FORMULATING THE - Analytic statement = one that is

always true
HYPOTHESIS - Contradictory statement = statements
with elements that oppose each other,
Hypothesis always false
- is the thesis, or main idea, of an - a hypothesis meets the definition of a
experiment synthetic statement when it can be stated
- a statement about a predicted in what is known as the "if...then..." form
relationship between at least 2 variables
- designed to fit the type of research Testable statements
design that has been selected - the means for manipulating
antecedent conditions and measuring
Nonexperimental hypothesis the resulting behavior must exist
- a statement of your predictions of how - untestable hypotheses are not
events, traits, or behaviors might be necessarily useless
related - not a statement about cause
and effect Falsifiable statements
- some nonexperimental designs, - must be disprovable by the research
particularly those that do not restrict findings
subjects' responses do not typically - hypotheses need to be worded so that
include a hypothesis failures to find the predicted effect
- other nonexperimental designs, such as must be considered evidence that the
correlational and quasi-experimental hypothesis is indeed false
studies, generally include hypotheses
about predicted relationships between Parsimonious statements
variables - a simple hypothesis is preferred over
one that requires many supporting
THE CHARACTERISTICS OF AN assumptions
EXPERIMENTAL HYPOTHESIS
- each experimental hypothesis is a Fruitful statements
tentative explanation of an event or - it leads to new studies
behavior
- a statement that explains the effects THE INDUCTIVE MODEL
of specified antecedent conditions on - the process of reasoning from specific
a measured behavior cases to more general principles
- by examining individual instances, we
Synthetic statements may be able to construct an overall
- are those that can be either true or explanatory scheme to describe them
false - research hypotheses often come from
- each experimental hypothesis must be the use of inductive reasoning
a synthetic statement so that there can - the basic tool of theory building
be some chance it is true and some
chance it is false
- Theory = is a set of general principles SERENDIPITY AND THE WINDFALL
that can be used to explain and predict HYPOTHESIS
behavior
- through induction, researchers Serendipity
construct theories by taking bits of - the knack of finding things that are
empirical data and forming general not being sought
explanatory schemes to - appeared in the work of Ivan Pavlov
accommodate those facts - can be useful in generating new
hypotheses only when we are open to
THE DEDUCTIVE MODEL new possibilities
- is the process of reasoning from - is not just a matter of luck; it is also
general principles to make a matter of knowing enough to use an
predictions about specific instances opportunity
- most useful when we have a well-
developed theory with clearly stated INTUITION
basic premises - may be defined as knowing without
reasoning
COMBINING INDUCTION AND - closest to phenomenology
DEDUCTION - we have a hunch about what might
- in practice, these approaches are not so happen in a particular situation, so we set
neatly separated up an experiment to test it
- through induction, we devise general - guides what we choose to study
principles and theories that can be - intuition is most accurate if it comes
used to organize, explain, and predict from experts
behavior until more satisfactory - intuition should not destroy objectivity
principles are found. Through
deduction we rigorously test the WHEN ALL ELSE FAILS
implications of those theories - one method is to pick a psychology
journal from your library's shelves and
BUILDING ON PRIOR RESEARCH just read through an issue
- the most useful way of finding - observation
hypotheses is by working from - turn your attention to a real-world
research that has already been done problem and try to figure out what
- sometimes, nonexperimental studies causes it
can suggest cause-and-effect
explanations that can be translated SEARCHING THE RESEARCH
into experimental hypotheses LITERATURE
- if you do not already have a specific
hypothesis in mind, you will find the Getting started
experimental literature useful in focusing - once you have decided on a hypothesis,
your thinking on important issues you will want to become familiar with
other published studies in your topic area
- Psychological journals = periodicals - Dependent variable (DV) = the
that publish individual research reports particular behavior we expect to change
and integrative research reviews because of our experimental treatment; it
- Meta-analysis = a statistical is the outcome we are trying to explain
reviewing procedure that uses data
from many similar studies to OPERATIONAL DEFINITIONS
summarize research findings about - Conceptual definition = used in
individual topics everyday language
- Operational definition = specifies the
Writing the report precise meaning of a variable within
- Introduction = consists of a selective an experiment; defines a variable in
review of relevant, recent research terms of observable operations,
- Discussion = integrate your procedures, and measurements
experiment into the existing body of
literature on your topic Defining the IV: Experimental
operational definitions
Finding the articles you need - Experimental operational definitions
- PsycINFO = explain the precise meaning of the
- PsycARTICLES independent variables
- SAGE Full Text - these definitions describe exactly
what was done to create the various
THE BASICS OF treatment conditions of the
experiment
EXPERIMENTATION
Defining the DV: Measured
INDEPENDENT AND DEPENDENT operational definitions
VARIABLES - Measured operational definitions =
- Variables = are measurable elements describe exactly what procedures we
that can vary or take on different values follow to assess the impact of
along some dimension different treatment conditions
- Independent variable (IV) = the - these definitions include exact
dimension that the experimenter descriptions of the specific behaviors
intentionally manipulates; it is the or responses recorded and explain
antecedent the experimenter chooses how those responses are scored
to vary
- Levels of the independent variable = Defining Constructs operationally
when an independent variable has - Hypothetical constructs = unseen
multiple experimental conditions (2 processes postulated to explain
levels = 2 diff. treatments, so on and so behavior (such as anxiety), cannot be
forth) observed directly
- in a true experiment, we test the - many psychological variables are
effects of a manipulated IV - not the hypothetical constructs
effects of different kinds of subjects
- we infer their existence from Test-retest reliability
behaviors that we can observe - comparing scores of people who
have been measured twice with the
Defining Nonconstruct variables same instrument
- operational definitions are equally - they take the test once, then they take
important when we are working with it again (after a reasonable interval)
variables that can be observed more
directly Interitem reliability
- the extent to which different parts of a
Defining scales of measurements questionnaire, test, or other
- many variables can be measured in instruments designed to assess the
more than one way same variable attain consistent
- the measurement alternatives differ in results
the degree of information they provide - scores on different items designed to
- when we have a choice between measure the same construct should be
different levels of measurement, highly correlated
researchers generally choose the
highest level possible because it 2 major approaches in evaluating
provides more information about a interitem reliability
variable - Split-half reliability = involves
- interval and ratio data are more splitting the test into 2 halves at
powerful than nominal and ordinal random and computing a coefficient
of reliability between the scores
EVALUATING OPERATIONAL obtained on the 2 halves
DEFINITIONS - Using statistical tests such as
Cronbach's alpha
RELIABILITY - both approaches measure the internal
- means consistency and consistency of the test items
dependability
- if we apply them in more than one VALIDITY
experiment, they ought to work in - principle of actually studying the
similar ways each time variables that we intend to study
- Manipulation check = providing
Procedures for checking the reliability evidence for the validity of an
of measurement techniques experimental procedure

Interrater reliability Face validity

Content validity
- depends on whether we are taking a
fair sample of the variable we intent to
measure
- "does the content of our measure fairly
reflect the content of the quality we are
measuring? are all aspects of the content
represented appropriately?"
- depends on the variable we want to
measure
- the more specific the variable, the
easier it will be

Predictive validity
- do our procedures yield information
that enables us to predict future
behavior or performance?

Concurrent validity
- compares scores on the measuring
instrument with an outside criterion
- comparative rather than predictive
- evaluated by comparing scores on
the measuring instrument with
another known standard for the
variable being studied

Construct validity
- deals with the transition from theory
to research application
- most important aspect of validity
- Convergent validity = how closely a
test is related with other tests that
measure the same or similar
constructs
- Discriminant validity = the extent to
which a test is not related to other
tests that measure different
constructs