Documenti di Didattica
Documenti di Professioni
Documenti di Cultura
PSYCHOLOGY
EXPERIMENTAL PSYCHOLOGY
Psychology is the science of behavior.
As Psychologists, we take a scientific approach to understanding behavior
Anamnesis
`
Science is Knowledge
Content (what we know - facts)
Process ( an activity, includes systematic ways in which we go about
gathering data, noting relationships, offering explanations)
Bad mood
Conflicts
Rivals
Celebration
Commonsense Psychology
Everyday non-scientific data gathering that shapes our expectations and
beliefs and directs our behavior towards others.
Nonscientific data gathering can leave us up in the air (hanging).
NON-SCIENTIFIC SOURCES OF
DATA
Friends
Relatives
People in Authority
People we admire
Media
Books read
** Often not a good source of obtaining valid information about Behavior.
** Nevertheless, they are typically accepted without question.
NON-SCIENTIFIC SOURCES OF
DATA
These beliefs tend to become stable due to rare testing, people do not
validate.
Once we believe on something, we tend to overlook instances that might
disconfirm our beliefs and we seek instead for confirmatory instances of
behavior.
** Ex. Full moon brings out psychotic behavior. You will recount/remember
instances when people acted abnormally while the moon was full and will ignore
the many instances in which no unusual behavior occurred.
Thus, the term Confirmation Bias
NON-SCIENTIFIC SOURCES OF
DATA
Assimilate some myths, superstitions, pop psychology explanations for
behavior is unlikely to be avoided.
Do you believe in the power of Crystals?
Amethyst
: Increase intuition
NON-SCIENTIFIC SOURCES OF
DATA
Research has shown that we are more likely to believe information if it
comes from certain kinds of individuals:
Popular People (celebrities, artists)
Attractive
High in Status
Seemingly Expert
Highly Confident Individuals (Promos: marketing, sales)
NON-SCIENTIFIC SOURCES OF
DATA
Researchers have discovered that we are not always privy to our own
decision-making processes (Nisbett & Wilson, 1977)
French music and German music played on alternative days in a
supermarket display featuring two French and German wines of similar
price and sweetness.
French wine outsold German wine when French music was played and viceversa
Clearly, the music had an influence on purchases.
When asked, only 1 out of 44 people mentioned the music.
NONSCIENTIFIC INFERENCE
One of the first and most important kinds of data we collect about others
comes in the form of Traits we assign to them.
Classmate with a new gadget
Friend in branded clothing
:
:
Conceited (mayabang)
Vain
NONSCIENTIFIC INFERENCE
SITUATIONS Classmate with a new gadget
Friend in branded clothing
: Vain
: Loud
: malambing, soft-toned
: Conceited (mayabang)
NONSCIENTIFIC INFERENCE
For decades, Psychologists have debated whether Traits or Situations are
better predictor of behavior.
There is evidence that both sides of the debate is probably correct some
of the time.
Traits might be more useful for predicting how someone will behave
over the long term.
Situations might be better predictors of momentary behaviors.
Gamblers Fallacy People are not good at using data as basis of
prediction
Overconfidence Bias
NONSCIENTIFIC INFERENCE
Psychologists believed it to be our brains own way of coping with too
much volume of information, shortcuts of managing the information
received.
Nonscientific Inference or the conclusions reached on the basis of
evidence and reasoning not following any systematic or objective process
is not enough source of information for an experiment or research.
Unless they are recorded and validated, it shall be considered as
nonscientific
SCIENTIFIC METHOD
Gather information
Verify information
Answer questions
Explain relationships
Communicate this information to others
EXPERIMENTAL PSYCHOLOGY
Psychology is the science of behavior.
As Psychologists, we take a scientific approach to understanding behavior
Anamnesis
EXPERIMENTAL PSYCHOLOGY
Nonscientific Inference
Trait, Situation, Stereotyping
Overconfidence Bias
The need for Scientific Method
THE CHARACTERISTICS
OF MODERN SCIENCE
DATA
SCIENCE
PRINCIPL
ES
LAWS
GOOD THINKING
Central feature of the scientific method.
Approach to the collection and interpretation of data should be
systematic, objective and rational.
The scientist (should) avoid letting private beliefs or expectations
influence observations or conclusions.
Good thinking includes being open to new ideas rather than being
closed-minded even when they contradict our prior beliefs or attitudes.
It also follows the rules of logic.
Conclusions will follow from the data, whether they are in agreement with
your predictions of not.
GOOD THINKING
Another important aspect, principle of parsimony, sometimes called
Occams razor.
William Occam: Entities should not be multiplied without necessity
Simplicity, precision and clarity of thought
Avoid making unnecessary assumptions to support an argument or
explanation.
When two explanations are equally defensible, the simplest explanation is
preferred until it is ruled out by conflicting data.
SELF CORRECTION
Modern scientists accept the uncertainty of their own conclusions.
Changes in scientific explanations and theories are an extremely important part of
scientific progress.
Experience favors a weight-of-evidence approach. The more evidence that
accumulates to support a particular explanation or theory, the more confidence we
have that the theory is correct.
**Link between media violence and aggressive behavior (Social Learning Theory)
**This theory asserted that people would perform the same kinds of aggressive
behaviors they had observed in films or on television by learning to imitate through
media models (actors, actress, etc.)
PUBLICIZING RESULTS
Modern science has become a highly public activity.
Scientists meet frequently through professional and special interest
groups and attend professional conferences to exchange information.
Scientific papers, organizational journals
REPLICATION
Another important part of scientific research.
We should repeat the same procedures and get the same results again if
we have gathered data objectively and if we have followed good thinking.
Findings obtained by one researcher have very limited scientific value.
