Researchers and policy makers posit that governance is one of the critical factors explaining the divergence in performance across developing countries. While supporting the governance model, researchers and policy makers have provided different good governance explanations to illuminate its meaning. The World Bank group defines governance as “the manner in which power is exercised in the management of the affairs of a nation and its relations with other nations” [12] . [13] explains that governance consists of the traditions and institutions through which authority in a country is exercised. [13] provides Worldwide Governance Indicators (WGI) along six dimensions: voice and accountability; political stability; government effectiveness; regulatory quality; rule of law and control of corruption. Four specific elements of governance are: accountability (officials being answerable for government behavior), participation, predictability and transparency [12] . For [6] , based on a long- standing research program at the World Bank, they capture six key dimensions of governance. Furthermore, over time, [6] identifies nine characteristics of good governance as participation, rule of law, transparency, responsiveness, consensus orientation, equity, effectiveness and efficiency, accountability and strategic vision. According to [8] [14] [15] , good governance is categorized into eight major attributes not very different from [12] classification above. To [14] , transparency means that decisions taken and their enforcement are done in a manner that follows rules and regulations. As for accountability, it involves designing effective avenues of recourse or redress to affected persons [15] .

In international affairs, analysis of good governance involves scrutiny of stakeholder relationships ranging between citizens to elected officials to governments and markets [12] [16] [17] . A variety of country level initiatives and international movements put emphasis on various types of governance reform based on their own needs and agendas [18] - [20] .

The [1] defines this concept as expanding the choices people have to lead lives that they value. The report identifies important choices to include knowledge, healthy life, education and decent standard of living, political freedom, guaranteed human rights and self-respect. The [21] stresses that education boosts peoples’ self-conf- idence, enables them find better jobs, make demands on government for health-care and mortality. People who are poor, unhealthy and illiterate simply have fewer choices in life. This means that failure to implement education policies adversely affects choices underpinning human development. Similarly, [22] argue that enjoying a long and healthy life, education, access to resources and participating in societal decision making enable citizens live better quality life.

Researchers and policy makers recognize the fact that despite the benefits currently manifesting in Northern Uganda, poverty in the country remains heavily regionalized [23] . The current [24] explicitly recognizes the need to integrate northern Uganda into the mainstream development of the country. The NDP does propose strategies for this region, including a commitment to the second Peace, Recovery and Development Plan (PRDP 2) for the north. PRDP 2 is expected to improve community participation and ownership of services, promote economic recovery and revamp the quality of life of the beneficiary communities in Northern Uganda [24] .

Until recently, scholars like [18] [25] have began to view that the literature on governance and/for sustainable development amounted to a great deal of discursive smoke but little in the way of empirical fire. [22] believes that governance and/for sustainable development need to be better described and accounted for in an empirical study. [26] argues that policy makers continue to confront long lists of things that must be done to achieve good governance, with little guidance about how to pick and choose among them as priorities. In this perspective, some researchers have attempted to establish linkages between good governance and human development. For instance, [26] , using panel data analysis, finds the evidence that good governance and past human development determines present human development in India. [4] , [18] ’s study in Nigeria found that control of corruption and voice and accountability have no significant negative impact on human development.

Investigating the empirical relationship between corruption and economic growth, [27] finds that a 1 percent increase in the corruption level reduces the growth by about 0.72 percent. [28] investigates the nexus between corruption and human development empirically by using a full information maximum likelihood approach and finds a negative correlation. Examining the impact of corruption on human development, [29] [30] reveals that there is a statistically significant negative relationship between corruption indexes and human development.

[6] shows that there is quantitative evidence that a positive correlation exists between measures of the quality of governance, broadly defined into six indicators, and various measures of economic development. For instance, this correlation finding is debated by [13] [31] [32] who find no significant differences in the scores on good governance of high-growth and low-growth developing countries. The studies above suggest that the relationship between good governance and human development are still inconclusive. This paper explores this relationship in developing country Uganda contextualized within a geographical region that had emerged from a devastating two decade conflict.

Based on the above, the following hypotheses were tested (Figure 1).

