EXPERIMENTAL PSYCHOLOGY

Anne Myers & Christine Hansen (2011) | Chapters 1-16

PSY2208N S.Y. ‘23 - ‘24 | BS PSYCH YEAR 2 - SEM 1 | FINALS

● As commonsense psychologists, the ability to gather data

systematically and impartially is constrained by two very
CHAPTER 1: Experimental Psychology and the important factors:
Scientific Method ○ the sources of psychological information
○ our inferential strategies
Psychology
● The science of behavior Nonscientific Sources of Data
Confirmation Bias
Psychological Science ● once we believe we know something, we tend to overlook
● Research about the psychological processes underlying instances that might disconfirm our beliefs, and we seek,
behavior instead, confirmatory instances of behavior
● Science ● if you believe that the full moon brings our psychotic
○ Comes from the Latin word scientia which simply behavior (the word lunacy comes from the Latin word for
means knowledge “moon”), you will notice and remember instances when
○ It has two meanings content (the facts we learn in our people acted abnormally while the moon was full and you
psychology or chemistry courses) and process (an will ignore the many, many more instances in which no
activity that includes the systematic ways in which we unusual behavior occurred
go about gathering data, noting relationships, and
offering explanations)
Nonscientific Inference
○ Explaining the process of psychological science is the
principal aim of this text One of the first and most important kinds of data we collect
about others comes in the form of traits we assign to them. All
Methodology commonsense psychologists are trait theorists.
● The scientific techniques used to collect and evaluate
psychological data (the facts and figures gathered in We tend to miss or ignore important information about the
research studies) situation.
Example: Stacy’s mother designs for the manufacturer with the
All areas of psychology rely on scientific methods. new designer outfit she is sporting (which might mean she is
● Researchers investigating perception vain) and gets Stacy’s clothes for almost nothing.
○ collect data in formal laboratory experiments
designed to provide the most precise information Our ability to make accurate predictions about someone’s traits
● Psychologists interested in understanding attitudes and increases with the length of acquaintanceship. We are
social behaviors generally more accurate when we know someone well than
○ sometimes gather data under controlled laboratory when we judge a stranger. But not always.
conditions
○ at other times, they conduct surveys in the community Perceiving others by their traits can be useful for predicting
or observe and record people’s behavior in natural their behavior but it can also lead to overestimations of the
settings likelihood that they will act in trait-consistent ways across a
● Psychologists studying human development wide variety of different situations.
○ might observe young children’s reactions under
different conditions in the laboratory or real-world Traits might be more useful for predicting how someone will
settings behave over the long term, whereas situations might be better
● Clinical psychologists predictors of momentary behaviors.
○ may collect data by administering psychological tests
or by observing personality functioning during The process of stereotyping illustrates a related problem of
sessions with patients nonscientific inference.
Example: Once we know that Carol is a librarian, we
1.1 The Need for Scientific Methodology automatically assume that she is probably also serious
because that characteristic is part of a librarian stereotype.

Commonsense Psychology Additional problems in prediction occur because people are

● the kind of everyday, nonscientific data gathering that not very good at using data to estimate the true probabilities of
shapes our expectations and beliefs and directs our events. The well-known “gambler’s fallacy” is a good example
behavior toward others of this problem. When we see that a certain slot machine has
● seems to work well enough for us most of the time not paid off in a long time, we tend to believe that it is overdue
● helps one become a better psychological scientist for a payoff. In reality, it makes no difference whether the
because it demonstrates the need for a more scientific machine has recently paid off or not (unless the machine is
approach to understanding and predicting behavior rigged). Each spin is independently random so your odds of a
jackpot this time are identical to your odds on any other try.
Overconfidence Bias predict behavior or other phenomena.
● a tendency of a person to overestimate their abilities ● the more a theory explain, the better it is; old theories are
● our predictions, guesses, and explanations tend to feel replaced by new theories with greater explanatory power
much more correct than they actually are, and the more
data we have available (accurate or not), the more Hypothesis
confidence we have in our judgments about behavior ● a testable prediction
● a phenomenon compounding our inferential shortcomings
Theory-Based Expectancies
Scientific Method
● can cause us to pay more attention to behavioral
● steps scientists take to gather and verify information,
information that is predicted by the theory and to overlook
answer questions, explain relationships, and communicate
nonpredicted behaviors
this information to others

1.2 The Characteristics of Modern Science 4. Good Thinking

Good Thinking
● essential to offset a predisposition to find only what we
1. Scientific Mentality are seeking
Scientific Mentality ● central feature of the experimental psychology scientific
● behavior must follow a natural order to be predicted method
● what Alfried North Whitehead call as the elementary ● our approach to the collection and interpretation of the
assumption which lies at the heart data should be systematic, objective, and rational

Determinism Parsimony
● there are specifiable causes for the way people behave, ● simplicity, precision, and clarity of thought; avoid making
and these causes can be discovered through research unnecessary assumptions to support an argument or
explanation
● another important aspect of good thinking; sometimes
2. Gathering Empirical Data
called Occam’s razor
Gathering Empirical Data ● William of Occam (a 14th-century philosopher) cautioned
● systematic observation and classification of naturally us to stick to a basic premise: “Entities should not be
occurring events multiplied without necessity”

Aristotle
5. Self-Correction
● assumed that order exists in the universe and can be
described systematically by collecting empirical data Self-Correction
● Empirical Data ● the ability to accept the uncertainty of their conclusion
○ observable or experienced data (commonsense data
collection) 6. Publicizing Results
○ can be verified or disproved through investigation Publicizing Results
● exchanging information through the data results to be
3. Seeking General Principles able to exchange failures as well as success
Seeking General Principles
● having a structuralized observation (theory and laws) 7. Replication
Replication
Laws ● repeating research procedures to verify that the outcome
● principles that have the generality to apply to all situations will be the same as before; a principal tool of the scientific
● general scientific principles that explain our universe and method.
predict events.
● are seldom determined outside the physical sciences, so
the behavioral sciences like psychology largely process 1.3 The Objectives of Psychological Science
by developing better theories

Theory 1. Description – a systematic and unbiased account of the

● interim explanation; pull together diverse sets of scientific
observed characteristics of behaviors
facts into an organizing scheme, such as a general
principle or set of rules that can be used to predict new
examples of behavior
● a set of general principles that attempts to explain and

PSY 2208N NOTES PAGE 2

2. Prediction – refers to the capacity for knowing in 3. the predicted outcome must be measurable.
advance when certain behaviors would be expected to
occur–to be able to predict them ahead of time–because we
1.5 Scientific Explanation in Psychological Science
have identified other conditions with which the behaviors
are linked or associated
Identifying Antecedent Conditions
3. Explanation – understanding what causes a behavior to Antecedent Conditions
occur once it is explained; includes knowledge of the ● are the circumstances that come before the event or
conditions that reliably reproduce the occurrence of a behavior that we want to explain; also called antecedents
behavior
Comparing Treatment Conditions
4. Control – refers to the application of what has been Treatments
learned about behavior ● specific sets of antecedent conditions
● we compare different treatment conditions so that we can
test our explanations of behaviors systematically and
Applied Research
scientifically
● research designed to solve real-world problems (e.g.
helping people to deal with grief or improve employee
morale) The Psychology Experiment
Psychology Experiment
Basic Research ● a controlled procedure in which at least two different
● research designed to test theories or to explain treatment conditions are applied to subjects whose
psychological phenomena in humans and animals behaviors are measured and compared to test a
hypothesis
Developmental Psychology
● devoted to bringing basic research on child violence Establishing Cause and Effect
together with public policy aimed at reducing it
Cause-and-Effect Relationship
1.4 The Scientific Method: Tools of Psychological ● the relation between a particular behavior and a set of
antecedents always preceding it.
Science
● If the XYZ set of antecedents always leads to a particular
behavior, whereas other treatments do not, we can infer
that XYZ causes the behavior.
1. Observation – the systematic noting and recording of
● Example: with all other factors constant, Baron and his
events; a principal tool of the scientific method
colleagues demonstrated that high concentrations of
negative ions caused changes in moods.
2. Measurement – the systematic estimation of an
observable event's quantity, size, or quality; the assignment Temporal Relationship
of numerical values to objects or events or their ● type of cause-and-effect relationship we establish through
characteristics according to conventional rules; a principal experiments because a time difference occurs in the
tool of the scientific method. relationship
● the treatment conditions come before the behavior, or,
Subject state somewhat differently, the cause precedes the effect.
● the scientific term for an individual who participates in
research.
Necessary vs. Sufficient Conditions

3. Experimentation – the process is undertaken to Necessary Conditions

● Examples of NC
demonstrate that particular behavioral events will occur
○ a snowmobile will run without fuel. Thus, fuel is a
reliably in certain specifiable conditions; a principal tool of
necessary condition for running a snowmobile
the scientific method.
Testable Sufficient Conditions
● capable of being tested; typically used about a ● Examples of SC
hypothesis. ○ cutting down on carbohydrates might be a sufficient
● Two requirements must be met for a testable hypothesis condition to produce weight loss is not a necessary
1. procedures for manipulating the setting must exist, condition since we could also lose weight by
2. the predicted outcome must be observable increasing our activity level or cutting calories

PSY 2208N NOTES PAGE 3

○ being in a good mood to increase helpfulness
CHAPTER 2: Research Ethics

1.6 From Pseudoscience to Psychological Science

2.1 Research Ethics
Psychology as an experimental science did not emerge until
the late 1800s. Research Ethics
● framework of values within which we conduct research
Wilhelm Wundt (1832-1926) is generally credited with being
the first experimental psychologist and the birth of Values
psychological science is usually dated from the opening of his ● the foundation of the ability to judge between right and
laboratory in Leipzig, Germany in 1879. wrong

Mental Philosophy Morals

● includes the study of consciousness and mental ● a system of belief (good or bad)
processes and was based on the premise that the human
mind begins as a blank slate, gaining knowledge of the Ethics
world through sensory experiences; primarily engaged in ● determines if behavior is right or wrong
the study of the five senses
Responsible Research
Pseudoscience ● aims to advance understanding of feelings, thoughts, and
● a field of study that gives the appearance of being behaviors that benefits humanity rather than satisfying idle
curiosity.
scientific but has no factual scientific basis
● pseudo is Greek for “false” — characterized any field of
Legal and Ethical Guidelines
study that gives the appearance of being scientific but
● to protect subjects of psychological research (Title 45,
has no true scientific basis and has not been confirmed
Section 46)
using the scientific method
Institutional Review Board (IRB)
1.7 The Organization of the Text ● review committee to evaluate studies before conducting
and ensure the safety and protection of participants.

