INFERENTIAL STATISTICS selecting to generalize

- A major portion of statistics deals with making

decisions, inferences, predictions, and
forecasts about populations based on results
L1: INTRODUCTION TO STATISTICS obtained from samples
CONTENTS: - Inferential statistics consists of methods that
- Nature of Statistics use sample results to help make decisions or
- Key Terms in Statistics predictions about a population
- Descriptive and Inferential Statistics - Statistical inference uses probability to
- Definition of Statistical Terms
determine how confident we can be that our
- Levels of Measurement
conclusions are correct
NATURE OF STATISTICS
If you will understand and appreciate the importance
- Statistics refers to numerical facts of Statistics, you can be more confident in the
- Statistics is a group of methods used to decisions you make in life
collect, analyze, present, and interpret data
KEY TERM IN STATISTICS
and to make decisions
- The science of statistics deals with the POPULATION universe
collection, analysis, interpretation, and - A population consists of all elements—
presentation of data individuals, items, or objects—whose
characteristics are being studied
DATA
- The population that is being studied is
- Refers to the information that has been
also called the target population
collected from an experiment, a survey, an
1. Theoretical – Entire Entity
historical record, etc. SAMPLE 2. Accessible – Researchers have access
to
o “Data” is plural and one piece of - A portion of the population selected for study
information is called a “datum” is referred to as a sample
- Representative of a population
IMPORTANCE OF STATISTICS
REPRESENTATIVE SAMPLE
- It helps in designing experiments, analyzing
- A sample that represents the characteristics of
and interpreting data
the population as closely as possible is called a
- It also contributes to making appropriate
representative sample. Inferences derived from
decisions in the light of the researcher's
a representative sample will be more reliable
findings
- Statistical science is a means to read the RANDOM SAMPLE
results of other research and the ability to - A sample drawn in such a way that each
distinguish good and stronger element of the population has a chance of
being selected is called a random sample
TYPES OF STATISTICS
SAMPLING
- Descriptive Statistics
- The process of selecting/drawing a sample or
- Inferential Statistics
a portion of the population
DESCRIPTIVE STATISTICS describe
STATISTICS AND PARAMETER
- Descriptive statistics are numbers that are
- A statistic is a number that represents a
used to summarize and describe data
property of the sample
- Consists of methods for organizing, displaying,
- A parameter is a number that is a property of
and describing data by using tables, graphs,
the population
and summary measures
- Organizing and summarizing data is called VARIABLE varies
descriptive statistics - A variable is a characteristic under study that
- Two ways to summarize data are by graphing assumes different values for different elements
and by using numbers such as finding an o The Quality of Life of Public-School
average Teachers in the Province of Cavite
During Covid-19 Pandemic
DATA QUANTITATIVE VARIABLES
- Data are the actual values of the variable - A variable that can be measured numerically
- They may be numbers or they may be words. - Also known as Numerical variables which take
Datum is a single value on values with equal units such as weight in
pounds and time in hours
SURVEY AND CENSUS
- The data collected on a quantitative variable
- The collection of information from a sample is
are called quantitative data
called a survey
o Types of Quantitative Variables
- A survey that includes every element of the
▪ Discrete
target population is called a census
▪ Continuous
SAMPLE SURVEY
DISCRETE QUANTITATIVE VARIABLE
- It is a survey conducted which selects a
- A variable whose values are countable. It can
sample and collect the required information
assume only certain values with no
from the elements included in that sample
intermediate values
- The technique of collecting information from
o Example: number of books in the
a portion of the population
library, student enrolment in DLSMHSI
WHY CONDUCT SAMPLE SURVEY?
CONTINUOUS QUANTITATIVE VARIABLE
- Often the target population is very large - A variable that can assume any numerical
- A census is rarely taken because it is value over a certain interval or intervals
expensive and time-consuming - These are variables that cannot be counted
- In many cases, it is even impossible to identify and they can assume any numerical value
each element of the target population between two numbers
o Example: weights of people, heights
ELEMENT OR MEMBER of students
- An element or member of a sample or
population is a specific subject or object (for QUALITATIVE OF CATEGORICAL VARIABLES
example, a person, firm, item, state, or
- Variables that cannot be measured
country) about which the information is
numerically but can be divided into different
collected
categories
DATA SET - A variable that cannot assume a numerical
- A data set is a collection of observations on value but can be classified into two or more
one or more variables nonnumeric categories
- The data collected on such a variable are
VARIABLE called qualitative data
- A variable is a characteristic under study that o Example: Status of an undergraduate
assumes different values for different CMIT student (freshman, sophomore,
elements junior, or senior)
- A variable, notated by capital letters such as X
and Y, is a characteristic of interest for each
person or thing in a population
o Relationship between Leadership
Style and Leadership Performance of
School Administrators in Selected HEIs
in Cavite

