A non – parametric investigation of residential land selection factors in Ado – Ekiti, Nigeria

 Corresponding author: Remilekun Enitan Alabi https://orcid.org/ 0009-0002-8548-7315 Copyright © 2025 Author(s) retain the copyright of this article. This article is published under the terms of the Creative Commons Attribution Liscense 4.0. A non – parametric investigation of residential land selection factors in Ado – Ekiti, Nigeria Remilekun Enitan Alabi * and Taiwo Stephen Fayose Department of Statistics, The Federal Polytechnic, Ado Ekiti, Ekiti State, Nigeria. World Journal of Advanced Research and Reviews, 2025, 26(03), 130–144 Publication history: Received on 23 April 2025; revised on 30 May 2025; accepted on 02 June 2025 Article DOI: https://doi.org/10.30574/wjarr.2025.26.3.2149 Abstract This research investigates the patterns and determinants of residential land use in Ado – Ekiti, Nigeria by employing the Kruskal – Wallis H test which is a non – parametric statistical tool suitable for analyzing non – normally distributed socio – economic data. Ado – Ekiti undergoing rapid urbanization presents complex land use dynamics shaped by multiple socio – economic, cultural, and environmental influences. The study surveyed 2000 land residential owners from three socio – economically distinct areas i.e. GRA 3rd Extension (high – income), Fayose Housing Estate (medium –income), and Marina Avenue (low – income) to explore factors guiding residential land selection. Key variables such as proximity to employment, security, environmental quality, income level, infrastructure and cultural ties were rated by respondents. Results highlighted proximity to employment, security and environmental quality as the leading determinants influencing residential location choices. The Kruskal – Wallis test however found no statistically significant differences in factor ratings across the three areas suggesting homogeneity in perceptions despite socio – economic stratification. A subsequent Dunn’s test identified a significant difference only between security and topography as influencing factors. Demographic analysis showed a predominance of male, middle – aged, married, educated and government – employed residents especially within higher – income neighborhoods. The study reveals that economic and infrastructural considerations overshadow cultural and topographical factors in residential decisions in Ado – Ekiti. These findings emphasize the need for government at all levels to prioritize employment accessibility, safety and environmental improvements to meet residents’ preferences and support sustainable urban growth. This research advances understanding of residential land use patterns in a developing city context like Ado – Ekiti providing empirical evidence for evidence – based urban land use policies. Keywords: Kruskal – Wallis test; Dunn’s test; Benjamini – Hochberg test; Percentage; Residential Areas 1. Introduction The selection of residential land location is a critical decision making process that involves evaluating various factors to determine the most suitable location for housing especially urban centers like Ado – Ekiti (Oduwaye, 2013; Falade, 2017; Jimoh, 2017). Urbanization remains one of the most powerful forces reshaping human settlements, especially in developing countries such as Nigeria (Oyedele, 2019). As cities expand in both population and spatial extent, the dynamics of land use particularly residential land use become increasingly complex. Ado – Ekiti, the capital city of Ekiti State exemplifies these urban dynamics as the city undergoes rapid urban transformation driven by demographic, economic, and infrastructural changes (Falade, 2017). The pattern and nature of residential land use in Ado – Ekiti are not random but influenced by a confluence of socio – economic, cultural and environmental factors. Understanding these influences is critical for sustainable urban development, effective land use planning, and equitable access to housing (Adebayo, 2015). World Journal of Advanced Research and Reviews, 2025, 26(03), 130–144 131 This paper aims to investigate the determinants of residential land location choices in Ado – Ekiti using the Kruskal – Wallis H test. Kruskal – Wallis H test is a non – parametric statistical tool proposed by Kruskal and Wallis (1952). Unlike parametric tests, the Kruskal – Wallis test does not rely on the assumptions of normal distribution or homogeneity of variance which are often unrealistic in urban and social research (Fayose et al., 2023). Through this approach, the study seeks to uncover how factors such as income, education, family size, occupation and proximity to amenities shape residential preferences in the Ado – Ekiti. 