ENGINEERING

DATA
ANALYSIS
DATA ANALYSIS

is a process of inspecting, cleansing, transforming and modeling data

with the goal of discovering useful information, informing conclusion
and supporting decision-making, (Wikipedia)

involves sorting through massive amounts of unstructured information

and deriving key insights from it. These insights are enormously
valuable for decision-making at companies of all sizes.
DATA ANALYSIS PROCESS

1. Data Requirement Gathering

 decide what to analyze and how to measure it

2. Data Collection
 collect your data based on requirements

3. Data Cleaning
 data should be cleaned and error free.
DATA ANALYSIS PROCESS
4. Data Analysis
 Use the data analysis tools and software which will
help you understand, interpret, and derive
conclusions based on the requirements

5. Data Interpretation
 Express or communicate your data analysis

6. Data Visualization
 They often appear in the form of charts and graphs
2 CORE AREAS OF DATA ANALYSIS:

Quantitative Data Qualitative Data

Analysis Methods Analysis Methods

Quantitative Data Analysis Methods

Quantitative data is defined as the value of data in the form of counts
or numbers where each data-set has a unique numerical value
associated with it.
Types of Quantitative Data:

Counter: Count equated with entities.

Measurement of physical objects: Calculating measurement of any
physical thing.
Sensory calculation: Mechanism to naturally “sense” the measured
parameters to create a constant source of information.
Projection of data: Future projection of data can be done using
algorithms and other mathematical analysis tools
Quantification of qualitative entities: Identify numbers to qualitative
information.
Qualitative Data Analysis Methods

Qualitative data analysis is made up of words, observations, images, and

even symbols

Measurement Scales:  Nominal Scale

 Ordinal Scale

 Interval Scale

 Ratio Scale
Measurement Scales:

NOMINAL SCALE numbers serve as “tags” or “labels” only, to identify

or classify an object

Examples:

Please select the degree of How would you describe

discomfort of the disease: your behavioral pattern?
1 – Mild E – Extroverted
2 – Moderate I – Introverted
3 – Severe A – Ambivert
M N
E O
A M
Could you please select an
S What is your gender? I
option from below to describe
U your hair color. M- Male N
R 1 – Black F- Female
A
E 2 – Brown L
M 3 – Burgundy dichotomous nominal scale
E S 4 – Auburn
5 – Other S
N C
C
T A
A
L
L
E
E
S
M
E ORDINAL SCALE reports the ranking and ordering of the data
A without actually establishing the degree of
S variation between them.
U Examples:
R
E “How satisfied are you with “How happy are you with
M our products?” the customer service?”
1- Totally Satisfied 1- Very Unhappy
E S
2- Satisfied 2- Unhappy
N C 3- Neutral 3- Neutral
T A 4- Dissatisfied 4- Unhappy
L 5- Totally Dissatisfied 5- Very Unhappy
E
S
M O
E The Likert scale is a variant of the ordinal scale that is used
to calculate customer or employee satisfaction R
A D
S I
U N
R A
E L
M
E S The frequency of occurrence – Questions such as “How
frequently do you have to get the phone repaired?” S
N C
Very often C
T A
Often A
L Not Often L
E Not at all E
S
M O
E R
A D
S Understanding preferences: If a marketer conducts surveys to I
U understand which laptop brand do their respondents do not prefer, they
can use the ordinal scale. Out of the five mentioned laptop brand, rate
N
R A
the order of preference –
E HP _____ L
M Apple _____
E S Lenovo _____
S
N C Dell _____
Acer _____ C
T A
A
L
L
E
E
S
M
E INTERVAL SCALE scale is a quantitative measurement scale where
A there is order, the difference between the two
S variables is meaningful and equal, and the
U presence of zero is arbitrary.
R
Examples: Time is also one of the
E
M most popular interval data
examples measured on an
E S interval scale where the
N C values are constant,
T A known, and measurable.
L
Temperature
E
S
M I
E Please state your annual income N
A •Below $40,000 T
•$40,000- $60,000 E
S
•$60,000- $80,000
U R
•$80,000- $100,000
R •Above $100,000 V
E A
M L
Age is also a variable that is
E S
measurable on an interval S
N C scale, like 1, 2, 3, 4, 5 years and
T A etc.
C
L A
E L
S E
M
E RATIO SCALE helps to understand the ultimate-order, interval,
A values, and the true zero characteristic is an essential
S factor in calculating ratios.
U Examples:
R
E What is your height in feet
M and inches?
E S  Less than 5 feet.
N C  5 feet 1 inch – 5 feet 5 inches
T A
 5 feet 6 inches- 6 feet
L
More than 6 feet
E