TYPES OF VARIABLES

- Quantitative Variables
- Qualitative Variables
 LEVELS OF MEASUREMENT LIMITATIONS OF ORDINAL MEASUREMENT
- The levels of measurement of data
- It fails to capture important information that
determines what type of analysis will be
will be present in the other scales
performed in a study
- In particular, the difference between two levels
- The categories are called "scale types," or just
of an ordinal scale cannot be assumed to be the
"scales"
same as the difference between two other levels
- Nominal - Interval
- In the satisfaction scale, for example, the
- Ordinal - Ratio
difference between the responses "very
NOMINAL dissatisfied" and "somewhat dissatisfied" is
- Measurements akin to mere categorization probably not equivalent to the difference
are nominal between "somewhat dissatisfied" and
- When measuring using a nominal scale, one "somewhat satisfied"
simply name or categorizes responses and
INTERVAL
they do not imply any ordering among the
- Measurements that have consistent distances
responses
between values
- Nominal scales embody the lowest level of
o Celsius and Fahrenheit temperature scales
measurement
having to are examples of interval measurements
do with o Gender, favorite color, and religion
- The difference between ordinal and interval
names are examples of variables measured
measurements is that the latter uses a
on a nominal scale
standard unit of measurement, or metric
CHARACTERISTICS OF IDENTIFIED CATEGORIES FOR o For temperature, the metric is degrees
NOMINAL DATA - The equidistant intervals between values
allows researchers to perform basic
- It must be exhaustive. Researchers must
mathematical operations such as addition and
categorize all cases appropriately
subtraction with interval measurements
o Example: Political Party: LAKAS-CMD,
- Variables operationalized at lower levels (i.e.,
PDP-LABAN, Nationalist People’s
nominal or ordinal) are not quantitative
Coalition, Liberal Party, Laban ng
Demokratikong Pilipino RATIO
- It must be mutually exclusive. It means that - Measurements that have an absolute zero
the researcher can classify a case only into a point are ratio
single category o Examples include income, Kelvin
o Example: Type of Place participants temperature scale, years married, and
currently live: urban, suburban, rural reaction time
- The ratio scale of measurement is the most
ORDINAL
informative scale
- Measurements that indicate the rank order
- It is an interval scale with the additional
of cases
property that its zero position indicates the
- With ordinal measurements, one can know
absence of the quantity being measured
the relative differences among the values and
- You can think of a ratio scale as the three
order them appropriately
earlier scales rolled up in one
- Ordinal scales allow comparisons of the
- Like a nominal scale, it provides a name or category
degree to which two subjects possess the
for each object (the numbers serve as labels)
dependent variable
- Like an ordinal scale, the objects are ordered
- For example, our satisfaction ordering makes
(in terms of the ordering of the numbers)
it meaningful to assert that one person is
- Like an interval scale, the same difference at
more satisfied than another ("very
two places on the scale has the same meaning
dissatisfied," "somewhat dissatisfied,"
"somewhat satisfied," or "very satisfied"
- The items in this scale are ordered, ranging
from least to most satisfied
NOTES:

STATISTICS

- “Status”
o political state
- Way of getting information from data

FUNCTIONS OF STATISTICS

1. Present facts in indefinite form

2. Simplify unimaginable/complex data
3. Used for making comparison
4. Enlarge individual experience
5. To provide guidelines in Formulation of
Policies
6. Enable measurement of magnitude of
phenomenon