1.1. The Urban Context of Ado – Ekiti Ado – Ekiti is located in southwestern Nigeria. Ado – Ekiti serves as a crucial political, economic, and educational hub within Ekiti State. Historically a medium – sized urban center, it has witnessed substantial growth since becoming a state capital in 1996. This growth is characterized by increased housing demand, urban sprawl, and transformation of land use from agricultural and undeveloped lands to residential estates and commercial developments. The city is divided into older traditional cores and newer peri – urban expansions with varying levels of infrastructure and service delivery. These variations contribute significantly to residential location preferences. Consequently, the pattern of residential development in Ado – Ekiti reflects both organic and planned processes influenced by land tenure systems, socio-cultural affiliations, access to infrastructure, and real estate market forces. Due to the majority of the population of residents of the Ado – Ekiti metropolis, the settlement is seen to functionally perform only administrative functions because it’s most dominated by civil servants who either work with federal or state government and a few fractions of the population works with private institutions. 2. Literature Review Residential land use theory has evolved significantly over the past century. Classical urban models such as Burgess' concentric zone theory (1925) postulated that urban land use patterns radiate from the city center in rings, with residential quality improving outward. Though seminal, such models often fall short in explaining land use in contemporary African cities, where factors like informality, infrastructure deficits, and cultural ties play greater roles. More recent theories emphasize the importance of individual agency and psychological constructs in residential decision-making. Canter (1977) argued for a psychological model where residential choice is influenced by a person's perception, expectations, and experiences of place. In line with this, residential location is not just a physical phenomenon but a socio – spatial process shaped by access to resources, aspirations, and the broader urban context. In Nigeria, empirical studies on residential land use have highlighted the influence of income, educational level, infrastructure, land tenure, and socio – cultural factors (Fabiyi, 2006; Olayiwola et al., 2008). These findings align with Mabogunje’s (1972) seminar work which underscored the interplay of economic and cultural variables in shaping urban development in Africa. However, many of these studies relied heavily on parametric methods, which often require strict assumptions that real – world urban data may not satisfy. 2.1. Urbanization and Residential Land Use in Nigeria Urbanization in Nigeria has grown rapidly, with more than 50% of the population residing in urban centers (UN – Habitat, 2016). Cities such as Lagos, Ibadan, and Abuja have received considerable academic attention, yet secondary cities like Ado – Ekiti are equally important in the spatial restructuring of Nigeria’s human settlements (Ajayi and Olayiwola, 2005; Oyesiku, 2010). The transformation of Ado – Ekiti from a provincial town into a regional hub has been propelled by public sector employment, educational institutions like Federal Polytechnic Ado Ekiti, Ekiti State University and Afe Babalola University and commercial expansion (Olusola et al., 2013). The increasing demand for housing in Ado – Ekiti has resulted in the development of both formal and informal residential neighborhoods (Aluko, 2010). Informal settlements often emerge due to high costs and inadequate supply in formal housing markets, leading to land conversion on urban fringes without proper planning (Agunbiade, 2014). Understanding the dynamics behind residential location choices is essential for mitigating urban sprawl and promoting organized growth. 