S
M
E R
A A
S How much time do you spend
What is your weight in kgs? T
U daily watching television?
I
R Less than 50 kgs
O

 Less than 2 hours
E  51- 70 kgs
 3-4 hours
M  71- 90 kgs
 4-5 hours S
E S 91-110 kgs
C

5-6 hours
N C 

More than 110 kgs

T A  More than 6 hours

A
L L
E E
S
SUMMARY of Measurement Scales:
POPULATION AND SAMPLE
Population (N) is a collection, or set of individuals, objects, or
measurements whose properties are to be analyzed. It is the totality of
the observation.

Sample (n) is a subset of a population. It is a smaller group representing

the population having identical characteristics from which it was taken.

𝑵 n = sample
Formula: 𝒏= N = population
𝟏 + 𝑵𝒆𝟐
E = marginal error

Marginal error is the probability of committing a mistake

Confidence level is the probability of getting the correct result
SAMPLE (n)
Example: In a population of 5,000 students, how many samples will you
take to get an accurate result of a study?

Solution: N = 5,000 students 𝑵

n=? 𝒏=
𝟏 + 𝑵𝒆𝟐

If e = 1% If e = 5% If e = 10%

5,000 5,000 5,000

𝑛= 𝑛= 𝑛=
1 + (5,000)(𝟎. 𝟎𝟏)𝟐 1 + (5,000)(𝟎. 𝟎𝟓)𝟐 1 + (5,000)(𝟎. 𝟏𝟎)𝟐

𝑛 = 3,333 𝑠𝑡𝑢𝑑𝑒𝑛𝑡𝑠 𝑛 = 370 𝑠𝑡𝑢𝑑𝑒𝑛𝑡𝑠 𝑛 = 98 𝑠𝑡𝑢𝑑𝑒𝑛𝑡𝑠

SAMPLING TECHNIQUES
SIMPLE
RANDOM CONVENIENCE
SAMPLING SAMPLING

CLUSTER PURPOSIVE
SAMPLING SAMPLING

PROBABILITY SAMPLING NON-PROBABILITY

SAMPLING SAMPLING
METHOD METHODS
METHOD
SNOWBALL
SYSTEMATIC SAMPLING
SAMPLING

STRATIFIED QUOTA
RANDOM SAMPLING
SAMPLING
SAMPLING TECHNIQUES

Sampling Techniques Or Sampling Plan is the process of gathering

sampling units from the population.

1. Probability Sampling Technique is a sampling technique wherein each of

the population unit has an equal chance of being included in the sample.

a. Random Sampling is also called “Lottery Sampling” or

“Raffle Sampling”.

Example: electronic raffle system, random picker, etc.

P
R b. Cluster Sampling . The process of randomly selecting intact groups,
not individuals, with the defined population sharing similar
O characteristics
B
A
c. Systematic Sampling in this technique, elements are numbered
B for identification and samples are selected at regular intervals.
S
I
A
L Example: If N = 300, e = 10% and n = 75, which of the
M
I 300 will be Included in the sample?
T P
𝑁 𝟑𝟎𝟎
L Use: 𝑖𝑛𝑡𝑒𝑟𝑣𝑎𝑙 = = =𝟒
Y 𝑛 𝟕𝟓
I
Therefore, from the population, take the
N elements tagged as the 4th, 8th, 12th, 16th,…
G
P
R d. Stratified Sampling in this technique, the set of interests is divided
O into groups or aggregates from which the actual sampling is done.
B
A Example:
B If e = 5%, n = 370. Using
S
I stratified sampling, how
A
L will you get the 370
M
I sample from the
T P 5,000population?
Y L
Use:
I
N # 𝑜𝑓 𝑠𝑡𝑢𝑑𝑒𝑛𝑡𝑠 𝑝𝑒𝑟 𝑐𝑜𝑙𝑙𝑒𝑔𝑒 × 𝑛
G 𝑁
2. Non-Probability Sampling Technique is a sampling technique wherein
each of the population unit has a n equal chance of being included in the
sample.

