Research Methods

Hindsight Bias is the tendency to believe, after learning the outcome that you knew all along.
❖​ Self Fulfilling Prophecy, and anecdotal evidence.

Overconfidence, we tend to think that we know more than we do

Confirmation Bias is the tendency to gather information that confirms pre-existing beliefs,
typically emphasizing or pursuing supporting evidence while dismissing or failing to seek
contradictory evidence.

Four Types of Research Designs are

Experimentation
❖​ Looking to prove Causal relationships, Cause equals effect
❖​ Your Hypothesis must be proved wrong.
❖​ It has to be able to express a relationship between two variables
➢​ A Variable is anything that can vary among participants in a study.
➢​ Ex: Participating in class leads to better grades than not participating.
❖​ The Independent Variable is being manipulated in the experiment, hopefully, the
independent variable brings change
➢​ If there’s a drug in an experiment, it almost always is considered the independent
variable.
❖​ The Dependent Variable would be the effect of the drug, it also is whatever is being
measured in the experiment, depending on the independent variable.
❖​ The Scientific Method
➢​ Question, Hypothesis, Experiment, Observe and Record, Analyze, Share Results.
■​ Results provided in published in Peer-reviewed journals
■​ Research must be able to be replicated
■​ Generalizability - the degree to which results can be applied to the general
population.
➢​ Operational Definitions, Explain what you mean in your hypothesis, and answer
how the variables will be measured in “real life” terms.
 ■​ How you operationalize the variables will tell us whether the study is
valid.
■​ Ex. If you say that chocolate increases violent behavior, what do you mean
by chocolate, and what do you mean by violent behavior?
❖​ Sampling
➢​ Identify the population you want to study. The sample must be representative of
the population you want to study. Always pick a Random population to study, to
avoid bias.
■​ Stratified Sample, is when you break up the population into more specific
groups
■​ Convenient Sampling leads to sampling bias.
■​ Beware of Confounding Variables
●​ The object of the experiment is to prove that A causes B. A
confounding variable is anything that could cause change in B, that
is not A.
●​ If I wanted to prove that smoking causes heart issues, what are
some confounding variables?
■​ Random Assignment, Once you have a random sample, randomly
assigning them into two groups helps control for confounding variables.
●​ Experimental Group vs Control Group.
■​ Single Blind, controls for the participant's bias, and Double Blind
controls for participants and the experimenter's Bias.
➢​ Experimenter Bias: another confounding variable, is often an unconscious act.
The double-blind procedure attempts to correct this. When researchers make
biased arguments, it's often considered the Halo Effect.
➢​ The placebo effect: is an observed effect simply based on the idea of receiving
any treatment, regardless of efficiency. This is what the single blind tries to limit.

Correlation Method
❖​ Correlation expresses a relationship between two variables but does not show causation.
➢​ That means that a third variable exists in this situation
❖​ There are two types of correlation methods, Positive and Negative
➢​ Positive correlation: This is when two variables go in the same direction like
studying improves grades.
➢​ Negative correlation: This is when two variables go in the opposite direction,
like heroin use and grades, which have a negative correlation.
❖​ Correlation Coefficient: A number that measures the strength of a relationship between
two variables
➢​ Often ranges from -1 to +1, The relationship between the two variables gets
weaker the closer you are to zero
 ➢​ No correlation is considered Illusory

Case Study
❖​ In depth investigation of an individual or small group who may have a highly unusual
trait.
➢​ Pros
■​ Details of subjects
■​ Unique quality or situation
■​ No unethical treatment on part of researcher
➢​ Cons
■​ No correlation data
■​ No generalizability
■​ Time-consuming

Meta Analysis
❖​ It looks at the statistical results of many studies of the same question. Gives us an idea of
size and consistency of a variable’s effects, allows for researchers to test the
generalizability of a variable’s effect across cultures, places and time, etc.

Survey/Questionnaires
❖​ A data collection method which can be used in any research method
➢​ It measures the correlation between two variables, is extremely fast and cheap, but
needs a good random sample.
➢​ Low response rate, and is extremely susceptible to self-report, and social
desirability biases.

Interviewing/Self-Reporting
❖​ Quantitative Structure: Respondents choose a predefined answer on a scale to express
their opinion on a statement providing numerical data that can be analyzed statistically,
like a likert chart.
❖​ Qualitative Structure: A set of predetermined questions that are asked in a structured
format, but the answers are always open-ended.

Naturalistic Observations
❖​ A data collection technique that can be used by any research method/design, which
promotes researchers to watch subjects in their natural environment, without any
manipulation of the environment.
APA Ethical Guidelines for Research
❖​ IRB- Internal Review Board, which is used for both Humans and Animals, this review
board helps organizations make proper and ethical decisions.
➢​ Human Research:
■​ Institutional Review, No coercion, which means that the study and
subjects must be voluntary, and one can stop at any point in time during
the study, and anonymity should be key, if you deceive the subject for the
purpose of the study, you must inform them after the study.
➢​ Animal Research:
■​ Clear purpose must be present, they should be treated in a humane way.
The animals must be acquired legally, and they should suffer the least
amount of pain.