The population of the study is the community, non-governmental and community−based organizations within four divisions of the municipality namely Pece, Laroo, Layibi and Bardege. The sample selection was convenience sampling-based and stratified random sampling according to community members, health workers, teachers, local leaders, employees in both non-governmental and community-based organizations. The community members, heath workers, teachers, local leaders, NGOs and CBOs were physically visited by the researchers and three assistants with an introductory letter from the faculty for a period of one month (January, 2015). All the sampled community members returned the questionnaires filled with the help of the researcher and assistants. The rest of the respondents were left to fill the questionnaire on their own and promised to have filled them by the next appointment visit. In all, 590 questionnaires were served to respondents out of which 387 were collected and found usable for the purpose of the study.

Both primary and secondary data were collected for this empirical research. Primary data was collected using a structured questionnaire while secondary data was collected from archival records of local government, NGOs and CBOs. The questionnaire was developed based on the literature on governance and human development [13] [14] [20] . All questionnaire items (27 for good governance and 11 for human development) were measured on a five-point scale (1 = strongly disagree, 5 = strongly agree) and pilot tested. The questionnaires were pre-tested on four faculty experts before being administered to respondents. Content Validity Index (CVI) of 0.8759 was calculated by tallying the results of the faculty experts based on the degree to which the experts agree on the relevance and clarity of the items. In addition, the instrument was piloted on ten people working in the local government in the municipality. These were conveniently approached by the research assistants, being familiar and born and raised in the municipality. The filled questionnaires were then coded, sorted, entered analyzed using SPSS Version 20. Reliability reflects the stability and consistency of an instrument measuring a concept [33] [34] . In view of the characteristics of the pilot test instruments used in this study, the inter-item reliability consistency alpha was used to measure its reliability. Chronbach’s alpha values of 0.805 and 0.795 for good governance and human development respectively were computed and presented in Table 1 below. This indicates that values are within the acceptable range of well above the 0.70 cut-off proposed by [34] - [36] . Hence the instrument items were appropriate for further analysis.

In order to assess the suitability of the data for factor analysis, we inspect the correlation matrix for coefficients of 0.3 and above, and calculating the Kaiser-Meyer-Olkin Measure of Sampling Adequacy (KMO) and Bartlett’s Test of Sphericity [35] . When the KMO exceeds 0.6 and Bartlett’s test of sphericity is large and significant, then factor analysis is suitable (Leech et al., 2005). Furthermore, to be able to apply principal component analysis in this study, both the KMO and Bartlett’s test were computed. In the pilot study, the KMO measure is 0.746, Sig. = 0.000 for good governance and 0.676, Sig. = 0.000 for human development. These measures are all well above the 0.60 proposed by [35] indicating that there were sufficient items for each variable. The principal

component analysis results suggested that the good governance item components accounted for 67.4% total variance explained while human development items accounted for 72.3% total variance explained using the Kaiser criterion [35] . The results from the pilot study suggested that the instrument was very reliable for further testing.

The results of the sample data are shown in Table 2.

As indicated in Table 2, 200 (51.7%) of the respondents were male and 187 (48.3%) were females. The table also shows that majority 134 (34.6%) respondents were certificate holders and 47.5% were aged below 30 years.

The independent variable is good governance measured by five attributes: accountability, government effectiveness, participation, control of corruption and bribery.

Accountability involves establishing effective avenues of recourse or redress to affected persons when stakeholders’ responsibilities and obligations are not fulfilled in ensuring successful outcomes [19] . Government efficiency means the quality of the public or civil service and the degree to which it is independent from the political process [13] . Control of corruption indicates the extent to which public power is used for private gain and the extent to which public power is controlled by private interests [13] . Bribery refers to the offering, giving, accepting or soliciting of an advantage as an inducement for an action which is illegal or a breach of trust [37] - [40] .

The dependent variable is human development measured by eight items. It is difficult to measure a firm’s human development level since it does not necessarily equal the sum of the constructs used in this study.

Good governance as an independent variable is measured by 26 items (see Appendix).

[37] suggests that the principal components analysis (PCA) is a data reduction technique used to reduce a large number of variables to a smaller set of underlying factors that summarize the essential information contained in the variables. The decision about which principal components to retain depends on the percentage of the variance accounted for by the variable, the absolute variance accounted for by each principal component (PC) and whether the component can be meaningfully interpreted [37] [38] . This research study uses varimax with Kaiser Normalization rotation method developed by [36] . In this method, the principal components with Eigen values greater than 1 are retained for further analysis [34] - [36] [41] .