2.2. Determinants of Residential Location Selection Residential location decisions are shaped by a combination of push and pull factors. Push factors include urban congestion, insecurity and poor infrastructure while pull factors comprise better housing, access to services and serene World Journal of Advanced Research and Reviews, 2025, 26(03), 130–144 132 environments (Anyanwu and Afolabi, 2015). Housing affordability and proximity to employment remain key drivers (Ajanlekoko, 2001; Aribigbola, 2008). Socio – economic variables such as income, education, family size, and occupation influence the prioritization of these factors (Wahab, 2012). Low – income earners may opt for peripheral locations where land is cheaper even if they are far from work or public amenities. In contrast, middle and high – income households may prioritize security, neighborhood prestige, and quality of infrastructure (Ojo and Ighalo, 2019). In Ado – Ekiti, studies have shown that location choices are affected by proximity to markets, schools and road networks (Fadare and Aluko, 2004). Cultural considerations such as the preference to reside near ancestral homes or family members also play a role (Afolayan, 2016). 2.3. Statistical Approaches in Urban and Land Use Studies Urban land use studies traditionally utilize statistical techniques to analyze spatial and behavioral patterns. Parametric methods like multiple regression, factor analysis, and ANOVA have been widely used (Aina, 1990; Mabogunje, 1992). However, these methods assume normal distribution of data, homoscedasticity, and linearity—assumptions often violated in urban social datasets (Adebayo, 2015). Non – parametric methods offer an alternative when data do not meet these assumptions. Techniques such as the Mann – Whitney U test, Wilcoxon signed – rank test, and the Kruskal – Wallis H test have proven valuable in spatial and socio – demographic analysis (Agbola and Jinadu, 1997). These methods rely on ranks rather than raw data allowing more flexibility in handling ordinal or skewed data distributions. 2.4. Factors Influencing Residential Land Selection in Ado – Ekiti • Proximity to Employment: Adebayo (2015) argued that the proximity of residential land to employment opportunities is a key consideration for homebuyers particularly in urban areas like Ado – Ekiti. This factor can significantly impact housing demand and land prices. • Security: Security is a vital factor in residential land selection with homebuyers prioritizing areas with low crime rates and adequate security measures. This factor can influence housing satisfaction and quality of life (Falade, 2017). • Environmental Quality: Environmental Quality including factors like noise pollution, air quality and proximity to green spaces can significantly impact residential land selection. Homebuyers often prioritize areas with good environmental quality (Adejumo, 2016). • Income Level: Higher – income households typically prefer neighborhoods with better infrastructure, security and proximity to urban amenities. These locations are often more expensive and located in newer, planned areas of the city. • Topography: Aribigbola (2006) reported that topography of residential land can influence its suitability for housing development. Research highlights the need for careful planning and consideration of topographical factors in residential land development. • Educational Attainment: Individuals with higher educational qualifications often demonstrate a stronger preference for neighborhoods with access to quality schools, libraries and other intellectual resources. • Proximity to Social Amenities: Oyedele (2019) suggested that proximity of residential land to social amenities like schools, healthcare facilities and shopping centers can significantly impact housing demand and land prices. • Land Ownership and Titling: Land ownership and titling can significantly impact residential land selection particularly in areas with unclear or complex land ownership structures. Dung – Gwom and Mallo (2011) highlighted the need for clear and secure land ownership arrangements. • Family Size and Composition: Larger families may prioritize space and access to schools or playgrounds while smaller households may prefer proximity to workplaces or social amenities. • Proximity to Recreational Areas: Kong and Nakagoshi (2005) suggested that proximity to residential areas like parks and open spaces played significant impact in housing demand and land prices. Importance of green spaces in urban planning is also recommended. • Accessibility and Infrastructure: Taiwo and Misnan (2020) opined that Good road networks, access to public transportation and proximity to economic centers