Setup Menus in Admin Panel

Data Interpretation and Statistical Methods

Analyze data using Spearman’s rank correlation coefficient (rho)

  • It measures the relationship between two logically related variables. Like the conventional correlation coefficient (r), Spearman rho can have any value between -1 to +1. A Spearman correlation of 1 results when the two variables being compared are monotonically related, even if their relationship is not linear. In a monotonic relationship, the variables tend to move in the same relative direction, but not necessarily at a constant rate. Monotonic relationship can be nonlinear. In a linear relationship, the variables move in the same direction at a constant rate.

CLICK HERE to refer to the diagrams of Linear, Nonlinear and monotonic relationships.

In Statistics, Correlation is largely a measure of an association between variables. In logically correlated data, the change in the magnitude of 1 variable is related to a corresponding change in the magnitude of another variable, either in the same (positive correlation: High-High, Low-Low) or in the opposite (negative correlation: Low-High or High-Low) direction. Very often, the term correlation is used in the context of a linear relationship between 2 continuous variables and expressed as Pearson product-moment correlation. For monotonically (nonlinear) distributed continuous ordinal data(indicate the order or rank of things) or for data with relevant outliers,  Spearman rank correlation can be used as a measure. Spearman rho refers to the ranked values rather than the original measurements.

Refer: IARS journal for more information

Determine how these two variables are correlated using Spearman’s rank correlation coefficient?

Variable A
(arranged in an ascending order-smallest to largest)
Rank (R1)
(assigned after arranging in order)
Variable BRank (R2)Difference in rank
(R1-R2)=d
200106.21+981
110095.12+749
190083.14+416
240074.63+416
310062.55+11
390051.57-24
420041.86-24
530030.410-749
590021.38-636
650010.59-864
Total d² = 320

Spearman rank (rho) = 1-( \frac{6 \sum d^2 }{n^3-n })

where n = number of paired observations

rho = 1-( \frac{6 \cdot 320 }{1000 - 10 })

= 1-( \frac{1920 }{990 }) = 0.939

There are (n) degrees freedom (10)

Significance Level

Compare your result to the critical values for Spearman's Rank correlation (ignoring + / -). If rho is greater than or equal to the critical value, then there is a significant correlation and the null hypothesis can be rejected.
Number of pairs of measurement
(n)
p = 0.05
(95%)
(+ or -)
p = 0.01
(99%)
(+ or -)
51
60.8861.000
70.7860.929
80.7380.881
90.6830.833
100.6480.818
110.6230.794
120.5910.780
130.5660.745
140.5450.716
150.5250.689
160.5070.666
170.4900.645
180.4760.625
190.4620.608
200.4500.591
250.4000.526
300.3640.478
350.3360.442
400.3140.413

Setup Menus in Admin Panel

Login

Create an Account Back to login/register
X