Risk/Benefit Analysis
1. Introduction – gives an overall orientation to the field of ● risks are outweighed by potential benefits or the
research methods, much as a literature review gives an knowledge to be gained
overall picture of the state of research in a particular
content area Psychologist Robert Rosenthal
3 reasons why poorly designed research can be unethical
● 1. Students, teachers, and administrators will waste time
2. Method – includes all the basic procedures used in that could be used for beneficial educational experiences.
conducting simple experiments, selecting subjects, and ● 2. This can lead to inaccurate and unwarranted
collecting data in a scientific way conclusions that may damage society.
● 3. Allocating time and money to poor quality science will
keep those finite resources from better-quality science.
3. Results – reviews the common statistical procedures
used to analyze data ● Assent
○ agreement of minor children ages seven and above as
a participation requirement.
4. Discussion – looks at the major issues involved in
drawing conclusions from data and communicating the ● Informed Consent
findings ○ give information on the experiment to make an
informed decision
○ subjects agree to participate after being aware of the
nature of the study.

● Mann (1994)
○ conducted an experiment that demonstrated that
most subjects/participants retain little information
about the consent forms they signed

PSY 2208N NOTES PAGE 4

The Evolution of Ethics in Research Classical Projection (Freud)
● the traits projected are those the person is not consciously
Code of 1947 aware of possessing.
● the basis of today’s ethical standards
2.3 Protecting the Welfare of Animal Subjects
Belmont Report
● a statement of government policy on research involving
human subjects Animal Welfare
● Three basic ethical principles: respect for the person, ● the humane care and treatment of animals (Animal
beneficence, and justice Welfare Act 1966)

Three Principles of Belmont Report Institutional Animal Care and Use Committee (IACUC)
1. Respect for Persons – every human is an autonomous ● animal research review board; to evaluate animal research
person with the right to make their own decisions about before it can be conducted
the research (informed consent)
2. Beneficence – an obligation to minimize the risk of harm Joseph Brady
and maximize benefits to individuals (risk/benefit ● studies the emotional behavior of Rhesus Monkeys.
analysis)
3. Justice – the fairness in both burdens and benefits of Objections to the Research
research; The subjects don’t get the right benefits for them 1. The use of restraining chairs is stressful for the animals.
2. The use of electric shock, which is also not specified
2.2 American Psychological Association (APA) Guidelines 3. The study was poorly designed with diminished benefit
because of the bias in the selection of monkeys
The American Psychological Association Guidelines Animal Rights
● applies to psychologists and students with the role of ● the idea that all sensate species are of equal values and
psychologists by engaging in research or practice. rights
Minimal Risk
● a risk that is no greater than encountered in daily life or
2.4 Fraud in Science
during a performance of routine physical or psychological
test
● Examples: observations of public behavior, anonymous Data Falsification
questionnaires, and some kinds of archival research ● a breach in the ethical principle stated in APA standard

At Risk Plagiarism
● the likelihood of a subject being harmed because of the ● common kind of fraud; to represent someone else’s ideas,
nature of the research; one who is more likely to be words, or written work as your own; a serious breach of
harmed in some way by participating in the research ethics that can result in legal action

Commitments of researchers to the subjects

● pay subjects, give credit, maintain confidentiality, and CHAPTER 3: Alternatives to Experimentation:
share results Nonexperimental Designs

Deception and Full Disclosure

Confederate 3.1 Describing Research Activities
● an accomplice to the research

Debriefing Traditional Psychology Experiments

● explaining the true nature and purpose of the study; the ● creating specific sets of antecedent conditions, or
principle of full disclosure at the end of the experiment treatments, to test a hypothesis about behavior.

Extensive Post - Experimental Debriefing Experimental Approach

● to remove the harmful beliefs resulting by the experiment ● a researcher must be able to set conditions for any
individual who ends up as a subject in the experiment.
Attributive Projection
● the process of projecting our traits onto another person Nonexperimental Approach
(the person is consciously aware of having the trait) ● used when an experiment is not practical or desirable.
Used whenever testing a hypothesis in a real-life situation
is necessary.

PSY 2208N NOTES PAGE 5

Internal Validity 3. They allow the study of rare phenomena.
● the degree to which a research design allows us to make 4. They provide exceptions or counter-instances to accepted
causal statements ideas, theories, or practices.
5. They have persuasive and motivational value
High Internal Validity
● we can demonstrate with certainty that the changes in Case Studies Limitations
behavior observed across treatment conditions were 1. working with only one or perhaps a few subjects
caused by treatment differences 2. not able to observe an individual directly all the time
3. subjects providing data for case studies might neglect to
External Validity mention important information
● generalizability or applicability to people and situations ○ Deviant Case Analysis – an extension of the
outside the research setting evaluative case study where deviant and ordinary
individuals is compared for significant differences
High External Validity ○ Retrospective Data – data collected in the present
● when observations can be generalized to others settings that are based on recollections of past events
and other people ○ Field Studies – used in the field or real-life settings
○ Focus Groups – a type of group interview, an
Two Major Dimensions organized discussion session with a small group of
people, usually led by a trained facilitator

Degree of Manipulation of Antecedent Conditions

3.4 Field Studies
● the extent to which the researcher set up the controlled
condition (high or low manipulation)
Field Studies
Degree of Imposition of Units ● are nonexperimental approaches used in the field or in
● the extent to which the researcher constraints, or limits, real-life settings
the responses a subject may contribute to the data.
Naturalistic Observation Studies
Five Nonexperimental Approaches Naturalistic Observation
1. Phenomenology ● the technique of observing behaviors as they occur
2. Case Studies spontaneously in natural settings
3. Field Studies ● Ethology
4. Archival Studies ○ to study behavior in the wild as well as in captivity
5. Qualitative Studies
Systematic Observation
● use of prearranged strategy for recording observations in
3.2 Phenomenology – the description of an which each observation is recorded using specific rules or
individual’s immediate experience guidelines so that observations are more objective
● INTERSECT
○ coding system developed for scoring various kinds of
Contemporary or Empirical Phenomenology teacher-student interactions.
● relies on the researcher’s experiences or experiential data ○ Has four types of responses:
from other sources. ■ positive reinforcement,
● It also relies on these sources of data: ■ acceptance
○ The researcher’s self-reflection on experiences ■ remediation
relevant to the phenomenon of interest ■ criticism
○ Participants’ oral or written descriptions of their
experiences of the phenomenon Strange Situation
○ Accounts of the phenomenon obtained from ● used to assess the quality of attachments that very young
literature, poetry, visual art, television, theatre, and children (usually between the ages of 1 and 2) have forged
previous phenomenological (and other) research with their parents or other caregivers
● Parts of Strange Situation
3.3 Case Studies – a descriptive record of an 1. The experimenter introduces the parent and baby to
individual’s experiences, behaviors, or both, kept the playroom and leaves.
by an outside observer 2. The parent sits while the baby plays.
3. A stranger enters, sits, and talks to the parent.
4. The parent leaves, and the stranger offers comfort if
Five Major Purposes of Case Studies the baby is upset.
1. They are a source of inferences, hypotheses, and theories. 5. The parent returns, greets the baby, and offers
2. They are a source for developing therapy techniques. comfort. The stranger leaves.

PSY 2208N NOTES PAGE 6

6. The parent leaves the room.
7. The stranger enters and offers comfort. 4.2 Constructing Surveys
8. The parent returns, greets the baby, comfort if
necessary, and tries to interest the baby in toys
Types of Survey Questions
○ Reactivity ● Closed Questions (also called structured questions)
■ the tendency of subjects to alter their behavior or ○ limited answers
responses when they are aware of an observer’s ○ Ex: Do you smoke?
presence ○ Ex: On a scale from 1 to 10, how much do you like
○ Unobtrusive Measures classical music?
■ behavioral indicators can be observed without
the subject’s knowledge ● Open-ended Questions (also called open questions)-
○ unlimited answers
○ Ex: Why do you prefer taking the train to fly?
Participant-Observer Studies ○ Ex: What are your feelings about animal rights?
Participant-Observer Study
● the researcher becomes part of the group being studied. ● Double-barreled (or compound)
○ two (or more) different ideas in the same question
3.5 Archival Study ○ Ex: Do you like strawberries and cream?