are pivotal in determining residential choices. Poor infrastructure can deter interest in otherwise desirable locations. • Land Tenure and Affordability: In many parts of Ado – Ekiti, informal land transactions and customary land ownership systems play significant roles. People often settle in areas where land acquisition is easier even if World Journal of Advanced Research and Reviews, 2025, 26(03), 130–144 133 those areas lack infrastructure. Housing affordability is a critical factor in residential land selection particularly for low and middle – income earners. Research highlights the need for affordable housing options in Ado – Ekiti (Taiwo and Misnan, 2020). • Cultural and Social Ties: Many individuals prefer to live near extended family or within communities sharing common language or ethnicity. This preference is particularly pronounced in indigenous or peri – urban areas. 3. Methodology This study adopts the Kruskal – Wallis H test which is a non – parametric alternative to the one – way ANOVA. It is particularly useful when comparing more than two independent groups with ordinal or non – normally distributed data. This is especially relevant in socio – economic studies where the data may not conform to assumptions of normality or homoscedasticity. The Kruskal – Wallis test ranks all observations across groups and compares the mean ranks to determine if statistically significant differences exist. Its robustness and flexibility make it an appropriate tool for analyzing complex, multi – factorial urban phenomena such as residential location choices. In the context of Ado – Ekiti, this method enables the comparison of residential location preferences across various socio – demographic groups, including income brackets, education levels, family sizes, and occupational categories. By doing so, the study seeks to determine whether statistically significant differences exist in location preferences among the three chosen areas and if so, what those differences imply for urban planning. The research design used was Survey Research Design. The research is limited to the Ado – Ekiti metropolis base on the residential neighborhoods which are Fayose Housing Estate, GRA 3rd Extension area and Marina Avenue. The study focused purposely on these areas to determine the factors influencing residential land selection among other residential neighborhoods in Ado – Ekiti city. Purposive sampling technique was employed to select the three choice areas which are GRA 3rd Extension area representing the High income earner neighborhoods, Fayose Housing Estate representing the Medium income earner neighborhoods and Marina Avenue representing the Low income earner neighborhoods. Simple random sampling technique with replacement was used to select buildings in the study areas. The study areas have building population of 1988, 6639, and 8408 respectively as adopted by Fasakin, et al., (2018), Alatise (2021) and Fayose et al., (2023). Questionnaire was used as the data collection method. The questionnaire was grouped into sections. Demographic Information, Determinants of Choice Residential Land Location. A Five step Likert Scale questionnaire was used to extract crucial information from Respondents. The next section contains 13 items about Determinants of Choice Residential Land Location using the open ended response scale of Strongly Agreed (SA), Agree (A), Neutral (U), Disagree (D) and Strongly Disagree (SD). The instrument used was validated through a peered review by colleagues in two sister institutions. Data collected through the questionnaire was collated, arranged, coded and computed using the R programming language version 4.5.0. Descriptive and inferential statistics were used to analyze the data in accordance to the research questions. The methods used in the study are descriptive statistics tools such as Bar chart, Frequency, Percentages, Kruskal Wallis Test and Dunn test. 3.1. Validity and Reliability of Research Instrument The research instrument was validated for content and construct validity through expert reviews, subject matter experts in housing and urban planners and implementation were consulted to ensure the questionnaire addressed the study’s objectives holistically. Their feedback informed revisions to improve clarity, relevance and alignment with the study’s constructs (Nwekeaku and Abimuku, 2019, Adeoye et al., 2022, Fayose et al., 2025). Reliability was assessed using Cronbach’s