DATA COLLECTION AND INSTRUMENTATION

Investigator - person who plans and conducts the

Data Collection - process of gathering and measuring statistical investigation independently or with the
information on variables of interest, in an help of others.
established systematic fashion that enables one to Respondent - person who answers/responds to the set
answer stated research questions, test hypotheses, of questions is called.
and evaluate outcomes. Enumerator - person who collects data by conducting
an enquiry or an investigation.
Tools for Data Collection Population Element - is the individual participant or
object on which the measurement is taken. It is the
1. Questionnaires - research instrument consisting unit of study.
of a series of questions for the purpose of - It may be a person, but it could also be any object
gathering information from respondents. of interest.
- a device for securing answers to questions by Population - total collection of elements about which we
using a form which the respondent will fill by wish to make some inferences.
himself. Sample Frame - the listing of all population elements
2. Interview - a two-way method which permits an from which the sample will be drawn.
exchange of ideas and information. - Here a small group is selected as representative
3. Schedules - tool or instrument used to collect of the whole population.
data from the respondents while interview is
conducted. Why use Sampling?
- contains questions, statements, and blank → Sampling provides:
spaces/tables for filling up the respondents. → Availability elements
4. Observation - described as a method to observe → Greater speed
and describe the behaviour of a subject. → Lower cost
- way of collecting relevant information and → Greater accuracy
data by observing.
5. Rating Scale - term applied to express opinion or WHAT IS A VALID SAMPLE?
judgment regarding some situation, object or
character. Accuracy - refers to how close or far off a given set of
- consists of close-ended questions measurements are to their true value.
- helps gain information on the qualitative and Precision - refers to how close or dispersed the
quantitative attributes. measurements are to each other.
Inaccuracy or Bias - refers to a systematic deviation
SAMPLING AND INSTRUMENTATION from the actual value.
Imprecision or Uncertainty - refers to the magnitude of
Survey - research method used for collecting data from scatter.
a predefined group of respondents to gain
information and insights into various topics of METHODS OF SAMPLING
interest.
- can have multiple purposes, and researchers ▪ Probability Sampling - selection of a sample from a
can conduct it in many ways depending on the population based on the principle of randomization or
methodology chosen and the study's goal. by chance.
- sampling method in which all members of the
TWO METHODS TO CONDUCT SURVEY population have an equal chance of participating
in the study.
Census Method (Parametric) - procedure of
systematically calculating, acquiring and recording Types of Probability Sampling
information about the members of a given
population. Simple random Sampling - an entirely Random
- deals with the investigation of the entire method of selecting the sample.
population. - sample selected from a population in such a
Sampling Method (Nonparametric) - selection of a manner that all members have an equal
subset of individuals from within a statistical chance of being selected.
population to estimate characteristics of the whole Systematic Sampling - Sample members from a
population. larger population are selected according to a
- a small group is selected as representative of the random starting point but with a fixed, periodic
whole population. interval.
 Sampling Interval - is calculated by dividing the Recommended minimum number of subjects for the
population size by the desired sample size. following types of studies:
Cluster sampling - e population is divided into smaller - 100 for a Descriptive Study
groups known as clusters then samples are - 50 for a Correlational Study
randomly selected among these clusters. - 30 in each group for Experimental and
- often used to study large populations, Causal-Comparative Study
particularly those that are widely
geographically dispersed. DETERMINATION OF SAMPLE SIZE
Stratified random Sampling - method of sampling that - Define population size.
involves dividing a population into smaller - Designate your margin of error.
groups–called strata. - Determine your confidence level.
- groups or strata are organized based on the - Predict expected variance.
shared characteristics or attributes of the - Finalize your sample size.
members in the group.
Stratification - process of classifying the population Sample Size Formula:
into groups

▪ Nonprobability Sampling - sampling method in which

not all members of the population have an equal
chance of participating in the study.

Types of Non-Probability Sampling

Convenience Sampling - sampling method where

research data are collected from a conveniently
available pool of respondents.
- most used sampling technique as it’s
incredibly prompt, uncomplicated, and
economical.
Purposive sampling - also known as judgmental,
selective, or subjective sampling.
- form of non-probability sampling in which
researchers rely on their own judgment
when choosing members of the population
to participate in their surveys. When to Use Larger Sample Sizes?
Judgment sampling - non-probability sampling → Number of subgroups
technique where the researcher selects units to → Population variance
be sampled based on his own existing → Desired precision
knowledge, or his professional judgment. → Confidence level
Quota sampling - refers to selecting participants that → Small error range
is a non-probabilistic version of stratified
sampling. Data - factual information (such as measurements or
Snowball sampling - researcher begins with a small statistics) used as a basis for reasoning,
population of known individuals and expands the discussion, or calculation.
sample by asking those initial participants to Variables (Data Item) - any characteristics, number, or
identify others that should participate in the study. quantity that can be measured or counted.