Content Analysis
Archival Study ● coding behaviors using systematic observational
● a descriptive research method in which already existing techniques.
records are reexamined for a new purpose ● It evaluates and categorizes the content of each answer.
● A system for quantifying responses to open-ended
3.6 Qualitative Research questions by categorizing them according to objective
rules or guidelines.
Qualitative Research
● relies on words rather than numbers for the data collected; Exhaustive
it focuses on self-reports, personal narratives, and the ● need to contain all possible options.
expression of ideas, memories, feelings, and thoughts.
● Paradigm 4.3 Measuring Responses
○ the set of attitudes, values, beliefs, methods, and
procedures that are accepted within a particular Level of Measurement
discipline at a certain point in time ● scale used to measure a response.
● four levels of measurement: Nominal, Ordinal, Interval,
CHAPTER 4: Alternatives to Experimentation Surveys Ratio
and Interviews
FOUR LEVELS OF MEASUREMENT

4.1 Survey Research Nominal Scale Simplest level of measurement.

It groups items together into

Survey Research categories that can be named
● a way of obtaining information about people’s opinions,
attitudes, preferences, and behaviors simply by asking. (nominal comes from the Latin word
for “name”).
Surveys
● allow gathering data about experiences, feelings, Also called the lowest level of
thoughts, and motives that are hard to observe directly measurement because it does not
and are useful for collecting data on sensitive topics provide magnitude.
because of anonymity.
Ordinal Scale the rank ordering of response items.
Most Common Survey Techniques
● Written questionnaires Interval Scale measures magnitude or quantitative
● Face-to-face interviews size using measures with equal
intervals between the values with no 0
points.

PSY 2208N NOTES PAGE 7

Buffer Items
● questions unrelated to either of the related questions
Ex: Fahrenheit and centigrade
temperature scales
4.5 Collecting Survey Data
Ratio Scale the highest level of measurement.

It has equal intervals between all Latent Content

values and a true zero point. ● the “hidden” meaning behind the question.

Ex: height and weight, and time. Response set

● a picture we want to create of ourselves. A tendency to
answer questions based on their latent content to create a
Selecting Levels of Measurement particular impression of ourselves.

Continuous Dimension Self-Administered Questionnaires

● traits, attitudes, and preferences dimension.
● using a written questionnaire, handing out the
● Ex: Introversion vs Extroversion an individual can rate
questionnaire in person, letting the subjects fill out the
relatively high, low, or somewhere in the middle.
questionnaire in private, anonymous collecting of
questionnaire data
4.4 Important Considerations for Survey Items
Mail Surveys
● include a polite and professional cover letter (and a
The question should have these characteristics:
stamped, self-addressed envelope).
1. Relevant to the central topic
● Also, include incentives and track mail returns to report.
2. Easy to answer
3. Interesting
4. Answerable by most respondents Computer and Internet Surveys
5. Closed format ● the use of the Internet for conducting surveys has
drawbacks since it could happen that one participant may
Response Styles answer multiple times
Response Styles
● tendencies to respond to questions or test items in Telephone Surveys
specific ways, regardless of the content. ● the most widely used method for conducting survey
research.
Different Response Styles ● Survey technique: random digit dialing.
1. Willingness to answer ● This has problems such as unresponsive phone numbers
○ comes whenever questions require specific and difficulty in generalizing conclusions
knowledge about facts or issues.
2. Position preference Interviews
○ the tendency to choose an option because of its
location. When in doubt about answers to ● the best way to gather high-quality survey data
multiple-choice questions, some people always select
a response in a certain position, such as answer c.
3. Manifest content 4.6 Evaluating Surveys and Survey Data
○ the plain meaning of the words that actually appear
on the page.
4. Yea-sayers Principles for Successful Interview
○ apt to agree with a question regardless of its manifest 1. rapport (i.e., winning the interviewee’s trust)
content (this is also called response acquiescence). 2. avoiding judgmental statements
5. Nay-sayers 3. knowing how to keep the interview flowing
○ tend to disagree no matter what they are asked (also
known as response deviation) Structured Interview
● the same questions are asked in precisely the same way
Context Effects each time.
● occur when the content of the questions preceding a
particular question influences how some respondents Unstructured Interviews
answer that question. ● free-flowing

PSY 2208N NOTES PAGE 8

Focus Groups Random Number Table
● another face-to-face technique used less often for ● a table of numbers generated by a computer so that every
collecting data about a particular topic number has an equal chance of being selected for each
● small groups of people with similar characteristics who position in the table.
are brought together by an interviewer or facilitator
● Facilitator Random Selection
○ who guides the group in a discussion of specific ● any member of the population has an equal opportunity to
issues be selected. A common assumption of the statistical tests
used most often to analyze data.
Three Qualities of a Reliable Survey
Reliability FOUR TYPES of PROBABILITY SAMPLES
● the extent to which the survey is consistent and
repeatable.
Simple the most basic form of probability sampling.
Three Qualities of a Reliable Survey Random A portion of the whole population is selected
1. Responses to similar questions in the survey should be Sample in an unbiased way.
consistent
2. The survey should generate very similar responses across Ex. 100 students (population)= 1/100
different survey-givers probability
3. A survey should generate similar responses if given to the
same person more than once. Systematic all population members are known and can be
Random listed unbiasedly; a researcher may select
Validity Sample every nth person from the population.
● refers to the extent to which a survey measures the
intended topic. Ex. 100 students (population), 10 students
(sample)= 100/10= 10 as n. Select every 10th
person from the unbiased list
4.7 Sampling
Stratified is used when the population is known to
Random contain distinct subgroups. Obtained by
Sampling Sample randomly sampling from people in each
● deciding who the subjects will be and then selecting them. subgroup in the same proportions as they
exist in the population.
Population
● consists of all people, animals, or objects with at least one Ex. school is made up of 10% teachers and
common characteristic. 90% students, sample of 100 people from
school, randomly select 10 teachers (10%)
Sample of Subjects and 90 students (90%)
● a group subset of the population of interest.
Cluster used when the population of interest is very
Representativeness Sample large. Participants are randomly selected, but
● how closely the sample mirrors the larger population whole groups are selected rather than
precisely, how closely the sample responses we observe individuals.
and measure reflect those we would obtain if we could
sample the entire population Ex. 3 million population, select clusters like
zip code areas, school districts, cities, or
Two General Sampling Approaches counties
● Probability Sampling
● Nonprobability Sampling
4.9 Nonprobability Sampling

4.8 Probability Sampling

Nonprobability Sampling
● the subjects are not chosen at random
Probability Sampling
● the preferred means of selecting research subjects. It is
FOUR TYPES of NONPROBABILITY SAMPLES
selecting subjects so that the odds of their being in the
study are known or can be calculated.
● Probability
Quota Sampling select samples through predetermined
○ refers to the study of the likelihood of events

PSY 2208N NOTES PAGE 9

quotas to reflect the population's
5.1 Correlational Designs
makeup.

Ex. A reporter is sent to a university to Correlational Designs

interview 40 students, half male, half ● used to establish relationships among pre existing
female, because the student body is behaviors and can be used to predict one set of behaviors
roughly half men, half women. from others (such as predicting your college grades from
scores on your entrance exams)
Convenience obtained by using any groups who ● can show relationships between sets of preexisting
Sampling happen to be available. Considered a antecedent conditions and behavioral effects
(accidental weak form of sampling because the
sampling) researcher exercises no control over the Quasi-Experiments
sample's representativeness. ● seem like actual experiments but lack manipulation of
antecedents or random assignment to treatment
Ex. a church choir, a psychology class, a conditions.
bowling league, or a supermarket ● Quasi — Latin means “seeming like.”
checkout line
Confounding
Purposive Nonrandom samples are selected ● inability to establish cause with certainty in research
Sampling because the individuals reflect a
specific purpose of the study. High External Validity and High the Imposition of Units
● Correlational Designs
Ex. The purpose of the study was to ○ low in manipulating antecedents and can be used
measure the success of a new training with laboratory and field data.
program for employees in two ● Quasi-experiments
departments (sales and human
resources) you select the employees of Low in Manipulation of Antecedents
those departments as your purposive ● Quasi-Experiments
sample to compare. ○ vary in the degree of manipulation of antecedents, but
without random assignment
Snowball used predominantly for sampling very ● Imposition of Units
Sampling small, uncommon, or unique ○ restrict, or limit, in the responses subjects can
populations when researchers do not contribute to the collected data.
know who the population members are ● Correlational Study
or how to contact them. The researcher ○ designed to determine the correlation, or degree of
locates people who fit the criterion and relationship, between two traits, behaviors, or events.
asks these people to lead them to ○ Variable – any observable behavior, characteristic, or
additional individuals event that can vary or have different values

