Explore chapters and articles related to this topic
Spatial-Temporal Patterns of Gender Inequalities in University Enrollment in Nigeria:
Published in Esra Ozdenerol, Gender Inequalities, 2021
Moses O. Olawole, Akanni I. Akinyemi, Olayinka A. Ajala
The Global Moran I statistic (Moran 1948), a measure of spatial autocorrelation, refers to the tendency that a value of a variable at a location is correlated to the values of the same variable at nearby locations Cliff and Ord (1973, 1981). It measures spatial autocorrelation based on both feature locations and features values simultaneously. The Global Moran’s I statistic is expressed as follows: I=N∑i=1N∑j=1Nwij×∑i=1N∑j=1Nwij(Xi−X¯)(Xj−X¯)∑j=1N(Xi−X¯)2
Spatial variability of trace elements with Moran’s I Analysis for shallow groundwater quality in the Lower Katari Basin, Bolivian Altiplano
Published in Yong-Guan Zhu, Huaming Guo, Prosun Bhattacharya, Jochen Bundschuh, Arslan Ahmad, Ravi Naidu, Environmental Arsenic in a Changing World, 2019
I. Quino, O. Ramos, M. Ormachea, J. Quintanilla, P. Bhattacharya
The Moran’s I statistic was used with LISA (Local Indicators of Spatial Association) method to know the spatial autocorrelation (SA) of each element. The global spatial dependence analysis (Global Moran’s I statistiscal test), the local spatial dependence (BiLISA Cluster Map) and the significant spatial test (BiLISA Significance Map) were made in the GeoDa 1.12.01 software.
Spatial Dependence in LUTI Models
Published in Rubén Cordera, Ángel Ibeas, Luigi dell’Olio, Borja Alonso, Land Use–Transport Interaction Models, 2017
Rubén Cordera, Luigi dell’Olio
The Moran I indicator takes values between −1 and 1. An indicator value equal to 0 means a complete absence of any kind of spatial autocorrelation, a value of 1 indicates perfect positive spatial autocorrelation and a value of −1 perfect negative spatial autocorrelation.
Spatial statistics for legal process
Published in Journal of Spatial Science, 2023
Riyajun Jannat, Mohammed Al-Amin
In many cases Ripley’s K and Moran’s I seem similar. However, Moran’s I is called a spatial autocorrelation coefficient and measures the autocorrelation between two closest neighbours or spatial points. Moran’s I shows the variables on both local and global scales. On the other hand, Ripley’s K suggests how the sample points are in a space. This means Ripley’s K shows the probable pattern of distribution of the sample points, e.g. the sample points are clustered, random, or systematic in consecutive specific distances. Overall, Ripley’s K estimates the presence of a probable number of neighbours surrounding a point comparing the number of neighbours present in multiple distances or radii, whereas Moran’s I measures only the spatial autocorrelation among the sample points not the presence of the neighbours. Thus, Ripley’s K predicts the number of neighbours and Moran’s I predicts the relationship among the neighbours. In Table 2, the Moran’s I value of 0.02 or 2% means the occurrence of a crime at a single point contributes 2% in occurrence of the other crimes in neighbouring locations. In Figure 7, Ripley’s K showed the probability of the presence of similar occurrences in the multi-distance neighbourhood. It showed spatial pattern of the individual or clustered crimes which can be clearly visualised in Figure 8. The normal distribution of crime spots in Figure 8 clarifies the clustered, random, and dispersed crimes which are shown by Ripley’s K value in Figure 7 and the relationship among the points measured by Moran’s I in Table 2.
Cluster analysis of the spatial distribution of pedestrian deaths and injuries by parishes in Kampala city, Uganda
Published in International Journal of Injury Control and Safety Promotion, 2023
Esther Bayiga Zziwa, Milton Mutto, David Guwatudde
Using ArcMap, a Geographic Information System analysis software (Booth & Mitchell, 2001), choropleth maps for fatal and serious pedestrian injury rates computed earlier using MS Excel were produced using the geometric interval classification method. The distribution pattern of serious and fatal pedestrian injury rates by parish based on the choropleth maps produced was then examined to determine if there was any significant clustering pattern. The global spatial autocorrelation statistical method was used to measure the correlation among neighbouring observations, and to find the patterns and levels of spatial clustering among neighbouring parishes (Goel et al., 2018). Cluster analysis using the spatial autocorrelation (Cliff & Ord, 1973) tool was then performed to determine whether the distribution pattern of the data was random, dispersed, or clustered. The Global Moran’s I statistic was used for spatial autocorrelation. It measures spatial autocorrelation based on feature locations and attribute values using the Global Moran’s I statistic and returns five values: the Moran’s I Index, Expected Index, Variance, z-score, and p-value.
Arsenic contamination and potential health risk to primary school children through drinking water sources
Published in Human and Ecological Risk Assessment: An International Journal, 2023
Jamil Ahmed, Li Ping Wong, Najeebullah Channa, Waqas Ahmed, Yan Piaw Chua, Muhammad Zakir Shaikh
The Local Moran’s I scatter plot was assessed by analyzing spatial clusters and outliers. Scatter plot divides the data into quadrants of: (i) Low- high; (ii) High-high; (iii) Low-low; and (iv) High- low values (Figure 6). Values in High-high and Low-Low quadrants represent the clusters of high and low contaminations. At the same time, high-low and low-high values represent outlier values. Local Moran’s I require the data to be normally distributed. The cluster is also affected by the weighting distance. The zero values were removed to further test the data and identify whether the cluster's data was normalized using Box-cox transformation. The study focus was on detecting contamination hotspots and fulfilling the condition of positive values for Box-cox transformation. Raw and normalized data were analyzed for distance bands of 12 km, 24 km, 36 km, and 48 km. The present study's findings elaborate that reduction in the lower Indus basin is the controlling phenomenon in the active flood plains, which is the major source of fresh groundwater in Sindh. pH-induced dissolution is the second phenomenon that occurs only in the irrigated area, especially at the boundary of the hotspots. This contradicts prior research that claimed an increased pH state was the primary cause of arsenic release throughout the Indus basin. (Podgorski et al. 2017).