Ch 2 Flashcards
Pitfalls of using personal experience to study behaviour
-Experience has no comparison group
-Cannot compare what would occur with and without the variable of interest
-Experience is confounded (other alternative explanations that cannot be ruled out)
Pitfalls of using intuition to study behaviour
- Intuition is biased
Eg. being swayed by a good story, persuaded by what easily comes to mind, focusing on evidence we like best, failing to think about what we cannot see
Rationalism
The idea that knowledge can be obtained through reasoning
Empiricism
The idea that knowledge can be obtained through experience and observation
Techniques to avoid biases
- Single blind tests: patients do not know which study group they are in
- Double blind tests: neither patients or researchers know which study group the patients are in
Goals of psychological research
- Measurement and description
- Understanding and description
- Application and control
Operational definition
Describes a variable in terms of specific procedures used to produce or measure it
Theory
Set of statements that describes general principles about how variables relate to one another
Hypothesis
Prediction about the outcomes of research based on theory
Law of parsimony (Occam’s Razor)
The fewer assumptions made by a hypothesis/theory, the less opportunities there are for it to be falsifiable
What makes a good theory
- Predictions are supported by research
- Conforms to law of parsimony
- Measurable
- Falsifiable (can be tested)
- Establish causality (relationship b/w cause and effect)
Descriptive Research
- Any means to capture, report, record, or describe a group
- Based on a single measured variable
- Includes naturalistic observation, surveys, case studies, and participant observation
Naturalistic Observation
- Observe behaviour without manipulation
- Most likely representative of real word behaviour
- No control over behaviour
- Hard to predict causes of behaviour
- High external validity
Participant Observation
- Researcher interacts with population of interest
- Insights from participant’s perspective
- May be subject to biases
- May not be repeatable
Case Study
- A report of a single person, group, or situation
- Very detailed
- NOT an experiment and NOT proof/evidence
- May be difficult to draw causal relationships
Survey
- Questions that extract information from a group of people
- Easy to administer and can gather lots of information
- Susceptible to biases from researchers and participants
Descriptive Statistics
Organizing and summarizing data in a useful way
Pros: can describe variables of interest
Cons: do not learn about relationships or causality and cannot manipulate measured variables
Inferential Statistics
Interpreting data and drawing conclusions
Measures of central tendency
Mode: most frequent value
Mean: average(center of dataset); can be skewed by outliers
Median: middle data point
Measures of variability
Range: subtract lowest from highest data value
Standard deviation: spread of data around the mean; sqrt of variance
Variance: average of squared deviation scores; (standard deviation)^2
Correlational Research
- At least two measured variables
- Looks at the relationship between two or more measured variables
Correlation does not equal causation
Correlation Coefficient (r)
- Describes the relationship between two variables (ranges from -1.0 to 1.0)
- Sign indicates the direction (positive or negative) and absolute value indicates the strength
Positive correlation: an increase in one variable relates to an increase in another variable
Negative correlation: an increase in one variable relates to a decrease in another variable
Zero correlation: the variables are not correlated with one another
Pros: used to make predictions about variables
Cons: can’t manipulate measured variables, shows association NOT cause, relationships may be due to confounding variables
Experimental Research
-At least one manipulated variable
-Manipulating a variable under controlled conditions so that resulting changes in another variable can be observed
-Goal is to detect cause-and-effect relationships
-Testing theories through controlled experiments (hypothesis driven)
Pros: conclusions about cause-and-effect can be drawn
Cons: artificial nature of experiments and ethical/ practical issues
Variables
Independent variable: the variable that is manipulated
Dependant variable:the variable that is affected by manipulation
Extraneous variable/confounding variable: uncontrolled events that can affect the dependant variable