An inspection of results from the PCA indicate that the Kaiser-Meyer-Olkin Measure of Sampling Adequacy (KMO) is 0.736 well above the 0.60 proposed cut off [35] and Barlett’s Test of Sphericity is large (χ² = 1521.677) and significant as shown in Table 3 below. These results satisfied the necessary conditions for use of PCA in this study.

Results in Table 4 show five components (see Table 5 and Table 6) with Eigen values greater than 1 accounted for 70.653% of the variance in good governance. The first component labeled government efficiency accounted for 31.516%; the second component labeled accountability accounted for 10.640%; the third component labeled control of corruption accounted for 10.379%; the fourth component labeled bribery accounted for 9.449% while the fifth component labeled participation accounted for 8.579% of the variance in good governance.

[35] suggests that you think carefully before including two variable constructs with a bivariate correlation of,

Source: Primary Data.

Source: Primary Data.

Extraction Method: Principal Component Analysis.

^aRotation converged in 7 iterations. Source: Primary Data.

Source: Primary Data.

say, 0.7 or more in the same analysis. The highest correlation in Table 8 is 0.793 suggesting a test of the normality assumption.

In fulfillment of normality assumption a data set is considered normal if the values of skewness fall within the range of +2 to −2 while kurtosis values do not exceed the range of +7 to −7 [35] . In regard to the above threshold for justifying the normality of data, all skewness values and kurtosis values in Table 7 fall within the acceptable range as advocated by [35] . Therefore, all observed data for items considered under this study are normal as indicated in Table 7 and all constructs are retained for analysis.

Correlation analysis was carried out to detect any autocorrelation between human development and each of the constructs of good governance. Such bivariate analysis was undertaken using Pearson correlation. In addition, correlation coefficients are also intended to attest the construct validity of the constructs of good governance and to check for multicollinearity [35] . The inter-variable correlations results are shown in Table 8.

From the results of correlations analysis as depicted in Table 8, control of corruption emerged as the feature with the best predictor on other good governance constructs (r = 0.585, p < 0.01) and on human development (r = 0.416, p < 0.01). All the constructs of good governance were positively statistically significant at all levels (p < 0.01 and p < 0.05) hence the hypotheses were achieved. All good governance constructs were significant and positively correlated with human development (p < 0.01) with the exception of bribery which, mostly, is not statistically significant at any reasonable level. Most importantly, bribery had a positive non-significant relationship with human development (r = 0.085, p > 0.01) but a negative statistically significant correlation with government efficiency (r = −0.200, p < 0.01). This implies that bribery is neither associated with human development nor any attributes of good governance. Moreover, bribery impacts negatively on government effici- ency.

Table 9 indicates that the independent variable (good governance) has adequate explanatory power on the dependent variable (human development) to the magnitude of 24.4% (Adj. R² = 0.244, P < 0.001).

Table 10 above presents the analysis of variance (ANOVA) results of F-test. One-way ANOVA test results reveal that there is significant difference between good governance indicators and human development (F = 25.823, p = 0.000).

Source: Primary Data.

^**Correlation is significant at the 0.01 level (2-tailed). ^*Correlation is significant at the 0.05 level (2-tailed). Source: Primary Data.

^aPredictors: (Constant), Bribery, Accountability, Participation, Corruption Absence, Government Efficiency. ^bDependent Variable: Human Development. Source: Primary Data.

^aPredictors: (Constant), Bribery, Accountability, Participation, Corruption Absence, Government Efficiency. ^bDependent Variable: Human Development. Source: Primary Data.

Inspection of the B coefficients in Table 11 shows that control of corruption is the most influential explanatory variable (B = 0.228, p = 0.000) and the least explanatory variable is government efficiency (B = 0.084, p = 0.139). Further, from Table 8, the regression coefficients indicated that the dimensions of control of corruption (B = 0.228, p = 0.00), accountability (B = 0.190, p = 0.00) and participation (B = 0.152, p = 0.009) exerted statistically significant influence on overall human development followed by the dimension of bribery (B = 0.093, p = 0.054). Thus hypotheses H2, H3, H4 and H6 were achieved. Government efficiency (B = 0.084, p = 0.139) was not statistically significant, hence hypothesis H1 is not met. This study suggests that greater levels of accountability, participation and control of corruption are associated with greater levels of human development. Most importantly, bribery had a positive non-significant relationship with human development (B = 0.093, p > 0.001) and a negative correlation with government efficiency. Hence, hypothesis H5 was not achieved. This implies that bribery is neither associated with human development nor any attributes of good governance.