INTRODUCTION AND INFORMATION TO 2.

DEPENDENT VARIABLES
STATISTICS
 A dependent variable relies on and can
DESCRIPTIVE AND INFERENTIAL be changed by other components.
 A grade on an exam is an example of a
 Descriptive statistics describes data (for
dependent variable because it depends
example, a chart or graph)
on factors such as how much sleep you
 Inferential statistics allows you to make
got and how long you studied.
predictions (inferences) from that data.
 Independent variables can influence
 With inferential statistics, you take data
dependent variables, but dependent
from samples and make generalizations
variables cannot influence independent
about a population.
variables.
VARIABLES  For example, the time you spent
studying (dependent) can affect the
 These are things you measure, grade on your test (independent) but
manipulate and control in statistics and the grade on your test does not affect
research. the time you spent studying.
 All studies analyze a variable, which can
describe a person, place, thing or idea. 3. INTERVENING VARIABLES
 A variable’s value can change between
 An intervening variable, sometimes
groups or overtime.
called a mediator variable, is a
 For example, if the variable in an
theoretical variable the researcher uses
experiment is a person’s eye color, its
to explain a cause or connection
value can change from brown to blue to
between other study variables usually
green from person to person.
dependent and independent ones. They
TYPES OF VARIABLES: are associations instead of
observations.
1. INDEPENDENT VARIABLES  For example, if wealth is the
 An independent variable is a singular independent variable, and a long-life
characteristic that the other variables in span is a dependent variable, the
your experiment cannot change. researcher might hypothesize that
 Age is an example of an independent access to quality healthcare is the
variable. intervening variable that links wealth
 Where someone lives, what they eat or and life span.
how much they exercise are not going 4. MODERATING VARIABLES
to change their age.
 Independent variables can, however,  A moderating or moderator variable
change other variables. changes the relationship between
 In studies, researchers often try to find dependent and independent variables
out whether an independent variable by strengthening or weakening the
causes other variables to change and in intervening variable's effect.
what way.  For example, in a study looking at the
relationship between economic status
(independent variable) and how
frequently people get physical exams
 from a doctor (dependent variable), age amounts. Examples might include
is a moderating variable. height, distance or number of
 That relationship might be weaker in items. Researchers can further
younger individuals and stronger in categorize quantitative variables
older individuals. into two types:
o Discrete: Any numerical variables
5. CONTROL VARIABLES
you can realistically count, such as
 Control or controlling variables are the coins in your wallet or the
characteristics that are constant and do money in your savings account.
not change during a study. o Continuous: Numerical variables
 They have no effect on other variables. that you could never finish
 Researchers might intentionally keep a counting, such as time.
control variable the same throughout
8. QUALITATIVE VARIABLES
an experiment to prevent bias.
 For example, in an experiment about  Qualitative, or categorical, variables are
plant development, control variables non-numerical values or groupings.
might include the amounts of fertilizer Examples might include eye or hair
and water each plant gets. These color. Researchers can further
amounts are always the same so that categorize qualitative variables into