EXPERIMENTAL
PSYCHOLOGY
PROF. MARIA LOURDES J. JAVIER, MIR
JUNE 14, 2016
QUIZ
1) Define Psychology
2) What is Psychological Science
3) What are the 2 components of Science? (2 points)
4) Why do we need scientific methodology or scientific method? (5
points)
5) What is Commonsense Psychology?
6) Sources of Nonscientific Data, give 5 examples. (5 points)
7) What is nonscientific inference, expound on 1 example. (2
points)
8) Define scientific mentality.
ALTERNATIVES TO
EXPERIMENTATION
The Primary purpose of an experiment is to establish a causal
relationship between a specified set of antecedent conditions
(treatments) and the subsequently observed behavior.
Antecedent preceding; prior
INTERNAL VALIDITY
Is the degree to which a research design allows us to make causal
statements.
Research is high on internal validity if it can demonstrate with
certainty that the changes in behavior observed across treatment
conditions were actually caused by differences in treatments.
Researchers like to use laboratory experiments because they are
potentially high in internal validity.
Potentially due to varied critics:
Artificial
Unrealistic
EXTERNAL VALIDITY
Generalizability or applicability to people and situations outside the
research setting.
When observations can be generalized to other settings and other people,
they are high in external validity.
Nonexperimental designs are often preferred because they may have
greater external validity
RESEARCH ACTIVITIES
The degree of manipulation of antecedent conditions
Low to high
Letting things happen as they will to setting up carefully
controlled conditions
COIN
PAPER
DIRECTION
LOW ANTECEDENT
HIGH ANTECEDENT
Deg High
ree
of
Medium
Imp
ositi Low
on
of
Unit
s
Low - High
Low Low
Medium - High
High - High
Medium - Medium
Medium Low
High - Low
Low
Medium
High
Degree of Manipulation of Antecedent
Conditions
5 COMMON NONEXPERIMENTAL
APPROACHES USED BY PSYCHOLOGISTS
1. Phenomenology
2. Case Studies
3. Field Studies
4. Archival Studies
5. Qualitative Studies
5 COMMON NONEXPERIMENTAL
APPROACHES USED BY PSYCHOLOGISTS
These approaches form an important source of data on human and
animal behavior.
They permit us to gather information and gain understanding when
experimentation is not desirable or possible.
PHENOMENOLOGY
Is the description of an individuals immediate experience.
Rather than looking at behaviors and events that are external to
us, we begin with personal experience as a source of data.
Focus is directed towards the subjects personal experience.
As a research approach, this falls near the low-low end of our
table of research activities.
Antecedents are not manipulated, and data may consist of any
immediate experience; no constraints are imposed.
PHENOMENOLOGY
Johannes Purkinje, 1787-1869 (Jan Evangelista Purkyne)
PHENOMENOLOGY
William James
Dealt with basic psychological issues, including habits, emotions,
consciousness, and the stream of thoughts.
He approached most ideas from the phenomenological perspective of
his own experiences.
One of his most appealing passages: difficulty in getting up in the morning
He pointed out that our resistance to getting up inhibits our movement.
While we concentrate on the pleasure of warm sheets and the dread of the cold
floor, we are paralyzed.
if we dont resist, we ought to be able to get out of bed without very much
effort
PHENOMENOLOGY
Since Phenomenology deals with immediate experience, its conclusions
are instantaneous.
They emerge at once and need not wait upon the results of calculations
derived from measurements.
They do not use statistics, since a frequency does not occur at an instant
and cannot be immediately observed.
There is no assurance that observing your own behavior is free of any
alteration in some way by your attention to it.
CASE STUDIES
This method also involves no manipulation of antecedent conditions.
This is a descriptive record of a single individuals experiences, or
behaviors, or both, kept by an outside observer.
Involves systematically recording experiences and behaviors as they have
occurred over time.
Used by the following disciplines:
Clinical Psychology
Forensic Psychology
Organizational Psychology
CASE STUDY
GROUP 1
OBSERV
ER
GROUP 2
CASE STUDY
Kazdin (2003) argued that case studies serve five major purpose:
1. They are a source of inferences, hypothesis, and theories
2. They are a source for developing therapy techniques
3. They allow the study of rare phenomena
4. They provide exceptions, or counterinstances, to accepted ideas,
theories, or practices
5. They have persuasive and motivational value
CASE STUDY
1. They are a source of inferences, hypothesis, and theories
A case study may be used to make inferences about the impact of life events,
the origin of disorders, or developmental process.
This approach provided the first systematic data on the development of
childrens motor, cognitive and linguistic abilities.
Extensive records of the behaviors of individual children, arrived at descriptions
of normal developmental sequences.
Freuds case study on Little Hans (Freud, 1933): Hans was afraid of horses
Freuds analysis of Hans conversation with his Father and dreams he
reported suggests that it was a symbol for Hanss fear of his father and
anxiety about castration.
Such theory led to Oedipus Complex
CASE STUDY
2. They are a source for developing therapy techniques
As you understand the impact of such events more fully, you may be
able to devise more appropriate treatment techniques, as well as
preventive measures.
The talking cure in psychotherapy, began as a result of treatment of
hysterical symptoms in one of Freuds early cases.
Behavioral therapy techniques was formed from Mary Cover Joness
case study of a 3-yr old boy who was afraid of rabbits.
CASE STUDY
3. They allow the study of rare phenomena
Case studies are perfect forum for investigating unique cases or rare
problems.
Some variants of paraphilia (cross-dressing and sexual masochism)
Influences of testosterone therapy on a female-to-male transgendered
individual (2-yr period):
Diary entries were much more likely to mention social interactions
and the individuals connections with others at times when
testosterone was low.
