78 for 12 items; Reliability – Split-Half: Part 1 = Value 0.73 (6 items), Part 2 = 0.68 (6 items), Spearman-Brown Coefficient [Equal Length = 0.62, Unequal Length = 0.62], Guttman Splitt-Half Coefficient = 0.61.

Appendix 3: Survey 2 - CR, AVE and Correlation Matrix of Constructs (N = 584)

Constructs/Factors CR AVE MSV MaxR(H) Civil and Patriotic Ethos

Personal Gratification Ethos

Democratic Ethos

Civil and Patriotic Ethos 0.65 0.50 0.30 0.79 0.71

Personal Gratification Ethos 0.68 0.35 0.30 0.86 0.55 0.59

Democratic Ethos 0.73 0.52 0.24 0.94 0.49 0.37 0.72