Reporting Samples CORRELATIONAL DESIGNS

● include gender, age, and other important demographic ● Correlational Study
characteristics (education level, racial and ethnic heritage, ○ designed to determine the correlation, or degree of
socioeconomic status, etc.) if they could be significant in relationship, between two traits, behaviors, or events.
interpreting the results. ● Simple Correlations
● Ex. The subjects were 60 undergraduates, 30 men and 30 ○ relationships between pairs of scores from each
women at the University of Oregon who responded to an subject
ad in the college newspaper
Pearson Product Moment Correlation Coefficient (r)
● most commonly used procedure for calculating simple
correlations
CHAPTER 5: Alternatives to Experimentation –
● three general outcomes are possible:
Correlational and Quasi-Experimental Designs ○ a positive relationship
○ a negative relationship, or
○ no relationship
2 Categories of Nonexperimental Research Methods:
● Correlational Designs Scatterplots (Scatter Graphs or Scattergrams)
● Quasi-Experimental Designs ● visual representations of the scores belonging to each
subject in the study

PSY 2208N NOTES PAGE 10

Regression Lines (Lines of Best Fit) ● used to predict or visualize a relationship between two
● lines drawn on the scatterplots which illustrate the different variables. The stronger the correlation, the better
mathematical equation that best describes the linear the prediction
relationship between the two measured scores.
Regression Equation
Subject Variable ● a formula for a straight line that best describes the
● characteristics of the subjects in an experiment or relationship between the two variables. It is an equation
quasi-experiment that the researcher cannot manipulate; for a straight line with a slope and an intercept.
sometimes used to select subjects into groups. ● Slope – the direction of the line
● Intercept – the value on the Y, or vertical, axis when X = 0
Positive Correlation ● Formula:
● if the computed value of r is positive and is also called a
direct relationship

Negative Correlation
● also called an inverse relationship

Features of the data: a nonlinear trend, range truncation, and

outliers Multiple Correlation and Multiple Regression

General Linear Model Multiple Correlations (R)

● used on statistical formulas for simple correlations, which ● to test the relationship of several predictor variables (X1,
assumes that the direction of the relationship between X X2, X3 …) with a criterion variable (Y)
and Y generally remains the same ● allows to use the information provided by two or more
measured behaviors to predict another measured behavior
Range Truncation when we have that information available
● an artificial restriction of the range of values of X or Y
With a limited range of data, it could show a range Partial Correlation
truncation of 0 or close to 0 Outliers can affect the ● measures the strength of a relationship between two
correlation coefficient because it disturbs the general variables while controlling for the effect of one or more
linear trend of the data other variables. (Measures the correlation between X and
Y, while controlling the Z)
3 Alternative Possibilities Whenever Two Behaviors Are
Strongly Correlated Multiple Regression Analysis
1. Normal Cause and Effect — Innate aggressiveness ● used to predict the score on one behavior from scores on
determines the preference for violent TV—not vice versa. the others. Used when more than two related behaviors
More aggressive people could gravitate toward more are correlated. Estimating the relationships between a
violent programs. dependent variable and one or more independent variables
2. Bidirectional Causation – refers to the inability to
legitimately deduce a cause-and-effect relationship Regression Equations
between two events or variables solely based on an ● determine the weight of each predictor to report the
observed association or correlation between them. weights (called beta weights) in a research report
Correlation does not imply causation.
3. Third Variable Problem – a confounding variable affects 5.2 Causal Modeling
both variables to make them seem causally related when
they are not. A third agent may be causing the two
behaviors to appear related. CAUSAL MODELING
● drawback of correlational designs is the problem of the
Coefficient of Determination (r2) direction of cause and effect
● estimates the variability in scores on one variable that can ● Path Analysis
be explained by the other variable—an estimate of the ○ used when subjects are measured on several related
strength of the relationship between them. behaviors. In path analysis, the researcher creates
○ Cohen (1988) – argued that r2 > or equal to 0.25 is models of possible causal sequences. uses beta
considered a strong association between variables weights to construct path models, outlining possible
causal sequences for the related behaviors
Linear Regression Analysis
● when two behaviors are strongly related, the researcher ● Beta Weights
estimates the score on one of the measured behaviors ○ partial coefficients that indicate the unique strength
from a score on the other. of the relationship between a predictor and criterion,
controlling for the presence of all other predictors.

PSY 2208N NOTES PAGE 11

■ Cross-lagged Panel Design - used to create looking for changes over time in the same group of
causal models uses relationships measured over subjects
time to suggest the causal path. Measures two
different variables at two points in time Natural experiment
● a quasi-experiment used to compare people exposed to a
SUMMARY OF CORRELATION and naturally occurring event with a comparison group (usually
unexposed people)
CORRELATION-BASED RESEARCH DESIGNS
Quasi-treatment groups
● formed based on the particular event, characteristic, or
TYPE PURPOSE
behavior whose influence we want to investigate
Correlation determines degree of relationship
Quasi-experimental design
between two variables
● used to estimate the causal impact of an intervention on
the target population without random assignment.
Multiple Correlation determines degree of relationship
between three or more variables Ex Post Facto Studies (after the fact)
● used when the effects of traits, behaviors, or naturally
Linear Regression uses regression equation to predict occurring events cannot or should not be manipulated by a
Analysis scores on one variable from scores researcher
on a second correlated variable ● researcher systematically examines the effects of subject
characteristics (often called subject variables) without
Multiple Regression uses regression equation to predict actually manipulating them
Analysis scores on one variable from scores ● Ex Post Facto Advantages
on sets of other variables ○ it deals with things as they occur
○ there is no manipulation of the conditions that
Factor Analysis determines subsets of correlated interest the researcher
variables within a larger set of ○ Low Internal Validity because there is the possibility
variables that other differences between the groups of subjects
were the actual cause of the effects
Path Analysis uses beta weights from multiple ○ High external validity because they focus on naturally
regressions analysis to generate occurring events
possible direction of cause and
effect from correlated variables Random assignment of subjects
● to create treatment groups in which any preexisting
Cross-Lagged Panel measures the same pair of differences in people are distributed evenly across all the
variables at two different points in treatment groups
time; looks at patterns of
correlations across time for Nonequivalent groups design
possible direction of cause and ● the researcher compares the effects of different treatment
effect conditions on pre existing groups of participants No
control over who gets each treatment because a random
assignment is not possible
● Ex Post Facto Disadvantages
5.3 Quasi-Experimental Designs
○ take place over periods of months, years, or even
decades
Quasi-Experimental Designs ○ The same sample must be tracked for a long time,
● seem like a real experiment, but lack one or more essential even as participants (and researchers) age, move
elements, such as manipulation of antecedents or random away, or become deceased
assignment to treatment conditions. ○ time-consuming and challenging to conduct
● This is used to explore the effects of different treatments
on pre-existing groups of subjects or to investigate the Longitudinal designs
same kinds of naturally occurring events, characteristics, ● used for studying human (and animal) growth and
and behaviors that we measure in correlational studies. development
○ Regression
Different Goals ■ a way of escaping the reality of a stressful
situation by reverting to more childlike patterns of
● Correlational Studies – look for relationships or behavior
associations between variables
● Quasi-Experiments – compare different groups of
subjects, looking for differences between them, or we are

PSY 2208N NOTES PAGE 12

Cross-Sectional Studies
● analyzes data from a population, or a representative
subset, at a specific point in time CHAPTER 6: Formulating the Hypothesis

Pretest/Posttest Design Ways of arriving at hypotheses suitable for experimental study:

● measure people’s level of behavior before and after the ● Induction
event and compare; used to assess the effects of naturally ● Deduction
occurring events when a true experiment is not possible ● Building on prior research
● Practice effects (pretest sensitization) ● Serendipity
○ influences on test results when a test is taken more ● Intuition
than once
○ To achieve maximum control over outside influences, Hypothesis
you would need some comparison groups: ● the thesis, or main idea, of an experiment. (Nonscientific
1. a nonequivalent control group—a group that took synonyms are speculation, guess, or hunch)
both the pretest and posttest but was not
exposed to the “treatment,” Nonexperimental designs
2. A group that received the treatment and took only ● used to demonstrate relationships between sets of
the post-test behaviors but is not used to infer a cause-and-effect
3. a posttest-only group relationship.
Nonexperimental hypothesis
● Solomon 4-group design ● a statement that predicts how events, traits, or behaviors
○ all four groups are included in the design are related and not about cause and effect.