IMPORTANCE IF STATISTICS

- Statistics plays a significant role in:

o Forecast
o Research
o Business market
o Medical field
o Finance

LIMITATIONS OF STATISTICS

1. Limited to numerical studies

2. Deal with population
3. Relies of estimation

2 WAYS TO SUMMARIZE DATA

1. Graphs
2. Numbers
 L2: SAMPLING 3. Avoid non-response errors
CONTENTS: a. This can become a problem if the non-
- Sampling respondents generally hold a view
- Probability and Non-Probability Sampling that is different from the respondents
- Types of Probability Sampling b. As a result, the final data will be
- Types of Non-Probability Sampling skewed toward the opinions of those
SAMPLING who responded
- It is the process of selecting the sample for 3 THINGS TO KEEP IN MIND DURING SAMPLING
estimating the population characteristics
- It is the process of obtaining information 1. Consistency
about an entire population by examining only a. If your sample is consistent, you can
a part of it be confident that any change in the
- Purpose of sampling is to provide an estimate data reflects real change across the
of the population parameter and to test the population, rather than change across
hypothesis a typical individual in the population
2. Diversity
ADVANTAGES OF SAMPLING a. Ensuring diversity of the sample is a
1. Save time and money tall order, as reaching some portions
2. Enable collection of comprehensive data of the population and convincing
3. Enable more accurate measurement as it them to participate in the survey can
conducted by trained and experienced be difficult
investigators 3. Transparency
4. Sampling remains the only way when a. It is imperative that researchers
population contains infinitely many members discuss limitations and maintain
transparency about the procedures
DISADVANTAGES OF SAMPLING followed while selecting the sample
1. Chances of bias and difficulties in selecting a TYPES OF SAMPLING
truly representative sample
2. Inadequate knowledge in the sampling 1. Non-Probability Sampling
technique 2. Probability Sampling
3. Units of the population may change hence, NON-PROBABILITY SAMPLING
units of the sample may be widely dispersed
4. Chances of sampling errors - Non-probability sampling technique is totally
based on judgement
SAMPLING ERROR - Non-probability sampling is a sampling
- The results of the sample differ from a target procedure that will not bid a basis for any
population quantity opinion of probability that elements in the
universe will have a chance to be included in
MINIMIZING SAMPLING ERROR the study sample
1. Avoid population specification errors TYPES OF NON-PROBABILITY SAMPLING
a. A population specification error
occurs when a critical segment of the 1. Quota Sampling
population is not included in the 2. Purposive (Judgmental) Sampling
sample 3. Self-Selection Sampling
2. Avoid sample frame error 4. Snowball Sampling
a. It occurs when a survey samples the 5. Accidental Sampling
wrong segment of the total 6. Expert Sampling
population, usually because the 7. Modal Instance Sampling
surveyor has missed a new trend or 8. Heterogeneity Sampling
change in their target population
QUOTA SAMPLING TWO STAGES OF SNOWBALL SAMPLING

- non-probability sampling method in which - Identification of a sample of respondents with

researchers create a sample involving characteristic x at the zero-stage
individuals that represent a population - Solicitation of referrals to other potentially
- Researchers choose these individuals eligible respondents believed to have
according to specific traits or qualities characteristic x at snowball stages

TYPES OF QUOTA SAMPLING CONVENIENCE SAMPLING

1. Proportionate - Haphazard Sampling or Accidental Sampling

2. Non-Proportionate - Nonprobability or nonrandom sampling where
members of the target population that meet
PROPORTIONATE QUOTA SAMPLING
certain practical criteria, such as easy
- Represent the characteristics of major
accessibility, geographical proximity, availability
population by sampling a proportional total
at a given time, or the willingness to participate
(Bala and Etikan, 2017)
are included for the purpose of the study
o Example if we are interested in
- Convenience samples are sometimes regarded
studying population of 40 percent of
as ‘accidental samples’ because elements may
females and 60 percent of males
be selected in the sample simply as they just
o If the sample size is 200, then we
happen to be situated, spatially or
need 80 females and 120 males for us
administratively, near to where the researcher
to hit our target (quota)
is conducting the data collection
NON-PROPORTIONATE QUOTA SAMPLING
SELF-SELECTION SAMPLING
- The nonproportional quota sampling is a
technique with small restriction of minimum - Appropriate when we want to allow units or
of sample number of units from each category cases, whether individuals or organizations to
- It’s not interested in having a number that will choose to take part in research on their own
match the proportions of the population accord
- The method is a non-probabilistic sampling - The key component is that research subjects
that typically used in assuring that small groups volunteer to take part in the research rather than
of samples are adequately represented being approached by the researcher directly

