EXPERIMENTAL PSYCHOLOGY 3.

Seeking General Principles

- in order to explain observations, general principles
CHAPTER 1: EXPERIMENTAL PSYCHOLOGY AND THE shall be seeked.
SCIENTIFIC METHOD
 Laws – general scientific principles that explain
PSYCHOLOGY AS SCIENCE our universe and predict events.

Science  Theory – a set of general principles that

- the systematic gathering of data to provide attempts to explain and predict behaviour and
descriptions of events taking place under specific other phenomena.
conditions, enabling researchers to explain, predict,
and control events. 4. Good Thinking
- came from the Greek word "scientia" meaning - organizer rational thought, characterized by open
"knowledge" mindedness, objectivity, and parsimony
- content and process - a principal tool of the scientific method

Scientific Method – steps scientists take to gather  Parsimony/Occam's Razor – the simplest
and verify information, answer questions, explain explanation is preferred until ruled out by
relationships, and communicate findings. conflicting evidence.

Methodology – the scientific technique used to 5. Publicizing Results

collect and evaluate psychological data. - science as a highly public activity
- exchange of information, publishing of scientific
Data – facts and figures gathered from observations papers, conferences, peer reviews
in research.
6. Replication – repeating research procedures to
Common Sense Psychology (Heider, 1958) verify that the outcome will be the same as before.
**Everyday, non-scientific collection of
psychological data are used to understand the 7. Self-correction – the content of science changes
social world and guide our behavior.** as we acquire new scientific information, and old
information is re-evaluated in light of new facts.
Characteristics of Modern Science
1. The Scientific Mentality Objectives of Psychological Science
- the goal of prediction rests on an important 1. Description
assumption: behavior must follow a natural order; 2. Prediction
therefore it can be predicted 3. Explanation
4. Control
* Determinisim – belief that there are specifiable
causes for the way people behave and that these Tools of Psychological Science
causes can be discovered through research. 1. Observation – the systematic noting and
recording of events.
2. Gathering Empirical Data – data that are
observable or experienced; capable of being 2. Measurement – the systematic estimation of the
verified or disproved through investigation. quantity, size, or quality of an observable event.
3. Experimentation CHAPTER 2: RESEARCH ETHICS
- the process undertaken to demonstrate that
particular behavioral events will occur reliably in RESEARCH ETHICS
certain specifiable conditions.
Research Ethics
 Testable - are a framework of values within which we
- capable of being tested conduct research
- typically used in reference to hypothesis - helps researchers identify actions we consider
- has three requirements: good and bad, and explain the principles by which
 procedures for manipulating the setting we make responsible decisions in actual situations
must exist
 predicted outcome must be observable Belmont Report's Three Principles (1979)
 measurable 1. Respect for persons – individuals have the right of
self-determination (basis of informed consent).

Scientific Explanation in Psychological Science 2. Beneficence – minimize harm and maximize

**In a scientific context, explanation means potential benefits (basis of risk/benefit analysis).
specifying the antecedent conditions of an event
or behavior.** 3. Justice – fairness in both the burdens and benefits
of research.
Antecedent Conditions – circumstances that occur
or exist before the event or behavior to be Informed Consent
explained. - a subject agrees in writing to the subject's
participation after relevant details of the
Treatment Conditions – a specific set of antecedent experiment have been explained
conditions created by the experimenter and - may include risks and benefits, but does not
presented to subject to test its effects on behaviors. extend to deception or the hypothesis

