Research Methods notes
The Scientific Method
-Origins of the scientific method is traced of philosophy known as empirism, which
holds that knowledge is gained through experience, observation, and experiment.
- Empirical: denote information referred to as Data because it can be measured and
- Empirical evidence differs from anecdotal evidence, which refers to impressions or
opinions of just one person.
- The scientific method is crucial to research because it minimized Bias by providing
the rules by which observations are collected and results are evaluated.
- Bias: familiar term that often indicates unfair practices that wrongly discriminate
Observation: Balance between focus and openness and serendipity effect.
What is a scientific question?
Questions that are researchable, some questions may deal with topics for which the
scientific method is moot. These include questions of religion, faith.
From theory to testable hypothesis
A theory generates hypothesis, which are evaluated empirically with the scientific
method. . A testable hypothesis is framed as a statement, often in the form of a
prediction that is made prior to the actual collection of Data.
Testable hypothesis is therefore described as a priori. Meaning that it exists before
experimentation or observation.
Variable and measurement
Variable: simply defined as any characteristics that can take on different values ot
can vary across research participants.
Deductive Research: Research strategy that involves reasoning from a theory to
Research circle: Theory->hypothesis->Date->Deductive research->Empirical
generalizations->back to theory. The substance of psychological research:
Independent variable: what is manipulated
- Levels=conditions of the independent variable.
- Control group=group that does not receive a treatment or manipulation
- Random assignment=each participant has an equal chance of being assigned
to any group.
Dependent Variable: Out come of interest
Control Variable: Held constant or statistically removed from different groups.
*Research I psychology is most usually evaluated using statistics.
*Statistical probabilities take form of p-values.
Problems with relying on statistical Significance: Statistical significance is
- The number of observations in the study
- In experiments: The magnitude of difference in the dependent variable
resulting form the manipulation
- In Correlational (association) Studies (Where there is no manipulation, but
measurement on more than 1 variable): The strength of association between
the measured variables.
- The priori (stipulated) significance (a) Level.
- Summary: Statistical significance is primarily dependent both on the “n” and
the “effect size”
- Effect size: The magnitude of difference resulting from the manipulation or
association between measured variables.
Significance can occur with:
- A large effect Size with a modest number of observations
- With small effect size and a vary large number of observations.
How to Read scientific Literature:
Evaluation a study:
- Consider the source: for scientific validity
- Ensure goals are clearly defined
- Consider warning signs of pseudoscience
- Consider how research was funded Empirical papers
Empirical paper will be organized into specific sections:
Belmont Report (1979)
- Established by the national commission for the protection of human subjects
of biomedical and behavioral research.
- Three Basic Principles: Respect, Beneficence, Justice
APA’s five general Principles:
- Beneficence and non-malfeasance
- Fidelity and responsibility
- Respect for people’s rights and dignity
Four Central Ethical Research Principles
- Achieving valid results
- Maintaining professional integrity
- Protecting subjects
- Encouraging appropriate applications of research
- Avoid harming research participants
- Participant must be competent
- Participant must fully voluntary
- Participant must be fully informed
- Participant must be able to comprehend
- Document is given to the participant.
- Avoid deception, except in limited circumstances
- Maintain privacy and confidentiality
Basic QEEG Concepts
There are two basic concepts that are employed in the analysis of the QEEG
The first refers to the type of activity that is occurring at a location in terms
of the 5 frequencies. Each frequency can be analyzed according to concepts such as microvolt,
relative power, peak frequency and peak amplitude.
In general, the important variables for cognitive success are the microvolt
and relative power measures.
The delta and theta frequencies are generally negatively related to cognitive
abilities and the beta variables are positively related.
Case study:Keyes (1997)
Constuct specified in theory to describe,explain, and predict a wide range of
empirical findings. Not directly observable. Must be translated into an operational
Measurement constructsAsignment of numerals to variables according to some
objective rule. Do our hypothetical constructs reflect objective reality?
Use of multiple methods of measurement is essential.
Levels of measurement:
Categorical or qualitative
No mathematical meaning
Must be mutually exclusive and exhaustive
Does not assume equal distance between values
All qualities of nominal ordinal scales
Equal distances between values
No true zero point
All qualities of nominal, ordinal, and interval scales
Numbers can be added, subtracted, multiplied, and divided.
Multiplication only works if there is an absolute zero on the scare
The Kelvin scale has a zero value that represents the total absence of heat in
the measured system.
Zero in the Fahrenheit or Celsius scales represents non-zero heat values. Choosing Level of Measurement
Important when considering statistical analyses
- Non-parametric statistics=generally used with nominal and ordinal scales
- Parametric statistics-generally used with interval and ratio scales
Psychometrics: Studies the application of psychological tests as objective measures
of the mind and mental processes.
Cultures often influences psychometric tests.
- People have certain levels of psychological traits
- Requires standardization:
Standardization sample= random and purportedly representative
selection of people drawn from defines population
- Calculate norms
Standard deviation=degree to which an individual score deviates from
- Stratified sample= formed by randomly selecting from relatively uniform
subpopulations called strata which is denied along specified, objective
- Many traits are distributed along the normal curve, which is the most scores
to fall near the center, with gradually fewer scores as you move forward
Evaluating measures: Reliability
- Test-retest= Scores should remain consistent over time
- Interobserver= degree to which raters code the same observed behavior
- Interitem=measures of the same construct should have a high correlation
- Alternate-forms= scores on alternate-forms of the same test should correlate;
used to avoid practice effects
Assessing inter-item reliability
Adding Valid Items increases inter-item reliability. Evaluating Measures: Validity
Looks like it measures what it intends to measure
Lacks a firm scientific basis
Items have sufficient breadth to capture the full range of a construct
test scores agree with an objective criterion
Concurrent=measured at the same time
Predictive= predicts an outcome in the future
Extent to which theoretical framework and empirical framework of the
construct are valid
Assessing Construct Validity
Using the idea of nomological network
Cronbach and Meehl