Research Methods Flashcards
Hindsight Bias
Belief that an outcome was foreseeable (after it has occurred)
- “knew it all along” effect
Why do we need to use the scientific methods to study psychology?
- The scientific method helps reduce bias.
- Allows us to study a phenomenon in a systematic way. We can build up evidence in support of a theory.
What are the steps in the Scientific Method?
- Theory
- Hypothesis
- Research
- Support or refute/fail to support theory
If support, refine with new hypothesis and research
If refute, discard, or revise your theory
Define Theory
A theory:
- Describes general principles about how variables relate to one another
- Organize and explain data that are observed
- Enable us to make predictions about new situations
Define Hypothesis
A hypothesis is a prediction
What makes a good theory?
Falsifiable
- There are multiple ways to show if a theory is false
What makes a good hypothesis?
- Specific and testable
Define correlational designs
Measure of how closely two variables vary together
Variables are measured and not manipulated
Define:
- Positive correlation
- Negative correlation
- Zero correlation
Positive correlation:
- Both variables increase or decrease together
Negative correlation:
- One variable increases when the other decreases
Zero correlation:
- One variable is not predictively related to the other
Operational
A description of a property in concrete measurable terms.
- For example, if you want to study internet use you would look at:
- Time spent on website
- Number of web pages visited
- Data used in gigabites
Limitations of correlational designs
Correlation cannot infer causation
Benefits of correlational designs
- Can sometimes be a more ethical approach
- Establishing relationships/making predictions can inspire experiments
Definition of a true experiment
We manipulate one variable and measure its effects on another vairable
Steps in Scientific Investigation
- Formulate a testable hypothesis
- Select the research method and design the study
- Collect the data
- Analyze the data
- Report the findings
Confounding variable
A confounding variable is an outside influence that changes the effect of a dependent and independent variable. This makes it difficult to determine the exact relationship between the independent and dependent variables.
Define independent variable
Variable that is manipulated
Define dependent variable
The behavior that is being measured
Defining the IV, what are the requirements?
- Must be manipulated
- Participants must be randomly assigned
- Must operationally define (what type of thing/manipulation are participants getting?)
Defining the DV, what are the requirements?
- Measure the DV
- Operational definition: type of measurement, scale of measurement.
How do we ensure groups are ‘equal’?
Random assignment
Define random assignment
assigning participants randomly to experimental conditions
How do we select participants for the study?
Random sampling
Define random sampling
Every person in the population has an equal chance of being selected
Define convenience sampling
This sample consists of people who are conveniently available for the study
Benefits of experimental designs
- Can infer causation
- Precise control over variables can eliminate alternative explations
Drawbacks of experimental design
- May be impractical or unethical to address certain types of research questions using an experimental design
- Highly controlled lab settings can make generalization of findings to the real world more difficult
Placebo Effect
When receiving special treatment or something new affects human behavior
How to avoid the placebo effect?
- Create a control for placebo effect:
- Control group (sugar pill vs. real pill)
- Participants must be blind to conditions
Define experimental bias
The experimenter’s expectations influence the outcome of a study
Example: Experimenters might see what they want to see or behave/treat participants differently
How to avoid experimenter bias?
Double-blind approach
Define double-blind
A double-blind experimental setup is a study design in which neither the participants nor the experimenters know who is receiving a particular treatment. This method is used to eliminate bias in research results.
Demand characteristics
Participants form an interpretation of an experiment’s purpose and subconsciously change their behavior
Quasi-Experimental Designs
- Similar to experimental research but there is no random assignment to conditions
- Relies on existing group membership (gender, married vs. single, age groups)
- It may be more difficult to infer causality
- Still have an IV but not directly manipulating it
Converging operations
A research strategy where a variety of research techniques are used to investigate or converge upon a particular experimental or research result.
Define replication
The process of repeating a study and finding similar outcome each time