a. Convenience Sampling in this technique, the samples are simple

those chosen to respond.

b. Purposive Sampling in this technique, the samples are chosen

based on specific objectives.

c. Snowball Sampling in this technique, the researcher can rely on

his initial respondents to refer him to the next respondents

d. Quota Sampling in this technique, the researcher characterizes

the population based on desired features.
DATA COLLECTION METHODS:

1. Direct or Interview – researcher prepares a set of questions

and respondents will answer verbally and directly.
D 2. Indirect or Questionnaire – researcher prepares a well- planned,
A written questions.
T
A C
O
L
L
E
C
T
I
O
N
D
A SURVEYS
T
A C
O A survey is a way to ask a lot of people a few well-constructed
questions. The survey is a series of unbiased questions that the
L
subject must answer.
L
E self-administered
C surveys
T
I
O
N
D Useful steps in designing a survey:
A
T
 Determine the goal of your survey
A C S
O  Identify the sample population U
L R
 Choose an interviewing method
L V
E  Decide what questions you will ask in what order, and E
C how to phrase them. Y
T
 Conduct the interview and collect the information.
I
O  Analyze the results by making graphs and drawing conclusions.
N
 Example of Planning and Conducting Surveys:
D
A We want to construct a survey that shows which sports students at
T your school like to play the most.
A C S
O (a) List the goal of the survey
U
L R
L  The goal of the survey is to find the answer to the question: V
E “Which sports do students at your school like to play the most?” E
C Y
(b) What population sample should we interview?
T
I  A sample of the population would include a random sample of the
O student population in your school. A good strategy would be to
N randomly select students
D (c) How should you administer the survey?
A
T  Face-to-face interviews are a good choice in this case. Interviews
A C will be easy to conduct since the survey consists of only one
S
O question which can be quickly answered and recorded, and asking U
L the question face to face will help eliminate non-response bias. R
L V
E (d) Create a data collection sheet that she can use to record her results
E
C Y
T  In order to collect the data to this simple survey you can design a data
I collection sheet .
O
N
 o Plenty of space is left for the tally
D marks.
A o Only one question is being asked.
T o Many possibilities are included,
A C but space is left at the bottom in S
O case students give answers that U
L you didn’t think of. R
L o The answer from each V
interviewee can be quickly
E E
collected and then the data
C collector can move on to the next Y
T person.
I
O (e) Display, analyze and present the data
N
 FOCUS GROUP DISCUSSION
D
A
FGD is an in-depth field method that brings together a
T
small homogeneous group (usually six to twelve persons) to
A C discuss topics on a study agenda.
O
L
General Principles to consider:
L
E
C  Standardization of questions
T The number of focus groups conducted
I Number of participants in a group
O Level of moderator involvement
N
D
A 3. Registration – document and records
T
A C
O
4. Observation – collects information on the characteristics of the
L units under study by actual measurement.
L
E
C
5. Experimental Method – makes trials and tests; it is used to
T describe any process that generates a set of data
I
O
N
 Planning and Conducting Experiments
D
A
T  An experiment is a series of tests conducted in a systematic
A C D
O manner to increase the understanding of an existing process or to
L explore a new product or process.
L O
E
 Design of Experiments (DOE), is a tool to develop an
C experimentation strategy that maximizes learning using a
T minimum resources. E
I
O
N
D
Stages of Design of Experiment
A
T
A C 1. Planning D
O
L 2. Screening
L O
E 3. Optimization
C
4. Robustness Testing
T E
I 5. Verification
O
N
 1. PLANNING
D
A  carefully plan for the course of experimentation before
embarking upon the process of testing and data collection.
T
A C D
O A team composed of individuals from different disciplines related to the
L product or process should be used to identify possible factors to
L investigate and the most appropriate response(s) to measure. O
E
C
T “Carefully planned experiments always lead E
I to increased understanding of the product
O or process.”
N
D 2. SCREENING
A
T
A C  are used to identify the important factors that affect the process D
O under investigation out of the large pool of potential factors.
L
L O
E 3. OPTIMIZATION
C
T  determine the best setting of these factors to achieve the desired E
I objective.
O
N
D 4. ROBUSTNESS TESTING
A
T
A C  to make the product or process insensitive to variations that are D
O likely to be experienced in the application environment.
L
L O
5. VERIFICATION
E
C
T  validation of the best settings by conducting a few follow-up
experimental runs to confirm that the process functions as desired
E
I and all objectives are met.
O
N
D Process Model for DOE
A
T
A C D
O
L
L O
E
C
T E
I
O
N
 Qualitative Research Quantitative Research
Methods: Methods:

