Bivariate Correlation with Spatial Data

Haining, Robert

doi:10.1111/j.1538-4632.1991.tb00235.x

Cited by 135 publications

(59 citation statements)

References 26 publications

Supporting

Mentioning

Contrasting

Order By: Relevance

“…Because the data had a non-normal distribution, we computed Spearman correlations. When spatial autocorrelation exists, the degrees of freedom in the conventional correlation tests of the significance may be incorrect, which can lead to misestimation of significance of effects [47][48][49] because spatial autocorrelation is a violation of the independence assumption. If necessary, we used the Clifford and Richardson adjustment method to account for spatial autocorrelation in the Spearman correlation coefficients, with six spatial lags used in generating correlation matrices based on the row-standardized first-order Queen contiguity weights matrix.…”

Section: Spatial Analysesmentioning

confidence: 99%

“…If necessary, we used the Clifford and Richardson adjustment method to account for spatial autocorrelation in the Spearman correlation coefficients, with six spatial lags used in generating correlation matrices based on the row-standardized first-order Queen contiguity weights matrix. 47,49 The sample size is adjusted, when using the Clifford and Richardson methodology, to account for the spatial dependence between observations. Based on the adjusted sample size, the corresponding t statistics and p values change.…”

Section: Spatial Analysesmentioning

confidence: 99%

See 1 more Smart Citation

The Geography of Recreational Open Space: Influence of Neighborhood Racial Composition and Neighborhood Poverty

et al. 2012

View full text Add to dashboard Cite

The geography of recreational open space might be inequitable in terms of minority neighborhood racial/ethnic composition and neighborhood poverty, perhaps due in part to residential segregation. This study evaluated the association between minority neighborhood racial/ethnic composition, neighborhood poverty, and recreational open space in Boston, Massachusetts (US). Across Boston census tracts, we computed percent non-Hispanic Black, percent Hispanic, and percent families in poverty as well as recreational open space density. We evaluated spatial autocorrelation in study variables and in the ordinary least squares (OLS) regression residuals via the Global Moran's I. We then computed Spearman correlations between the census tract socio-demographic characteristics and recreational open space density, including correlations adjusted for spatial autocorrelation. After this, we computed OLS regressions or spatial regressions as appropriate. Significant positive spatial autocorrelation was found for neighborhood sociodemographic characteristics (all pvalue00.001). We found marginally significant positive spatial autocorrelation in recreational open space (Global Moran's I00.082; pvalue00.053). However, we found no spatial autocorrelation in the OLS regression residuals, which indicated that spatial models were not appropriate. There was a negative correlation between census tract percent non-Hispanic Black and recreational open space density (r S 0−0.22; conventional pvalue00.005; spatially adjusted p value00.019) as well as a negative correlation between predominantly non-Hispanic Black census tracts (960 % non-Hispanic Black in a census tract) and recreational open space density (r S 0−0.23; conventional pvalue00.003; spatially adjusted p value00.007). In bivariate and multivariate OLS models, percent non-Hispanic Black in a census tract and predominantly Black census tracts were associated with decreased density of recreational open space (pvalueG0.001). Consistent with several previous studies in other geographic locales, we found that Black neighborhoods in Boston were less likely to have recreational open spaces, indicating the need for policy interventions promoting equitable access. Such interventions may contribute to reductions and disparities in obesity.

show abstract

Section: Spatial Analysesmentioning

confidence: 99%

Section: Spatial Analysesmentioning

confidence: 99%

The Geography of Recreational Open Space: Influence of Neighborhood Racial Composition and Neighborhood Poverty

et al. 2012

View full text Add to dashboard Cite

show abstract

“…To account for possible spatial dependence in both x and y a modified version of the t-test (Clifford et al, 1989) was used, which estimates the (reduced) effective sample size for spatially autocorrelated processes. The test of Clifford et al (1989) was initially developed for Pearson's correlation coefficient, but has been shown also to be valid for Spearman's ρ (Haining, 1991). The effective sample size m eff is estimated, following the notation of Dale and Fortin (2009), as…”

Section: Catchment Characteristics and Climatic Conditionsmentioning

confidence: 99%

Low-frequency variability of European runoff

Gudmundsson

Tallaksen

Stahl

et al. 2011

Hydrol. Earth Syst. Sci.

View full text Add to dashboard Cite

Abstract. This study investigates the low-frequency components of observed monthly river flow from a large number of small catchments in Europe. The low-frequency components, defined as fluctuations on time scales longer than one year, were analysed both with respect to their dominant space-time patterns as well as their contribution to the variance of monthly runoff.The analysis of observed streamflow and corresponding time series of precipitation and temperature, showed that the fraction of low-frequency variance of runoff is on average larger than, and not correlated to, the fraction of lowfrequency variance of precipitation and temperature. However, it is correlated with mean climatic conditions and is on average lowest in catchments with significant influence of snow. Furthermore, it increases (decreases) under drier (wetter) conditions -indicating that the average degree of catchment wetness may be a primary control of low-frequency runoff dynamics. The fraction of low-frequency variance of runoff is consistently lower in responsive catchments, with a high variability of daily runoff.The dominant space-time patterns of low-frequency runoff in Europe, identified using nonlinear dimension reduction, revealed that low-frequency runoff can be described with three modes, explaining together 80.6 % of the variance. The dominant mode has opposing centres of simultaneous variations in northern and southern Europe. The secondary mode features a west-east pattern and the third mode has its centre of influence in central Europe. All modes are closely related to the space-time patterns extracted from time series of precipitation and temperature.Correspondence to: L. Gudmundsson (lukas.gudmundsson@geo.uio.no) In summary, it is shown that the dynamics of lowfrequency runoff follows well known continental-scale atmospheric features, whereas the proportion of variance attributed to low-frequency fluctuations is controlled by catchment processes and varies with mean climatic conditions. The results may have implications for interpreting the impact of changes in temperature and precipitation on river-flow dynamics.

show abstract

“…We used bivariate spatial association analysis or spatial correspondence analysis (Haining, 1991) to measure relations between the intensity of TB clusters and their distances to the four types of transport infrastructures (i.e. provincial roads, national roads, highways, and railways) at the same scales as stage 1.…”

Section: Stagementioning

confidence: 99%