STEPS IN SAMPLING DESIGN TWO GENERAL TYPES OF DATA

- What is the target population?
- What are the parameters of interest? ▪ Qualitative Data - data that is describing qualities
- What is the sampling frame? or characteristics.
- What is the appropriate sampling method?
- What size sample is needed? Two General Types of Qualitative Data
Nominal Data (nominal scale) - data is classified
SAMPLE SIZE - samples should be as large as a without a natural order or rank. Ex. (gender,
researcher can obtain with a reasonable religion, country)
expenditure of time and energy.
 Ordinal Data (ordinal scale) - data is classified with Skewness – measure of asymmetry of the data
a natural order or rank. Ex. (educational level, distribution.
satisfaction level, political position) Kurtosis – measure of “tailedness” of the data
distribution.
▪ Quantitative Data - data that can be counted or Frequency Distribution – tabular or graphical
measured in numerical values. representation of data that shows the
number of times each value or category occurs in a
Two General Types of Quantitative Data dataset.
Continuous Data - data that can be measured on an - How often each unique data appears.
infinite scale. Data that can take any value. Ex. Histogram - graphical representation of data that
(height, weight, temperature, length) displays the frequency of values within specified
Discrete Data - data that can be measured on a finite intervals or bins.
scale. Data that can take certain value. Ex. (no. Frequency Table - tabular summary of data that
of students in a class, no. of patients in the shows the count or frequency of values in
hospital) various categories or interval.

INFERENTIAL STATISTICS CALCULATION

STATISTICS
Hypothesis Testing – comparing sample statistics to
population parameters to determine if observed
Statistics - discipline that concerns the collection,
differences or relationships are statistically
organization, analysis, interpretation, and presentation
significant.
of data.
Analysis of Variance (ANOVA) - partitioning the total
- used to describe and draw conclusions about the
variation.
data.
Chi Square Test – used for categorical data.
- used to estimate the likelihood that some factor
TWO MAJOR TYPES OF STATISTICS
other than chance accounts to the observed
relationship.
Descriptive Statistics - a brief descriptive coefficients
- comparing the expected frequency and observed
that summarize a given data set.
frequency.
- can be broken down into measures of central
Regression Analysis - calculating the probability of
tendency and measures of variability (spread).
occurrence of any phenomenon or for predicting the
Inferential Statistics - used to draw inferences or
phenomenon or relationship between different
conclusions about the characteristics of the
variables.
samples or population, and to decide how certain
- estimating the coefficient of regression model.
they can be of the reliability of those conclusions.
DESCRIPTIVE STATISTIC
Statistical Inference - process through which
Mean
inferences about a population are made based on
certain statistics calculated from a sample of data
drawn from that population.

DESCRIPTIVE STATISTIC CALCULATION

Median
Measures Of Central Tendency – statistical value that
describes the central or typical value of data set.
Mean – arithmetic average of dataset.
Median – middle value a dataset, ordered from
lowest to highest.
Range
Mode – most frequently occurring value.
Range = Max(x)-Min(x)
Measures of Variability – measure of dispersion or
Variance
spread.
Range – difference between maximum and minimum
values in dataset.
Variance – measure how data points vary from the
mean.
Standard Deviation
Standard Deviation – square root of variance.
Interquartile Range – range between quartile 1 &
quartile 3.
Measures of Distribution Shape – how data points are
spread or clustered.
Quartile - a type of quantile which divides the number evidence in favor of the alternative
of data points into four parts, or quarters, of more- hypothesis).
or-less equal size. - If the p-value is greater than α, you fail to
- data must be ordered from smallest to reject the null hypothesis (indicating
largest to compute quartiles. insufficient evidence to support the
Coefficient of Determination - statistical measurement alternative hypothesis)
that examines how differences in one variable can G. Draw Conclusions - draw conclusions about
be explained by the difference in a second variable, the null hypothesis and the research question.
when predicting the outcome of a given event.