SUMMARY OF QUASI-EXPERIMENTAL RESEARCH True experimental hypothesis

DESIGNS ● a statement that predicts the cause-and-effect relationship
of variables

TYPE PURPOSE EXAMPLES OF NONEXPERIMENTAL HYPOTHESES

Quasi-Experiment investigate differences in

preexisting groups of subjects; TYPE OF DESIGN HYPOTHESIS
group differences on some variable
may be explored or different Correlational The amount of TV viewing will be
treatments given to preexisting directly related to vocabulary size
groups may be compared
Ex post facto Repressors will report fewer
Ex post factor explores characteristics, behaviors, treatment-related side effects than
or effects of naturally occurring will nonrepressors
events in preexisting groups of
subjects Non-equivalent Incandescent lighting (in Company
groups A) will produce better performance
Nonequivalent compares the effects of different than fluorescent lighting (in
groups treatment conditions on preexisting Company B)
groups of subjects
Longitudinal Firstborn children will show
Longitudinal investigates changes across time imitative behaviors after birth of a
by measuring behavior of same sibling
group of subjects at different points
in time Cross-section Firstborns with a 4-month-old
sibling will show more imitative
Cross-section investigates changes across-time behavior than firstborns with a
by comparing groups of subjects 1-month-old sibling
already at different stages at a
single point in time

Pretest/Posttest explores the effects of an event (or 6.1 The Characteristics of an Experimental Hypothesis
treatment) by comparing behavior
before and after the event or Experimental Hypothesis
treatment ● a tentative explanation of an event or behavior

PSY 2208N NOTES PAGE 13

Synthetic Statements 6.4 Combining Induction and Deduction
● statements that can be either true or false

Nonsynthetic Statements have two categories: Three simple propositions (Walster, Walster, and Berscheid)
● analytic ● Individuals try to optimize outcomes (outcomes ¼ rewards
● contradictory minus costs).
● When individuals believe they are in an inequitable
Analytic Statements relationship, they feel distress in direct proportion to the
● statement that is always true and can explain all possible perceived degree of inequity.
outcomes ● The more distress they feel, the harder they work to
● Ex. “I am pregnant or I am not pregnant” restore equity

Contradictory statements 6.5 Serendipity and the Windfall Hypothesis

● statements with elements that are contradicting and
always false. The Three Princes of Serendip (18th-century tale by Horace
● Ex. “I have a brother and I do not have a brother” Walpole)
● describes the adventures of three princes who found
Testable Statements valuable things they were not seeking
● Testable
○ the means for manipulating antecedent conditions Serendipity
and measuring the resulting behavior must exist ● finding things that are not being sought.
● Testability
○ extent to which truthness or falseness of a
6.6 Intuition
hypothesis/declaration can be tested, based on the
accessibility of evidence
Intuition
Falsifiable Statements ● knowing without reasoning
● falsifiable (disprovable) by the research findings ● The more we know about a topic, the better our intuitive
● Falsifiability hypotheses are likely to be
○ the capacity of hypothesis to be proven wrong. ● cannot be valid until translated into hypothesis and
subjected to empirical tests
Parsimonious Statements
● parsimony means the simplest explanation is preferred 6.7 When All Else Falls

Fruitful Statements
● it leads to new studies Causal attributions
● forming hypotheses about antecedent conditions that
affect people’s behavior
6.2 The Inductive Model
6.8 Search the Research Literature

Inductive model (of formulating a hypothesis)

● the process of reasoning from specific cases to more Getting Started
general principles ● Psychological Journals
● Taking empirical data and form general explanatory ○ periodicals that publish individual research reports
schemes to construct theories and integrative research reviews (up-to-date
summaries of what is known a topic).
Theory
● set of general principles used to explain and predict ● Meta-analysis
behavior ○ a statistical procedure that uses data from similar
studies to summarize research findings about
individual topics.
6.3 The Deductive Model ○ uses statistical procedures to measure how strong
the cause-and-effect relationship between the
antecedent conditions and measured behaviors
Deductive model (of formulating a hypothesis)
● the process of reasoning from general principles to predict Writing the Report
specific instances. ● Introduction
○ consists of selective review of relevant, recent
research.

PSY 2208N NOTES PAGE 14

○ select articles that are directly related to the research Some Research Examples
hypothesis
○ studies cited in Introduction should provide empirical
background and guide the reader toward research Schachter (1959)
hypothesis ● If people are anxious, then they will want to affiliate, or be,
● Discussion with others.
○ implications of your findings or what the results mean ● The more anxious group would be more likely to want to
wait with others
Finding the Articles, You Need ● anxiety was the independent variable in this experiment as
● PsycINFO and PsycARTICLES – example webs to look for set by Schachter
articles ● dependent variable was affiliation that was resulted by the
anxiety of the group

CHAPTER 7: The Basics of Experimentation Hess (1975)

● tested the hypothesis: large pupils make people more
attractive
● the independent variable in this experiment was pupil size
7.1 Independent and Dependent Variables at 2 levels (large and small pupils)
● the dependent variable is attractiveness
Variables Example:
● measurable elements that can vary or take on different ● Independent Variable is word orientation with two
values along some dimension treatment conditions: horizontal orientation and vertical
orientation
Experimental Hypothesis ● Dependent Variable is rate of learning
● expresses relationship between two kinds of variables

Independent Variable (IV)

7.2 Operational Definitions
● variable that the experimenter manipulates; Values are
created by the experimenter and are not affected by
anything else Operational definition
● Also referred to as: treatments, manipulations, ● meaning of a variable within an experiment: It defines a
interventions, conditions variable in terms of observable operations, procedures,
and measurements
Levels of the Independent Variable ● statements of operating procedures, sets of instructions
● the possible values in every experiment that tell others how to carry out an experiment
● Ex. a professor gives tests on either blue or yellow paper ● It is called operational definition because it describes the
to see if the color of the paper influences the scores. Blue operations involved in manipulating or measuring the
and yellow represent two levels of the one IV, color. variables in an experiment

Quasi-Experiments Experimental operational definitions

● researcher selects rather than creates levels of IV by ● explain the meaning of the independent variables; they
assigning subjects to treatment groups on the basis of describe what was done to create various treatment
subject variable conditions. It includes all the steps followed to set up each
value of the independent variable
Ex post facto study
● individuals in treatment groups differ on subject variable Measured operational definitions
because the subject variable is the one being tested ● (of the dependent variable) describe what procedures we
follow to assess the impact of different treatment
True Experiment conditions
● effects of a manipulated independent variable is tested
not the effects of different kinds of subjects “Scores on the Culture Fair Intelligence Test”
● We identify the test by the name of this. If we are using
Dependent Variable (dependent measures) (DV) scores on a standardized test to measure our dependent
● particular behavior expected to change because of variable
experimental treatment
● the outcome we are trying to explain. Defining Constructs Operationally
● Measured to determine whether the independent variable
had an effect Hypothetical Constructs
● Also referred to as: measures, effects, outcomes, results ● concepts are unseen processes postulated to explain
behavior

PSY 2208N NOTES PAGE 15

● Researchers formulate different definitions for the same Internal Validity
construct variable when it is used in different experiments ● the certainty that the changes in behavior observed across
● Labeling the variable “anxiety” would be too vague without treatment conditions in the experiment were actually
a precise operational definition because it varies on each caused by the independent variable
person’s experience.
Extraneous Variables
● a variable other than an independent or dependent
7.3 Evaluating Operational Definitions variable; a variable that is not the focus of an experiment
but can produce effects on the dependent variable if not
controlled
Reliability
● means consistency and dependability Confounding Variables
● an error that occurs when the value of an extraneous
Interrater Reliability variable changes systematically along with the
● the agreement between their measurements; used in a independent variable in an experiment; an alternative
content analysis when raters must score the qualitative explanation for the findings that threatens internal validity
content of subjects’ responses
History Threat
Test-Retest Reliability ● a threat to internal validity in which an outside event or
● reliability of measures can also be checked by comparing occurrence might have produced effects on the dependent
scores of people who have been measured twice with the variable
same instrument
● they take the test once, then they take it again (after a Maturation Threat
reasonable interval) ● a threat to internal validity produced by internal (physical
or psychological) changes in subjects
Interitem Reliability
● the extent to which different parts of a questionnaire, test, Testing Threat
or other instruments designed to assess the same variable ● a threat to internal validity produced by a previous
to attain consistent results administration of the same test or other measure

Validity Instrumentation Threat

● formulating definitions is stating definitions that are valid ● a threat to internal validity produced by changes in the
● refers to the principle of actually studying the variables measuring instrument itself
that we intend to study
● Manipulation Check Statistical Regression Threat
○ providing evidence for the validity of an experimental ● a threat to internal validity that can occur when subjects
procedure are assigned to conditions on the basis of extreme scores
on a test; upon retest, the scores of extreme scorers tend
Face Validity to regress toward the mean even without any treatment
● the degree to which a manipulation or measurement
technique is self-evident Selection Threat
● a threat to internal validity that can occur when
Content Validity nonrandom procedures are used to assign subjects to
● the degree to which the content of a measure reflects the conditions or when random assignment fails to balance
content of what is being measured out differences among subjects across the different
conditions of the experiment
Predictive Validity
● the degree to which a measuring instrument yields Subject Mortality Threat
information allowing prediction of actual behavior or ● A threat to internal validity produced by differences in
performance dropout rates across the conditions of the experiment

Concurrent Validity Selection Interaction

● the degree to which scores on the measuring instrument ● A family of threats to internal validity produced when a
correlate with another known standard for measuring the selection threat combines with one or more of the other
variable being studied threats to internal validity; when a selection threat is
already present, other threats can affect some
Construct Validity experimental groups but not others
● the degree to which an operational definition accurately
represents the construct it is intended to manipulate or
measure

PSY 2208N NOTES PAGE 16

Experimenter Bias
CHAPTER 8: Solving Problems: Controlling Extraneous ● any behavior of the experimenter that can create
Variables confounding in an experiment

