MATH 403 - Engineering Data Analysis
Reviewer Notes
CHAPTER 1. OBTAINING DATA
Data Collection and Conducting Surveys ● Documents
● Records
Methods of Data Collection ● Tests
● Examinations
DATA COLLECTION - refers to the data
gathering — a systematic method of Significant persons involved in data
collecting and measuring data from different gathering:
sources of information in order to provide
answers to relevant questions. ● INVESTIGATOR - person who
conducts the inquiry
POPULATION - is the entire set of items
from which you draw data for a statistical ● ENUMERATOR - person who helps
study. It serves as the study of your interest in collecting information
where you draw conclusions from. Example
of a study: “Academic Performance of the ● RESPONDENT - the source of data
college students studying in Batangas State in a form of a person
University”. The population/study of interest
would be the college students studying in Three Basic Methods of Collecting Data:
BSU in the timeframe of data collection.
1. RETROSPECTIVE STUDY - use the
SAMPLE - is defined as a smaller and more population or sample of the historical
manageable representation of a larger data which had been archived over
group. In addition, it’s a subset of a larger some period of time.
population containing the characteristics of
the same population. This is usually used if 2. OBSERVATIONAL STUDY - is
there is a large population. For example: where the process or population is
Consider the same study and population observed and the quantities of
earlier, a sample can be the 4th year interests are recorded
students in BSU.
3. DESIGNED EXPERIMENT - has
Data can be gathered from the following: deliberate or purposeful changes in
the controllable variables of the
● Respondents system or process is done. In
● Observations addition, it includes manipulation of
● Published Literatures (dictionaries, processes in order to get desired
books, encyclopedias) results.
● Surveys
● Experimentation
Planning and Conducting Surveys judgment of the researcher rather
than random selection.
SURVEYS - method of asking respondents
some well-constructed questions. It is an A. Convenience Sampling -
efficient way of collecting information and where samples are selected
easy to administer wherein a wide variety of from the population only
information can be collected. This comes in because they are
different forms like face-to-face interviews, conveniently available to the
virtual interviews, questionnaires, google researcher. There is no
forms, and more. Moreover, it is usually pattern whatsoever in
used in qualitative research. acquiring these respondents
— they may be recruited
When designing a survey, the following merely asking people who
steps are useful: are present in the street, in a
public building, or in a
1. Determine the objectives of your workplace, and asking
survey through devices, for
2. Identify the target population sample example.
3. Choose an interviewing method
4. Decide what questions will you ask B. Purposive Sampling - is a
in what order, and how to phrase predetermined sampling
them where researchers choose
5. Conduct the interview and collect the only those people who they
information deem fit to participate in the
6. Analyze the results by making research study. It involves
graphs and drawing conclusions selecting participants
because the researcher
Sampling believes that they might
contribute something to the
SAMPLING - the selection process of a analysis. For example, from
sample in a target population used in the the same study entitled
study “Academic Performance of
the college students studying
Two types of Sampling: in Batangas State
University”, the researcher
1. NON-PROBABILITY SAMPLING might choose students that
(AKA Judgment or Subjective excel academically.
Sampling) - is a convenient and
economical method but the C. Quota Sampling - is a
inferences made based on the non-probability sampling
findings are not so reliable. It is method that relies on the
biased because the researcher non-random selection of a
selects samples based on subjective predetermined number or
proportion of units called
“quota”. The researcher units you want in
selects participants based on each category — not
specific characteristics, requiring numbers to
ensuring they represent match the proportions
certain attributes in in the population.
proportion to their prevalence
in the population. It’s like - a minimum number
stratified but without the of sampled units in
random sampling since each category is
researchers manipulate the specified and not
quota. concerned with
having numbers that
● Proportional Quota match the proportions
Sampling - the major in the population
characteristics of the
population are 2. PROBABILITY SAMPLING - is
represented by where every member of the
sampling them in population is given an equal chance
regards to their to be selected as a part of sample
proportion in the
population of study. A. Simple Random Sampling -
Researchers not only is the basic sampling
ensure participants technique where a group of
represent certain subjects (a sample) is
attributes selected for study from a
proportionally, but larger group (a population)
also control for the
order in which they B. Stratified Sampling - is
are selected. obtained by taking samples
from each stratum or
- the major sub-group of a population.
characteristics of the Researchers divide subjects
population by into subgroups called strata
sampling a based on characteristics that
proportional amount they share (e.g., race,
of each is gender, educational
represented. attainment). Once divided,
each subgroup is randomly
● Non-proportional sampled using another
Quota Sampling - a probability sampling method,
less restrictive quota choosing individuals in each
sampling where you strata to form the sample.
specify the minimum
number of sampled
C. Cluster Sampling - is a - Create a model like the IPO
sampling technique where (Input-Process-Output)
the entire population is flowchart
divided into groups/clusters
and a random sample of 4. COLLECT
these clusters are selected. - Generate and record after
conducting the experiment
5. FIT
- Validate the results and
analyze
6. PREDICT
Planning and Conducting Experiments:
Introduction to Design of Experiments
DESIGN OF EXPERIMENTS (or DOE) - is
a tool to develop an experimentation
strategy that maximizes learning using
minimum resources.
- a technique needed to identify the vital
need factors in the most efficient manner
and then directs the process to its best
setting to meet the ever-increasing demand
for improved quality and increased
productivity
Six Stages to be carried out for the design
of experiments:
1. DESCRIBE
- Identify the objectives and
factors
2. SPECIFY
- Specify the best setting and
methodology in
accomplishing the objectives
3. DESIGN