The Scatter Plot Solving Algorithm

Summary

The Scatter Plot Solving Algorithm

Final Checklist: The 6-Step Approach

1

Decode the Axes

Identify X and Y variables, units, and what a single point represents.

2

Assess the Global Trend

Determine Direction, Form, and Strength.

3

Identify Subgroups

Look for clusters or distinct colors/shapes. Analyze each independently.

4

Spot Anomalies

Identify outliers and determine if they skew the overall trend line.

5

Compare (If Applicable)

Compare slopes and intercepts of different trend lines or subgroups.

6

Execute with Constraints

Interpolate, don't extrapolate. Avoid causal assumptions unless justified.

The Time Series Solving Algorithm

Summary

The Time Series Solving Algorithm

Final Checklist: The 6-Step Approach

1

Decode the Axes and Units

Identify the X and Y axes. Check if the scale is in absolute numbers, thousands, or millions.

2

Identify the Core Trend

Look past the short-term noise. Is the long-term direction upward, downward, or stationary?

3

Choose the Right Metric

To compare volume of growth use Absolute Difference. To compare rate of growth use Percentage Change or CAGR.

4

Beware the Base Effect

If a percentage growth is exceptionally high, immediately check the initial base value to see if it is an illusion.

5

Use Visual Shortcuts

For "maximum absolute growth", find the steepest line segment. For "overtaking", find the exact intersection.

6

Apply Logical Constraints

Never assume a past exponential trend continues forever. Look for market saturation or capacity limits.

The Candlestick Decoding Algorithm

Final Checklist: The 5-Step Approach

1

Verify the Legend:

Confirm which color or shading means Bullish (Close > Open) and which means Bearish (Close < Open).

2

Extract the Extremes (Wicks):

Identify the very top of the upper wick $\to$ High. Identify the very bottom of the lower wick $\to$ Low.

3

Map the Core (Body):

If Bullish: Bottom edge = Open, Top edge = Close. If Bearish: Top edge = Open, Bottom edge = Close.

4

Compute the Required Metric:

For Volatility / Range: $High - Low$ . For Absolute Change: $Close - Open$ . For Percentage Change: $(Close - Open) / Open$ .

5

Execute Visual Scans:

Tallest total vertical length $\to$ Maximum Volatility. Largest body relative to its starting base $\to$ Maximum Percentage Change.

The Multi-Graph Decoding Checklist

Final Checklist for Relationship and Well-being Sets:

1

Map the Entities:

Ensure you know exactly which dot or bar in Graph 2 corresponds to which entity in Graph 1.

2

Verify the Axes:

Check if the X and Y variables change between graphs. Note the units and scales.

3

Identify the Shape:

Is the relationship linear, a plateau, or a threshold? This dictates how you calculate marginal changes.

4

Distinguish the Metrics:

Never mix up life satisfaction (cognitive) with momentary affect (emotional).

5

Watch the Language:

Reject causal conclusions. Stick to correlational observations.

6

Hunt for Outliers:

The most interesting exam questions usually revolve around the data points that break the general trend.

Graphs, Scatter Plots and Trend Analysis: Solved Questions with Step-by-Step Explanations (2 Problems)

Question 1 · Quantitative Aptitude and Data Interpretation (QA & DI) MCQ

Refer to the hypothetical dataset describing 12 regions. Graph A plots "Per Capita Healthcare Spend ( $) " o n X - a x i s v s " L i f e E x p ec t an cy (Y e a r s) " o nY - a x i s . G r a p h B pl o t s " P er C a p i t a H e a l t h c a r e S p e n d ($ )" on X-axis vs "Patient Satisfaction Score (0-10)" on Y-axis.

Observations:

In Graph A, points form a steep positive curve for spends up to $3, 000, t h e n f l a tt e n co m pl e t e l y in t o ah or i z o n t a l l in e f or s p e n d s >$ 3,000.
In Graph B, points show a steady linear upward trend across the entire spend range ( $500 t o$ 8,000).

Based ONLY on these graphs, which inference is MOST valid?

A.
Increasing healthcare spend beyond $3,000 yields no additional benefit to patients.
B.
Life expectancy is biologically capped, whereas patient satisfaction is subjective and unbounded by physiological limits.
C.
For regions spending > $3,000, marginal gains in life expectancy are negligible, but marginal gains in satisfaction persist.
D.
Regions with highest satisfaction scores necessarily have the highest life expectancy.

Correct Answer:

C

Step-by-Step Solution

Key idea: This is a multi-graph synthesis question. It tests the ability to reconcile conflicting/divergent trends across two related visualizations and avoid over-generalization.

Step 1: Interpret Graph A (Life Expectancy).

Shape: Logarithmic/Saturation. Steep rise then plateau.

Meaning: Marginal utility of spend on LE drops to zero after $3k.

Step 2: Interpret Graph B (Satisfaction).

Shape: Linear. Constant positive slope.

Meaning: Marginal utility of spend on Sat remains constant/positive even at high spend.

Step 3: Synthesize.

At spend > $3 k :$ \Delta LE \approx 0 $, b u t$ \Delta Sat > 0$.

This directly supports Option C.

Step 4: Reject distractors.

A: False. Says "no additional benefit." Ignores Satisfaction benefit shown in Graph B.