Mentions of social connections were much less frequent at times
when testosterone was high
**increased testosterone can produce more anger, dominance, or
aggression which is not present in these transgendered subjects
CASE STUDY
4. They provide exceptions, or counterinstances, to accepted ideas,
theories, or practices
Case studies can provide evidence that casts doubt on theories or
generally accepted practices.
The psychoanalysis notion that it could be harmful to treat overt
symptoms of dysfunction without treating their base causes was
negated as counterinstances accumulated.
Simply treating outward symptoms with behavioral therapy
techniques is often very effective.
CASE STUDY
5. They have persuasive and motivational value
Case studies are sometimes a dramatic way to illustrate abstract
concepts.
to see is to believe
Advertisers frequently use case studies to sell products.
CASE STUDY:
LIMITATIONS
1. Working with only one or perhaps a few subjects.
Does not guarantee that they represent the general population.
2. Inability to observe the subject directly all the time.
Actual and available observation may not guarantee that it covers all
the relevant aspects of the subject.
CASE STUDY:
LIMITATIONS
4. Frequently rely on retrospective data.
Retrospective data are data collected in the present that are based on
recollections of past events.
Individuals may not accurately remember all that had happened
Memories become altered or reconstructed over time by the cognitive
system
Retrospective data has an element of bias due to the situation or
venue, mood of data provider.
FIELD STUDY
Nonexperimental approach used in the field or in real-life settings.
Often combined with various types of data gathering to capitalize on the
richness and range of behavior found outside the laboratory.
Antecedent conditions are not manipulated but the degree of constraint
on responses varies considerably from study to study.
Includes observational studies, studies on child development, market
research and studies of organizations.
FIELD STUDY:
DIFFERENT TYPES
1. Naturalistic Observation Studies
Is the technique of observing behaviors as they occur spontaneously
in natural settings.
Descriptive method, like phenomenology and the case study method,
no manipulation of antecedent conditions.
Subjects responses are free to vary.
This approach as been extensively used in animal behavior research
called ethology to study behavior in the wild as well as in captivity.
Researchers attempt to remain inconspicuous (stay out of sight).
Requires flexibility and adaptability to vast array of responses,
unanticipated and unconventional responses.
Requires decisiveness on who and when to observe and what to
FIELD STUDY:
DIFFERENT TYPES
Requires a lot of expertise not only in ethnography, but also in the
field that is being studied.
Dealing with specific samples of time may or may not contain the
behaviors we want to observe.
Multiple forms of data may need to be collected and this can be
difficult to compile and interpret.
Difficult to translate the findings of qualitative field studies into
design.
FIELD STUDY:
DIFFERENT TYPES
2. Participant-Observer Studies
The researcher becomes part of the group being studied.
Sometimes, this is the only method that can be used to study a group
especially when voluntary cooperation in not possible in a research
investigation (e.g. social experiments)
Group members are not told they part of the study to ensure objective
observations.
Does not usually include systematic observation or measurement
techniques.
Data gathered tend to be qualitative, impressions by researcher are
merely described.
FIELD STUDY:
DIFFERENT TYPES
Possibility of altering subjects behavior by mere presence of observer
in unknown ways.
Observers at times find difficulty to remain objective and unbiased.
At times, special relations are formed like friendship which makes it
more difficult to remain an objective scientist.
May contribute to some ethical concerns in the group, community.
**Behaviors are observed and recorded as they occur in the natural
setting, and subjects interviewed in the wild where the contaminating
effects of a laboratory setting are absent.
ARCHIVAL STUDY
Descriptive research method in which already existing records are
reexamined for a new purpose.
Vast amount of data collected by the government, private agencies,
hospitals, businesses, schools, institutes.
Information about crime, death rates, education levels, salaries, housing
patterns, disease rates are accessible to any researcher.
** Job mismatch: Product of Education vs. Available Industry Work
QUALITATIVE RESEARCH
Relies on words rather than numbers.
Focus on self-reports, personal narratives, and expression of ideas,
memories, feelings and thoughts.
Without understanding qualitative methods, one cannot fully appreciate
the breadth or research methodology.
Used to study phenomena that are contextual.
CONSTRUCTING SURVEYS
1. Map out your research objectives (as specific as possible)
Attitudes of psych. students toward group activities
Which group activity?
Activities done outside? or inside the classroom?
Sit-down activities? or Mobile activities?
EXAMPLES
Demographics
Census
Epidemiology
Health surveys
Marketing
Customer satisfation
Politics
Polls
Psychology
Attitudes, Emotions
Sociology
Social trends
68
PSYCHOLOGICAL TOPICS
IN SURVEY RESEARCH
Attitudes and opinions : e.g., older people's feelings about
extending
retirement age
Behaviours
Motivations
Emotions
Satisfaction
MEASURING RESPONSES
Because statistics deals with data and data are the result of
measurement, thus, the need to discuss measurement of responses.
Open-ended questions illicit answers in narrative form, rather than
selecting from a choice of responses.
Closed questions, on the other hand, use a number of different formats to
illicit a response from a choice of responses.
The type of format will largely determine the kinds of statistical analyses
we can conduct on the data.
Different types of formats comprise different levels of measurement.
MEASURING RESPONSES:
LEVEL OF MEASUREMENT
Is the kind of scale used to measure a response.
Different statistical tests are required for different levels of measurement
or measuring scale.
Theoretically, a measuring scale can have one or more of the following
mathematical attributes: magnitude, an equal interval between adjacent
units, and an absolute zero point:
Magnitude - the magnitude of a mathematical term tells you how big that
term is. In math, this means how far away the math term is from zero.
Equal Interval Between Adjacent Units - means that there are equal amounts
of the variable being measured between adjacent units on the scale.