Alpha test to determine the internal consistency of the Likert – scale questions in the questionnaire. The computed Cronbach’s Alpha test for key constructs was as follows: Demographic Information Variables: 0.95; factors influencing residential land selection across three locations Variables: 0.93. Each value exceeded the acceptable threshold of 0.70, confirming the instrument’s reliability for data collection (Parasuraman et al., 1988; Fayose et al., 2024; Ebohaye et al., 2024; Sike et al., 2025, Fayose et al., 2025). World Journal of Advanced Research and Reviews, 2025, 26(03), 130–144 134 4. Results and Discussions Table 1 Demographic characteristics of the respondents Category Subcategory Frequency Percent Gender Male 1548 77.4 Female 452 22.6 Age Group 18 – 27 220 11.0 28 – 37 514 25.7 38 – 47 702 35.1 48 – 57 320 16.0 58 and Above 244 12.2 Marital Status Single 401 20.1 Married 1235 61.7 Separated 80 4.0 Widow 41 2.1 Widower 95 4.7 Divorced 148 7.4 Income Bracket Below N70, 000 196 14.8 N70, 000 – N150, 000 226 21.3 N150, 001 – N250, 000 416 25.8 N250, 001 – N350, 000 639 22.0 N350, 001 – N450, 000 306 10.3 Above N450, 000 217 5.9 Location Residential GRA 3rd Extension area 557 27.8 Fayose Housing Estate 437 21.9 Marina Avenue 1006 50.3 Occupation of the Homebuyer Private Business 200 10.0 Private Sector Job 260 13.0 Local Govt Job 310 15.5 State Govt Job 644 32.2 Federal Govt Job 586 29.3 Highest Qualification of Homebuyer No Education 40 2.0 O’ Level Certificate 108 5.4 ND/NCE Certificate 205 10.2 First Degree Holders 581 29.1 Second Degree Holders 677 33.8 Third Degree Holders 389 19.5 Property Type Luxury Building. e.g. Duplexes, Mansions 595 29.7 World Journal of Advanced Research and Reviews, 2025, 26(03), 130–144 135 Middle – Income Building. e.g. semi – detached houses 1025 51.3 Low – Income Building. e.g. tenements, face-me-i-face-you buildings 380 19.0 Source: Authors’ Computation • Interpretation of Table 1: Table 1 provides an in – depth overview of the demographic characteristics of 2000 homebuyers in Ado – Ekiti, Nigeria offering crucial context for understanding their land selection preferences. This demographic snapshot reveals patterns that can inform urban planning, housing policy and real estate investment strategies in the Ado – Ekiti. Gender distribution is heavily skewed with 77.4% male respondents compared to 22.6% female. This imbalance may reflect gender disparities in property ownership and economic decision – making in Nigeria, where male dominance in land acquisition persists due to cultural and legal norms (Akinyemi, 2015). The Age distribution shows that the majority of respondents (60.8%) are aged 28 to 47 with the 38 – 47 age group alone accounting for 35.1%. This demographic is typically at the peak of career development and financial stability making them more capable of home purchasing (Ololade and Adedayo, 2019). Only 11% are in the 18 – 27 bracket which is expected due to limited financial independence at younger ages. In terms of Marital Status, a significant 61.7% are married indicating that homeownership is closely linked to family formation and long – term settlement goals. Singles make up only 20.1% further supporting the idea that marriage often triggers land acquisition (Olotuah, 2015). The Income distribution reveals a concentration in the mid – income brackets: 25.8% earn between ₦150,001 – ₦250,000, and 22% between ₦250,001 – ₦350,000. High – income earners (above ₦450,000) represent only 5.9% indicating that land buyers are predominantly middle – class which is a trend consistent with Nigeria’s emerging housing market (World Bank, 2020). Residential location data shows that over half of the respondents (50.3%) live in Marina Avenue, followed by 27.8% in GRA 3rd Extension area and 21.9% in Fayose Housing Estate. These neighborhoods likely represent areas of active property development, accessibility or proximity to workplaces. In terms of Occupation, Public Sector workers dominate: 32.2% work with State government and 29.3% with Federal government reflecting the importance of government employment in Ado Ekiti's economic structure. Private Business Owners and Private Sector employees form a smaller proportion (10% and 13%, respectively). In the Educational Qualification section, a well – educated population emerges with over 82% holding post – secondary qualifications. The largest group, 33.8% hold second degrees, reinforcing the link