they do not affect the plants' growth. three types:
o Binary: Variables with only two
6. EXTRANEOUS VARIABLES
categories, such as male or female, red
 Extraneous variables are factors that or blue.
affect the dependent variable but that o Nominal: Variables you can organize in
the researcher did not originally more than two categories that do not
consider when designing the follow a particular order. Take, for
experiment. example, housing types: Single-family
 These unwanted variables can home, condominium, tiny home.
unintentionally change a study's results o Ordinal: Variables you can organize in
or how a researcher interprets those more than two categories that follow a
results. particular order. Take, for example,
 Take, for example, a study assessing level of Satisfaction: Unsatisfied,
whether private tutoring or online neutral, satisfied.
courses are more effective at improving
students' Spanish test scores. 9. CONFOUNDING VARIABLES
 Extraneous variables that might  A confounding variable is one you did
unintentionally influence the outcome not account for that can disguise
include parental support, prior another variable's effects.
knowledge of a foreign language or  Confounding variables can invalidate
socioeconomic status. your experiment results by making
7. QUANTITATIVE VARIABLES them biased or suggesting a
relationship between variables exists
 Quantitative variables are any data when it does not.
sets that involve numbers or
  For example, if you are studying the o Interviews.
relationship between exercise level  A primary source is collected directly
(independent variable) and body mass from the original source. It is not
index (dependent variable) but do not clouded with someone else’s views or
consider age's effect on these factors, it judgments.
becomes a confounding variable that
SECONDARY DATA
changes your results.
 It is a data that gathered from studies,
10. COMPOSITE VARIABLES
surveys, or experiments that have been
 A composite variable is two or more run by other people or for other
variables combined to make a more research. Examples of secondary data
complex variable. include:
 Overall health is an example of a o Encyclopedias
composite variable if you use other o Essays
variables, such as weight, blood o Newspaper opinion pieces
pressure and chronic pain, to determine o Reviews
overall health in your experiment. o Textbooks
DATA  Typically, a researcher will begin a
project by working with secondary data.
 Facts that can be analyzed or used in an This allows time to formulate questions
effort to gain knowledge or make and gain an understanding of the issues
decisions; information. being dealt with before the more costly
 From the Latin datum, meaning what is and time-consuming operation of
given. collecting primary data.
 Data can take the form of characters
such as letters, numbers, punctuation NOMINAL SCALE
marks, mathematical operators, and  A nominal scale is a scale (of
control characters. Data also can take measurement) that uses labels to
the form of photographic display classify cases (measurements) into
elements, such as pixels. classes.
TYPES OF DATA:  Some examples of variables that use
nominal scales would be religious
PRIMARY DATA affiliation, sex, the city where you live,
etc.
 It is a data that is collected by a
researcher from first-hand sources, ORDINAL SCALE
using methods like:
o Creative works (paintings, movie reels,  Ordinal scale is measurement that
music etc.) reports the ranking and ordering of the
o Diaries data without actually establishing the
degree of variation between them.
o Experiments performed by you, the
Ordinal level of measurement is the
researcher
second of the four measurement scales.
o Letters
o Surveys and censuses
  Example:

“How satisfied are you with our products?”

1- Totally Satisfied
2- Satisfied
3- Neutral
4- Dissatisfied
5- Totally Dissatisfied

INTERVAL SCALE

 It is defined as a numerical scale where

the order of the variables is known as
well as the difference between these
variables.
 Interval indicates distance between two
entities, which is what Interval scale
helps in achieving.
 Example: 80 degrees is always higher
than 50 degrees and the difference
between these two temperatures is the
same as the difference between 70
degrees and 40 degrees.

RATIO SCALE

 It is defined as a variable measurement

scale that not only produces the order
of variables but also makes the
difference between variables known
along with information on the value of
true zero.
 Example:

What is your weight in kilograms?

 Less than 50 kilograms

 51- 70 kilograms
 71- 90 kilograms
 91-110 kilograms
 More than 110 kilograms

DATA COLLECTION
  Data collection is a process of collecting 1. Focus group
information from all the relevant
 One popular research interview method
sources to find answers to the research
is conducting a focus group interview,
problem, test the hypothesis (if you are
which involves a group of individuals
following deductive approach) and
interviewed at the same time.
evaluate the outcomes.
 Data collection methods can be divided 2. Structured interview
into two categories: secondary methods
of data collection and primary methods  Structured interviews are another
of data collection. option. Typically, structured interviews
comprise closed-ended questions,
SECONDARY DATA COLLECTION METHODS which are questions that respondents
can answer with "yes" or "no."
 Secondary is a type of data that already
been published in books, newspaper, 3. Unstructured interview
magazines, journals, online portals etc.
 An unstructured interview, also called
PRIMARY DATA COLLECTION METHODS an informal interview, is the opposite of
a structured interview. In unstructured
 Primary data is the type of data that has
interviews, the interviewer doesn't ask
not been around before. Primary data is
standardized questions of each
unique findings of your research.
interviewee
 Primary data collection methods can be
divided into two groups: quantitative 4. Semi-structured interview
and qualitative.
 You can also use a semi-structured
TWO GROUPS IN PRIMARY DATA COLLECTION interview method, which combines
pieces of both structured and
 Quantitative data is expressed in
unstructured interviews. Although
numbers and graphs and is analyzed
interviewers might follow a general
through statistical methods
plan and set of questions, they often
 Qualitative data is expressed in words
have the flexibility to make changes.
and analyzed through interpretations
and categorizations. 5. Personal interview
INTERVIEW METHODS  A personal interview takes place in
person as a one-on-one interaction
 Interview methods in research are
between an interviewer and an
different approaches you can use to
interviewee. Personal interviews are
conduct effective research interviews.
ideal if you want to speak directly to an
 This is the process of interview b/w
individual and cater your questions to
interviewer and interviewee.
them.

6. Phone interview
7 INTERVIEW METHODS IN RESEARCH
  You can also conduct interviews over  Unsuitable for people with limited
the phone. Phone interviews can be an literacy or verbal skills
easy way to gather responses.  Susceptible to a nonresponse bias
(most people invited may not complete
7. Online interview
the questionnaire)
 Online interviews are another research  Biased towards people who volunteer
interview option. Online interviews can because impersonal survey requests
involve surveys or video chat often go ignored
applications.
RESEARCHER-ADMINISTERED
QUESTIONNAIRE METHOD QUESTIONNAIRES

 A questionnaire is a list of questions or  Researcher-administered

items used to gather data from questionnaires are interviews that take
respondents about their attitudes, place by phone, in-person, or online
experiences, or opinions. between researchers and respondents.
 Questionnaires can be used to collect
Researcher-administered questionnaire can:
quantitative and/or qualitative
information.  Help you ensure the respondents are
 Questionnaires can be self- representative of your target audience
administered or researcher-  Allow clarifications of ambiguous or
administered. Self-administered unclear questions and answers
questionnaires are more common  Have high response rates because it’s
because they are easy to implement harder to refuse an interview when
and inexpensive, but researcher- personal attention is given to
administered questionnaires allow respondents
deeper insights.
But researcher-administered questionnaires can
SELF-ADMINISTERED QUESTIONNAIRES be limiting in terms of resources. They are:

 Self-administered questionnaires can be  Costly and time-consuming to perform

delivered online or in paper-and-pen  More difficult to analyze if you have
formats, in person or through mail. qualitative responses
 Likely to contain experimenter bias or
Self-administered questionnaires can be:
demand characteristics
 Cost-effective  Likely to encourage social desirability
 Easy to administer for small and large bias in responses because of a lack of
groups anonymity
 Anonymous and suitable for sensitive
OPEN-ENDED VS CLOSED-ENDED QUESTIONS
topics
 Self-paced CLOSE-ENDED QUESTIONS

