from extremely large to very small

PSYCHOLOGICAL STATISTICS depending on how the researcher defines

the population.
WHY STUDY STATISTICS?
STATISTICAL RESEARCHES MUST BE S.M.A.R.T
We will gain the ability to comprehend psychology research articles.
SAMPLE (n) set of individuals selected from a population,
Essential for understanding and conducting your own research. usually intended to represent the population in a
Respondents research study
Enhances your analytical and critical thinking skills.
intended to be representative of its population,
To see the norms using statistics and standardize test according to the result and a sample should always be identified in terms
of the population from which it was selected. (Ex:
To be able to use it in assessing the needs of other people (needs of 10 participants for each college program
assessment)
To answer the research question, a researcher
GOAL OF PSYCHOLOGY studies a sample and then generalizes the results
from the sample to the population.
DESCRIBE (organize your data) - PREDICT (hypothesize) - EXPLAIN
(cause and effect) - CONTROL (experimental) PARAMETER a value, usually a numerical value, that describes
a population.
STATISTICS set of mathematical procedures for organizing,
summarizing, and interpreting information. usually derived from measurements of the
individuals in the population.
TWO GENERAL PURPOSES OF STATISTICS
Every population PARAMETER has a
1. used to organize and summarize the information so that the corresponding sample statistic, and most research
researcher can see what happened in the research study and studies involve using statistics from samples as
can communicate the results to others. the basis for answering questions about
population parameters.
2. help the researcher to answer the questions that initiated the
research by determining exactly what general conclusions are STATISTIC value, usually a numerical value, that describes a
justified based on the specific results that were obtained. sample
Numerical value
POPULATION (N) entire set of individuals of interest for a usually derived from measurements of the
of sample
particular research question. individuals in the sample
The population need not consist of people—it
could be a population of rats, corporations, parts
produced in a factory, or anything else a
researcher wants to study.
● The population being studied should
always be identified by the researcher.
● Populations can obviously vary in size
 DATA SET is a collection of measurements or observations.

DESCRIPTIVE are statistical procedures used to summarize,

STATISTICS organize, and simplify data.
> only organized the data, you cannot make
inferences
> you can predict the factors, but can’t see what
really causes the change in the variable
> can help identify statistical tool to use

INFERENTIAL consist of techniques that allow us to study

STATISTICS samples and then make generalizations
(justifications) about the populations from which
they were selected.
Typically, researchers use sample statistics as the
basis for drawing conclusions about population
parameters.

VARIABLE is a characteristic or condition that changes or

has different values for different individuals. (Ex:
height, weight, gender, personality, temperature,
and time of the day etc.)
> we control the independent variable (can cause
change to dependent variable)

RESPONDENTS Quantitative

PARTICIPANTS Qualitative

VALUES possible number of categories that a score can

have.
(ex: 0-63 in Beck depression scale, male and
female, stress scale)

SCORES/DATUM raw scores is a particular value on a variable.

