Introduction to Behavioral Research Methods Leary
Chapter 1 Research in the Behavioral Sciences
Goals of behavioral research
- Basic research: to increase our knowledge and to understand psychological
processes without regard for whether the knowledge is immediately
applicable.
- Applied research: to find solutions for certain problems
- Evaluation research: to assess the effects of social or institutional programs on
behavior
Describing behavior
- Survey researchers conduct studies of randomly selected respondents to
determine what people think, feel, and do.
Predicting behavior
- Personnel psychologists try to predict employees’ job performance from
employment tests and interviews. Developing ways to predict job performance,
school grades, or violent tendencies requires considerable research.
Explaining behavior
- We don’t really understand something until we explain it. Until we can explain
why something is this and something is that, the picture is not complete.
Behavioral science and common sense
- Common sense: things that we all know already
o Common sense can interfere with scientific progress; scientists’ own
commonsense assumptions about the world can blind them to find
alternative ways of thinking.
Scientific approach
- Systematic empiricism: empiricism is the practice of relying on observations to
draw conclusions about the world. Conclusions are based on what can
objectively be observed and not on assumptions, hunches, unfounded beliefs,
or the products of people’s imaginations. By measuring the results instead of
only observing, it is systematic empiricism.
- Public verification: research must be conducted in such a way that the findings
of one researcher can be observed, replicated, and verified by others
o This ensures that the phenomena are real and observable
o And this makes science self-correcting
- Solvability: scientist can only investigate those questions that are answerable
given the current knowledge and research techniques. Many questions fall
outside the realm of scientific investigation.
Theory
- A set of propositions that attempts to explain the relationships among a set of
concepts. A good theory: proposes casual relationships, is clear,
, straightforward, logical, and consistent, parsimonious, generates testable
hypotheses, stimulates other researchers, and solves an existing theoretical
question.
Model
- Describes only how concepts are related, whereas a theory specifies both how
and why. A model tries to describe the hypothesized relationships and a
theory tries to explain them.
Post hoc explanations: explanations that are made after the fact.
A priori hypotheses: constructing a hypothesis before collecting data
Deduction: a process of reasoning from a general proposition (the theory) to
specific implications of that proposition (the hypothesis).
Hypotheses: if a then b
Induction: abstracting a hypothesis from a collection of facts.
Empirical generalizations: hypotheses that are based on previously observed
patterns of results.
Falsification: the characteristic that distinguishes science from other ways of
seeking knowledge, such as philosophical argument, personal experience, or
religious insight
Methodological pluralism: using many different methods and designs as
theories are tested.
Strategy of strong inference: depending on how results turn out, the data will
confirm one of the theories while disconfirming the other.
Conceptual and operational definitions
- Conceptual: like the definition we find in a dictionary: hunger=having a desire
for food
- Operational: the defining of a concept by specifying precisely how the concept
is measured or induced: hunger = being deprived of food for 12 hours
Proof and disproof
The impossibility of proof: theories cannot be proved because obtaining empirical
support for a hypothesis does not necessarily mean that the theory from which the
hypothesis was derived is true.
The impossibility of disproof: disproof is a logically valid operation. Null findings:
results showing that certain variables are not related to behavior. File-drawer
problem: the failure to publish studies that obtain null findings
Scientific progress: the failure to obtain support in many studies provides evidence
that the theory has problems, as well as a theory whose hypotheses are corroborated
by research is considered supported by the data.
The scientific filter
All ideas > Filter 1 > Filters out nonsense >
Initial research projects > Filter 2 > Filters out dead ends, fringe topics >
Research programs > Filter 3 > Filters out methodological biases and errors,
unimportant contributions >
, Published research > Filter 4 > Filters out non-replication, uninteresting and non-
useful stuff > established knowledge
Behavioral research
Descriptive Research: describes the behavior, thoughts, or feelings of a particular
group of individuals. For example, opinion polls.
Correlational Research: investigates the relationships among various psychological
variables. For example, is there a relationship between self-esteem and shyness?
Experimental research: determining whether a certain variable changes in behavior.
The researcher manipulates one variable (the independent variable) to see whether
changer occur consequently (the dependent variable).
Quasi-experimental research: when a researcher is not able to manipulate the
independent variable or control all other factors. The researcher either studies the
effects of some variable or the event that occurs naturally.
Chapter 2 Behavioral Variability and Research
Schema: a cognitive generalization that organizes and guides the processing of
information. For example, leadership; through experience you have observed what
good leadership is or big cities; when you hear ‘New York’ you think of crowded and
dangerous or on the other hand interesting and exciting.
Five ways in which variability is central to the research process
1. Psychology and other behavioral sciences involve the study of behavioral
variability; understanding behavior and mental processes really means
understanding what makes behavior, thought, and emotion vary.
o Behavior can vary across situations, among individuals, and over time.
2. Research questions in all behavioral sciences are questions about behavioral
variability.
3. Research should be designed in a manner that best allows the researcher to
answer questions about behavioral variability. Flaws in the design of a study
can make it impossible for a researcher to determine why participants
behaved as they did.
4. The measurement of behavior involved the assessment of behavioral
variability. We would like the variability in the numbers we assign to various
participants, to correspond to the actual variability in participants’ behaviors,
thought, emotions, or physiological reactions.
5. Statistical analyses are used to describe and account for the observed
variability in the behavioral data. Inferential statistics: used to draw
conclusions about the reliability and generalizability of one’s findings.
Conceptual and statistical explanation of variance
Range: subtract the smallest from the largest score above.
Mean: the average, the sum of the set divided by the number of scores you have
Deviation score: subtracting the mean from each score.
Deviation score squared: squaring each of the deviation scores (to get rid of the -)
Total sum of squares: the sum of the deviations scores squared
