1.3.3
Intuition Badi Picture kya hai?Graphs data ke baare mein visual stories hote hain. Ek line graph dikhata hai ki koi cheez time ke saath kaise badlti hai (jaise har saal tumhari height badhna), jabki ek scatter plot dikhata hai ki kya do cheezein related hain (jaise "kya lambe log ke paon bhi bade hote hain?"). Sabse zaroori skill yeh hai ki graph jo keh raha hai use padh sako: trends, patterns, outliers, aur relationships.
Ek line graph data points ko line segments se jodta hai taaki continuous variable (usually time) ke across trends dikhaye. X-axis aksar independent variable (time) dikhati hai, aur y-axis dependent variable dikhati hai (jo tum measure kar rahe ho).
Purpose : Time-ordered categories mein change, growth, decline, ya patterns ko visualize karna.
Ek scatter plot (ya scatter diagram) individual data points ko dots ki tarah coordinate plane par display karta hai, jisme ek variable ek axis par hoti hai. Yeh do quantitative variables ke beech relationships ya correlations reveal karta hai.
Purpose : Identify karna ki do variables saath chalti hain (positive correlation), ulti chalti hain (negative correlation), ya koi pattern nahi dikhata (no correlation).
###1. Identify the Axes
WHY? Axes batate hain ki kya measure ho raha hai aur kaunsi units matter karti hain.
HOW:
X-axis label padho (horizontal): usually time ya ek ordered sequence hota haiY-axis label padho (vertical): jo quantity measure ho rahi haiScale note karo: Kya values evenly spaced hain? Units kya hain?
WHAT: Trend woh general direction hai jisme data ja raha hai.
WHY? Trends ka jawab dete hain: "Kya yeh upar ja raha hai? Neeche? Flat reh raha hai? Direction badal raha hai?"
HOW:
Increasing trend : Line upar ki taraf slope karti hai (x badhne par y-values badhti hain)Decreasing trend : Line neeche ki taraf slope karti hai (y-values girती hain)Constant : Horizontal line (koi change nahi)Fluctuating : Line baar baar upar-neeche jaati hai
Worked example Worked Example 2: 8 Weeks mein Plant ki HeightWeek 1: 2 cm, Week 2: 4 cm, Week 3: 7 cm, Week 4: 11 cm, .., Week 8: 35 cm
Step 1 : Points plot karo (1, 2), (2, 4), (3, 7), (4, 11), ..., (8, 35)
Step 2 : Line segments se connect karo
Observation : Line consistently upar slope karti hai → increasing trend (plant grow kar rahi hai)
Rate of change : Week 1 se 4 tak, height 3 weeks mein 9 cm badhi = 3 cm/week. Week 4 se 8 tak, 4 weeks mein 24 cm badhi = 6 cm/week → growth rate accelerate ho rahi hai .
WHY this step? Rate calculate karna quantify karta hai ki change kitni tezi se ho raha hai, na sirf ki ho raha hai.
Worked example Worked Example 3: 10 Dino mein Stock PriceDay 1: ₹50, Day 2: ₹52, Day 3: ₹48, .., Day 7: ₹62 (peak), .., Day 10: ₹55
Maximum : Day 7 par ₹62
Minimum : Day 3 par ₹48
Range : 62 − 48 = ₹14
Interpretation : Stock us period mein ₹14 fluctuate hua. Day 7 par peak shayad us din ki achhi news indicate karta hai, jiske baad correction aaya.
WHY? Har dot ek observation hai jisme do measurements hain.
HOW: Agar (hours studied, test score) plot kar rahe ho, toh point (3, 70) matlab ek student ne 3 ghante padha aur 70% score kiya.
Worked example Worked Example 4: Hours Studied vs. Test ScoreData: (1, 45), (2, 55.5, 60), (3, 70), (4, 75), (4.5, 80), (5, 85)
Har point : Ek student ka data
x-coordinate: hours studied
y-coordinate: test score (%)
Jo student 4 ghante padha woh point (4, 75) hai, matlab usne 75% score kiya.