Psychology Experiment **Perhaps the most important principle built into

- a controlled procedure in which at least two ethics code is the right of a participant to refuse to
different treatment conditions are applied to be in the study or discontinue participation.**
subjects whose behavior are then measured and
compared to test a hypothesis about the effects of **Ethical researchers, therefore, cannot coerce
treatments on behavior participants to agree to be in the study or present
- at least two different treatments so that we can participants from discontinuing the study.**
compare behavior under valid conditions
Deception
**Successful experiments relies heavily on the - may be used when it is the best way to obtain
principle of control, while other factors that could information
be producing the effect we want to explain are - may not be used to minimize the participant's
carefully controlled.** perception of risk or exaggerate their perception of
potential benefits
 THE APA GUIDELINES 2nd: Replication – is where researchers attempt to
reproduce the findings of others. (Dito tamang
APA Guidelines on Deception gaya lang ng study na na-publish na pero siyempre
- APA ethical guidelines apply to psychologists and may difference naman.)
students when they assume the role of psychologists
during research or practice. 3rd: Competition – the need to compete with other
researchers/colleagues for scarce resources.
Steps in Using Deception (Siyempre dito di lang ikaw researcher may
1. Subjects must be allowed to withdraw from the kaagaw ka tamang pa-competetive/competent
experiment at any time and should never force lang kaya ginagalingan to the point na
coercion to remain. mandadaya na hahaha)
2. The experimenter should provide debriefing (full
disclosure) after either their personal participation Plagiarism
of the completion of the entire study. - is misrepresenting someone's "ideas, words, or
written work" as your own
Debriefing - is a form of fraud, in which an individual claims to
- is an essential component of good experimental false credit for another's ideas, words, or written
research work
- a full explanation of the study's reason for use of
deception to the participants
- involves explaining the true nature and purpose of ETHICAL REPORTS
the experiment
Ethical Issues involved in Research Reports
Confederate 1. Authorship credit should only be given to those
- is an experimenter's accomplice who made major contribution to the research or
- is deceptive because subjects are led to believe writing.
that the confederate is another subject, 2. Researchers should not take credit for the dame
experimenter, or bystander when he or she is research more than once.
actually part of the experimental manipulation

CHAPTER 3: ALTERNATIVES TO EXPERIMENTATION –

FRAUD IN SCIENCE NON-EXPERIMENTAL DESIGNS

Scientific Fraud DESCRIBING RESEARCH ACTIVITIES

- involves falsifying or fabricating data
- is done by researchers due to factors like Non-Experimental Approaches
graduation, tenure, promotion, funding, or - do not create levels of an independent variable
reputation nor randomly assign subjects to these levels
- they are used where experiments are not ethical
Main Lines of Defense against Fraud or possible, or where we want to test hypotheses in
1st: Peer Review – process filters submitted realistic conditions
manuscripts so that only 15-20% of articles are
printed. (Basta dito yung mga submitted na Internal Validity – is the degree to which a
manuscprits sasalain muna para yung mga researcher can establish a causal relationship
magkakapareho ng topic ekis na sis) between the independent and dependent
variables. (Eto tungkol sa gano ka-strong yung
relationship ng IV tsaka DV.)
  Cluster Sampling
External Validity – is the degree to which research
findings can be generalized to other settings and 2. Non-probability Sampling
individuals. (Eto naman, kung may ginawa kang - main methods:
study tas participants mo HAU, tas ginawa mo yung  Quota Sampling
same study and method sa AUF, dapat same results  Convenience Sampling
yung makukuha mo. Kumbaga yung findings mo  Purposive Sampling
generalizable siya for all hahaha)  Snowball Sampling

SURVEY RESEARCH
QUALITATIVE RESEARCH
Survey Research
Qualitative Research - obtains data about opinions, attitudes,
- obtains data consisting of words instead of preferences, and behaviours using questionnaires or
numbers interviews.
- this information is obtained through self-reports, - allows researchers to study private experience,
personal narratives, and expression of ideas, which cannot be directly observed
memories, feelings, and thoughts. - can efficiently collect large amounts of data
- can increase the accuracy of answers to sensitive
Paradigm Shift – a change in attitudes, values, questions especially if done anonymously
beliefs, methods, and procedures accepted during - allows drawing inferences about the causes of
a specific time period. behaviour and can complement laboratory and
field experiments
- does not allow us to test hypotheses about causal
CHAPTER 4: ALTERNATIVES TO EXPERIMENTATION – relationships because we do not manipulate
SURVEYS AND INTERVIEWS independent variables and control extraneous
variables

SAMPLING
CONSTRUCTING SURVEYS
Population – consists of all people, animals, or
objects that share at least one characteristic. Steps in Constructing Surveys
1. Identify specific research objectives.
Sample – is a subset of the population of interest 2. Decide on the degree of imposition of units
(the population we are studying). (degree of response restriction).
3. Decide how you will analyse the survey data.
Two Types of Sampling Methods
1. Probability Sampling Closed Questions (Structured Questions) – can be
- more likely to represent the population than a answered using a limited number of alternatives
non-probability sample and have a high imposition of units. (Answerable
- knows the exact odds of members of the lang to ng yes or no ganern.)
population being included in the sample which tells
whom the sample represents Open-ended Questions (Open Questions) – require
- main methods: participants to respond with more than a yes or 1-10
 Simple Random Sampling rating and have a low imposition of units. (Bukod sa
 Systematic Random Sampling yes or no na sagot, may explanation pa siya basta
 Stratified Random Sampling mahaba ganern may pa-follow-up questions haha)
Guide on Constructing Surveys Context Effects
1. Keep items single and unambiguous, and avoid - are changes in question interpretation due to their
double negatives. position within a survey
2. Avoid double-barrelled (compound) questions - is likely to happen when two questions are related
that require responses about two or more unrelated and not separated by buffer items (unrelated
ideas. questions)
3. Use exhaustive response choices.