Rosenthal Effect
● the phenomenon of experimenters treating subjects
8.1 Physical Variables
differently depending on what they expect from the
subjects
Physical Variables ● also called the Pygmalion effect
● aspects of the testing conditions that need to be
controlled Double-Blind Experiment
● an experiment in which neither the experimenter nor the
Elimination subjects know which treatment condition the subjects are
● a technique to control extraneous variables by removing in; used to control experimenter bias
them from an experiment
8.3 Personality Variables
Constancy of Conditions
● a control procedure used to avoid confounding; keeping
Personality Variables
all aspects of the treatment conditions identical except for
● the personal characteristics that an experimenter or
the independent variable that is being manipulated
volunteer subject brings to the experimental setting
Balancing
8.4 Context Variables
● a technique used to control the impact of extraneous
variables by distributing their effects equally across
treatment conditions Context Variables
● extraneous variable stemming from procedures created
8.2 Social Variables by the environment, or context, of the research setting

Social Variables CHAPTER 9: Basic Between-Subjects Designs

● the qualities of the relationships between subjects and
experimenters that can influence the results of an
experiment Experimental Design
● the general structure of an experiment (but not its specific
Demand Characteristics content)
● the aspects of the experimental situation itself that
demand or elicit particular behaviors; can lead to Between-Subjects Design
distorted data by compelling subjects to produce ● a design in which different subjects take part in each
responses that conform to what subjects believe is condition of the experiment
expected of them in the experiment
9.1 Selecting and Recruiting Subjects
Single-Blind Experiment
● an experiment in which subjects are not told which of the Effect Size
treatment conditions they are in; a procedure used to ● a statistical estimate of the size or magnitude of the
control demand characteristics treatment effect

Placebo Effect 9.2 One Independent Variable: Two-Group Designs

● the results of giving subjects a pill, injection, or other
treatment that actually contains none of the independent
variable; the treatment elicits a change in subjects’ Two-Group Design
behavior simply because subjects expect an effect to ● the simplest experimental design, used when only two
occur treatment conditions are needed

Cover Story
● a plausible but false explanation of the procedures in an
experiment told to disguise the actual research
hypothesis so that subjects will not guess what it is

PSY 2208N NOTES PAGE 17

9.3 Two Independent Groups Range Matching
● creating pairs of subjects whose scores on the matching
variable fall within a previously specified range of scores
Two-Independent-Groups Design
● an experimental design in which subjects are placed in
Rank-Ordered Matching
each of two treatment conditions through random
● creating matched pairs by placing subjects in order of
assignment
their scores on the matching variable; subjects with
adjacent scores become pairs
Random Assignment
● the technique of assigning subjects to treatments so that
9.5 Multiple Groups
each subject has an equal chance of being assigned to
each treatment condition
Multiple-Groups Design
Block Randomization ● a between-subjects design with one independent
● a process of randomization that first creates treatment variable, in which there are more than two treatment
blocks containing one random order of the conditions in conditions
the experiment; subjects are then assigned to fill each
successive treatment block Multiple-Independent-Groups Design
● the most commonly used in multiple-groups design in
Experimental Condition which the subjects are assigned to different treatment
● a treatment condition in which the researcher applies a conditions at random
particular value of an independent variable to subjects
and then measures the dependent variable; in an Pilot Study
experimental group-control group design, the group that ● a mini-experiment using only a few subjects to pretest
receives some value of the independent variable selected levels of an independent variable before
conducting the actual experiment
Experimental Group
● the subjects in an experimental condition
CHAPTER 10: Between-Subjects Factorial Designs
Control Condition
● a condition in which subjects receive a zero value of the
independent variable
10.1 More Than One Independent Variable
Control Group
● the subjects in a control condition
Factorial Designs
Placebo Group ● an experimental design in which more than one
● in drug testing, a control condition in which subjects are independent variable is manipulated
treated exactly the same as subjects who are in the
experimental group, except for the presence of the actual Factors
drug; the prototype of a good control group ● an independent variable in a factorial design

Two-Experimental-Groups Design Two-Factor Experiment

● a design in which two groups of subjects are exposed to ● the simplest factorial design, having two independent
different levels of the independent variable variables

9.4 Two Matched Groups Main Effect

● the action of a single independent variable in an
experiment; the change in the dependent variable
Two-Matched-Groups Design produced by the various levels of a single factor
● an experimental design with two treatment conditions and
with subjects who are matched on a subject variable Interaction
thought to be highly related to the dependent variable ● the effect of one independent variable changes across the
levels of another independent variable; can only be
Precision Matching detected in factorial designs
● creating pairs whose subjects have identical scores on
the matching variable

PSY 2208N NOTES PAGE 18

Higher-Order Interactions ● Significant Main Effect for Sex of Subject
● an interaction effect involving more than two independent
variables

10.2 Laying Out A Factorial Design

Shorthand Notation
● a system that uses numbers to describe the design of a
factorial experiment

10.3 A Research Example

Understanding Effects from Factorial Design

● No Significant Main Effect or Interaction
● Two Significant Main Effects

● Significant Main Effect for Partner’s Sex

● Significant Subject Sex x Partner Sex Interaction

PSY 2208N NOTES PAGE 19

● Maximum Significant Subject Sex x Partner Sex
Interaction 10.4 Choosing a Between-Subjects Design

Hypothetical Mean Level of Aggression as a Function of Ion

Level and Type A-Type B Dimension

● A Significant Subject Sex Main Effect, A Significant

Partner Sex Main Effect, and a Significant Subject Sex x
Partner Sex Interaction

CHAPTER 11: Within-Subjects Design

Within-Subjects Design
● .a design in which each subject takes part in more than
one condition of the experiment; also called a
repeated-measures design

Power
● the chance of detecting a genuine effect of the
independent variable

11.1 Within-Subjects Factorial Design

Within-Subjects Factorial Design

● a factorial design in which subjects receive all conditions
in the experiment

11.2 Mixed Designs

Mixed Design
● a factorial design that combines within-subjects and
between-subjects factors

11.3 Advantages & Disadvantages of

Within-Subjects Designs

Order Effect
● change in subjects’ performance that occurs when a
condition falls in different positions in a sequence of
treatments

PSY 2208N NOTES PAGE 20

11.4 Controlling Within-Subjects Designs Latin Square Counterbalancing
● a partial counterbalancing technique in which a matrix or
square, of sequences is constructed so that each
Counterbalancing
treatment appears only once in any order position
● a technique for controlling order effects by distributing
progressive error across the different treatment conditions
Carryover Effects
of the experiment; may also control carryover effects
● the persistence of the effect of a treatment condition after
the condition ends
Fatigue Effects
● changes in performance caused by fatigue, boredom, or
Balanced Latin Square
irritation
● a partial counterbalancing technique for constructing the
matrix, or square, of sequence in which each treatment
Practice Effects
condition (1) appears only once in each position in a
● change in subjects’ performance resulting from practice
sequence and (2) precedes and follows every other
condition an equal number of times
Progressive Error
● changes in subjects’ responses that are caused by testing
in multiple treatment conditions, includes order effects, CHAPTER 12: Within-Subjects Designs: Small N
such as the effects of practice for fatigue

Subject-by-Subject Counterbalancing
● a technique for controlling progressive error for each 12.1 Small N Designs
individual subjects by presenting all treatment conditions
more than once Large N Designs
● a design in which the behavior of groups of subjects is
Reverse Counterbalancing compared
● a technique for controlling progressive error for each
individual subject by presenting all treatment conditions Small N Designs
twice, first in one order, then in the reverse order ● a design in which just one or a few subjects are used;
typically, the experimenter collects baseline data during
Block Randomization an initial control condition, applies the experimental
● a process of randomization that first creates treatment treatment, then reinstates the original control condition to
blocks containing one random order of the conditions in verify that changes observed in behavior were caused by
the experiment; subjects are then assigned to fill each the experimental intervention
successive treatment block
Baseline
Across-Subjects Counterbalancing ● a measure of behavior as it normally occurs without the
● a technique for controlling progressive error that pools all experimental manipulation; a control condition used to
subjects’ data together to equalize the effects of assess the impact of the experimental condition
progressive error for each condition
12.2 ABA Designs
Complete Counterbalancing
● a technique for controlling progressive error using all
possible sequences that can be formed out of the AB Design
treatment conditions and using each sequence the same ● a design in which a baseline condition (A) is measured
number of times first, followed by measurements during the experimental
intervention (B); there is no return to the baseline
Partial Counterbalancing condition
● a technique for controlling progressive error by using
some subset of the available sequences of treatment ABA Design
conditions ● a design in which a baseline condition (A) is measured
first, followed by measurements during the experimental
Randomized Partial Counterbalancing condition (B), followed by a return to the baseline
● the simplest partial counterbalancing procedure in which condition (A) to verify that the change in behavior is linked
the experimenter randomly selects as many sequences of to the experimental condition
treatment conditions as there are subjects for the ● also called a reversal design
experiment

PSY 2208N NOTES PAGE 21

ABAB Design Discrete trials designs using more than one subject typically
● a design in which a baseline condition (A) is measured include statistics to evaluate treatment effects.
first, followed by measurements during a treatment
condition (B), followed by a return to the baseline
CHAPTER 13: Why We Need Statistics
condition (A), to verify that the change in behavior is
linked to the experimental condition, followed by a return
to the treatment condition (B)
13.1 The Argument for Using Statistics
ABABA Design
● a design in which a baseline condition (A) is measured
first, followed by measurements during a treatment Statistics
condition (B), followed by a return to the baseline ● quantitative measurements of samples; quantitative data
condition (A), followed by a return condition (B) and a final
baseline measurement condition (A) to verify that the 13.2 Statistical Inference: An Overview
change in behavior is linked to the experimental condition
Statistical Inference
● meeting the set criterion for significance; the data do not
12.3 Multiple Baseline Design support the null hypothesis, confirming a difference
between the groups that occurred as a result of the
Multiple Baseline Design experiment
● a small N design in which a series of baselines and
treatments are compared; once established, however, a Null Hypothesis (H0)
treatment is not withdrawn ● a statement that the performance of treatment groups is
so similar that the groups must belong to the same
population; a way of saying that the experimental
12.4 Changing Criterion Designs manipulation had no important effect