SAMPLING ERROR the naturally occuring discrepancy, or error, that
DATUM (singular) is a single measurement or observation exists between a sample statistic and the
and is commonly called a score or raw score. corresponding population parameter.
> make sure your sample is a representative of
DATA (plural) are measurements or observations. your population in the effect size to actually see
the representation
 IS YOUR VARIABLE AN INTERNAL CHARACTERISTICS? Usually discrete and categorize observations, but do not make
any quantitative distinctions between
CONSTRUCTS also known as hypothetical constructs, are observations
internal attributes or characteristics that cannot
Not tangible be directly observed but are useful for describing ex: type of fruits, gender (male, female,
and explaining behavior. (ex: hunger, anxiety, non-binary),
intelligence etc.). blood types (A, B, AB, O), or car brands (Toyota,
Ford, BMW).
OPERATIONAL identifies a measurement procedure (a set of
DEFINITION operations) for measuring an external behavior Categorical — Labels ONLY
(represent the construct/ manifestation) and uses No Ranking — No Arithmetic Operations
the resulting measurements as a definition and a
measurement of an internal construct. ORDINAL SCALE consists of a set of categories that are organized
in an ordered sequence. Measurements on an
Note that an operational definition has two components: MAGNITUDE ordinal scale rank observations in terms of size or
1. It describes a set of operations for measuring construct. (more-ness) magnitude.
(ex: intelligence - intelligent behaviors - IQ scores)
2. It defines the construct in terms of the resulting measurements ex: t-shirt sizes, socioeconomic status, religion,
civil status > one item likert scale
NATURE OF VARIABLE
Ordered Categories
DISCRETE consists of separate, indivisible categories. No Direction of difference (e.g., customer satisfaction
VARIABLE values can exist between two neighboring ratings: dissatisfied, neutral, satisfied).not size of
categories the difference.
ex: class attendance, gender, college programs Unequal Intervals
Limited Arithmetic Operations
CONTINUOUS there are an infinite number of possible values
VARIABLE that fall between any two observed values. INTERVAL SCALE consists of ordered categories that are all
- is divisible into an infinite number of intervals of exactly the same size. Equal
fractional parts. MAGNITUDE differences between numbers on the scale reflect
ex: Height of individuals (e.g., 170.5 cm, 182.2 cm), (more-ness) equal differences in magnitude. However, the zero
Temperature (e.g., 36.6°C, 98.4°F), Time taken to point on an interval scale is arbitrary and does
run a marathon (e.g., 2.75 hours, 3.1 hours) EQUAL INTERVAL not indicate a zero amount of the variable being
(equal distance) measured.
LEVELS OF MEASUREMENTS
ex: temperature, IQ scores, calendar years,
LEVELS OF A classification that describes the nature of standardized test scores, bet depression scale
MEASUREMENT yielded scores or information by a variable. This > 10 item likert scale - examining their differences
can suggest which scale of measurement should in the responses (interval)
be used in statistics.
Ordered Data with Equal Intervals
NOMINAL SCALE consists of a set of categories that have different No True Zero Point (e.g., temperature in Celsius or
names. Measurements on a nominal scale label Fahrenheit).
 Meaningful Differences/equal differences: (e.g., ● Variable/s are described by descriptive statistics
the difference between 30°C and 40°C is the ● May use category and/or numerical variables (ex: study time
same as between 70°C and 80°C). and academic performance of 2nd year college students in
psychology)
RATIO SCALE interval scale with the additional feature of an
absolute zero point. With a ratio scale, ratios of DESCRIPTIVE statistical procedures used to summarize,
MAGNITUDE numbers do reflect ratios of magnitude. STATISTICS organize, and simplify data. (ex: survey research
(more-ness) to obtain data such as preference, opinions,
EQUAL INTERVAL ex: basketball scores, height, weight, age, attitudes of the respondents)
allowance, test score
ABSOLUTE ZERO CORRELATIONAL Two different variables are observed to
Ordered Data with Equal Intervals METHOD determine whether there is a relationship
True Zero Point between them.
Meaningful Differences and Ratios
All arithmetic operations are possible, including RELATIONSHIP Simply measure the two (or more) variables for
addition, subtraction, multiplication, and division. BETWEEN each individual.
VARIABLES Data structure use to determine the relationship
between variables

DATA STRUCTURE II: ONE GROUP WITH ONE OR MORE VARIABLES

MEASURED FOR EACH INDIVIDUAL

● A single (one) group of participants is observed, with two

variables measured for each individual.
● The aim is to describe the type and strength of the relationship
between these variables.
● Patterns in the data are analyzed to identify any existing
relationships
● A non-experimental study method.
● ex: The relationship of Academic performance (GPA) with study
time, attendance, and participation of 2nd-year college students
1. Level of psychological statistics anxiety - ordinal
2. Psychological statistics anxiety scores - interval ATTENDANCE ——————— ACADEMIC PERFORMANCE (GPA)
3. Scores in bet depression scale - interval X Y
4. Psychological statistics quiz - ratio
⬆️ Y ⬆️
⬆️ Y ⬇️
5. Volleyball shirt number - nominal TYPES OF POSITIVE CORRELATION - X
CORRELATION NEGATIVE CORRELATION - X
DATA STRUCTURE AND QUANTITATIVE RESEARCH METHOD NO CORRELATION

DATA STRUCTURE I: ONE GROUP WITH ONE OR MORE VARIABLES STRENGTH OF Strong
MEASURED FOR EACH INDIVIDUAL RELATIONSHIP Moderate
Weak
● One or more variables measured per individuals
 CORRELATION DOES NOT MEAN CAUSATION actually measured.
You should realize that this is different from a
Therefore, we cannot conclude that solely attendance either improves or correlational study, in which both variables are
negatively impacts academic performance. measured and the data consist of two separate
scores for each individual.

LIMITATIONS OF It can demonstrate the presence of a relationship

CORRELATIONAL between variables
METHOD: Does not provide explanation for the
relationship
Does not imply a cause-and-effect connection.