 Closed-ended, or restricted-choice,
questions offer respondents a fixed set
of choices to select from. Closed-ended
But they may also be:
 questions are best for collecting data on o Neither satisfied nor dissatisfied
categorical or quantitative variables. o Somewhat satisfied
o Very satisfied
EXAMPLES OF CLOSED-ENDED QUESTIONS FOR
DIFFERENT VARIABLES OPEN-ENDED QUESTIONS
NOMINAL VARIABLES – include categories that  Open-ended, or long-form, questions
can’t be ranked, such as race or ethnicity. This allow respondents to give answers in
includes binary or dichotomous categories. their own words
Examples: Examples:
What is your race? 1. How do you feel about open science?
o White 2. How would you describe your personality?
o Black of African American
3. In your opinion, what is the biggest obstacle
o American Indian or Alaska native
for productivity in remote work?
o Asian
RGISTRATION METHOD
Are you satisfied with the current work-from-
home policies?  Respondents provide information in
compliance with certain laws, policies,
o Yes
or standard practices
o No
Examples: marriage contract, birth certificate,
ORDINAL VARIABLES – include categories that
cedula, etc.
can be ranked. Consider how wide or narrow a
range you’ll include in your response items, and EXPERIMENTAL METHOD
their relevance to your respondents.
 The experimental method involves
Examples: manipulating one variable to determine
if changes in one variable cause
What is your age?
changes in another variable.
o 15 or younger
OBSERVATION METHOD
o 16-35
o 36-60  The observation method is described as
o 61-75 a method to observe and describe the
o 76 or older behavior of a subject. As the name
suggests, it is a way of collecting
LIKERT-TYPE QUESTIONS - collect ordinal data relevant information and data by
using rating scales with 5 or 7 points. observing.
Examples: OBSERVATION METHOD IN DATA COLLECTION
How satisfied or dissatisfied are with your CAN BE:
online shopping experience today?  Structured observation method – This
o Very dissatisfied is a systematic observation method
o Somewhat dissatisfied where data is collected as per a pre-
defined schedule. The specific variable
 is used in this method for data
collection.

 Unstructured observation method –

The unstructured observation method
is conducted in a free and open manner
without using any pre-determined
objectives, schedules or variables.
DETERMINING THE SAMPLE SIZE: SLOVIN’S Sample computation:
FORMULA
 Determine the error of tolerance
SAMPLING
5% or 0.05
 Sampling is the process of selecting few
 Identify the population size
(a sample) from a bigger group a
(population) to become the basis for 186 Grade 11 ABM students
predicting the prevalence of an
unknown piece of information,  Compute for the sample population
situation, or outcome, regarding the Formula:
bigger group.
 “The Larger the sample size the more n = N / (1 + Ne²)
accurate the findings” where:
SLOVIN’S FORMULA n = Number of samples ?
Slovin’s Formula is used to compute for the N = Total population 186
sample size in a study given the total population
and margin of error: E = Error tolerance (level) 0.05

Formula:

n = 186 / (1) + (186) (0.05²)

n = 186 / (1) + (186) (0.0025)

Formula:
n = 186 / (1) + (0.465)
n = N / (1 + Ne²)
n = 186 / 1.46
where:
n = 126.962
n = Number of samples
n = 127
N = Total population
Suppose that a group of 1,000 city government
E = Error tolerance (level) employees needs to be surveyed to find out
which tools are best suited to their jobs. For this
TOLERANCE LEVEL
survey a margin of error of 0.08 is considered
 It is the level of confidence that indicate sufficiently accurate.
that data is reflective of the entire
Formula:
population.
n = 1000 / (1) + (1000) (0.08²)
How do you compute for the sample size?
n = 1000 / (1) + (1000) (0.0064
 Determine the error of tolerance;
 Identify the population size; n = 1000 / (1) + (6.4)
 Compute for the sample population.
n = 1000 / 7.4

n = 135.135
SAMPLING TECHNIQUE 2. SYSTEMATIC SAMPLING

 Population is the entirety of the group o Systematic sampling draws a random

including all the members that forms a sample from the target population by
set of data. selecting units at regular intervals
 Sample contain a few members of the starting from a random point.
population. They were taken to o This method is useful in situations
represent the characteristics or traits of where records of your target population
the whole population. already exist, such as records of an
agency’s clients, enrollment lists of
PROBABILITY SAMPLING
university students, or a company’s
 In Probability sampling, every member employment records.
of the population has the chance of
being selected. It involves principles or
randomization or chance.

TYPES OF PROBABILITY SAMPLING:

1. SIMPLE RANDOM SAMPLING

o Simple random sampling gathers a

random selection from the entire
population, where each unit has an 3. STRATEFIED SAMPLING
equal chance of selection. o Stratified sampling collects a random
o To compile a list of the units in your selection of a sample from within
research population, consider using a certain strata, or subgroups within the
random number generator. population. Each subgroup is separated
Examples: from the others on the basis of a
common characteristic, such as gender,
 You are researching the political views race, or religion.
of a municipality of 4,000 inhabitants.
 You have access to a list with all 4,000
people, anonymized for privacy
reasons.
 You have established that you need a
sample of 100 people for your research.