COLLECTING SURVEY DATA

MEASURING RESPONSES
Social Desirability Response Set – is representing
Continuous Dimension – is represented by ourselves in a socially appropriate fashion when
psychological variables like traits, attitudes, and responding to a question’s latent content
preferences. (underlying meaning).

**The best type of scale depends on the variable CHAPTER 5: ALTERNATIVES TO EXPERIMENTATION –
you are studying and the level of precision you QUASI-EXPERIMENTAL DESIGNS
desire.**
CORRELATIONAL DESIGNS
**Several levels of measurement “fit” equally well.**
Quasi – means “seeming like.”
**When working with variables like sociability,
psychologists often select the highest scale since it Quasi-Experiments
provides more information and allows analysis using - superficially resemble experiments, but lack their
more powerful statistics.** required manipulation of antecedent conditions
and/or random assignment to conditions.
Important Considerations for Survey Items - should be used when we cannot or should not
1. Subjects decide to refuse to answer surveys manipulate antecedent conditions
during the start or first few questions. - could study the effect of spouse abuse on the
2. Engage subjects from the start by asking frequency of child abuse
interesting questions they will not mind answering.
3. The first survey question should be: Properties of Correlation
 Relevant to the survey’s central topic Pearson Correlation Coefficient
 Easy to answer - is used to calculate simple correlations between
 Interesting two variables and may be expressed as:
 Answerable by most respondents r (50) = +.70, p = .001
 Closed format
where;
4. Whenever possible, use commonly used response r = determines how strong or weak the correlation is
options p = determines the significance of the correlation
5. Avoid value-laden questions that might make a
response seem embarrassing. Correlation Coefficient
- has four properties:
Response Styles – are tendencies to respond to  Linearity – means how the relationship
questions or test items without regard to their actual between x and y can be plotted as line (linear
wording. relationship) or a curve (curvilinear
relationship).
  Sign – refers to whether the correlation CHAPTER 6: FORMULATING THE HYPOTHESIS
coefficient is positive or negative.
CHARACTERISTICS OF AN EXPERIMENTAL HYPOTHESIS
 Magnitude – is the strength of the correlation
coefficient, ranging from -1 to +1. Hypothesis – is an explanation of a relationship
between two or more variables.
 Probability – is the likelihood of obtaining a
correlation coefficient of this magnitude due Experimental Hypothesis
to chance. - is a tentative explanation of an event, or a
behavior
Scatterplots - a statement that predicts the effect of an
- are a graphic display of pairs of data points on the independent variable on a dependent variable
x and y axis
- illustrates linearity, sign, magnitude, and
probability (indirectly) of a correlation. Non-Experimental Hypothesis - predicts how
variables (events, traits, or behaviors) might be
**Regression Line correlated, but not causally related
- eto yung line na dino-drawing sa scatterplot para
madetermine yung direction tsaka kung saan Synthetic Statements – hypotheses that has a
madaming matatamaan na points (variables). property of being true or false which is either
- “line of best fit” supported or contradicted by data gathered.

Factors Affecting Correlation Coefficients Testable Statements – an experimental hypothesis

1. Range Truncation that is assessed by manipulating an IV and
- is an artificial restriction of the range of X and Y measuring the results on the DV.
that can reduce the strength of a correlation
coefficient. Parsimonious Statements
- removes outliers - means that we prefer a simple hypothesis over
one requiring many supporting assumptions
2. Outliers – are extreme scores that usually affect - allows us to focus attention on the main factors
correlations by disturbing the trends in the data. that influence our dependent variable

Coefficient of Determination (r2) – estimates the

amount of variability that can be explained by a References:
predictor variable. Myers, A. and Hansen C. (2011). Experimental
Psychology
Correlation
- does not prove causation
- does not create multiple levels of an independent Compiled by: Bryle Zyver R. Pineda |
variable and randomly assign subjects to conditions @brylezyver
in which in turn cannot establish causal relationships

Why doesn’t correlation prove causations?

1. Casual direction
2. Bidirectional causation
3. The third variable problem