Absolute Zero Point may refer to absence, no value or lowest point
increase
in age from 21 to 22 would be
(depending on the variable)
An
the same as an increase in age from 60 to
MEASURING RESPONSES:
4 LEVELS OF MEASUREMENT
1. NOMINAL SCALE
A nominal scale is the lowest level of measurement and is most often
used with variables that are qualitative (categorical) in nature rather
than quantitative (continuous).
When a nominal scale is used, the variable is divided into its several
categories.
These categories comprise the units of the scale, and objects are
measured by determining the category to which they belong.
Measurement with a nominal scale really amounts to classifying the
objects and giving them the name (hence, nominal scale) of the
category to which they belong.
MEASURING RESPONSES:
4 LEVELS OF MEASUREMENT
It is important to note that because the units of a nominal scale are
categories, there is no magnitude relationship between the units of a
nominal scale. LES
That is, there is no quantitative relationship between the categories.
A fundamental property of nominal scales is that of equivalence.
All members of a given class are the same from the standpoint of the
classification variable.
An operation often performed in conjunction with nominal
measurement is that of counting the instances within each class.
MEASURING RESPONSES:
4 LEVELS OF MEASUREMENT
The frequency allows us to compare the number of items within each
category.
A nominal scale does not posses any of the mathematical attributes of
magnitude, equal interval, or absolute zero point. It merely allows
categorization of objects into mutually exclusive categories.
MEASURING RESPONSES:
4 LEVELS OF MEASUREMENT
2. ORDINAL SCALES
An ordinal scale represents the next higher level of measurement.
It possesses a relatively low level of the property of magnitude.
We rank-order the objects being measured according to whether they
posses more, less, or
Allows determination of whether A > B, A = B, or A < B.
Although this scale allows better than, equal to, or less than
comparisons, it says nothing about the magnitude of the difference
between adjacent units on the scale.
MEASURING RESPONSES:
4 LEVELS OF MEASUREMENT
The ordinal scale does not have the property of equal intervals
between adjacent units.
The scale does not tell the absolute level of the variable (e.g., they all
could be high or they all could be low). the same amount of the
variable being measured.
MEASURING RESPONSES:
4 LEVELS OF MEASUREMENT
3. INTERVAL SCALES
The interval scale represents a higher level of measurement than the
ordinal scale.
It possesses the properties of magnitude and equal intervals between
adjacent units but does not have an absolute zero point.
The interval scale possesses the properties of the ordinal scale and
has equal intervals between adjacent units.
The phrase equal intervals between adjacent units means that there
are equal amounts of the variable being measured between adjacent
units on the scale.
MEASURING RESPONSES:
4 LEVELS OF MEASUREMENT
Since with an interval scale there are equal amounts of the variable
between adjacent units on the scale, equal differences between the
numbers on the scale represent equal differences in the magnitude of the
variable.
It also follows logically that greater differences between the numbers on
the scale represent greater differences in the magnitude of the variable
being measured, and smaller differences between the numbers on the
scale represent smaller differences in the magnitude of the variable being
measured.
In addition to being able to determine whether A > B, A = B, or A < B, and
interval scale allows us to determine whether A B = C D, A B > C D,
or A B < C D.
MEASURING RESPONSES:
4 LEVELS OF MEASUREMENT
4. RATIO SCALES
The next, and highest, level of measurement is called a ratio scale.
It has all the properties of an interval scale and, in addition, has an
absolute zero point.
Without an absolute zero point it is not legitimate to compute ratios
with the scale readings.
A zero on the ratio scale represents the complete absence of
We cannot compute ratios with the interval scale. That is, we cannot say
for example that 20o Celsius is twice as hot as 10o Celsius.
However, on the ratio scale we can indicate ratios. For example, on the
Kelvin scale 20o is twice as hot as 10o.
5
4
3
Pasteurized
Milk
5
4
3
Free of skin
5
4
3
2
1
Order of
Brand
blemish
Preference
Good value
5
4
3
2
1
Please indicate
1
Rambo your views on Balkan Olives by ticking the appropriate
2
R.I.P.
Killalot
4 Succulent
D.O.A.
5 Freshness Bugdeath
Freedom from skin blemish
Value for money
Attractiveness of packaging
responses below:
Excellent
Very Good
Good
Fair
Poor
IMPORTANT CONSIDERATIONS
FOR SURVEY ITEMS
1. According to Czaja and Blair (1996), the first question should have these
characteristics:
a. Relevant to the Central Topic
b. Easy to Answer
c. Interesting
d. Answerable by most respondents
e. Close format
IMPORTANT CONSIDERATIONS
FOR SURVEY ITEMS
2. The first few questions should be ones that:
a. Subjects do not have to think about a great deal (no open-ended
questions)
b. Subjects are able to answer without saying I dont know
c. What subjects will think of are relevant to the topic of survey
3. Consider placing demographic information (e.g. age, vital statistics,
income, race, etc.) at the end of the questionnaire. People tend to be lazy,
sensitive.
a. Create questions with care and sensitivity if and when it is vital to
the study.
IMPORTANT CONSIDERATIONS
FOR SURVEY ITEMS
What is your age?
18 24
25 44
45 64
65 years and over
IMPORTANT CONSIDERATIONS
FOR SURVEY ITEMS
4. Make sure your questions are not value laden. Do not word your
questions in ways that would make a positive (or negative) response
seem embarrassing or undesirable:
Version 1:
Do you believe doctors should be allowed to kill unborn
babies during the first trimester of pregnancy?
Version 2:
Do you believe doctors should be allowed to terminate a
pregnancy during the first trimester?
