Testing for Bivariate Categorical Analysis

Answer

Testing for Bivariate Categorical Analysis

Scenario 1: Democracy and Trust in Police by Citizens

It is hypothesized that trust in the police is an integral part of any democracy. To confirm this hypothesis, the test of association of the categorical variables is done as articulated by Frankfort-Nachmias and Leon-Guerrero (2017).  The research question being answered in this test is: Is trust in the police by citizens influenced by the democracy of the country?

The null hypothesis tested states that there is no association between democracy and trust in the police.

Table 1: Bivariate table of police trust and democracy.

The table indicates that there is a higher in a democratic country, a majority (77.5%) have at least a little trust in the police, with a relatively high 26% agreeing to have a lot of trust in police. On the other hand, citizens from non-democratic countries are less likely to trust in the police, with a high proportion of respond

Table 2: Association test between Trust in police and Democracy.

Table 3: Table of strength of association.

The χ² = 1242.17, p < 0.05 indicates that there is a significant association between democracy and citizens’ trust in the police (reject the null hypothesis). This affirms the hypothetical thought that trust in the police is an integral part of any democracy. Further, the association is significantly strong (p-value < 0.05). This study finds meaning in enhancement of policing and social democracy. Therefore, it is true that trust in the police is important for democracy in Africa.

Scenario 2: Trust in Police and Residence

The hypothetical thought that there is a relationship between citizen trust in the police and whether respondents reside in rural, urban or semi-urban. The categorical data analysis design is used to answer the research question, ‘Is there an association between citizen trust in the police and whether respondents reside in rural, urban or semi-urban?’ The null hypothesis states that there is no relationship between citizens’ trust in the police and residence block.

Table 4: Cross-tabulation of trust in the police and residence.

A very little proportion of the urban populace (20%) have a lot of trust in the police, which is relatively low compared to 31% of Rural and 32% of the Semi-Urban. Those who do not trust in the police at all are composed of 27% proportion of the urban residents, 20% of the rural populace, and 16% of the semi-urban. The bivariate tabulation indicates that there is variation in level of police trust based on residence.

Table 5: Test of Association table

Table 6: Strength of association analysis

The results of the test indicate that there is a significant association between citizens’ trust in the police and residence (χ² = 877.48, p < 0.05). Therefore, the null hypothesis is rejected. The Phi and Crammer’s V test indicate that the association between the two variables is indeed strong. These results affirm the hypothesis of relationship between the variables. The social meaningfulness of this study is found in the assessment of policing around regions and aligning of police duties. Among the possible array of reasons for this are media influence (based on media reports of police misconduct) and political ideologies based on exposure (Wang & Sun, 2018).

Current Economic Conditions and Extent of a Democracy

The study is set to furnish the research question – Is there any relationship between citizen’s perceptions of a country’s current economic conditions and the extent of democracy? To sufficiently answer this, the null hypothesis that there is no association between a citizen’s perceptions of a country’s current economic conditions and the extent of democracy is tested.

Fig 1: Bivariate bar-graph of perception of present economic condition and extent of democracy.

ents (40%), indicated to be having no trust at all in the police.

Table 7: Test of Significance Association

The bivariate tabulation indicates that there is variation in citizens’ perception of current economic status of the country based on extents of democracy. Further, the χ² = 2562.47, p < 0.05 results indicate that there is a significant relationship between citizen’s perceptions of a country’s current economic conditions and the extent of democracy (null hypothesis is rejected). Moreover, the association is proved to have significant strength by the Phi and Cramer’s V tests (p < 0.05). This affirms that there is a significantly strong association between citizen’s perceptions of a country’s current economic conditions and the extent of democracy. This aids in the socio-economic discipline to ascertain level of citizen satisfaction with government policies and development. Among the possible reasons for this phenomenon are visible results of delegation of authority and regulations of social conflicts, education, healthcare, higher human capital accumulation, lower inflation, lower political instability, and higher economic freedom (Song, Berger, & Kim, 2017).