Skip to main content

Spearman's Rank Correlation — Previous Year Questions

📊 Topic Overview​

MetricValue
Papers Appeared19 of 27 (70.4%)
Priority🔴 VERY HIGH
Section DistributionSec A: ~14
Most Repeated"Compute Spearman's Rho for the following data"
ProbabilityVery High — primary Section A computation in ~2 of every 3 papers

🎯 Must-Know Concepts​

  1. Spearman's rho (ρ) is a non-parametric measure of rank correlation
  2. Formula: rho = 1 − [6Σd²/ n(n²−1)]
  3. d = difference between ranks for each pair
  4. Range: −1 to +1 (same as Pearson's r)
  5. Used when data is ordinal, or Pearson's r assumptions are violated
  6. Tied ranks: assign average of tied ranks

🔥 Core Computation Procedure​

Step-by-Step Spearman's Rho​

Step 1: Assign ranks to each variable Rank the scores for Variable X from 1 (lowest) to n (highest). Do the same for Variable Y. For tied scores, assign the average rank.

Step 2: Compute d (rank differences) For each pair, d = Rank of X − Rank of Y

Step 3: Compute d² Square each d value.

Step 4: Sum the d² values ΣD² = sum of all squared rank differences

Step 5: Apply the formula ρ=1−6ΣD2n(n2−1)\rho = 1 - \frac{6\Sigma D^2}{n(n^2 - 1)}

Where n = number of pairs.

Computation Table:

SubjectXYRank XRank YD (RX−RY)D²
1
2
...
ΣD² =

Step 6: Interpret

  • rho = +1: Perfect positive rank correlation
  • rho = −1: Perfect negative rank correlation
  • rho = 0: No rank correlation
  • Compare with critical r at df = n − 2

📝 Conceptual Questions​

Spearman's Rho vs Pearson's r (Section B — ~250 words)​

AspectSpearman's Rho (ρ)Pearson's r
TypeNon-parametricParametric
Data scaleOrdinal (ranked data)Interval or ratio
DistributionNo normality assumptionAssumes normality
Based onRank differencesActual values
Formula1 − 6ΣD²/n(n²−1)Σ(X−X̄)(Y−Ȳ)/nSDxSDy
OutliersLess sensitiveSensitive
UseWhen ranks are natural, or data non-normalWhen interval/ratio data, normal distribution

When to use Spearman's rho:

  1. Data measured on ordinal scale (e.g., rankings, ratings)
  2. Data does not meet normality assumptions
  3. Small samples
  4. Presence of outliers that would distort Pearson's r

Often questions are phrased as: "Define outliers. Compute Spearman's Rho for the following data." — prepare a definition of outliers: extreme values that deviate markedly from other data points, which can distort correlation coefficients.

Linear and Non-linear Relationship (Section A pairing — ~250 words)​

Questions often pair Spearman's Rho computation with "Explain linear and non-linear relationship":

Linear relationship: When scores on two variables change at a constant rate — a straight line fits the scatter plot. The relationship can be positive (both increase together) or negative (one increases as the other decreases).

Non-linear relationship: When the relationship between variables is not constant — the scatter plot follows a curved pattern. Pearson's r is not appropriate for non-linear relationships; Spearman's rho may better capture monotonic non-linear relationships.

🧠 Memory Aid​

Spearman Steps (RRDS-F):

  • Rank X variable
  • Rank Y variable
  • Difference (d = Rank X − Rank Y) and D²
  • Sum the D² values
  • Formula: rho = 1 − 6ΣD²/n(n²−1)

📚 Related Content​

✅ Self-Assessment​

  1. Write the formula for Spearman's rho.
  2. What is the value of n(n²−1) when n = 8?
  3. If ΣD² = 24 and n = 6, what is rho?
  4. When would you choose Spearman's rho over Pearson's r?
  5. How do you handle tied ranks in Spearman's rho computation?

💡 Exam Tips​

  • Show the full table with columns: Subject, X, Y, Rank X, Rank Y, D, D².
  • Verify: the sum of ranks for X should equal n(n+1)/2.
  • For tied ranks: both scores get the average of the two ranks (e.g., if 2 scores tie for ranks 3 and 4, both get rank 3.5).
  • Questions often come with a theory part ("explain outliers" or "linear vs non-linear") before the computation — prepare both.