IMPORTANT CONSIDERATIONS
FOR SURVEY ITEMS
RESPONSE STYLES
Are tendencies to respond to questions or test items in specific ways,
regardless of the content (Cronbach, 1950; Rorer, 1965).
People differ in response styles, such as Willingness to Answer, position
preferences, across countries, nationalities, languages.
Unwillingness to answer is often a problem in survey research (e.g.
unsure, blank answers, take a guess, omit items, etc.).
Position Preference is another concern (safe answer: c, maybe, uncertain).
Manifest Content is the plain meaning of the words that actually appear
on the page (answerable by true false yes no).
IMPORTANT CONSIDERATIONS
FOR SURVEY ITEMS
RESPONSE STYLES
Questions needs to be pretested before actual data collection.
Before pretesting, read each question out loud to ensure correctness, clarity and easy to
understand.
Solicit any problems or difficulties that made a question hard to answer.
Modify any question/s that seemed difficult for the volunteers and pretest the new
questions until you are satisfied that they are reasonably easy to most people to
answer.
Pretesting can identify Context Effects. Sometimes the position of a question can
influence how the question is interpreted. This is likely to happen when two questions
are related.
Sometimes it helps to separate related items by Buffer Items. These are questions
unrelated to either of the related questions.
DIMENSIONS OF RESEARCH
Antecedent Manipulation
Treatments
Independent variable (IV)
Imposition of Units
Behavioral measures
Dependent variable (DV)
CORRELATIONAL AND
QUASI-EXPERIMENTAL DESIGNS
CORRELATIONAL DESIGNS
The investigator focuses attention on two or more variables to
determine whether they are related
When two things are correlated, changes in one are associated
with changes in another.
Antecedents are preexisting
Degree of imposition of units high
Tend to be higher in external validity
Values is between -1.00 and + 1.00. The sign (-,+) tells us the
positive or negative direction of the relationship.
CORRELATIONAL DESIGNS
STEPS:
1. Selected traits or behaviors of interest are measured (first).
2. Numbers (e.g. scores, ratings, etc.) are recorded that represent the
measured variables.
3. The degree of relationship or correlation, between the numbers is
determined through statistical procedures.
EXAMPLE:
CORRELATIONAL DESIGN
Study / Experiment:
To determine whether obesity and high blood pressure are related in adults
older than 30 years,
The researcher might measure the fat level and blood pressure of
individuals in a sample of adults older than 30.
The researcher would then analyze the results to see whether a relationship
exists between these variables; that is,
Do individuals with low fat levels also have low blood pressure?
Do individuals with moderate fat levels have moderate blood pressure?
Do individuals with high fat levels have high blood pressure?
CORRELATIONAL DESIGNS
Imposition of Units
High
Low/
High
Medium
Low
Low
Medium
High
Antecedent Manipulation
QUASI-EXPERIMENTAL DESIGNS
Can seem like an experiment, but subjects are not
randomly
assigned to treatment conditions.
Antecedent control varies
Degree of imposition of units - high
Tend to be higher in external validity
QUASI-EXPERIMENTAL DESIGNS
Imposition of Units
High
Low /
High
Meduim /
High
Low
Medium
Medium
Low
High
Antecedent Manipulation
EXAMPLE OF A QUASIEXPERIMENT
Lighting condition
: Fluorescent vs Incandescent
Subjects
: From company A (fluorescent lights) or B
(incandescent)
Performance measure : productivity
Can cause-effect be established with confidence?
HYPOTHETICAL RELATIONSHIPS
B.
Negative r = -.72
Positive r = +.69
Variable Y
Variable Y
A.
Variable X
SCATTERPLOTS
Visual representations of the scores belonging to each subject in a
study.
Each dot = two scores (x,y) from one subject.
One score places the dot along the horizontal axis (x) and the other
score places it along the vertical (y) axis.
Regression Lines (of best fit) represent the mathematical equation
that best represents the relationship between the two measured
scores.
Positive Correlation - when the computed value of r is positive.
Negative Correlation - when computed value of r is negative.
PREJUDICE PROFILING
19
7
15
6
22
9
12
6
9
4
19
7
16
8
21
9
24
5
13
5
10
7
12
4
17
8
23
9
19
10
23
10
18
5
11
6
10
4
19
8
24
8
22
7
r=
0.640658
LINEAR CORRELATIONS
Correlations that follow a line
CHARACTERISTICS OF A
RELATIONSHIP
A correlation is a numerical value that describes and measures three
characteristics of the relationship between X and Y. These three
characteristics are as follows:
1. The Direction of the Relationship. The sign of the correlation, positive or
negative, describes the direction of the relationship.
a. In a Positive Correlation, the two variables tend to change in the
same direction: As the value of the X variable increases from one
individual to another, the Y variable also tends to increase; when the
X variable decreases, the Y variable also decreases.
b. In a Negative Correlation, the two variables tend to go in opposite
directions. As the X variable increases, the Y variable decreases.
That is, it is an inverse relationship.
CHARACTERISTICS OF A
RELATIONSHIP
2. The Form of the Relationship. In the preceding coffee and beer
examples, the relationships tend to have a linear form; that is, the
points in the scatter plot tend to cluster around a straight line. We have
drawn a line through the middle of the data points in each figure to help
show the relationship. The most common use of correlation is to
measure straight-line relationships. However, other forms of
relationships do exist and there are special correlations used to measure
them.
CHARACTERISTICS OF A
RELATIONSHIP
3. The Strength or Consistency of the Relationship. For a linear
relationship, for example, the data points could fit perfectly on a
straight line. Every time X increases by one point, the value of Y also
changes by a consistent and predictable amount.
