Six Sigma MEASURE phase

Data Types and

Parameters
TYPES OF DATA
Types of Data

Descriptive
Orientation “Gappiness”

Response Discrete
(dependant variable) (obtained bycounting)

Predictor Continuous
(independent variable) (obtained bymeasuring)
TYPES OF DATA
Whether a variable is to be described or be described
Types of Data by other variables:

Response (y) (or DependentVariable)

➢ Variable under investigation is described in terms of
Descriptive other variables
Orientation
Predictor (x) (or Independent Variable)
➢ Variable is used in conjunction with other variables
to describe a given response
Response (dependant
variable)

Predictor
Y = f (x)
(independent Dependent Independent
variable) Variable Variable
TYPES OF DATA
Whether there are gaps between successively observed
Types of Data values of a variable:

Discrete Data
▪ Gaps exist between observations
▪ Obtained bycounting
“Gappiness”

Discrete Variable Discrete Data

Discrete
(obtainedby Lines
Characteristics
counting)
Line Charac teristi c D U OP TOP DPU DPO DPMO Shift Z.B
1 Type A 81 83 95 7,8850.976 0.0103 10,2731.50 3.82 Defects
2 Type B 67 584 20 11,6800.115 0.0057 5,736 1.50 4.03 Units
3 Type C 19 225 38 8,5500.084 0.0022 2,222 1.50 4.35
4 Type D 33 884 57 50,3880.037 0.0007 655 1.50 4.71 Opportunities
5 Type E 67 774 37 28,6380.087 0.0023 2,340 1.50 4.33 TotalOpportunities
6 Type F 27 669 91 60,8790.040 0.0004 444 1.50 4.82
NO-GO GO

1 2 3 4 5 6

2 3 4 5 6 7

Continuous
1

2 3 4 5 6 7 8

(obtained by 3 4 5 6 7 8 9

measuring)
4 5 6 7 8 9 10

5 6 7 8 9 10 11

6 7 8 9 10 11 12

_001
TYPES OF DATA
Whether there are gaps between successively observed
Types of Data values of a variable:

ContinuousData
▪ No gaps exist between observations
▪ Obtained bymeasuring
“Gappiness”
Continuous Data Continuous Variable
g X1 X2 X3 X4 X5
1 1.242 1.239 1.239 1.242 1.240
2 1.240 1.241 1.240 1.239 1.242

Discrete 3
4
1.239
1.241
1.239
1.240
1.239
1.240
1.239
1.240
1.240
1.241
Caliper

(obtainedby 5 1.240 1.241 1.240 1.238 1.241

counting) B 6
7
1.241
1.237
1.240
1.240
1.240
1.240
1.240
1.237
1.239
1.238
8 1.240 1.242 1.240 1.240 1.238
9 1.240 1.239 1.240 1.239 1.242
10 1.239 1.239 1.241 1.239 1.240 Dimension
11 1.239 1.238 1.242 1.238 1.240
12 1.239 1.241 1.239 1.239 1.242 “B”
Continuous 13
14
1.239
1.240
1.242
1.239
1.239
1.240
1.239
1.239
1.240
1.241
Time

(obtained by 15 1.241 1.240 1.240 1.240 1.240

measuring) 16
17
1.24
1.24
0 1.239 1.240 1.240
1.240
1.240
1 1.239 1.238 1.240
18 1.23 1.239
9 1.239 1.241 1.241 TEMPERATURE
19 1.24 1.242
0 1.239 1.240 1.238
20 1.241 1.240 1.241 1.239 1.240 Thermometer
STATISTICS OVERVIEW
Statistics

Descriptive Inferential
Statistics Statistics

Graphical Numerical Parameter Hypothesis

Presentations Measures Estimation Testing

Charts Location Point Estimate Parametric

Methods

Non
Tables Dispersion Interval
parametric
Estimate
Methods

Shape
STATISTICS OVERVIEW
DESCRIPTIVE STATISTICS VERSUS INFERENTIAL STATISTICS

Descriptive Statistics
Concerned with collecting and describing a set of data so as to yield
meaningful information

Inferential Statistics
Concerned with the analysis of sample data and generalizing results back
to the larger population, i.e. draw predictions or inferences about the
entire population
STATISTICS OVERVIEW Descriptive
Statistics

Graphical Numerical
Presentations Measures

Charts Tables Location Dispersion Shape

Frequency
Dot Plot Distribution Mean Range Skewness

Standard
Box Plot Median Deviation Kurtosis

Histogram Mode Variance

Stem & Leaf Interquartile

Quartiles
Diagram Range

Bar Chart

Trend Chart
STATISTICS OVERVIEW
Descriptive
Statistics
Measure of Location 
 
(Central Tendency)

Numerical
Measures Where does the data point


Location

Dispersion
Shape typically concentrated

Mean Range Skewness 

Measure of Dispersion
(Variation)
Standard
Median Kurtosis How wide or small the data
Deviation
points are scattered?

Mode Variance

Quartiles
Interquartile
Range
 Measure of Shape
(Distribution)
What shape do the data
take?
STATISTICS OVERVIEW
Measures of Location - QUARTILES

When an ordered set of data is divided into four equal parts,

the division points are called quartiles:

▪ The first or lower quartile Q1 is a value that has

approximately 25% of the observations below its value

▪ The second quartile Q2 is a value that has approximately

50% of the observations below in value. It is also called
the median

▪ The third or upper quartile Q3 is a value that has

approximately 75% of the observations below its value
STATISTICS OVERVIEW
Measures of Dispersion - RANGE

▪ The sample range is defined as the difference between the largest and
smallest observations in the sample, i.e.
▪ The range is the simplest measure of dispersion or variation of the data.
▪ However, it ignores all the information in the sample between the smallest
and the largestobservations.

Range (r) = rmax - rmin

STATISTICS OVERVIEW
Measures of Dispersion – VARIANCE and STANDARD DEVIATION

▪ The sample variance and the sample standard deviation are the most
important measuresof dispersion.
▪ The units of measurement for:

i. Variance - square of the original unit of the variable

 (x − x )
n 2

=
2 i=1 i
s
n −1
ii. Standard deviation - original unit of the variable

 (x − x )
n 2

s = i=1 i

n −1

▪ A smaller value of s² (and s) implies less variability.

STATISTICS OVERVIEW
Measures of Dispersion – SKEWNESS
▪ Skewness measures the degree of asymmetry of a distribution around its
mean.
▪ Positive Skewness implies an asymmetric tail (the longer tail) extending
towards higher values. Sometimes referred to as right-handed skew or skew
to the right.
▪ Negative Skewness implies an asymmetric tail extending towards lower
values. Sometimes referred to as left-handed skew or skew to the left.
STATISTICS OVERVIEW
Measures of Dispersion – KURTOSIS

▪ Kurtosis characterizes the relative

peakness or flatness of a distribution
compared to a normal (mesokurtic)
distribution.
▪ Positive kurtosis indicates a “sharper
peak” (leptokurtic) distribution compared
to the normal distribution.
▪ Negative kurtosis indicates a “flatter
peak” (platykurtic) distribution compared
to the normal distribution.
▪ Kurtosis is relevant only for symmetrical
distributions.