Pearson Coefficient of Correlation

Inferential statistics provide:

Test for Difference - to test whether a significant
difference exists between groups
Tests for Relationship - to test whether a significant
relationship exist between a dependent variable
(Y) and independent variable (X) PROBABILITY

Key Steps in Hypothesis Testing

Probability - is an index which measures the chance or
A. Formulate Hypotheses likelihood that an event resulting from a statistical
- Null Hypothesis (H0): This is the default or experiment will occur.
status quo hypothesis. It typically represents - is expressed as the ratio of number of outcomes
a statement of no effect, no difference, or no pertaining to the desired event to the total number
relationship in the population. of possible outcomes.
- Alternative Hypothesis (Ha): This is the Statistical Experiment - is a process of using sample
hypothesis that the researcher is trying to information to draw conclusions about certain
support or demonstrate. It often represents characteristics of a population that is under study.
the research question or the statement that
there is an effect, difference, or relationship IMPORTANCE OF PROBABILITY AND STATISTICAL
in the population. INFERENCE
B. Collect Data - Gather data from a sample of the
population that is relevant to the research → The study of probability and statistical
question or hypothesis. inference aids in the translation of sample
C. Choose the Significance Level, Alpha - information into something conclusive or
significance level, often denoted by the symbol inconclusive about the scientific system under
α, is a pre-defined threshold that represents study.
the probability of making a Type I error → It allows us to quantify the strength or
(rejecting a true null hypothesis). “confidence” that we have in our conclusions.
- Common significance levels include 0.05 More importantly, the conclusion drawn about
(5 %) and 0.01 (1 %) the scientific system being studied helps an
D. Perform a Statistical Test - select an individual in the decision-making process.
appropriate statistical test based on the
research question, data type, and EXPERIMENT - is a process which generates an
assumptions. observation, or a measurement commonly referred
E. Calculate the p-value - probability value, or the to as raw data or information.
observed or computed significance level. An experiment is considered to be a random or
- smaller p-values indicate stronger evidence chance process if:
against the null hypothesis. - all possible outcomes are known before the
F. Make a Decision - Compare the calculated p- experiment is performed.
value to the chosen significance level (α). - no particular outcome can be predicted with
- If the p-value is less than or equal to α, you certainty before the performance of the trial or
reject the null hypothesis (indicating experiment.
SAMPLE SPACE AND SAMPLE POINT
Terms used to imply intersection of events:
Sample Space (S) - is a list of all possible outcomes of ▪ either event A or event B will occur
a statistical experiment defined in such a way that ▪ one or the other will occur
exactly one of the elements will occur. ▪ at least one of the two events will occur
Sample Point (X) - also known as member or element ▪ one of the two events will occur
in the sample space. ▪ a/an event will occur
- refers to each possible outcome in the sample Let S = {a, b, c, d, e}
space. A = {a, b, c} B = {b, c, d, e}
- the sample space has a finite number of Then, A U B = {a, b, c, d, e}
elements, we may list the members separated by
commas and enclosed in braces. Intersection, A ∩ B - set whose members are
*The sample space S, of possible outcomes when a elements of both A and B.
coin is flipped, may be written as: S = {H, T} E - means that both events will happen at the same
time
THREE WAYS OF DEFINING SAMPLE SPACE Terms used to imply intersection of events:
Listing or Roster Method - is done by enumerating all ▪ both A and B will happen
the elements of the sample space. ▪ neither event A nor B will happen
Tree Diagram Method - uses a “tree” to express the Let S = {a, b, c, d, e}
sequence of events of an experiment in A = {a, b, c} B = {b, c, d, e}
chronological order. Then, A ∩ B = {b, c}
Defining Property or Rule Method - is done by choosing
a property or characteristic common to all sample Disjoint Event - two events cannot happen at the
points and then using this common characteristic to same time, they are said to be mutually exclusive
define the sample space. events or disjoint events.
Null Set, Φ or { } - also known as empty set or
*Format: S = { x | x is (statement) } where | is read “such impossible event.
that”. (Sample space for larger number of possible Complement, A’ - set of all elements not found in A.
outcome) Let S = {1, 2, 3, 4, 5, 6}
* Format: S = {}. “phi” (sample space for nothing) A = {1, 3, 4}
Then, A’ = {2, 5, 6}
Event - a subset of a sample space
Simple event - an event which consists of only one Several results that follow from the foregoing
outcome. definition, which may easily be verified by means of
Compound Event - an event which consists of more Venn diagrams, are as follows:
than one outcome and may be decomposed into ▪ A∩Φ=Φ
simple events ▪ AUΦ=A
*Roll a die and observe the number appearing on the ▪ A ∩ A’ = Φ
top face. ▪ A U A’ = S
The sample space, S, is given as S = {1, 2, 3, 4, 5,6}. ▪ S’ = Φ
The following are some of the possible events ▪ Φ’ = S
▪ (A’)’= A
▪ (A ∩ B)’ = A’ U B’
▪ (A U B)’ = A’ ∩ B’

Events A and B are known as compound events.

Events C to H are simple events

Operations on Events
Venn Diagram - representations of the union between
two events, intersection between two events,
mutually exclusive events, and the complement of
an event.

Union, A U B - set whose elements belong to A or B or

both includes the possibility that only event A will
happen or only event B will happen, or both A and
B will happen at the same time.