Statistically Significant
Changing Criterion Design ● meeting the set criterion for significance; the data do not
● a design used to modify behavior when the behavior support the null hypothesis, confirming a difference
cannot be changed all at once; instead, the behavior is between the groups that occurred as a result of the
modified in all increments, and the criterion for success experiment
changes as the behavior is modified
Alternative Hypothesis (H1)
● a statement that the data came from different
12.5 Discrete Trials Designs populations; the research hypothesis, which cannot be
tested directly
Discrete Trials Design
● a design that relies on presenting and averaging across 13.3 Applying Statistical Inference: An Example
many, many experimental trials; repeated applications
result in a reliable picture of the effects of the independent Directional Hypothesis
variable ● a statement that predicts the exact pattern of results that
will be observed, such as which treatment group will
12.6 When to Use Large N and Small N Designs perform the best.

Whether to choose a small or large N design depends on Normal Curve

several factors. Both practical and methodological ● the distribution of data in symmetrical, bell-shaped curve
considerations come into play.
Significance Level
A small N design can be used to study the behavior of a single ● the statistical criterion for deciding whether to reject the
individual in depth, or it can be used when subjects are scarce. null hypothesis or not, typically p < .05
For example, psychologists who want to study the effects of a
therapeutic intervention on a single client with a rare Experimental Errors
psychopathology could use a small N design. A small N design ● variation in subjects’ scores produced by uncontrolled
requires fewer subjects, and it is less time-consuming when extraneous variables in the experimental procedure,
each human or animal subject requires extensive training. experimenter bias, or other influences on subjects not
related to effects of the independent variable

PSY 2208N NOTES PAGE 22

Type 1 Error Descriptive Statistics
● an error made by rejecting the null hypothesis even ● the standard procedures used to summarize and describe
though it is really true; stating that an effect exists when it data quickly and clearly; summary statistics reported for
really does not an experiment, including mean, range, and standard
deviation
Type 2 Error
● an error made by failing to reject the null hypothesis even Measures of Central Tendency
though it is really false; failing to detect a treatment effect ● summary statistics that describe what is typical of a
distribution of scores; include mean, median, mode
13.4 The Odds of Finding Significance
Mode
● the most frequently occurring score in a distribution; a
Critical Regions
measure of central tendency
● portion in the tail(s) of the distribution of a test statistic
extreme enough to satisfy the researcher’s criterion for
Median
rejecting the null hypothesis—for instance, the most
● the score that divides a distribution in half, so that half the
extreme 5% of a distribution where p <.05 is the chosen
scores in the distribution fall above the median, half
significance level
below; a measure of central tendency
Two-Tailed Test
Mean
● a statistical procedure used when nondirectional
● an arithmetical average computed by dividing the sum of
prediction has been made; the critical region of the
a group of scores by the total number of scores; a
distribution of the test statistic is divided over both tails of
measure of central tendency
the distribution
Variability
Nondirectional Hypothesis
● fluctuation in data; can be defined numerically as the
● a statement that predicts a difference between treatment
range, variance, or standard deviation
groups without predicting the exact pattern of results
Range
One-Tailed Test
● the difference between the largest and smallest scores in
● a statistical procedure used when a directional prediction
a set of data; a rough indication of the amount of
has been made; the critical region of the distribution of
variability in the data
the test statistic is measured in just one tail of the
Variance
distribution
● the average squared deviation of scores from their mean;
a more precise measure of variability than the range
13.5 Test Statistics
Standard Deviation
Inferential Statistics ● the square root of the variance; measures the average
● statistics that can be used as indicators of what is going deviation of scores about the mean, thus reflecting the
on in a population; also called test statistics amount of variability in the data

Test Statistics
● statistics that can be used as indicators of what is going
on in a population and can be used to evaluate results
● also called inferential statistics

13.6 Organizing and Summarizing Data

Raw Data
● data recorded as an experiment is run; the responses of
individual subjects

Summary Data
● descriptive statistics computed from the raw data of an
experiment, including the measures of central tendency
and variability

PSY 2208N NOTES PAGE 23

CHAPTER 14: Analyzing Results 14.2 The Chi-Square Test

Critical Value
14.1 Which Test Do I Use? ● the value of the test statistic that must be exceeded to
reject the null hypothesis at the chosen significance level

Chi-Square (x2)
● a nonparametric, inferential statistic that tests whether the
frequencies of responses in our sample represent certain
frequencies in the population; used with nominal data

Degrees of Freedom (df)

● the number of members of a set of data that can vary or
change value without changing the value of a known
statistic for those data

F Ratio
● a test statistic used in the analysis of variance; the ratio
between the variability observed between treatment
groups and the variability observed within treatment
groups

Cramer’s coefficient phi

● an estimate of the degree of association between the two
categorical variables tested by x2

14.3 The t Test

T Test for Independent Groups

● a statistic that relates differences between treatment
means to the amount of variability expected between any
two samples of data from the same population; used to
analyze the results of a two-group experiment with
independent groups of subjects

T Test
● a statistic that relates differences between treatment
means to the amount of variability expected between any
two samples of data from the same population; used to
analyze the results of a two-group experiment with one
independent variable and interval or ratio data

Robust
● a term describing a statistical test that can be used
without increasing the probability of Type 1 or Type 2
errors even though its assumptions (e.g., the population is
normally distributed and has equal variances) are violated

PSY 2208N NOTES PAGE 24

Confidence Intervals Mean Square (MS)
● a range of values above and below a sample mean that is ● an average squared deviation; a variance estimate used in
likely to contain the population mean with the probability analysis-of-variance procedures and found by dividing the
level (usually at 95% or 99%) that the mean of the sum of squares by the degrees of freedom
population (the true mean) would actually fall somewhere
in that range Mean Square Within Groups (MSw)
● the variance (or average squared deviation) within a single
T Test for Matched Groups treatment group; produced by the combination of sources
● a statistic that relates differences between treatment called error
means to the amount of variability expected between any
two samples of data from the same population; used to Mean Square Between Groups (MSn)
analyze two-group experiments using matched-subjects ● the variance (or average squared deviation) across
or within-subjects designs. Also called a within-subjects different treatment groups produced by error and
t-test treatment effects; found by dividing the sums of squares
by the degrees of freedom
14.4 Analysis of Variance
Post Hoc Tests
● statistical test performed after the overall analysis
Analysis of Variance (ANOVA)
indicates a significant difference; used to pinpoint which
● the statistical procedure used to evaluate differences
differences are significant
among three or more treatment means by breaking the
variability in the data into components that reflect the
A Priori Comparisons
influence of error and error plus treatment effects
● statistical test between specific treatment groups that
was anticipated, or planned, before the experiment was
Within-Groups Variability
conducted
● the degree to which the scores of subjects in the same
● also called planned comparison
treatment group differ from one another (that is, how
much subjects vary from others in the group); an index of
14.7 One-Way Repeated Measures Analysis of
the degree of fluctuation among scores that is attributable
Variance
to error.

Between-Groups Variability One-Way Within Subjects for Repeated Measures ANOVA

● the degree to which the scores of different treatment ● statistical procedure used to evaluate within-subjects
groups differ from one another (that is, how much experiments with three or more levels of a single
subjects vary under different levels of the independent independent variable
variable); a measure of variability produced by treatment
effects and error
CHAPTER 15: Drawing Conclusion: The Search for the
14.5 Sources of Variability Elusive Bottom Line

Errors
● the variability within and between treatment groups that is 15.1 Evaluating the Experiment from the Inside:
not produced by changes in the independent variables; Internal Validity
variability produced by individual differences,
experimental error, and other extraneous variables.
Manipulation Check
● an assessment to determine whether the independent
14.6 A One-Way Between-Subjects Analysis of
variable was manipulated successfully
Variance
Statistical Conclusion Validity
One-Way Between-Subjects Analysis of Variance ● the degree to which conclusions about a treatment effect
● statistical procedure used to evaluate a between-subjects can be drawn from the statistical results obtained
experiment with three or more levels of a single
independent variable

PSY 2208N NOTES PAGE 25

15.2 Taking a Broader Perspective: The Problem of CHAPTER 16: Writing the Research Report
External Validity
Research Reports
● written report of psychological research, which contains
Aggregation
four major sections: Introduction, Method, Results, and
● the grouping together and averaging of data gathered in
Discussion
various ways, including aggregation over subjects, over
stimuli and/or situations, over trials and/or occasions, and
16.1 The Written Report: Purpose and Format
over measures.

