Scribd
Upload
12 views

Lecture 2 - Mean, Median and Mode

Uploaded by

kritikagarg98
Copyright
© © All Rights Reserved
We take content rights seriously. If you suspect this is your content, claim it here.
Available Formats
Download as PPTX, PDF, TXT or read online on Scribd
12 views

Lecture 2 - Mean, Median and Mode

Uploaded by

kritikagarg98
Copyright
© © All Rights Reserved
We take content rights seriously. If you suspect this is your content, claim it here.
Available Formats
Download as PPTX, PDF, TXT or read online on Scribd
You are on page 1/ 9

Centre of a circle, triangle, large data?

550 200 500 700 400 500 250 450 500


400 550 600 500 500 500 350 550 400
450 425 400 350 350 350 400 450
360 600 500 450 525 300 400 800
500 600 1000 600 400 600 400 450
650 400 300 400 450 400 375 500

Centroid, Circumcentre, Arithmetic Mean (Average), Median, Mode,


Incentre, Orthocentre .... Geometric Mean, Harmonic mean…
Central tendency measures for raw data

1. Arithmetic Mean, AM (average) = sum of the values/number of observations.


1 3 4 8 9 AM = (1+3+4+8+9)/5= 25/5= 5. AM = /N

2. Median- value of middle observation- after sorting the data in ascending order.
1 3 4 8 9 Median value= 4.
11 22 30 40 50 60 700000 Median value= 40.
1 3 4 8 9 12, Even number of observations Median value= 6 = (4+8)/2.

3. Mode- value of most frequently occurring observation in the data.


1 2 2 3 4 5 Mode value= 2.
11 22 30 40 50 50 50 50 66 66 77 Mode value= 50. 2
When to use Median?

1. If mean cannot be computed then median can be used, see the


table.
2. Median is used in 5-number summary and the Boxplot (later in this
chapter) to spot whether a frequency distribution is symmetric.
3. Median is preferred when mean is highly ‘distorted’ by few extreme
values as in the data below (Mean = Sum/N = 1566/9 = 174, a poor
measure of centre of the data).

2 5 6 7 8 9 11 18 1500
4. Distribution of wealth is highly skewed- few individuals have high
wealth, as in above data. Hence median wealth is preferred over
mean wealth. Distribution of income and salary are also skewed,
hence median income and median salary are preferred over mean
income and mean salary.
Men's 5000 meters 2016 Summer Olympics

3
When to use Mode?

 Mode- Most frequently occurring value in the data.

1 2 2 3 4 5 Unique modal value, 2.

1 2 2 3 4 5 5 6 Two modal values, 2 and 5.


1 2 3 4 5 6 No modal value.

 Usage of Mode as a measure of central tendency is


rare. It is mostly used in voting- candidate who secures
largest number of votes is declared the winner.

4
Central tendency measures of raw data
Sorted data: Spending, GradSurvey dataset
200 400 400 500 550
250 400 425 500 600
300 400 450 500 600
300 400 450 500 600
350 400 450 500 600
350 400 450 500 600
350 400 450 500 650
350 400 450 525 700
360 400 500 550 800
375 400 500 550 1000

Mean 470.7
Median 450
Mode 400 11 times

5
Measures of dispersion (What is the
spread in the data)

6
Central Tendency Measures-
Raw vs. Grouped data
Value 500 will be considered in
Raw data: Sorted Grouped data Class Interval 300-500.
200 400 400 500 550 Class Interval, CI Freq. Cum. Freq.
250 400 425 500 600 100-300 4 4
300 400 450 500 600 300-500 33 37
300 400 450 500 600
350 400 450 500 600
500-700 11 48
350 400 450 500 600 700-900 1 49
350 400 450 500 650 900-1100 1 50
350 400 450 525 700
Sum 50
360 400 500 550 800
375 400 500 550 1000

Mean 448.0
Median 427.3
11 times Mode 413.7

Notice that the differences ≠ 0.


7
Mean of Grouped data- Computation

Class Interval, CI Freq. MidPoint of CI, MP MP*Freq.


100-300 4 200 800
300-500 33 400 13200
500-700 11 600 6600
700-900 1 800 800
900-1100 1 1000 1000
Total 50 Sum 22400
Mean =
448.0
Sum/Total Freq.

Computation of Mean of Grouped data is in the syllabus. Computation


of Median and Mode of Grouped data is not in the syllabus. 8
Measuring variation Either lower than Q1-1.5 IQR, or higher than Q3 + 1.5 IQR

All the observations that have z score of less than -3 or more than +3

1. Range = Maximum-Minimum
2. Variance, population = σ2 = 1/N * ∑ (xi-Mean)2
3. Standard deviation, population =σ
4. Coefficient of Variation, population = σ/Mean
5. Variance, sample = s2 = 1/(N-1) * ∑ (xi-Mean)2
6. Standard deviation, sample =s
7. Coefficient of Variation, sample = s/Mean
8. Mean absolute deviation = 1/N * ∑ |xi-Mean|
9. Z score = [xi-Mean]/ σ
10. Quartiles (Q , Q , Q )
1 2 3 Smallest 25%, 50%, 75% observations.
11. Inter-quartile range = Q 3 – Q1
12. 5-number summary (Min., Q , Q , Q Max.)
1 2 3, Minimum, 3 Quartiles, Maximum
13. Boxplot (Called Box and Whisker chart in MS Excel) Plot of 5-number summary

You might also like