- Ethnography - Descriptive Research

- Grounded Theory - Correlational Research
- Phenomenology - Experimental Research
- Quasi-Experimental
Research
Presentation of Data

1. Textual – Data gathered are presented in paragraph form

Example:

Of the 150 sample interviewed, the following

complaints were noted: 27 for lack of books in the
library, 25 for a dirty cafeteria, 20 for lack of
laboratory equipment, 17 for a not well
maintained university building.
P
R 2. Tabular – using a statistical table where data is systematically
E organized in columns and rows
S
E
Parts of a statistical table
N
T Title Body
A Heading Footnotes
T Stubs Source Notes
Box head
I
O
N
P
Example: 20 applicants were given a performance
R
evaluation appraisal.
E
S The data set is:
E
N
T
A
T
I
O
N
P
R
E Performance Evaluation Appraisal
S
E CLASS FREQUENCY PERCENT
N
T HIGH 7 35
A AVERAGE 8 40
T LOW 5 25
I
O
N
P
R 3. GRAPHICAL – using a statistical table where data is systematically
organized in columns and rows
E
S
E Types
N Bar Graph
T Pie or Circle Graph
Line Graph
A
Pictograph
T Etc.
I
O
N
P Bar Graph:
R
E Selected Causes of Death in the Philippines:
S
E
N
T
A
T
I
O
N
P Pie Chart:
R
E Three Leading Causes of Child Mortality
S Among Filipinos Ages 5- 9
E
N
T
A
T
I
O
N
P Line Graph:
R
E Distribution of Enrolment at a Day Care, 1999-2006
S
E
N
T
A
T
I
O
N
P
R
E Pictogram:
S Number of
E Persons Who
N have Excessive
T Depression by
A Cluster
T
I
O
N
 Frequency Distribution Table
P
R
E Frequency tells you how often something happened. The frequency
S of an observation tells you the number of times the observation
E occurs in the data.
N
T Example: Let’s say you did a survey on number of households
A to find out how many pets they own
T
3, 0, 1, 4, 4, 1, 2, 0, 2, 1
I 2, 0, 2, 0, 1, 3, 1, 2, 1, 3
O
N
P
R
E Step 1: Construct the table. Write the Categories:
S
E
N
T
A
T
I
O
N
P 3, 0, 1, 4, 4, 1, 2, 0, 2, 1, 2, 0, 2, 0, 1, 3, 1, 2, 1, 3
R
E Step 2: Tally the numbers (raw data)
S 0, 0, 0, 0, 1, 1, 1, 1, 1, 1, 2, 2, 2, 2, 2, 3, 3, 3, 4, 4
E
N
T
A
T
I
O
N
P
R
Step 3: Write the data into numerical frequencies
E
S
E
N
T
A
T
I
O
N
P Step 4: Determine the percentage
R
E
S
E
N
T n=20
Frequency of
A the class
T Percentage Formula:
I
O
N Percentage Total number
of values
P GROUPED Frequency Distribution Table
R
E Construction of a Grouped Frequency Distribution Table
S
Example:
E
N Jake measured the lengths of leaves from a certain tree (to the nearest cm):
T
A 9,16,13,7, 8,4,18,10,17, 18,9,12,5,9, 9,16,18,17,
T 1,10,5,9,11, 15,6,14,9,1, 12,5,16,4,16, 8,15,14,17
I
O
N
P
R Construction of a Grouped Frequency Distribution Table
E
Step 1: Put the numbers in order, then find the smallest and largest
S
values in your data. Calculate the Range.
E
N 1, 1, 1, 4, 4, 5, 5, 5, 6, 7, 8 ,8 ,8 ,9 ,9 ,9 ,9 ,9 ,9 ,10, 10, 11, 12,
T 12, 13, 14, 14, 15, 15, 16, 16, 16, 16, 17, 17, 17, 18, 18
A
T
Range = largest value – smallest value
I
O
Example: Range =18 cm – 1 cm = 17 cm
N
P
R Step 2: Calculate the approximate number of classes K.
E
S K = 1 + 3.322 Log N , N is the number of values
E
N Example: K = 1 + 3.322 Log 38 = 6.24 = 6
T
A Step 3: Determine the class size C.
T
I C = R/K
O Example: C = 17/6.33 = 2.72 = 3
N
P
R
Step 4:
E
S Starting at 1 with a class size
E of 3 we get:
N 1, 4, 7, 10, 13, 16
T
Write down the groups,
A include the end value of each
T group (must be less than the
I next group)
O
N
P Step 5: Write the frequency for each group
R
E
S
E 1,1,1,4,4,5,5,5,6,7,8,8,8,
N 9,9,9,9,9,9,10,10,11,12,
T 12,13,14,14,15,15,16,
A 16,16,16,17,17,17,18,18
T
I
O
N
P Step 6: Write the relative frequencies (rf)
R
E
S
E
N
T
A 10  38 ≈ 0.2631
T
I
O
N
P Step 7 : Write the percentage (%f)
R
E
S
𝑓
𝑥100
E 𝑛
N
T
A
T
(10  38) x 100%
I ≈ 26.31
O
N
P Step 8 : Determine the cumulative frequencies (cf)
R
E
S
E = 3+6 = 9
= 9 + 10 = 19
N
T
A
T
I
O
N
P
R
Step 9:
E 1+3
= =2
S Compute the
2
4+6
= =5
E midpoint or class 2