Multivariate Designs Scientific Writing Style

● research design or statistical procedure used to evaluate ● a concise, impersonal, and unbiased form of writing used
the effects of many dependent variables in combination, in research reports
including multiple correlation, factor analysis, and
multivariate analysis of variance 16.2 Major Sections

Multivariate Analysis of Variance (MANOVA)

Title
● the statistical procedure used to study the impact of
● the name of the report, which describes what the report is
independent variables on two or more dependent
about; typically includes the variables tested and the
variables; and the extension of analysis of variance
relationship between them
Reactivity
Abstract
● the tendency of subjects to alter their responses or
● a brief summary of the report (approximately 150-250
behaviors when they are aware of the presence of an
words), which precedes the four major sections
observer
Introduction
Unobtrusive Measures
● beginning section of a research report that guides the
● a procedure used to assess subjects’ behaviors without
reader toward your research hypothesis; includes a
their knowledge; used to obtain more nonreactive data
selective review of relevant, recent research
Field Experiments
Method
● an experiment conducted outside the laboratory that is
● the section of a research report in which the subjects and
used to increase external validity, verify earlier laboratory
experiment are described in enough detail that the
findings and investigate problems that cannot be studied
experiment may be replicated by others;
successfully in the laboratory.
● it is typically divided into subsections. such as:
○ Participants
Naturalistic Observation
○ Stimuli or Materials
● a descriptive, nonexperimental method of observing
○ Procedures
behaviors as they occur spontaneously in natural settings
Results
● the section of a research report in which the findings are
15.3 Handling a Nonsignificant Outcome
described and the results of statistical tests and summary
data are presented
There is also the practical dilemma of dealing with
nonsignificant findings. Experimental procedures must always Discussion
be reviewed for sources of internal invalidity, such as ● the fourth major section of the research report is used to
uncontrolled extraneous variables. Low statistical power and draw conclusions and integrate the experimental findings
weak manipulation of the independent variable must be into the existing body of knowledge
considered. Finally, the thinking that led to the hypothesis
should be reviewed to ensure the predictions were reasonable. References
● a list of books and articles cited in the research report; is
Ceiling Effect placed at the end of the report
● occurs when a large percentage of participants achieve
the highest score on a test

PSY 2208N NOTES PAGE 26

Method
16.3 Looking at a Journal Article ● The Method section describes in detail how the
experiment was carried out. Notice that the authors have
adapted the format of the subsections to fit the kind of
A SAMPLE JOURNAL ARTICLE
information they need to present; each subsection is
Title
clearly labeled. The Participants subsection includes
● the name of the report, which describes what the report is
important information about the subjects: It is also
about; typically includes the variables tested and the
important to describe what compensation subjects
relationship between them
received, if any. Relevant information is included to give
the reader a clear picture of the sample for purposes of
Names and Affiliation
assessing external validity and comparing results across
● The authors' names are given as they would ordinarily be
different experiments.
written; titles (Dr., Mr., Ms.) are not stated. The university,
● Manipulation of the independent variables was embedded
agency, or business affiliation of the authors is also listed.
within a written message that subjects were instructed to
read. The authors described the procedures of the
Abstract
experiment and the operationalization of the independent
● In published journal articles, the Abstract is conspicuously
variables in a single subsection labeled Procedure and
indented and set off by a smaller type. Notice that the
materials. This subsection begins with a statement about
Abstract summarizes what was done and what was
random assignment of subjects to one of the four
found; all the main points of the article are presented.
experimental conditions, and the design is described in
shorthand notation form as a 2 (message frame: gain vs.
Author Notes
loss frame) x 2 (temporal context: short-term vs. long-term
● Author Notes always include contact information for the
consequences) between-subjects design, allowing us to
"corresponding author," the person to contact about the
visualize the four conditions of the factorial design quite
article. Author Notes also provide information about
easily. So that the reader can see the messages given to
special circumstances. For instance, if the results were
the four groups of subjects, the actual words used in each
present elsewhere or were part of a dissertation or thesis
condition are shown in Table 1 on the next journal article
project. The author notes also acknowledge special
page.
contributions such as support from granting organizations
● In addition, the procedures they used in the study for both
or individuals who facilitated the research.
session one and session two were described in detail
here. Because not all subjects (<80%) returned for
Introduction
session two, the authors wisely considered the potential
● In a published article, the Introduction begins immediately
for a subject mortality threat (from Chapter 7). To eliminate
after the Abstract.
this possibility, they reported statistical analyses which
● The Introduction is labeled in some psychology journals
showed that (1) dropout rates did not differ across the
(as it is in this one), but not in others. In this section, the
four experimental conditions and (2) subjects who
authors present the logic and background research that
completed the second session did not differ in age,
suggested this study. The general problem area is
gender, ethnicity, or year in college from those who failed
mentioned in the opening paragraphs, both independent
to complete session two.
variables are introduced. This is followed by a brief review
● Finally, the authors utilized a subsection called Measures
of relevant published findings for message framing along
to describe each of the measuring instruments (surveys)
with the theory that the authors use to explain them. In
used in the experiment. Two surveys were used in the
this case, the authors have chosen "prospect theory,"
session one given before the message requesting
which has been used by other researchers to explain
demographic information (gender, age, etc.) and one after
differences in the effects of gain-framed vs. loss-framed
the message asking about their perceptions of the
health messages.
message and how they felt about it. The final survey was
given in session two. This survey asked about their
Running Head
drinking behavior since session one (frequency, quantity,
● A running head is an abbreviated title (with a maximum of
and binge drinking).
50 characters) that is printed above the pages of the
article to help readers identify it in a journal containing
Results
other articles. In their typed manuscript, which can be
● Before they reported statistical analyses testing their
found in Appendix C, Gerund and Cullen used "Message
hypotheses, the authors used a section called
Framing and Temporal Context" as their running head.
Manipulation checks to verify that the two factors, framing
Most psychology journals will use your running head, but
and temporal context, were successfully manipulated in
this journal has a different style and uses the author's
the messages given to subjects.
name and journal information instead.
● Notice how the authors used statistics to convey to the
reader that the messages successfully expressed the two

PSY 2208N NOTES PAGE 27

desired levels of framing and the two desired levels of the best tactic. They also suggest that messages
temporal context. Actual statistical values were provided stressing the short-term benefits of reducing problem
along with means and standard deviations. Gerald and drinking behaviors could easily be incorporated into
Cullen also provided statistical evidence showing that the existing intervention programs.
alcohol messages were not different on any of several ● The authors include a subsection called Study limitations
other important dimensions (e.g., level of difficulty, and directions for future research. Here, they discuss the
affective responses, etc.) that might have confounded the limitations of the study, such as the self-reports of
experiment if they had not been evaluated. drinking behaviors, which constituted the dependent
● Next, the authors reported the statistical results of their measures. They also discuss the possibility that framing
hypothesis tests in the subsection titled Effects of effects might be influenced by individual difference
message framing and temporal context on drinking variables, such as differences in the way individuals
behavior. (To see this subsection, you will need to return perceive near-term vs. future consequences. They
to the previous journal article page.) They first described conclude this section by advocating future research to
the dependent variables, the hypotheses, and the investigate the generalizability of the current findings to
statistical analyses used to test the hypotheses. Next, different areas of risky health behaviors and also to
they analyzed the effects of message framing and behaviors completely outside of the health arena. Finally,
temporal context on the quantity of drinking, number of the authors opted for a Conclusion subsection at the end,
drinking occasions, and frequency of binge drinking since which restates the major results of the experiment and the
the first session. For these analyses, they used a 2 important implications of their findings.
(message framing) x 2 (temporal context) ANOVA. They
also used a planned contrast (from Chapter 14) to References
evaluate their hypothesis that the most effective message
would be the gain-framed, short-term consequences
message.

Discussion
● Gerund and Cullen begin the Discussion section by
summarizing their major findings: Considering the benefits
of avoiding risky behavior, like drinking, produces greater
avoidance than consideration of the costs of the risky
behavior, but only when the benefits occur in the near
future (rather than in the distant future). If replicated, these
effects could have important consequences for the design
of intervention strategies for reducing many kinds of risky 16.4 Preparing Your Manuscript: Procedural Details
health behaviors.
● Next, the authors discuss how the results of their study fit
into the context of what is already known about framing Each section must contain certain kinds of information. There
persuasive messages with a goal of producing healthier is a set format for types of manuscripts, which must be
behaviors. They first explain how the results of the current followed exactly. The published and typed versions of the
study are consistent with previous framing research on report contain the same information, but they look quite
the promotion of healthy behaviors and the reduction of different.
unhealthy behaviors. Then they explain how their study
advances our knowledge by demonstrating that
interventions focusing on near-term gains are likely to be 16.5 Making Revisions
the most effective. They also explain an important
boundary condition on framing effects. Considerable past Reports are written in stages; the first stage is a draft, which is
research has shown that, in general, gain messages are then revised and polished to give the experiment the best
more effective than loss messages, but the current study possible presentation. Grammatical, spelling, or typographical
has shown that this may only be true when the errors will greatly detract from the credibility of your research
consequences are near-term rather than in the distant report, so proofread your manuscript carefully and learn to use
future, an important limitation to the benefits of positive the spelling and grammar checkers available in most
framed interventions. word-processing programs.
● Next, the authors discuss the practical implications of
their research as it applies to the problem of excessive
drinking among college students. The authors argue that
many existing intervention programs on college campuses
stress the negative consequences of problem drinking
and focus on long-term costs, neither of which might be

PSY 2208N NOTES PAGE 28