4. CLUSTER SAMPLING

o Cluster sampling is the process of

dividing the target population into
groups, called clusters. A randomly
selected subsection of these groups
then forms your sample.
 o Cluster sampling is an efficient
approach when you want to study
large, geographically dispersed
populations. It usually involves existing
groups that are similar to each other in
some way (e.g., classes in a school).

There are two types of cluster sampling:

TYPES OF NON-PROBABILITY SAMPLING
 Single (or one-stage) cluster sampling,
when you divide the entire population 1. CONVENIENCE SAMPLING
into clusters
 Multistage cluster sampling, when you o Convenience sampling is primarily
divide the cluster further into more determined by convenience to the
clusters, in order to narrow down the researcher.
sample size This can include factors like:

 Ease of access
 Geographical proximity
 Existing contact within the population
of interest

2. QUOTA SAMPLING

o In quota sampling, you select a

predetermined number or proportion
MULTI-STAGE SAMPLING of units, called a quota. Your quota
should comprise subgroups with
o Multi-stage sampling is a more complex specific characteristics (e.g., individuals,
form of cluster sampling, in which cases, or organizations) and should be
smaller groups are successively selected selected in a non-random manner.
from larger populations to form the
3. SNOWBALL SAMPLING
sample population used in your study.
o Snowball sampling is used when the
NON-PROBABILITY SAMPLING
population you want to research is hard
o Non-probability sampling is a sampling to reach, or there is no existing
method that uses non-random criteria database or other sampling frame to
like the availability, geographical help you find them. Research about
proximity, or expert knowledge of the socially marginalized groups such as
individuals you want to research in drug addicts, homeless people, or sex
order to answer a research question. workers often uses snowball sampling.

4. PURPOSIVE (JUDGEMENTAL) SAMPLING

o Purposive Sampling is a non –

probability technique where the
 researcher relies on the discretion to  All people that play football who are
choose for the sample population. older than 45

GATHERING DATA

 Gathering data is the first step in

statistical analysis.
 Say for example that you want to know
something about all the people in
France.
STATISTIC INTRODUCTION

WHAT IS STATISTICS USED FOR?

 The population is then all of the people
 Statistics is used in all kinds of science
in France
and business applications.
 It is often much easier to collect a
 Statistics gives us more accurate
smaller group of that population and
knowledge which helps us make better
analyze that. This is called a sample.
decisions.
 Statistics can focus on making A REPRESENTATIVE SAMPLE
predictions about what will happen in
 The sample needs to be similar to the
the future.
whole population of France.
 It can also focus on explaining how
 It should have the same characteristics
different things are connected.
as the population.
TYPICAL STEPS OF STATISTICAL METHODS  If you only include people named
Jacques living in Paris who are 48 years
The typical steps are:
old, the sample will not be similar to the
 Gathering data whole population.
 Describing and visualizing data  A good representative sample is crucial
 Making conclusions for statistical methods.

HOW IS STATISTICS USED? Note: Data from a proper sample is often just as
good data from the whole population, as long
 Statistics can be used to explain things
as it is representative
in a precise way. You can use it to
understand and make conclusions DESCRIBING DATA
about the group that you want to know
DESCRIPTIVE STATISTICS
more about. This group is called the
population.  The information (data) from your
sample or population can be visualized
A population could be many different kinds of
with graphs or summarized by
groups. It could be:
numbers.
 All of the people in a country  This will show key information in a
 All the businesses in an industry simpler way than just looking at raw
 All the customers of a business data. It can help us understand how the
data is distributed.
EXAMPLES OF GRAPHS INCLUDES:  Probability theory is used to calculate
the certainty that those statistics also
 Histograms
apply to the population.
 Pie charts
 Bar graphs CONFIDENCE INTERVALS
 Box plots
 Confidence intervals are numerical
ways of showing how likely it is that the
true value of this statistic is within a
certain range for the population.
QUARTILES
HYPOTHESIS TESTING – is another way of
 Quartiles are the data split into four checking if a statement about a population is
equal size parts, or quarters. true.

