Scientific method

 The goal is to obtain better-quality answers.

Introduction to Research Method: Method of Tenancy

Why take a research Method course?
 To learn how to apply science to the study of Human Behavior  Information accepted as true because it has always been believed or
 To understand and interpret other’s research because superstitions support it.
 To evaluate claims in everyday media  Limitations: potential inaccuracies: no methods for correcting
 To discover how to acquire knowledge and make decisions by erroneous ideas.
applying the scientific method.
Method of Intuition
Methods of Knowing and Acquiring  Information accepted on the basis of a hunch
Knowledge  Limitation: no method for separating accurate from inaccurate
knowledge.
are ways in which a person can know things or discover answers
to questions. Methods of Authority
Nonscientific approaches
 Relies on information or answers from an expert in the field
 Method of tenacity  Good starting point – often the quickest and easiest way to obtain
 Method of Intuition answers
 Method of authority  Some limitations:
 Rational Method o Does not always provide accurate information
 Method of Empiricism o Often accepts expert’s statement as fact (method of faith)
o Not all “experts” are experts
The Rational Method Summary of Nonscientific Methods of
Acquiring Knowledge
 Also known as the Rationalism
 Seeking answers by logical reasoning
 Premise statements: describe facts or assumptions that are
presumed to be true
 Argument: a set of premise statements that are logically combined
to yield a conclusion.

Limitation of Rational Methods

 Logical conclusions are only valid if the premise statement is true

 “The rational method “assumes” the premise statement is true” The scientific Method
People are not very good at logical reasoning. “Sometimes the
arguments do not logically support the conclusions. An approach to acquiring knowledge
 Involves formulating specific questions and then systematically
finding answers
The Empirical Method  Combines several different methods of acquiring knowledge.

 Also known as empiricism: answering questions by direct The Steps of the Scientific Method
observation or personal experience
 Based on the philosophy that all knowledge is acquired through the
1. Observe behavior or other phenomena
senses.
2. Form a tentative answer or explanation (a hypothesis)
 Limitation:
3. Use your hypothesis to generate a prediction
o Observations can be misinterpreted
4. Evaluate the prediction by making systematic, planned
o Sensory experience can be swayed by belief
observations
o Time–consuming and sometimes dangerous.
5. Use the observations to support, refute, or refine the original
hypothesis.
Inductive and Deductive Reasoning Variables and Hypothesis

What is Induction? Variable: Characteristics or conditions that change or have different

 A small set of specific observations is the basis for forming a general values for different individuals
statement about a larger set of possible observation  Examples: weather, health status and age
 Induction = Increase What is a hypothesis?
What is deduction  A statement that describe or explains a relationship between
 A general statement is the basis for reaching a conclusion about a variables: “best guess”
specific example  A hypothesis can lead to several different observable and
 Deduction = decrease measurable predictions

Example of Induction and Deduction Three Important Principles of the

Scientific Method
Empirical
 Answers are obtained by making structured or systematic
observations.
 Results will provide clear support for the hypothesis or will clearly
refute the hypothesis.
Public
 Observations are available for evaluation by others
 Replication is key to the scientific method
 Objective: The outcome is not skewed by bias.
The process of Scientific Inquiry The research process
Quantitative research produces numerical scores
 Submitted to statistical analysis for summary and interpretation
Qualitative research is based on making observation
 Summarized and interpreted in a narrative report.

The Steps in the Research Process

1. Find a research idea: select a topic and search the literature to
find an unanswered question
2. Form a hypothesis
3. Determine how you will define and measure your variable
4. Identify the participants or subjects for the study, decide how
they will be selected and plan for their ethical treatment
5. Select a research strategy
6. Select a research design
7. Conduct the study
Science versus Pseudoscience 8. Evaluate the data
9. Report the results
Scientific Research Pseudoscience 10. Refine or reformulate your research idea
Return to step 2 and continue
Evidence is gathered from careful, Lacks some of the key components Important: there are no final answer
systematic, and objective essential to scientific research
observation
Example: aromatherapy, astrology,
and intelligent design.
 Sources of Research Ideas

What are possible sources?