TEMPERATU
RE
SALES
14.2
$215
16.4
$325
11.9
$185
15.2
$332
18.5
$406
22.1
$522
19.4
$412
25.1
$614
23.4
$544
18.1
$421
22.6
$445
17.2
$408
HOW TO CALCULATE
(PEARSON'S CORRELATION)
Let us recall the two sets of data "x" and "y" (in our case Temperature is x
and Ice Cream Sales is y):
Step 1: Find the mean of x, and the mean of y
Step 2: Subtract the mean of x from every x value (call them "a"), do the
same
for y (call them "b")
Step 3: Calculate: a b, a2 and b2 for every value
Step 4: Sum up a b, sum up a2 and sum up b2
Step 5: Divide the sum of a b by the square root of [(sum of a2) (sum of
b2)]
AS A FORMULA:
Where:
is Sigma, the symbol for "sum up"
STUDE
NT
AGE
STUDE
NT
AGE
17
14
16
17
15
16
16
16
19
ITE
M
16
17
19
HANKY
15
18
17
CP
19
19
17
NECKLACE
17
20
17
RING
19
21
18
ID LACE
19
22
18
WATCH
10
15
23
17
11
15
24
17
12
18
25
17
13
19
26
17
CORRELATION
Correlation coefficients can be strongly affected by several
features of the data:
1. No Linear Trend (nonlinear trend)
Relationships between some psychological variables,
such as the relationship between test anxiety and
test performance, do not fit a linear model; instead,
their relationship appears curvilinear like an inverted
U
CORRELATION
2. Range Truncation
An artificial restriction of the range of values of X or Y.
In statistics, truncation results in values that are limited
above or below, resulting in a truncated sample
** Correlation between childrens age and shoe size (4-16):
. As children get older, their shoe size increases. (r = +.67)
. What if focus was limited to 8 and 9 yrs. Old children.
. There will be no strong positive trend. Area on the box looks more like
CORRELATION
3. Outliers
Extreme scores
an outlier is an observation point that is distant from other
observations
An outlier may be due to variability in the measurement or it
may indicate experimental error; the latter are sometimes
excluded from the data set
Presence may dramatically reduce the size of the correlation
coefficient because it disturbs the general linear trend of the
data.
Causes of Outliers:
Outliers can have many anomalous causes.
A physical apparatus for taking measurements may have
suffered a transient malfunction.
There may have been an error in data transmission or
transcription.
Due to changes in system behaviour, fraudulent behaviour,
human error, instrument error or simply through natural
deviations in populations.
A sample may have been contaminated with elements from
outside the population being examined.
Alternatively, an outlier could be the result of a flaw in the
assumed theory, calling for further investigation by the
researcher.
Additionally, the pathological appearance of outliers of a certain
Bidirectional Causation
The behaviors could affect each other (TV violence and Aggressive Behavior)
CORRELATION
Coefficient Determination (r)
Estimates the amount of variability in scores on one variable
that can be explained by the other variable.
An estimate of the strength of the relationship between them.
E.g., if r = .56, then r 2 = .31.
31% of the variability in scores on variable X can be accounted
for by variable Y.
An r 2 .25 (Cohen, 1988) can be considered a strong
association.
MULTIPLE REGRESSION
MULTIPLE CORRELATION
Multiple Correlation
=>
LINEAR REGRESSION
Y = Y + r [Sy / Sx] (X
X)
Sx = standard deviation of x
X = value of X
X = mean score of X
I WATCH TELEVISION.
Factor 1 : Pass Time
Factor 2 : Entertainment
Factor 3 : Information
Factor 4 : Escape
Factor 5 : Relaxation
Factor 6 : Status Enhancement
** 2 5 questions per category
FACTOR ANALYSIS
Common correlational procedure used when individuals are measured on a large number
of items.
FA allows to see the degree of relationship among many traits or behaviors at the same
time. (*computers, for statistical data analysis)
Commonly used in personality research.
Warm/cold, shy/sociable, dominant/submissive
Sociability, agreeableness, conscientiousness, emotionality, culture
FACTOR ANALYSIS
FA determines the factors and sorts items according to their groupings in
the different factors, allowing the researcher to identify items belonging in
each factor.
Items with high factor loadings are grouped together, showing that
subjects scores on these items are all strongly interrelated.
FA also sorts the factors in order of importance: Factor 1 is the most
important because it accounted for more variation in all of the subjects
scores.
Another important use, data reduction. When a researcher create a new
questionnaire or scale, they often begin by testing man more items than
what will be used (actual). FA helps determine which items seem to be
measuring similar qualities, thus, eliminating redundant items.
MULTIPLE CORRELATION
Combined correlation (R) between Years of Education and Motivation with a
Salespersons Annual Sales
Predict/explain a persons current compensation based on a number of
employee characteristics
Predict/explain how much money a salesperson will bring in based on
education and score on motivation assessment scale
Serve as guide in hiring people
Serve as guide in promoting people
MULTIPLE REGRESSION
When more than 2 related behaviors are correlated.
Can be used to predict the score on one behavior from scores on the
others.
Y = b1 X1 + b2 X2 + a
ORY = a + b1 X1 + b2 X2
There were 4 applicants with information below. What would be the best
prediction of how much sales (money) can a person bring in on an annual
basis.
EDUCATION
MOTIVATION
(YRS)
SCORE
10
30
13
49
11
55
15
25
3. Calculated correlations:
Highest year of school completed with Motivation
Highest Year of School Completed with Dollars in Sales
Motivation with Dollars in Sales
CORRELATIONS
1. Correlation between Highest Year of School and Motivation
(r X1, X2) = 0.968
2. Correlation between Highest Year of School and Annual Sales
(r Y, X1) = 0.880
3. Correlation between Motivation and Annual Sales
(r Y, X2) = 0.772
R = .9360
MULTIPLE REGRESSION
FORMULA
Y
= b 1 X1 + b 2 X2 + a
ORY
= a + b1 X1 + b2 X2
= prediction constant
Note that X1 and X2 has no assigned values. These are the places
where you indicate the values you are going to use to make a
prediction. In this case, X1 = 13 and X2 = 49.