Some examples of statements or questions that

can be checked with hypothesis testing:

o People in the Netherlands taller than

people in Denmark
o Do people prefer Pepsi or Coke?
o Does a new medicine cure a disease?

Note: Confidence intervals and hypothesis

SUMMARY STATISTICS testing are closely related and describe the
same things in different ways. Both are widely
These are individual 'statistics’. Some important used in science.
examples are:
CASUAL INFERENCE
 Mean, median and mode
 Range and interquartile range Causal inference is used to investigate if
 Quartiles and percentiles something causes another thing.
 Standard deviation and variance Example:
Note: Descriptive statistics is often presented as o Does rain make plants grow?
a part of statistical analysis. Descriptive
statistics is also useful for guiding further Note: Good experimental design is often
analysis, giving insight into the data, and difficult to achieve because of ethical concerns
finding what is worth investigating more or other practical reasons.
closely. PREDICTION AND EXPLANATION
STATISTICAL INFERENCE PREDICTION
 Statistics from the data in the sample is  Some statistical methods are not
used to make conclusions about the focused on explaining how things are
whole population. connected. Only the accuracy of
 This is a type of statistical inference. prediction is important.
  Some types of machine learning let PARAMETERS AND STATISTICS
computers do the hard work, but the
Parameter
way they predict is difficult to
understand.  A number that describes something
about the whole population.
Note: Predictions about future events are called
forecasts. Not all predictions are about the Sample statistics
future.
 A number that describes something
about the sample.

EXPLANATION

 Different statistical methods are often

used for explaining how things are
connected. These statistical methods
may not make good predictions.
 If these methods accurately explain
how all the relevant things are
MEAN, MEADIAN AND MODE
connected, they will also be good at
prediction. But managing to explain  are different types of averages (typical
every detail is often challenging. values in a population).
 Sometimes we are specifically
interested in figuring out if one thing For example:
causes another. This is called causal  The typical age of people in a country
inference.  The typical profits of a company
Note: Making conclusions about causality  The typical range of an electric car
should be done carefully. VARIACE AND STANDARD DEVIATION
POPULATIONS AND SAMPLES  are two types of values describing how
Population spread out the values are.

 Everything in the group that we want to STUDY TYPES

learn about.  A statistical study can be a part of the
Sample process of gathering data.
 There are different types of studies.
 A part of the population. Some are better than others, but they
might be harder to do.
Examples:
MAIN TYPES OF STATISTICAL STUDIES

 The main types of statistical studies are

observational and experimental studies.
 Experimental studies are generally
better than observational studies for
 investigating this, but usually require  Quantitative Data
more effort.
Information about something that is described
 An observational study is when observe
by numbers.
and gather data without changing
anything. MEASUREMENT LEVELS
EXPERIMENTAL STUDIES  Nominal Level
 Ordinal level
 In an experimental study, the
 Interval Level
circumstances around the sample is
 Ratio Level
changed.
 Experimental studies can allow us to
investigate causal relationships.
 A well-designed experimental study can
be useful since it can isolate the
relationship we are interested in from
other effects.

SAMPLE TYPES

 A study needs participants and there

are different ways of gathering them.
 Some methods are better than others,
but they might be more difficult.

DIFFERENT TYPES OF SAMPLING METHODS

 Random Sampling
 Convenience Sampling
 Systematic Sampling
 Stratified Sampling
 Clustered Sampling

DATA TYPES

 There are two main types of data:

 Qualitative (or 'categorical') and
quantitative (or 'numerical’).
 These main types also have different
sub-types depending on their
measurement level.

DIFFERENT TYPES OF DATA

 Qualitative Data

Information about something that can be

sorted into different categories that can't be
described directly by numbers.