 Personal interest and curiosities
The research process o Pick a topic about which you want to learn more
 Casual observation
Steps o Behavior of people or animals may generate a general
Step 1. Find a research idea research topic
o Involves selecting a topic and searching the literature to find an  Report of others’ observation
unanswered question o Idea might come from informal sources (e.g. magazines
Step 2: Form a hypothesis or newspapers) or academic materials.
 Pick a topic in which you are interested  Practical problems or questions
o Examples: a particular group of individual. o Stemming from daily life (e.g. placement of audio controls
Behavior and general topic from less distracted driving)
 Do your homework: collect background information General categories of research
o Read books and journals articles o Applied research is directed toward solving practical
o Do not try to read everything problems
 Read enough to gain a good grasp of your topic. o Basic research involves studies that are intended to solve
 Keep an open mind theoretical issue.
o Begin with a general topic and let background reading lead you  Behavioral theories
to a more specific topic o Testing predictions that are part of a theory
o Be critical: ask questions as you read
o Maintain a degree of flexibility Finding Background Literature:
 Focus, Focus, Focus
- Develop one research question and find
Conducting a literature search
relevant background information.
Take one step at a time The literature: the mass of published information worldwide
Your Job:
 Start with a general idea
 Search the existing literature to gather background information
  “Find” a specific research questions
How new research grows out of old
Primary and Secondary Sources

Primary sources: firsthand reports in which the authors describe their

own observations.
- Example: Empirical journal articles/theses
Secondary sources: secondhand reports in which the authors discuss
someone else observe
- Examples: textbooks/introductory sections of research reports.

The purpose of a literature search

Two basis goals:

1. To gain a general familiarity with the current research in your
specific in your specific area of interest Preparing for a literature Search
2. To find a small set of research studies to serve as the basis for
your research idea.  Narrow down your general idea to a specific research design
 Find a set of published research reports defining the current  Find all published information to document and support
state of knowledge that question
o Identify a gap in that knowledge base that your study will  A good starting point: recently published secondary sources
attempt to fill.  A textbook, chapter headings, and subheading will help
focus your search on more narrowly defined area.
 PsycARTICLES is the American Psychological Association (APA)
Make Notes of the following items definitive source of full-text, peer-reviewed scholarly and scientific
articles in psychology
 Contains more than 163,000 articles from more than 80
Subject words journals published by the APA, the Educational Publishing
o List correct terms or subject words. Foundation (EPF), and from allied organization.
 These words identify and describe the variables in
the study and the characteristics of the
participants. Information about ERIC DATABASE
Author names
o The same authors’ names encountered repeatedly are ERIC (Education Resource Information Center)
likely the current leading researchers in the field.  Provides access to education literature and resources
o Information from journals includes in the current index of
journals in educations and resources in Education Index.
Using Online Databases  More than 1.3 million records and links to more than
323,000 full-text documents dating back to 1366.
 A typical database contains about one million publications
o Cross-referenced by subject words and author names Information about MEDLINE Database
o Search the database by subject word or author name
o Articles listed by the search may be MEDLINE with Full Test
 Brief summaries Provides authoritative medical information on medicine, nursing,
 Abstracts dentistry, veterinary medicine, the healthcare system and pre-clinical
 Full-text sciences.
Feature full test for more than 1470 journals indexed in
Information about PsycINFO Database MEDLINE

PsycINFO is the American Psychological Association’s (APA) definitive

resource for abstracts for scholarly journal articles, book chapters,
books, and dissertations.
- Largest resources devoted to peer-reviewed literature in
behavioral science and mental health.
 o Authors sometimes point out limitations or problems in
their own studies
Screening Articles During a Literature o Create a new idea from two existing results
Search o Frame questions based on differing results.

Ways to weed out irrelevant materials: Find suggestion for Future Research
 Use the title as your first basis for screening
 Use the abstract as your second screening  When you are familiar with the current research in an area, the
 Look for a link to a full – text version idea for the next study simply involves extending the current
 If the articles still look, relevant, then read it carefully research one more step.
 Use references from articles you have already found to expand  This may not be as simple as it sound
your literature search.  Finding new research ideas is easily done by
looking for them as explicit statements I the journal
Ending a Literature Search articles you already have.