Definition Correlation Types
Positive correlation : Jaise x badhta hai, y bhi badhta hai (dots ka upward trend)Negative correlation : Jaise x badhta hai, y ghadta hai (downward trend)No correlation : Dots randomly scattered, koi clear pattern nahiStrong vs. Weak : Strong = dots ek line ke paas; weak = dots phele hue hain
WHY correlation matter karta hai? Yeh batata hai ki do variables related hain ya nahi aur ek se doosre ko predict karne mein help karta hai.
Worked example Worked Example 5: Ice Cream Sales vs. TemperatureDaily temperature (°C) vs. ice cream sales (₹) ka scatter plot:
(25, 500), (28, 650), (30, 800), (22, 400), (26, 550), (31, 900), (20, 300)
Plotting : Temperature x-axis par, sales y-axis par
Pattern : Jaise temperature badhti hai, sales badhti hai → positive correlation
Strength : Points roughly ek upward line follow karte hain → moderately strong
Interpretation : Garam din → zyada ice cream sales. Hum predict kar sakte hain ki 27°C ke din sales lagbhag ₹600 hogi interpolation karke.
WHY this step? Correlation pehchanna humein predictions karne aur cause-effect relationships (ya kam se kam associations) samajhne deta hai.
Ek outlier woh data point hai jo overall pattern mein fit nahi hota — doosre points se bahut door hota hai.
WHY outliers identify karein? Yeh errors, special cases, ya interesting exceptions represent kar sakte hain jo investigate karne layak hain.
Worked example Worked Example 6: Age vs. Reaction TimeScatter plot: (20, 0.25), (25, 0.26), (30, 0.28), (35, 0.30), (40, 0.33), (45, 0.35), (50, 0.40), (55, 0.45), (30, 0.60)
Pattern : Zyatar points umar ke saath reaction time (seconds) mein thodi badhot dikhate hain
Outlier : (30, 0.60) — ek 30-saal ka insaan jiska reaction time 0.60 sec hai, us umar ke doosron se bahut slow
Possible reasons : Thakaan, distraction, medical condition, measurement error
Action : Investigate karo ya agar error confirm ho toh remove karo
Abhi calculate karna zaroori nahi, lekin scatter plot mein ek line of best fit (trend line) average relationship dikhati hai.
WHY? Yeh diye gaye x ke liye predict karne deti hai.
HOW (conceptual): Ek line kheencho taaki zyatar points uske paas hon, aur roughly barabar numbers uske upar aur neeche hon.
y = m x + c
jahan m c
Interpretation ka derivation : Agar "hours studied vs. score" mein m = 2
"Tuesday ko temperature kya thi?" → X-axis par Tuesday dhundho, line tak trace karo, y-value padho.
"Month 1 se Month 3 tak sales kitni badhi?" → Subtract karo: Month 3 value − Month 1 value
"Kis period mein growth tez hui?" → Alag-alag intervals ke liye Δ x Δ y
"Agar x = 4 aur x = 5 par data hai toh x = 4.5 par value estimate karo." → Dono points ke beech mentally ek line draw karo; x = 4.5 par y-value dono ke beech halfway hogi.
"Kya scatter plot koi relationship dikhata hai?" → Trend dekho; agar dots ek pattern banate hain (upward, downward), toh haan.
Common mistake Common Error #1: Correlation ko Causation se Confuse KarnaGalat idea : "Ice cream sales aur drowning deaths correlated hain, toh ice cream drowning cause karta hai!"
Kyun sahi lagta hai : Jab do cheezein saath hoti hain, hum instinctively sochte hain ek doosre ko cause karta hai.
Fix : Correlation matlab variables saath chalte hain , yeh nahi ki ek doosre ko cause karta hai . Yahan dono ek teesre factor se cause hote hain: hot weather (zyada swimming + zyada ice cream). Hamesha poochho: "Kya koi hidden third variable ho sakta hai?"
Common mistake Common Error #2: Scale ko Ignore KarnaGalat idea : Ek line graph dekhna jahan y-axis 0 ki jagah 90 se start hoti hai, ek chhota sa rise dramatic lagta hai.