B: Invalid. Introduces external biological/psychological theory not present in data. Data shows correlation, not cause/mechanism.

D: False. Correlation between Sat and LE is not established. High spend regions have high Sat but flat LE. They might have same LE as mid-spend regions.

Answer: C

Question 2 · Quantitative Aptitude and Data Interpretation (QA & DI) NAT

A scatter plot consists of 10 data points. The summary statistics for these 10 points are:

$\sum X = 50$ , $\sum Y = 110$ , $\sum X^{2} = 310$ , $\sum X Y = 730$ .

The line of best fit (OLS) for these 10 points is $Y = 2 X + 1$ .

One point is identified as an outlier and removed. The new line of best fit for the remaining 9 points is $Y = 3 X - 2$ .

What is the $Y$ -coordinate of the removed outlier point?

Correct Answer:

-7

Step-by-Step Solution

Key idea: This is a reverse engineering question using the algebraic properties of OLS summary statistics.

Step 1: Set up the equations for the means

Let the removed point be $(x_{0}, y_{0})$ .

For the 10 points, $\overset{ˉ}{X} = 5, \overset{ˉ}{Y} = 11$ .

For the 9 points, the new means $\overset{ˉ}{X}^{'}, \overset{ˉ}{Y}^{'}$ must lie on the new line $Y = 3 X - 2$ .

$\overset{ˉ}{X}^{'} = \frac{50 - x _{0}}{9}$ , $\overset{ˉ}{Y}^{'} = \frac{110 - y _{0}}{9}$ .

Substituting into the new line equation:

$\frac{110 - y _{0}}{9} = 3 (\frac{50 - x _{0}}{9}) - 2$

$110 - y_{0} = 150 - 3 x_{0} - 18$

$110 - y_{0} = 132 - 3 x_{0} ⟹ y_{0} = 3 x_{0} - 22$ .

Step 2: Use the slope condition for the 9 points

The slope for the 9 points is $m^{'} = \frac{S _{X Y}^{'}}{S _{X X}^{'}} = 3$ .

$S_{X X}^{'} = \sum_{9} X^{2} - \frac{( \sum _{9} X ) ^{2}}{9} = (310 - x_{0}^{2}) - \frac{( 50 - x _{0} ) ^{2}}{9}$ .

$S_{X Y}^{'} = \sum_{9} X Y - \frac{( \sum _{9} X ) ( \sum _{9} Y )}{9} = (730 - x_{0} y_{0}) - \frac{( 50 - x _{0} ) ( 110 - y _{0} )}{9}$ .

Step 3: Simplify and solve for $x_{0}$

Substitute $110 - y_{0} = 132 - 3 x_{0}$ :

$S_{X X}^{'} = \frac{2790 - 9 x _{0}^{2} - ( 2500 - 100 x _{0} + x _{0}^{2} )}{9} = \frac{290 + 100 x _{0} - 10 x _{0}^{2}}{9}$ .

$S_{X Y}^{'} = 730 - x_{0} (3 x_{0} - 22) - \frac{( 50 - x _{0} ) ( 132 - 3 x _{0} )}{9}$

$= \frac{6570 - 27 x _{0}^{2} + 198 x _{0} - ( 6600 - 282 x _{0} + 3 x _{0}^{2} )}{9}$

$= \frac{- 30 + 480 x _{0} - 30 x _{0}^{2}}{9}$ .

We need $S_{X Y}^{'} = 3 S_{X X}^{'}$ :

$- 30 + 480 x_{0} - 30 x_{0}^{2} = 3 (290 + 100 x_{0} - 10 x_{0}^{2})$

$- 30 + 480 x_{0} - 30 x_{0}^{2} = 870 + 300 x_{0} - 30 x_{0}^{2}$ .

The quadratic terms cancel out perfectly!

$180 x_{0} = 900 ⟹ x_{0} = 5$ .

Step 4: Find $y_{0}$

$y_{0} = 3 (5) - 22 = 15 - 22 = - 7$ .

Answer: -7

Graphs, Scatter Plots and Trend Analysis Short Notes for XAT: Concepts, Formulas, Worked Examples & Practice

The Scatter Plot Solving Algorithm

The Scatter Plot Solving Algorithm

Decode the Axes

Assess the Global Trend

Identify Subgroups

Spot Anomalies

Compare (If Applicable)

Execute with Constraints

The Time Series Solving Algorithm

The Time Series Solving Algorithm

Decode the Axes and Units

Identify the Core Trend

Choose the Right Metric

Beware the Base Effect

Use Visual Shortcuts

Apply Logical Constraints

The Candlestick Decoding Algorithm

The Candlestick Decoding Algorithm

The Multi-Graph Decoding Checklist

The Multi-Graph Decoding Checklist

Graphs, Scatter Plots and Trend Analysis: Solved Questions with Step-by-Step Explanations (2 Problems)

Step-by-Step Solution

Step-by-Step Solution

More short notes in this unit

A question from this chapter

Login to view the complete short notes

Personalised first. High quality throughout.

Built around you, not around a syllabus PDF

Revision that hits your weak spots

Questions calibrated to the real exam

Notes written for recall, not for volume

One place for everything

Honest progress

Unlock the whole course