The search is completed when you: Check Out the Discussions Sections
Feel comfortable with your knowledge about the topic idea
Have found a few recent research studies that are particularly  The discussion sections often end with a set of suggestions for
relevant to your own interests. future research
 The authors usually point out questions that remain unanswered
How many articles is enough? o You may use these suggestions as ideas for your own
The key criterion is that the study (or studies) you find provides research
some justification for new research.  Authors may also point out limitations or problems with their own
study
Finding an Idea for a Research Study from o You can design a study that fixes these issues.
a Publishing Research Articles
 How can existing publications lead to an idea for a new research
study?
o Authors generally make explicit statements regarding
unanswered questions
The components of Research Articles Using a Hypothesis to create a Research
Critical Reading Study
 The research process- steps 3 and 4:
o Step 3: specify how the variable will be defined and
measured
o Step 4: Identify the individual who will participate in the
study, describe how they will be selected, and provide for
their ethical treatment
 Different research studies can be developed from the same
hypothesis.

Using a Research Idea to form a

Hypothesis and Create a Research Study

What are the characteristics of a good hypothesis?

 Logical: the logical conclusion of a logical argument
 Testable: All of the variables, events, and individuals can be
defined and observed.
 Refutable: can be demonstrated to be false
 Positive: Must make a positive statement about the existence of
something.
 Theories and Constructs

 Theory: a set of statements:

- About mechanisms underlying particular behaviors
- Theories help organize and unify different observations
An Introduction to Measurement related to the behavior, and good theories generate
predictions about the behavior.
First step in the research process  What are constructs (hypothetical constructs)
Find an unanswered question that will serve as a research Idea o Hypothetical entities created from theory and speculation
Second Step  Cannot be seen, but are assumed to exist
Form a hypothesis  Help explain and predict behavior in a theory
Next Step: Use the Hypothesis
Develop an empirical (observable) research study that will either Operational Definitions
support or refute the hypothesis.

Variable  Researchers often refer to the process of using an operational

definition as operationalizing a construct
Characteristics or conditions  An operational definition specifies a measurement procedure (a
Change or have different values for different individuals set of operations) for measuring an external, observable
Evaluating differences or changes in variables behavior, and uses the resulting measurements as a definition
Must be able to define and measure the variable being studied. and a measurement of the hypothetical construct
 Methods of defining and measuring constructs
o Variables that cannot be observed or measured directly
Construction and Operational Definitions  Example: intelligence can be operationally
defined as a score on an IQ test: and
Types of Variable involved in research “hunger” can be operationally defined as
Well-defined, easily observed and easily measured. the number of hours of good deprivation
Example: height and weight
Intangible, abstract attributes
Example: motivation or self-esteem
Measurement is more complicated.
Limitation of Operational Definition Relationship Among Variables

 An Operational definition is not the same as the construct itself Determine the consistency of a relationship by computing a
 Concerns about the quality of operational definitions and correlation between the two measures.
measurements produced. - Consistent positive relationship: a correlation near +1.00
- it is easy for operational definitions to leave out important - Consistent negative relationship: a correlation near -1.00
components - Inconsistent relationship: a correlation near zero
- Operational definitions often include extra components that are
not part of the construct being measured. Scatter Plot Showing Different
Relationships
Validity and Reliability of Measurement

