📈

Research Methodology - 11
MODULE XI: DATA ANALYSIS

A. MEASURES OF CENTRAL TENDENCY

1. Mean (Arithmetic Average)

Definition: Sum of all observations divided by the number of observations.

Formula: xˉ=1n∑i=1nxi\bar{x} = \tfrac{1}{n}\sum_{i=1}^n x_i

Example: Monthly sales (₹ 000s) for 5 months: 12, 15, 18, 20, 25

Sum = 12+15+18+20+25 = 90

Mean = 90 ÷ 5 = 18

Key Point: Sensitive to extreme values.

2. Median (Middle Value)

Definition: The middle observation when data are ordered. If n is even, it’s the average of the two central values.

Example (Odd n): Sales above → ordered [12, 15, 18, 20, 25], median = 18 (3ᵈ value).

Example (Even n): Add one more month, say 30 → [12, 15, 18, 20, 25, 30], median = (18+20)/2 = 19

Key Point: Unaffected by outliers; best for skewed distributions.

3. Mode (Most Frequent Value)

Definition: Observation that occurs most often. A distribution may be unimodal, bimodal, or multimodal.

Example: Customer visits per week: [2, 3, 3, 4, 5, 3, 4] → 3 occurs 3 times → mode = 3

Key Point: Only measure defined for nominal data; may not exist (all unique) or not be unique (multiple modes).

B. MEASURES OF DISPERSION
Quantify the spread or variability in the data.

1. Range

Definition: Difference between the maximum and minimum values.

Formula: Range = xmax⁡–xmin⁡x_{\max} – x_{\min}

Example (Sales data): Max = 25, Min = 12 → range = 13

Research Methodology - 11 1
 Key Point: Simple but highly sensitive to outliers; ignores intermediate values.

2. Standard Deviation (SD)

Definition: Average distance of observations from the mean.

Formula:
s=1n−1∑i=1n(xi−xˉ)2 s = \sqrt{\frac{1}{n-1}\sum_{i=1}^n (x_i - \bar{x})^2}

Example: Sales [12, 15, 18, 20, 25], mean=18

1. Compute deviations: –6, –3, 0, +2, +7

2. Square them: 36, 9, 0, 4, 49 → sum = 98

3. Divide by n–1 = 4: 98/4 = 24.5

4. SD = √24.5 ≈ 4.95

Key Point: Uses all data; common basis for further analysis (e.g., z-scores).

C. CONCEPT OF CORRELATION & REGRESSION

1. Correlation

Purpose: Measures the strength and direction of the linear relationship between two variables X and Y.

Pearson’s r formula:
r=∑(xi−xˉ)(yi−yˉ)∑(xi−xˉ)2 ∑(yi−yˉ)2 r = \frac{\sum (x_i - \bar{x})(y_i - \bar{y})}{\sqrt{\sum(x_i - \bar{x})^2
\;\sum(y_i - \bar{y})^2}}

Example: Hours studied vs. exam score for 5 students:

X: [2, 4, 6, 8, 10] Y: [50, 60, 70, 80, 90]

Perfect linear ↑ → r = +1

Interpretation:

r ≈ +1 strong positive

r ≈ 0 no linear relationship

r ≈ –1 strong negative

2. Simple Linear Regression

Purpose: Predict Y from X via the line Y=a+bXY = a + bX.

Slopes & Intercept:

b=∑(xi−xˉ)(yi−yˉ)∑(xi−xˉ)2,a=yˉ−b xˉ b = \frac{\sum (x_i - \bar{x})(y_i - \bar{y})}{\sum(x_i - \bar{x})^2},

Example (continued):

With perfectly linear data above, b = 4, a = 42

Regression equation: Score = 42 + 4 × (Hours studied)

Predict: study 5 hrs → score = 42 + 4×5 = 62

Key Points:

b = change in Y per one‐unit change in X.

a = predicted Y when X = 0.

R² (coefficient of determination) = r² measures % variance in Y explained by X.

INTEGRATED EXAMPLE

A small bakery tracks daily advertising spend (₹000s) and daily sales (₹000s) over 7 days:

Day Ad Spend (X) Sales (Y)

1 2 20

Research Methodology - 11 2
 2 3 30

3 5 45

4 4 40

5 6 60

6 7 70

7 8 80

Mean X = 5; Mean Y = 50

Range sales = 80 – 20 = 60

SD (X) ≈ 2.16; SD (Y) ≈ 21.6

r ≈ +0.99 (very strong positive)

Regression line:

b = Cov(X,Y)/Var(X) = (Σ (x−5)(y−50)) / Σ(x−5)² = (280) / (28) = 10

a = 50 – 10×5 = 0

Y = 10 X (each ₹1 000 ad spend yields ₹10 000 sales)

Research Methodology - 11 3