PREDICTION
Job Applicant:
13 years of education
With 49 motivation
score (motivation scale)
Best Prediction of
Earning:
$ 685,881.74
CAUSAL MODELING
1. Path Analysis
the cause
XY
YX
X Y
X2 X Y
XY
YX
X Y
X1 : Years of
PATH ANALYSIS
Education
X2 : Motivation
Path Analysis is one sophisticated statistical technique that
Score
can be used with correlational data (Baron & Kenny, 1986;
Holmbeck, 1997).
PATH ANALYSIS
Path analysis involves the identification of mediator variables and
moderator variables.
A mediator variable is a variable that is used to explain the correlation
between two variables.
A moderator variable is a variable that affects the direction or strength of
the correlation between two variables.
TH
A
P
POVERTY
PA
TH
PATH A
PSYCHOLOGIC
AL DISTRESS
** The final step in path analysis is to show that when the correlations between
Paths B and C are taken into account using a statistical procedure, the
correlation observed initially for path a (between poverty and distress) becomes
zero.
This is exactly what Evans and his colleagues found. Their path analysis
allowed them to say that the relationship between poverty and childrens
distress can be explained by, or is mediated by, the degree of chaos in the
home.
Several analysts have argued that two waves are insufficient to obtain
much information about how processes unfold over time.
Age 8
Time watching
TV
r = .20
r = .07
Size of
Vocabulary
r = .14
Time watching
TV
r = -.59
r = .41
r = .05
Size of
Vocabulary
PURPOSE
Correlation
Multiple Correlation
Factor Analysis
Path Analysis
THAN
NONEQUIVALENT GROUPS
A manipulation is carried out but subjects are not randomly assigned
to groups
Internal validity can be increased by controlling extraneous variables
after careful consideration of potential confounds.
Ex.
Comparing 8-year-old children and 10-year-old children
People with an eating disorder and those with no disorder
Comparing children from a single-parent home and those from a
two-parent home.
NONEQUIVALENT GROUPS
Because this type of
research compares
preexisting
groups, the researcher
cannot control the
assignment of
participants to groups and
cannot
ensure equivalent groups.
LONGITUDINAL DESIGNS
Measure the behavior of the same group of subjects across time.
A form of within-subject design
the same sample of respondents is surveyed more than once
Important for studying growth and development and aging
Because of the correlational nature of survey data, it is difficult to identify
the causes of individuals changes over time.
As people drop out of the study over time (attrition), the final sample may
no longer be comparable to the original sample or represent the
population.
Time consuming and hard to conduct
LONGITUDINAL DESIGNS
1. Changes in life satisfaction before and after divorce in an 18-year
longitudinal study of German households that began in 1984
(Lucas, 2005)
2. Changes in attitudes and behaviors related to eating during the
transitions from college to early adulthood and from early
adulthood to middle years (Heatherton, Mahamedi, Striepe, Field,
& Keel, 1997; Keel, Baxter, Heatherton, & Joiner, 2007).
3. Behavioral changes in firstborn children after the birth of a
second child (Stewart, Mobley, Van Tuyl, Salvador, 1987).
1 month before birth; 1,4,8,12 months after the birth.
CROSS-SECTIONAL STUDIES
Investigates changes across time by comparing groups of subjects
already at different stages at a single point in time.
Typically requires more subjects than the longitudinal study.
Subjects may differ in ways other than those being studied (similar
to Ex post facto).
CROSS-SECTIONAL STUDIES
1. Life satisfaction after divorce.
Comparison could have been based between divorced and undivorced
subjects
Comparison could have been based between those planning to divorce and
divorced subjects
PRETEST/POSTTEST DESIGN
Investigates the effects of a
treatment by comparing
behavior before and after the
treatment.
Practice effects (pretest
sensitization)
Outside influences cannot be
ruled out
Low internal validity
FORMULATING HYPOTHESIS
Is the thesis, or main idea of an experiment.
It is a statement about a predicted relationship between at least two
variables (speculation, guess, hunch).
Some nonexperimental designs are used to demonstrate relationships
between sets of behaviors, but they may not be used to infer a causeand-effect relationship between them (true experiments).
Nonexperimental hypothesis is a statement of predictions of how events,
traits, or behaviors might be related, not a statement on cause-andeffect.
Each experimental hypothesis is a tentative explanation of an event or
behavior
RESEARCH ETHICS
1. The APA Guidelines
8. Research and Publication
8.01 Institutional Approval
8.02 Informed Consent to
Research
8.03 Informed Consent for
Recording Voices and Images in
Research
8.04 Client/Patient, Student, and
Subordinate Research Participants
8.05 Dispensing With Informed
Consent for Research
8.06 Offering Inducements for
RESEARCH ETHICS
2. Fraud in Science
Researchers publishing false data.
Peer review is an important part of the reporting process. It
ensures merit of submission, identifies problems and suggests
recommendations for improvements.
The reviewers and the editor are the first line of defense against
data falsification.
Replication is the second line of defense. Being able to replicate
a research or study is an indication that it is truthful.
RESEARCH ETHICS
3. Plagiarism
Is a common kind of fraud committed unintentionally,
intentionally and carelessly.
We must be careful in the way we draw on the work of others in
our presentations.
To plagiarize means to represent some elses ideas, words or
written work as your own.
It is not merely borrowing facts and figures, it includes using it
without proper credit.
