You are on page 1of 3

Statistics 1 Chapter 2 Revision

Data Types
There are two types of data:
1. Qualitative data: non-numeric data. (Favorite color, favorite animal, etc)
2. Quantitative data: numeric data.
- Continuous data: can take on any value within a range (weight, height, etc)
- Discrete data: can only take on specific values within a range (number of children in a family)

Grouped Frequency Tables


Finding class boundaries
• If there are gaps between class ranges, class boundaries lies halfway between two ranges.
• If there are NO gaps between class ranges, class boundaries are the same as class ranges.

Finding class width


If the class range is a ⩽ x < b, class width is b − a.

Finding class midpoint


a +b
If the class range is a ⩽ x < b, class midpoint is .
2

Measures of Central Tendency (Averages)


Mean
• Denoted as x̄.
∑x
• The formula is x̄ = , where n is the number of items in the dataset.
n
∑ fx
• For frequency tables, x̄ = , where f denotes frequency.
∑f
∑mf
• For grouped frequency tables, x̄ = ∑ f , where m is the midpoint of each class.

• If set A, of size n1, has mean x̄1 and set B, of size n 2, has mean x̄ 2 then the mean of the combined set of A
n1 x̄1 + n 2 x̄ 2
and B is x̄ =
n1 + n 2
Median
• First find the rank by divide the number of items with 2.
n
• If is a whole number, take nth and (n + 1)th values, and add them and divide by 2 to get median.
2
• If n is not a whole number, round up and take the nth value to get median.

Mode
• Mode is the most frequently occurring data in the dataset.
Other Measures of Location (Quartiles and Percentiles)
Quartiles
• Lower Quartile (Q1) is one-quarter of the way through the dataset.
• Median (Q2) is halfway through the dataset.
• Upper Quartile (Q3) is three-quarter of the way through the dataset.

To find rank r of Quartiles,


1 1 3
r1 → n r2 → n r3 → n
4 2 4

• If r is a whole number, take r-th and (r + 1)-th values, and add them and divide by 2 to get the
corresponding quartile.
• If r is not a whole number, round up and take the r-th value to get median.

Percentiles
• Percentiles split the data into 100 parts.
• The 10-th percentile lies one-tenth of the way through the dataset.

Quartiles and percentiles in grouped frequency table (Linear Interpolation)


1. Find the rank r.
2. Find the class in which r falls into.

Rank rbegin r rend

Value vbegin v vend

r − rbegin v − vbegin
=
rend − rbegin vend − vbegin

Where,
• v is the percentile or the quartile we want, and r is the rank.
• rbegin and rend are the class boundaries (first column).
• vbegin and vend are the cumulative frequencies of the classes (second column).

Assumption made in using linear interpolation


• Linear interpolation gives an estimate of quartiles and percentiles, not the actual values.
• Linear interpolation assumes that the data are evenly distributed.
Measures of Spread (Measure of dispersion)
Range, IQR and IPR
• Range is the difference between the smallest and the highest values in the dataset.
• Interquartile range (IQR) is the difference between upper and lower quartile.
• Range is effected by extreme values.
• IQR is not effected by extreme values.
• Interpercentile range is the difference between two given percentiles.

Variance and standard deviation


2
∑ (x − x̄ )2 ∑ x2
( n )
2
Sx x ∑x
Variance (σ ) = = = −
n n n
2
∑ (x − x̄ )2 ∑ x2
( n )
Sx x ∑x
Standard deviation (σ) = = = −
n n n

(x − x̄ )2 is known as summary statistic.



Where, Sx x =

Choosing the Average and Measure of Spread


• If the data is qualitative, use mode.
• If the data is quantitative and skewed, use median and IQR, because they are not effected much by outliers.
• If the data is quantitative and symmetric, both mean and σ or median and IQR can be used.

Coding
• Coding is a way of simplifying statistical calculation.
x −a
• A frequently used coding method is to transform each value x into y = . Then
b
x̄ = b ȳ + a
σx = bσy

Comment on the…
When the question asks to comment on the data between two values (for example, “comment on the mean
between two datasets),
1. Say which one is higher,
2. State what that means.

You might also like