 Researchers have developed two general criteria for evaluating the

quality of any measurement procedures
- Validity and Reliability

Consistency of Relationship

 Demonstrating consistency
- Two scores (one score from the new measure and one score
from a well-established measure) are represented for each person a
scatter plot
o Positive Relationship: The two measurements change
together in the same direction
o Negative Relationship: The two measures change in
opposite directions
Validity of Measurement as the variable itself, then we
have established the
To establish validity Convergent Validity involves creating two different
- Measurement procedure must accurately capture the variable methods to measure the same
that it is supposed to measure construct, then showing a strong
relationship between the
measures obtained from the two
Six common Definition of Validity methods.
Divergent Validity involves demonstrating that we
Face validity Simplest and least scientific are measuring one specifi c
construct and not combining two
= Whether a measure different constructs in the same
superficially appears to measure measurement process
what it claims to measure.
Concurrent validity The validity of a new
measurement is established by Reliability of Measurement
demonstrating that the scores
obtained from the new
measurement technique are  Stability or the consistency if the measurements produced by a
directly related to the scores specific measurement procedure.
obtained from another, better-  Concept of reliability
established procedure for o Inconsistency of a measurement comes from error
measuring the same variable. o Measured Score = True Score + Error
Predictive validity When the measurements of a
construct accurately predict Source of Error
behavior (according to the
theory), the measurement
 Observer error
procedure is
o The individual who makes the measurements can
Construct Validity we can demonstrate that
introduce simple human error
measurements of a variable
 Environmental changes
behave in exactly the same way
o It is difficult to attain the ideal of identical circumstance
  Participant changes The Relationship Between Reliability and
o The participants can change between measurements
(e.g., degree of focus)
Validity

Type of Measures of Reliability  Reliability is a prerequisite for validity

o A measurement procedure cannot be valid unless it is
reliable
Successive measurements Test-retest reliability  It is not necessary for a measurement to be valid for it to be
compares scores of two reliable
successive measurements of o Consistency of measurement is no guarantee of validity
the same individuals and
correlates the scores
Simultaneous measurements Inter-rater reliability:
Scale of Measurement
agreement between two
observers who simultaneously  Measurement is a procedure for classifying individuals
record measurements of the o Scale of measurements is the set of categories used for
behaviors classification
Internal consistency Split-haft reliability: splits the  The process of measurements involves two components
test in half, and then o Set of categories
calculating a separate score o Procedures for assigning individuals to categories.
for each half, and then
calculating the degree of Four Type of Measurements Scales
consistency between the two
scores for a group of Nominal Scale Represents qualitative
participants. differences in the variable
measured
Ordinal Scale Represents differences in a
series of ranks
Interval Consist of series of equal
intervals like the inches on a ruler

Ratio A scales has a true zero value

Dealing with Equivocal Measurements - Provides a direct - Ex: galvanic skin - Provide researchers
measure: validity is response (GSR) and with a vast number of
dubious MRI’s options
 Interval or ratio scale scores are: - Provides objective - Possible to selects
o Compatible with basic arithmetic measures the behavior(s) that
o Suitable for statistical analysis (e.g mean and variance) - Require expensive seems to be best for
o Allow hypotheses testing equipment: may not defining and
 Scores from an ordinal scale are not provide valid measuring the
o Compatible with basic arithmetic or appropriate for measures of construct
statistical analysis constructs - Behaviors may be
 Distinction of a measure as an interval and ratio is not always only a temporary or
clear out situational indicator of
an underlying
construct.
Selecting a Scale of Measure
Other Aspects of Measurement
 What is the scale’s ability in comparing different measurements?
o Nominal: reveals whether a difference exists  Multiple measures provide more confidence in the validity of the
o Ordinal: indicates the direction of the differences (Which measurements
is more and which is less) o Can introduce problems
o Interval: determines the direction and the magnitude of a  Complex statistical analysis and interpretation
difference  Desynschrony: Lack of agreement between two
 Affects the ability to describe relationships between variables. measures
o Method for limiting problems
Modalities of Measurement  Combine the measures into a single score or each
individual
Self – report Physiological Behavioral measures
measurement ask measures are based are based on overt
participants direct on manifestations of behaviors that can be
questions to study the underlying observed and
behaviors construct measured.
Sensitivity and Range Effects

 Range effect: a measurement that is not sensitive enough to

detect a difference
o Ceiling effect: clustering of scores at the high end of a
measurement scale
 Allowing little or no possible of increases in value
o Floor effect: clustering of scores at the low end of a
measurement scale
 Allowing little or no possibility of decreases in
value

Artifacts: Experimenter Bias and

Participant Reactivity
 Artifact: a non-natural feature accidentally introduced into
something being observed
o Experimenter bias: measurements are influenced by the
experimenter’s expectations regarding the outcome of the
study.

Limiting Experimenter Bias

 Standardize or automate the experiment
 Single-blind study
o The research study is conducted by an experimenter who
does not know the expected results
 Double-blind study
o Neither the experimenter nor the participants know the
expected results of the study