DATA STRUCTURE III: COMPARING TWO OR MORE GROUPS OF

SCORES
INDEPENDENT The variable that is manipulated by the
● One variable defines the groups, VARIABLE researcher.
● While scores are measured on a second variable. In behavioral research, it usually consists of the
● This structure is used in both experimental and non-experimental two (or more) treatment conditions/levels to
studies. which subjects are exposed.
It consists of the antecedent conditions that are
EXPERIMENTAL ● One variable is manipulated while another manipulated prior to observing the dependent
METHOD variable is observed and measured. variable.
● To establish a cause-and-effect
relationship between the two variables, an TREATMENT CONDITIONS / LEVELS
experiment attempts to control all other
variables to prevent them from influencing EXPERIMENTAL Individuals in the experimental condition do
the results. GROUP receive the experimental treatment
An experimental study evaluates the relationship
between two variables by manipulating one CONTROL GROUP Individuals in a control condition do not receive
variable (the independent variable) and the experimental treatment. Instead, they either
measuring one variable (the dependent receive no treatment or they receive a neutral,
variable). placebo treatment.
Note that in an experiment only one variable is The purpose of a control condition is to provide a
baseline for comparison with the experimental
 condition. a result, there are a number of other research
designs that compare groups of scores but are
DEPENDENT variable that is observed to assess the effect of not true experiments.
VARIABLE the treatment.
Control condition QUASI- There are also nonexperimental studies that
EXPERIMENTAL compare two or more groups of scores.
EXPERIMENTAL METHOD: ESSENTIALS ELEMENTS IN EXPERIMENT DESIGN Take note: there is a fine line between
correlational methods and quasi experimental
MANIPULATION The researcher manipulates one variable by (association vs. comparison).
changing its value from one level to another. A ex: academic stress and test scores of 2nd year
second variable is observed (measured) to psychology students
determine whether the manipulation causes
changes to occur. can seem like an experiment, but they lack one or
more of its essential elements, such as
RANDOM a process used in experimental research to manipulation of the IV or random assignment to
ASSIGNMENT allocate participants or subjects to different treatment conditions.
groups (such as treatment or control groups) in a as the word implies “quasi” - “as if” or “almost”
random manner. This method is crucial for
minimizing bias and ensuring that each QUASI - In a nonexperimental study, the “independent”
participant has an equal chance of being INDEPENDENT variable that is used to create the different
assigned to any group. VARIABLE groups of scores

WHY WE RANDOMLY ASSIGN OUR SUBJECTS? NONEQUIVALENT The researcher compares the effect of different
GROUP DESIGN treatment conditions to a PRE-EXISTING
SUBJECT These are characteristics such as age, gender, GROUPS.
VARIABLES and intelligence that vary from one individual to To evaluate the effects of an intervention by
another. comparing groups.

CONTROL OF The researcher must exercise control over the

EXTRANEOUS research situation to ensure that other,
VARIABLES extraneous variables do not influence the
relationship being examined.

ENVIRONMENTAL These are characteristics of the environment such

VARIABLES as lighting, time of day, and weather conditions

NON - Take notes: Correlational studies are also

EXPERIMENTAL examples of nonexperimental research. There are
METHODS also nonexperimental studies that compare two or
more groups of scores.
In particular, a real experiment must include
manipulation of an independent variable and
rigorous control of other, extraneous variables. As
 PRETEST/ The researcher assess whether the occurrence of VARIABLE (4 hours, 6 hours, 8 hours).
POSTTEST an event increases or decreases the existing level
DESIGN of a behavior. DEPENDENT The participants' reaction time, measured in
To assess the impact of an intervention by VARIABLE milliseconds during the computer-based test.
measuring changes over time.

EX-POST FACTO The researcher systematically examines the

DESIGN (AFTER effects of pre-existing subject characteristics
THE FACT) (variables) by forming groups on the basis of
naturally occurring differences that already
exist. (ex: introverts and extroverts’ motivation)
To explore potential causal relationships after the
fact.

LONGITUDINAL The same groups of subjects are followed and

DESIGN measured at different points in time.
Looks for changes over time.
ex: attachment styles

A health researcher is interested in understanding the effect of different

amounts of sleep on reaction time. She conducts an experiment where
participants are assigned to sleep for 4 hours, 6 hours, or 8 hours per
night over a week. At the end of the week, she measures each
participant's reaction time using a computer-based test.

INDEPENDENT The amount of sleep participants get each night