Skip to main content

Table 6 Spearman’s rank correlations between measures of the accessibility of hospitals by aggregation method

From: The approaches to measuring the potential spatial access to urban health services revisited: distance types and aggregation-error issues

Accessibility measure

Aggregation method

Accessibility measures using Euclidean distance

Accessibility measures using shortest network time (on foot)

CTCa

WDAb

WBLc

CTCa

WDAb

WBLc

Minimum distance

CTCa

–

  

–

  

WDAb

0.994

–

 

0.992

–

 

WBL1c

0.992

0.999

–

0.988

0.997

–

WBL2d

0.991

0.998

0.999

0.988

0.997

0.999

Average distance to all hospitals

CTCa

–

  

–

  

WDAb

0.999

–

 

0.999

–

 

WBL1c

0.999

1.000

–

0.999

1.000

–

WBL2d

0.999

1.000

1.000

0.999

1.000

1.000

Average distance to three closest hospitals

CTCa

–

  

–

  

WDAb

0.998

–

 

0.997

–

 

WBL1c

0.998

1.000

–

0.995

0.999

–

WBL2d

0.997

1.000

1.000

0.995

0.999

1.000

Average distance to five closest hospitals

CTCa

–

  

–

  

WDAb

0.999

–

 

0.998

–

 

WBL1c

0.998

1.000

–

0.998

1.000

–

WBL2d

0.998

1.000

1.000

0.998

0.999

1.000

Number of hospitals within 500 m or 10 min

CTCa

–

  

–

0.792

0.749

WDAb

0.753

–

 

0.792

–

0.935

WBL1c

0.716

0.940

–

0.749

0.935

–

WBL2d

0.704

0.923

0.985

0.740

0.921

0.962

Number of hospitals within 1000 m or 20 min

CTCa

–

  

–

  

WDAb

0.856

–

 

0.899

–

 

WBL1c

0.819

0.954

–

0.875

0.967

–

WBL2d

0.809

0.942

0.984

0.878

0.963

0.985

Number of hospitals within 2000 m or 30 min

CTCa

–

  

–

  

WDAb

0.956

–

 

0.950

–

 

WBL1c

0.950

0.992

–

0.940

0.983

–

WBL2d

0.949

0.991

0.998

0.935

0.982

0.994

Potential gravity model (friction parameter = 1)

CTCa

–

  

–

  

WDAb

0.995

–

 

0.996

–

 

WBL1c

0.996

0.999

–

0.995

0.999

–

WBL2d

0.995

0.999

1.000

0.995

0.999

0.999

Potential gravity model (friction parameter = 1.5)

CTCa

–

  

–

  

WDAb

0.988

–

 

0.980

–

 

WBL1c

0.989

0.998

–

0.975

0.993

–

WBL2d

0.988

0.997

0.998

0.981

0.995

0.996

Potential gravity model (friction parameter = 2)

CTCa

–

  

–

  

WDAb

0.972

–

 

0.952

–

 

WBL1c

0.979

0.990

–

0.931

0.960

–

WBL2d

0.981

0.986

0.995

0.962

0.984

0.974

Two-step floating catchment area (2000 m or 30 min)

CTCa

–

0.977

0.976

–

  

WDAb

0.977

–

0.996

0.883

–

 

WBL1c

0.976

0.996

–

0.863

0.976

–

WBL2d

0.976

0.996

0.999

0.864

0.978

0.992

Enhanced two-step floating catchment area with a slow step-decay function

CTCa

–

  

–

  

WDAb

0.918

–

 

0.908

–

 

WBL1c

0.908

0.986

–

0.897

0.976

–

WBL2d

0.908

0.986

0.997

0.889

0.972

0.991

Enhanced two-step floating catchment area with a fast step-decay function

CTCa

–

  

–

  

WDAb

0.916

–

 

0.905

–

 

WBL1c

0.906

0.986

–

0.894

0.975

–

WBL2d

0.907

0.986

0.997

0.887

0.971

0.991

Enhanced two-step floating catchment area with a gradient function

CTCa

–

  

–

  

WDAb

0.917

–

 

0.907

–

 

WBL1c

0.908

0.987

–

0.895

0.975

–

WBL2d

0.909

0.987

0.997

0.887

0.972

0.991

  1. aAggregation method based on census tract centroid (the least accurate method)
  2. bAggregation method based on the population-weighted mean of the accessibility measure for dissemination areas within census tracts
  3. cAggregation method based on the population-weighted mean of the accessibility measure for blocks within census tracts
  4. dAggregation method based on the population-weighted mean of the accessibility measure for block centroids (adjusted with the land use map) within census tracts (the most accurate method)