PURPOSIVE/JUDGEMENTAL SAMPLING EXPERT SAMPLING

- The sampling design is based on the - The researcher seeks for the consent of those
judgement of the researcher as to who will that are expert or known expert in the area of
provide the best information to succeed for study, and begin the process of collecting
the objectives study information directly from individual or group
- It is the deliberate choice of a participant due of respondents
to the qualities the participant possesses - The reasons for using expert sampling are to
- It is a nonrandom technique that does not have a better way of constructing the views of
need underlying theories or a set number of individuals that are expert in a definite area
participants - It is also used in providing confirmation of validity
- The researcher decides what needs to be to another approach of a selection of sampling
known and sets out to find people who can
MODAL INSTANCE SAMPLING
and are willing to provide the information by
virtue of knowledge or experience - The purpose of modal instance sampling is to
sample the most typical members of a population
SNOWBALL SAMPLING
- The term modal comes from the mode, which
- Chain referral sampling is the most common item in a set
- Non-probability method of survey sample - One problem with modal instance sampling is
selection that is commonly used to locate rare identifying what is the most “typical” member
or difficult to find populations of a set
HETEROGENEITY SAMPLING SYSTEMATIC RANDOM SAMPLING

- Sampling for Diversity - It is a probability sampling method where the

- This is used when one wants to include all researcher chooses elements from a target
opinions or views, and are not concerned population by selecting a random starting
about representing these views point and selects sample members after a
proportionately fixed ‘sampling interval.’
- In many brainstorming or nominal group - The sampling interval is the standard distance
processes (including concept mapping), one between the elements
would use some form of heterogeneity sampling
STEPS IN SYSTEMATIC RANDOM SAMPLING
because the primary interest is in getting broad
spectrum of ideas, not identifying the “average” - Determine the population of interest
or “modal instance” ones - Compute for the sample size (n)
- Assign a number to every member of the
PROBABILITY SAMPLING
population
- Probability sampling is any sampling scheme - Define the interval of the sample
in which the probability of choosing each - Randomly choose the starting member of the
individual is the same population and add the interval to the
- These are also called random sampling random number to keep adding members
- They require more work, but are much more - Systematic Sampling formula for interval (i)
accurate o i = N/n
▪ N – population size
TYPES OF PROBABILITY SAMPLING
▪ n – sample size
- Simple Random Sampling
- Systematic Random Sampling
- Stratified Random Sampling
- Cluster Sampling or Multi-Stage Sampling

SIMPLE RANDOM SAMPLING PROS AND CONS OF SYSTEMATIC RANDOM

SAMPLING
- In this technique, each member of the
population has an equal chance of being PROS
selected as subject - Spreads the sample more evenly over the
- The entire process of sampling is done in a population
single step with each subject selected - Easier to conduct than a simple random sample
independently of the other members of the CONS
population - If the sampling technique coincides with the
periodicity (regularly recurrent) of the trait,
PROS AND CONS OF SIMPLE RANDOM SAMPLING the sampling technique will no longer be
PROS random and representativeness of the sample
- Ease of assembling the sample is compromised
- Representativeness of the population STRATIFIED RANDOM SAMPLING
- It is reasonable to make generalizations from
the results of the sample back to the - A method of sampling that involves the
population division of a population into smaller groups
CONS known strata
- The need of a complete list of all the - In stratified random sampling, the strata are
members of the population formed based on members shared attributes
- Elected independently of the other members or characteristics
of the population - A random sample from each stratum is taken
in a number proportional to the stratum’s size
when compared to the population
- These subsets of the strata are then pooled to
from a random sample
 PROS AND CONS OF STRATIFIED RANDOM
SAMPLING

PROS
- It provides us with a sample that is highly
representative of the population being studied,
assuming that there is limited missing data
- It allows us to make generalizations (i.e.
statistical inferences) from the sample to the
population
CONS
- It not useful when the population cannot be
exhaustively partitioned into disjoint
subgroups

CLUSTER RANDOM SAMPLING

- Use when the total area of the research is too

large
- In this sampling technique, the researcher
divides the area (clusters) in to smaller part of
the same or equal and then select randomly
from the smaller units
- One of the advantages of using the cluster
sampling is economical in reducing cost by
concentrating on the selected clusters

MULTI-STAGE RANDOM SAMPLING

- Multistage sampling has to with the
combination of the various methods of
probability sampling in most effective and
efficient approach

PROS AND CONS OF CLUSTER OR MULTI-STAGE

RANDOM SAMPLING

PROS
- Economy
- Reduced variability
- Feasibility
CONS
- Biased sampling
- Sampling error

Total Enumeration Sampling

- Too few to be able to include the total

number of populations