Send message to Contributor

Contribution short abstract:

The research involves the development of a partisan disconfirmation model (PDM), which builds on Oliver's (1971) expectancy disconfirmation model (EDM) on citizen satisfaction and its application in citizen satisfaction with public services (Van Ryzin, 2004, 2006, 2013).

Contribution long abstract:

The partisan expectancy disconfirmation model states that citizens' evaluation of economic performance depends on citizens' identification with either the ruling or opposition party. That is, the interaction between partisanship and perceived economic performance results in citizens formulating satisfaction judgements with the ruling party's performance. If perceived economic performance is higher than citizens’ alignment with the ruling political party, it leads to positive disconfirmation and, hence, higher satisfaction with the incumbent president’s performance. Similarly, when citizens alignment with an opposition political party is higher than perceived economic performance, it leads to negative disconfirmation and hence lower satisfaction with the ruling party or president’s performance, making up the two confirmation hypotheses. On the other hand, when citizens’ identification with either the ruling (supporters) or opposition (sympathisers) party is weaker, they are more likely to give honest evaluations of the economic performance as well as honest satisfaction judgements of the ruling party's performance, thereby yielding the two disconfirmation hypotheses where we find ruling party supporters evaluating economic conditions as bad and dissatisfied with the ruling party’s performances, as well as some opposition party sympathisers who eventually evaluate economic conditions as good and expressing satisfaction with the ruling party’s performance. This model is tested using Afrobarometer survey data from an African country. The Chi2 statistic confirms the existence of the association between partisanship and economic performance, and the Cramer's V ascertains a moderately strong relationship between the two variables, hence the need to test the model's disconfirmation process using advanced econometric analyses.