ITPC 106: QUANTITATIVE METHODS

INTRODUCTION TO DESCRIPTIVE STATISTICS Ordinal - Variables with an inherent rank or order,

e.g. mild, moderate, severe. Can be compared for
Basics of Statistics equality, or greater or less, but not how much
Definition: Science of collection, presentation, greater or less.
analysis, and reasonable interpretation of data.
Interval - Values of the variable are ordered as in
- Statistics presents a rigorous scientific method for Ordinal, and additionally, differences between
gaining insight into data. values are meaningful, however, the scale is not
- Statistics can give an instant overall picture of absolutely anchored. Calendar dates and
data based on graphical presentation or numerical temperatures on the Fahrenheit scale are
summarization irrespective to the number of data examples. Addition and subtraction, but not
points. multiplication and division are meaningful
- Statistics make inference and predict relations of operations.
variables.
Ratio - Variables with all properties of Interval plus
A Taxonomy of Statistics an absolute, non-arbitrary zero point, e.g. age,
weight, temperature (Kelvin). Addition,
subtraction, multiplication, and division are all
meaningful operations.

Some Definitions
Distribution - (of a variable) tells us what values
the variable takes and how often it takes these
values.
- Unimodal - having a single peak
- Bimodal - having two distinct peaks
- Symmetric - left and right half are mirror images.

Statistical Description of Data Data Presentation

Statistics describes a numeric set of data by its 1. Graphical Presentation: We look for the overall
- Center pattern and for striking deviations from that
- Variability pattern. Overall pattern usually described by
- Shape shape, center, and spread of the data. An
individual value that falls outside the overall
Statistics describes a categorical set of data by pattern is called an outlier.
- Frequency, percentage or proportion of each
category Bar diagram and Pie charts are used for categorical
variables. (Lists the categories and presents the
Some Definitions percent or count of individuals who fall in each
Variable - any characteristic of an individual or category)
entity. A variable can take different values for
different individuals. Variables can be categorical Histogram(Overall pattern can be described by its
or quantitative. Per S. S. Stevens… shape, center, and spread. The following age
distribution is right skewed. The center lies
Nominal - Categorical variables with no inherent between 80 to 100. No outliers), stem and leaf and
order or ranking sequence such as names or Box-plot(Describes the five-number summary) are
classes (e.g., gender). Value may be a numerical, used for numerical variable.
but without numerical value (e.g., I, II, III). The only
operation that can be applied to Nominal variables
is enumeration.
The Shape of the Histogram  The number of peaks in the histogram
determines whether a distribution is classed as
Unimodal, Bimodal or Multimodal.
 In addition to this classification we can further
classify UniModal distributions as to whether
they are symmetric or not.
 A unimodal distribution is defined to be
Symmetric if there is a vertical line of
symmetry through the middle of the
distribution such that the distribution to the
left of this line is the mirror image of the
distribution to the right of this line.
 The right part of a unimodal distribution is
called the Upper Tail of the distribution while
the left part is called the Lower Tail:
 A Unimodal distribution which is not
symmetric is called skewed, there are two
types of skewness.
 Positive Skew: If the upper tail of the
distribution stretches out more than the lower
tail then the distribution is said to be positively
skewed.
 Negative Skew: If the Lower tail of the
distribution stretches out more than the upper
tail then the distribution is said to be
negatively skewed.

Symmetric
Distribution

Positively Skewed Negatively Skewed

Numerical Presentation
A fundamental concept in summary statistics is
that of a central value for a set of observations and
the extent to which the central value characterizes
the whole set of data. Measures of central value Which measure is the best?
such as the mean or median must be coupled with  If the distribution is symmetric then, Mean =
measures of data dispersion to indicate how well Median
the central value characterizes the data as a whole.  If the distribution is Positively Skewed (to the
right) then, Median < Mean
Methods of Center Measurement  If the distribution is Negatively Skewed (to the
(Measure of central tendency) left) then, Mean < Median
- Center measurement is a summary measure of  So the difference between the mean and
the overall level of a dataset median can be used to measure the skewness
of a dataset.
Commonly used methods are:  Note: The presence of outliers affects the
Mean: Summing up all the observation and mean but not the median.
dividing by number of observations
Median: The middle value in an ordered sequence
of observations.
Mode: The value that is observed most frequently.
Geometric mean etc.

Mean or Median?
The median is less sensitive to outliers (extreme
scores) than the mean and thus a better measure
than the mean for highly skewed distributions.

Mean or Median or Mode?

Measure of Central Tendency
Ungrouped Data
Example:

X
15
16  Arithmetic Mean =
17 164
x= =18.22
18 9
18  Median
19 18+18
x=18( ) For even
20 2
x=16 For odd
25
 Mode
164 x=18

Grouped Data
Example:

Classe f (n) M fM d <cf

s
15-19 2 17 34 -3 2
20-24 3 22 66 -2 5
25-29 5 27 135 -1 10
30-34 7 30 224 0 17
35-39 5 37 185 1 22
40-44 3 40 126 2 25
n = 25 Σ fM =770 Σ fd =6
  Mean:
770
 Mid-point Method: = 25 = 30.80
Σfd −6
 Unit-deviation/Coded Method: x=¿ x0 +( )i = 32+( )5=30.8 0
n 25

n
− cfp
 Median: Lb 2 12.5 −10
+( )i=29.5+( )5=31.29
fmd 7
d1 2
 Mode: Lb +( )i=29.5+( ) 5=32.00
d 1+d 2 2+2

Measure of Variability/Spread
Example:

x x-x (x - x)2
15 -3.22 10.3684
16 -2.22 4.9284
16 -2.22 4.9284
17 -1.22 1.4884
18 -0.22 0.0484
18 -0.22 0.0484
19 0.78 0.6084
20 1.78 3.1684
25 6.78 45.9684
2
Σ( x − x) =71.5556

R (Ungouped Data) = HV - LV = 25-15 = 10

R (Grouped Data) = UBH - LBL = 44.5 - 14.5 = 30

2
Σ( x− x)
S2 = =8.94
n− 1

S = √ S 2=2.99