N mark, x for each

T class
A 𝑳𝑳 + 𝑼𝑳
T 𝒙=
𝟐
I
O
N
P
Step 10: Class boundaries (Real Limits), less than cumulative frequency,
R greater then cumulative frequency
E
S
E
N
T
A
T
I
O
N
P
R
E
Step 11: Present the distribution graphically
S
E
 Frequency polygon (x, f) – line graph
N
T  Histogram
A  Frequency Polygon Superimposed on a histogram
T  Ogive – cumulative frequencies (LCB, CF)
I
O
N
P
Graphical Presentation
R
E
S
E
N
T
A
T
I
O
N
 Graphical Presentation
P
R
E
Length in f x
S
cm
E 1–3 3 2
N 4–6 6 5
T 7–9 10 8
A 10 – 12 5 11
T 13 – 15 5 14
I 16 – 18 9 17
O 38

N
P
Graphical Presentation
R
E
Length in f x
S
cm
E 1–3 3 2
N 4–6 6 5
T 7–9 10 8
A 10 – 12 5 11
T 13 – 15 5 14
I 16 – 18 9 17
O 38

N
 Graphical Presentation
P
R
E Class <cf >cf
S Boundaries
E 0.5 - 3.5 3 38
N 3.5 - 6.5 9 35
T 6.5 - 9.5 19 29
A 9.5 - 12.5 24 19
T 12.5 - 15.5 29 14
I 15.5 - 18.5 38 9
O
N
HOME WORK #1

Please click on CLASSWORKS Tab in our Google

classroom to open HW#1.