RESEARCH ETHICS
4. Ethical Reports
Ethical Principles of Psychologists and Code of Conduct provides
further guidance concerning ethics and research reports.
Section 8.12:
Psychologists take responsibility and credit, including authorship
credit, only for work they have actually performed or to which
they have substantially contributed.
EXPERIMENTAL METHOD
The Basics of Experimentation
1. Independent Variable (IV)
Is the dimension that experimenter intentionally manipulates.
It is the antecedent the experimenter chooses to vary.
This variable is independent in the sense that its values are
created by the experimenter and are not affected by anything else
that happens in the experiment.
E.g. Lightning (bright or dim), noise levels (loud or soft); difficulty
(easy, hard), meaningfulness (nonsense syllables vs real words);
anxious vs unanxous, happy vs sad, succeeding vs failing.
EXPERIMENTAL METHOD
2. Dependent Variable
Is the particular behavior we expect to change because of our
experimental treatment.
It is the outcome we are trying to explain.
In an experiment, we are testing effects of IV on the DV. Because we
manipulate the IV and measure its effects on the DV, dependent
variables are sometimes called dependent measures.
EXPERIMENTAL METHOD
Independent Variable
Treatments
Manipulations
Interventions
Conditions
Dependent Variable
Measures
Effects
Outcomes
Results
EXPERIMENTAL METHOD
The Problem
1. Solvable Problem
Is one that posses a question that can be answered with the use
of our normal capacities.
2. Unsolvable Problem
Raises a question that is unanswerable.
Often concern supernatural phenomena or questions about ultimate
causes (e.g. what caused the universe to exist studies in religion and
classical philosophy)
SOLVABLE PROBLEM
1. The Proposed Solution is Testable.
Not all questions can be answered by science.
A problem can qualify for scientific study only if it is solvable.
A problem is solvable if it can be empirically answered in a yes
or no fashion.
A problem is solvable if , and only if, one can empirically test its
tentative solution (hypothesis).
SOLVABLE PROBLEM
2. The Proposed Solution is Relevant to the Problem.
A hypothesis that is relevant to a problem is one that can solve the
problem if it is true.
What is a testable hypothesis?
A hypothesis is testable if, and only if, it is possible to determine that
it is either true or false.
SOLVABLE PROBLEM
2. The Proposed Solution is Relevant to the Problem.
THE HYPOTHESIS
THE CHARACTERISTICS OF AN
EXPERIMENTAL HYPOTHESIS
1. Synthetic Statements
Are those that can be either true or false.
Terminology was borrowed from the filed of Logic, neither true of false, yet to
be proven.
Each experiment should be a synthetic statement so that there can be some
chance it is true and some chance it is false.
An analytic statement is one that is always true (e.g. I am pregnant, weight of
dieters will fluctuate)
Contradictory statements are statements with elements that oppose each other
(false statements).
I have a brother and I do not have a brother;
If. then form/statements (if 1st born children will be followed by another
birth, then there will changes in behavior)
THE CHARACTERISTICS OF AN
EXPERIMENTAL HYPOTHESIS
2. Testable Statements
The means for manipulating antecedent conditions and measuring the
resulting behavior must exist.
If dogs display muscle twitches and vocalizations during sleep, then they
must be dreaming
manipulation of a number of antecedents to encourage sleep (warm vs.
room temperature food, milk vs. water, active play vs. relaxed activity
during the day, etc.)
Some might sleep more than others
Some might bark and twitch more that others while asleep
How do we know they are dreaming? Ask them?
THE CHARACTERISTICS OF AN
EXPERIMENTAL HYPOTHESIS
3. Falsifiable Statements
Disprovable by the research findings.
Hypothesis need to be worded so that failures to find the
predicted effect must be considered evidence that the
hypothesis is indeed false.
If you read this book carefully enough, then you will be able
to designs a good experiment
Subjects did not read the book carefully enough.
THE CHARACTERISTICS OF AN
EXPERIMENTAL HYPOTHESIS
4. Parsimonious Statements
Parsimony simplest explanation is preferred.
A simple hypothesis is preferred over one that requires many
supporting assumptions.
(+) If you look at an appealing photograph, then your pupils will
dilate
() If you look at an appealing photograph, then your pupils will
dilate if it is a warm Saturday in June.
THE CHARACTERISTICS OF AN
EXPERIMENTAL HYPOTHESIS
5. Fruitful Statements
It leads to new studies.
It is often difficult to know in advance which hypotheses will be the
most fruitful.
Classical Conditioning (fear of otherwise neutral objects could be
acquired through learning):
If a child is repeatedly exposed to a loud, cry-inducing noise in the
presence of a harmless furry animal, then the child will begin to cry at
the sight of the animal alone.
Confirmation of this study lead to the multitude of studies on classical
conditioning in human subjects until today.
INDUCTIVE MODEL
(OF FORMULATING A
HYPOTHESIS)
INDUCTIVE MODEL
(OF FORMULATING A
HYPOTHESIS)
INDUCTIVE MODEL
(OF FORMULATING A
HYPOTHESIS)
conclusions
The more we are exposed to something, the more we
like it.
(advertising and political campaigns)
DEDUCTIVE MODEL
(OF FORMULATING A
HYPOTHESIS)
DEDUCTIVE MODEL
(OF FORMULATING A
HYPOTHESIS)
INTUITION
Using intuition is not necessarily unscientific; rather, the inferences drawn
from intuition can sometimes violate scientific criteria
Intuition may be defined as knowing without reasoning. As such, it is
probably closest to phenomenology.
We acquire phenomenological knowledge simply by attending to our own
experience.
We have a hunch, a guess, a feeling about what will happen in a
particular situation, so we set up an experiment to test it.
Intuition guides what we choose to study.
GOOD LUCK!