Kyun sahi lagta hai : Humara brain actual numbers ki jagah visual slope par focus karta hai.
Fix : Hamesha y-axis range check karo. 90-100 scale par 95 se 97 ka change sirf +2 hai, lekin agar graph zoomed in hai toh bahut bada lagta hai .
Common mistake Common Error #3: Data se Bahar Extrapolate KarnaGalat idea : Agar ek plant 8 weeks tak 5 cm/week badhti hai, toh ek saal mein (52 weeks) 260 cm tall hogi.
Kyun sahi lagta hai : Observed data mein trend consistent tha, toh continue kyun nahi?
Fix : Trends aksar observed range ke bahar badal jaate hain (plant ek growth limit hit karti hai). Sirf data range ke andar ya thoda sa bahar predict karo (interpolation extrapolation se safer hai).
Recall Ek 12-Saal ke Bacche ko Explain KaroSoch lo tum apne video game scores ek hafte track kar rahe ho. Tum har din ka score likhte ho: Monday = 200, Tuesday = 250, Wednesday = 300, aur aage. Agar tum un points ko lines se connect karo, tumhe ek line graph milta hai. Tum ek nazar mein dekh sakte ho: "Oh, main har din better ho raha hoon!" ya "Hmm, Thursday ko mera bura din tha."
Ab soch lo tum wonder karte ho: "Kya mere class ke lambe bacche tez daurte hain?" Tum sabki height aur 100-meter time measure karte ho, phir har bacche ke liye ek dot banate ho (height ek taraf, time doosri taraf). Yahi scatter plot hai. Agar dots neeche-daayein jaate hain (lambe → tez), toh height aur speed connected hain. Agar dots sab jagah hain, toh koi connection nahi.
Graphs tumhare data ki tasveer ki tarah hain — yeh tumhe boring numbers ki table ghoore bina patterns spot karne dete hain!
Mnemonic LINE = "Look for Increase, Notice Extremes"LINE graph padhte waqt:
L ook at the axes and labels first (pehle axes aur labels dekho)I ncrease, decrease, or flat? Identify the trend (badhna, ghadna, ya flat? Trend identify karo)N ote the maximum, minimum, and range (maximum, minimum, aur range note karo)E xplain what the pattern means in context (context mein pattern ka matlab explain karo)
SCATTER plots ke liye: "S pot the C orrelation, A lert for T rends, T est for E xceptions, R emember no causation!"
#flashcards/maths
Line graph ka primary purpose kya hai? :: Dikhana ki ek variable time ke saath ya ek ordered sequence mein kaise badlta hai (trends aur patterns visualize karna).
Scatter plot par har point kya represent karta hai? Do measurements ke saath ek observation — har axis ek variable se correspond karti hai.
Scatter plot mein positive correlation define karo :: Jaise ek variable badhta hai, doosra variable bhi badhta hai (dots ka upward pattern).
Line graph se range kaise calculate karte hain? Range = Maximum value − Minimum value.
Scatter plot mein outlier kya hota hai? Ek data point jo doosre points ke overall pattern se door hota hai, shayad error ya special case ki wajah se.
Kyun correlation ko causation se confuse nahi karna chahiye? Correlation matlab do variables saath chalte hain, lekin ek necessarily doosre ko cause nahi karta; ek third factor dono ko affect kar sakta hai.
Line graph padhte waqt, line ki slope kya indicate karti hai? Slope rate of change indicate karti hai — steeper slope matlab tez change.
Interpolation vs. extrapolation kya hai? Interpolation data range ke andar estimate karna hai; extrapolation us se bahar predict karna hai (zyada risky, kyunki trends badal sakti hain).
Line graph mein decreasing trend kaise identify karte hain? :: Line left se right ki taraf neeche slope karti hai (x badhne par y-values ghadti hain).
Scatter plot mein "no correlation" ka kya matlab hai? Dots randomly scattered hain jisme koi discernible pattern nahi; do variables related nahi hain.
Graphs as visual data stories
Relationship between two variables
Trends: up, down, flat, fluctuating