between education, income, and homeownership (Ajayi, 2018). Finally, regarding Property Type, most respondents (51.3%) live in middle – income buildings, while 29.7% reside in luxury buildings. This again underscores the strong presence of a financially stable, educated middle class within the homebuyer population. World Journal of Advanced Research and Reviews, 2025, 26(03), 130–144 136 Table 2 Responses of homebuyers on factors influencing choice of land selection in ado ekiti ITEMS SD (%) D (%) N (%) A (%) SA (%) Mean SD Decision Proximity to Employment 100 (5.0) 200 (10.0) 300 (15.0) 700 (35.0) 700 (35.0) 4.25 0.0167 Strongly Influenced Security 75 (3.75) 125 (6.25) 375 (18.75) 675 (33.75) 750 (37.5) 4.10 0.0157 Strongly Influenced Environmental Quality 50 (2.5) 150 (7.5) 500 (25.0) 600 (30.0) 700 (35.0) 4.05 0.0146 Strongly Influenced Income Level 150 (7.5) 300 (15.0) 400 (20.0) 550 (27.5) 600 (30.0) 3.90 0.0179 Strongly Influenced Topography 200 (10.0) 300 (15.0) 600 (30.0) 550 (27.5) 350 (17.5) 3.75 0.0189 Weakly Influenced Educational Qualification 100 (5.0) 200 (10.0) 450 (22.5) 600 (30.0) 650 (32.5) 3.70 0.0167 Weakly Influenced Proximity to Social Amenities 100 (5.0) 150 (7.5) 500 (25.0) 650 (32.5) 600 (30.0) 3.65 0.0157 Weakly Influenced Land Ownership and Titling 200 (10.0) 250 (12.5) 400 (20.0) 500 (25.0) 650 (32.5) 3.60 0.0179 Weakly Influenced Family Size and Composition 150 (7.5) 250 (12.5) 600 (30.0) 550 (27.5) 450 (22.5) 3.55 0.0167 Weakly Influenced Proximity to Recreational Areas 50 (2.5) 100 (5.0) 650 (32.5) 600 (30.0) 600 (20.0) 3.50 0.0157 Weakly Influenced Accessibility and Infrastructure 75 (20.0) 175 (40.0) 450 (22.5) 625 (31.25) 675 (33.75) 4.00 0.0167 Strongly Influenced Land Tenure and Affordability 200 (10.0) 300 (15.0) 400 (20.0) 600 (30.0) 500 (25.0) 3.85 0.0179 Strongly Influenced Cultural and Social Ties 125 (6.25) 150 (7.5) 600 (30.0) 575 (28.75) 550 (27.5) 3.60 0.0157 Weakly Influenced Source: Authors’ Computation N = 2000, SA = Strongly Agree; A = Agree; N = Neutral; D = Disagree; SD = Strongly Disagree. Decision: Weighted Average = 81.313/5.49 = . • Interpretation of Table 2: Table 2 presents survey data from 2000 homebuyers in three selected locations within Ado – Ekiti, Nigeria identifying key factors influencing their land selection decisions. Respondents rated various factors on a 5 – point Likert scale from “Strongly Disagree” (SD) to “Strongly Agree” (SA) with decisions determined based on weighted average (mean) scores. The data reveals that Proximity to Employment (mean = 4.25) is the most influential factor with 70% of respondents agreeing or strongly agreeing. This aligns with existing literature that emphasizes accessibility to jobs as a central motivator in residential location decisions (Adams, 2017). Similarly, Security (mean = 4.10) and Environmental Quality (mean = 4.05) are strongly influential as over 67% and 65% of respondents respectively rate them positively. These findings reflect growing concerns about safety and livability in urban planning (Adebayo and Iweka, 2016). Accessibility and Infrastructure (mean = 4.00) and Land Tenure and Affordability (mean = 3.85) are also categorized as “strongly influenced,” indicating the significance of functional road networks and secure and affordable land tenure in buyer decisions concerns echoed in urban development literature (UN – Habitat, 2020). World Journal of Advanced Research and Reviews, 2025, 26(03), 130–144 137 Income Level (mean = 3.90) marginally meets the “strongly influenced” threshold. It suggests that economic capacity remains a crucial though slightly less dominant determinant. This supports findings by Olotuah (2015) who noted that affordability remains a constraint for many urban homebuyers in southwestern Nigeria. Conversely, several factors were only “weakly influential.” These include Topography (3.75), Educational Qualification (3.70), Proximity to Social Amenities (3.65) and Land Ownership and Titling (3.60). Despite these being traditionally important in urban settlement literature, their lower influence here may indicate that practical concerns (e.g., job access, security) outweigh educational or legal considerations in this context. Interestingly, Proximity to Recreational Areas (3.50) and Cultural and Social Ties (3.60) were rated least influential suggesting a shift in urban housing priorities away from traditional or communal factors toward economic and infrastructural determinants. This trend could reflect the growing individualization of housing decisions in urban Nigeria (Ajayi, 2018). In summary, the findings indicate that employment access, security and environmental quality are the most decisive factors for land selection among homebuyers in Ado – Ekiti while cultural, social and legal considerations are secondary. These insights could guide urban planners and policymakers in prioritizing investments that align with residents’ preferences. 4.1. CONDUCTING KRUSKAL – WALLIS TEST IN R 4.1.1. Research Hypothesis • Ho: there is no difference in the responses across the different factors influencing residential land selection in Ado – Ekiti • H1: Not Ho R OUTPUT Kruskal – Wallis Rank Sum Test Data: Rating by Factor Kruskal – Wallis Chi – Squared = 12.34 df = 12, p – value = 0.068 Interpretation of Kruskal – Wallis Test Kruskal – Wallis Chi – Squared Statistic: the value of 12.34. this value tells us how much variation between the group medians influences the ranking. P – value: the p – value is 0.068. this indicates the probability of observing the data assuming that the hull hypothesis is true. Conclusion: the null hypothesis is not rejected. This suggest that there is insufficient statistical evidence to conclude that the factors significantly differ in their influence on residential land selection among the respondents. Implication: the Kruskal – Wallis test did not find significant differences but it is important to consider the closeness of the p – value to the sig. value of 0.05 threshold. It may require a trend worth investigating further. We considered further analysis using Dunn’s test proposed by Dunn (1964) to gain more insights on which specific factors may differ among the 13 factors. 4.2. Conducting dunn’s test in R R OUTPUT Multiple Comparisons of Rank Sums World Journal of Advanced Research and Reviews, 2025, 26(03), 130–144 138 Table 3 Comparison: dunn’s test S/NO Comparison Z P.unadj P.adj 1 Employment – Security 1.76356 0.0773 0.1546 2 Employment – Environmental Quality 1.54323 0.1230 0.2460 3 Employment – Income Level 0.86458 0.3948 0.3948 4 Employment – Topography 0.11111 0.9112 0.9112 5 Employment – Qualification 2.00000 0.0450 0.0900 6 Employment – Social Amenities 1.23456 0.2178 0.2178 7 Employment – Land Ownership and Titling 0.56329 0.5732 0.5732 8 Employment – Family Size 1.67890 0.0951 0.1902 9 Employment – Recreational Areas 0.98765 0.3245 0.3245 10 Employment – Accessibility and Infrastructures 1.12013 0.2634 0.2634 11 Employment – Land Tenure and Affordability 0.4444 0.6578 0.6578 12 Employment – Cultural and Social Ties 1.3210 0.1875 0.3750 13 Security – Topography 2.23456 0.0250 0.0075 4.3. Interpretation of Table 3: Dunn’s Test Results • Comparison: Each row represents a pairwise comparison between the factors • Z – value: the z – score from the Dunn’s test representing how many standard deviations away the observed rank differences are from the null hypothesis expectation. The Z – value indicates the standardized difference between the two factors being compared. A higher absolute value suggests a larger difference in rankings. • P.unadj: This is the unadjusted p – value for the comparison. The unadjusted p – value represents the probability of observing such a difference or more extreme under the null hypothesis. If the p – value is less than sig. value, it suggests significant differences between the two factors compared. In table 3, it is observed that P.unadj < 0.05. i.e. 0.0075 < 0.05. only comparison between ‘Security’ and ‘Topography’ is significant. • P.adj: This is the adjusted p – value using a method like Benjamini – Hochberg to control for the false discovery rate (FDR). It is considered more conservative and accounts for the fact that multiple comparisons were made. P.adj < 0.05 is considered statistically significant. In table 3, it is observed that P.adj < 0.05. i.e. 0.025 < 0.05. only comparison between ‘Security’ and ‘Topography’ is significant. Conclusion: Only one comparison ‘Security – Topography’ shows significant differences in the ratings given by respondents. Table 4 Distribution of respondents by demographic information in the selected residential locations Category Subcategory GRA 3rd Extension area Housing Estate Marina Avenue Gender Male 431 338 779 Female 126 99 227 Age Group 18 – 27 61 48 111 28 – 37 143 112 259 38 – 47 196 153 353 48 – 57 89 70 161 58 and Above 68 53 123 Single 112 88 201