Chapter 2 Flashcards
LO: What are the three primary goals of science?
1) description (what)
2) prediction (when, where)
3) explanation (why)
Data
Measurements gathered during the research process
Scientific Method
Scientific procedure for conducting research
Theory
- ideas or concepts
- empirical evidence
- makes predictions about a phenomenon
- explanation based on observation
- best theories generate a lot of hypotheses
- simple
- typically been around for a while
- not the same as opinions or beliefs
Hypothesis
- prediction based on theory
- testable = falsifiable (can be proven wrong)
- you can’t prove a hypothesis, you can only support it
Research
- scientific process
- careful collection, analysis, and interpretation of data
- tests hypothesis
- can be duplicated (replicable)
- either supports or fails to support the theory
Supports: strengthens theory
Fails to support: revise or discard
Q: Which scientific goal is fulfilled by a project seeking to understand the causes of teen vaping?
Explanation (why)
SQ: Describe the 7 steps used in the Scientific Method
1) Frame a RESEARCH QUESTION
2) Conduct a LITERATURE REVIEW
3) Form a HYPOTHESIS
4) DESIGN a study
5) CONDUCT the study
6) ANALYZE the data
7) REPORT RESULTS
1) Research Question
- specific, basic
- testable using data
2) Literature Review
- why is it helpful?
- search for what’s already been discovered on your topic
- helps you guide your research: inspiration
3) Hypothesis
- testable research predictions
- design tests
4) Design Study
- decide which research method to use
5) Conduct Study
- recruit participants
- measure responses
Operational Definition
- why is it important?
a definition that qualifies (describes) and quantifies (measures) a variable so it can be easily understood
- helps with replicability
Variable
something that can vary and that a researcher can manipulate, measure, or both
6) Analyze the Data
- describe findings
- look for conclusions
7) Report Results
- why is it important
- how does it support science
- submit to research journals
- peer review
- benefits society, supports scientific culture, inspires others
Q: In the scientific method, what do you call a specific, testable prediction?
Hypothesis
Replication
- why do some studies not replicate?
Repetition of a study to confirm or contradict results
- theory is wrong, false positives, questionable research methods
False Positive
results when there is no real effect but a study produces statistically significant results by chance
Questionable Research Methods
Practices that unintentionally make research less replicable
What are 4 common questionable research methods?
HARKing
p-hacking
Small samples
Underreporting null effects
- HARKing
- why is it a problem?
Hypothesizing After Results are Known
- leads people to believe the study was conducted for that one outcome when many other outcomes could have been possible by chance
- p-hacking
Testing the same hypothesis until finding a significant result
What are 3 practices that can help avoid questionable research methods?
- preregistration
- meta-analysis
- larger samples
- Preregistration
documenting a study’s hypotheses, methods, and analysis plan ahead of time and publishing it on a time-stamped website
- Meta-analysis
“study of studies”
- combines findings of multiple studies to arrive at a conclusion
Q: How does preregistration prevent HARKing?
- holds researchers accountable
- forces them to submit their hypothesis publicly before conducting their study
What are the 3 types of studies used in psychological research?
1) descriptive
2) correlational
3) experimental
- Descriptive Studies
Observing and describing behaviour
- answers “what” questions
Includes: case studies, observational studies, and self-report studies
1.a) Case Study
- what is the goal?
Descriptive research method that involves intense examination of an atypical person or organization
Goal: describe events or experiences that led up to or resulted from the exceptional feature
1.b) Observational Studies (2)
- what is the goal?
1) Participant Observation
- researcher involved in situation
2) Naturalistic Observation
- researcher as a passive observer
Goal: assessing and coding overt behaviour
1.c) Self-reports and interviews
- what is the goal?
Data collection through surveys or questionnaires
Goal: gather large amounts of data in short periods of time
- Correlational Studies
- used for making predictions
- examine how variables are naturally related in the real world
- can describe some types of relationships between variables but can’t make claims about causal relationships
Directionality Problem
Researchers find a relationship between variables but they can’t determine which variable is causing changes in the other
(ex. does wealth cause happiness or does being a happy person prompt success?)
Third-Variable Problem
There might be a third variable that causes both A and B to happen
Maybe A and B don’t actually have any correlation
(ex. what if success and happiness are determined by intelligence and resilience and don’t actually impact each other at all?)
Scatterplot
Graphic depiction of the relationship between two variables
Correlation Coefficient
A descriptive statistic that indicated the strength and direction between two variables
- Experimental
Tests causal hypothesis by manipulating and measuring variables
Q: Suppose a study finds that hair length has a negative correlation with body weight: People with shorter hair weigh more. Why should you avoid acting on these results by growing your hair to lose weight?
Because correlation does not indicate causation
Independent Variable
Manipulated
Dependent Variable
Measured (y axis)
Experimental Group
participants who receive treatment
Control Group
participants who receive no intervention or one that is unrelated to the independent variable being manipulated
Random Assignment
Placing research participants into an experiment in a way that each participant has an equal chance of being assigned to any group
Confound
Anything that affects a dependent variable (unintentional third variable)
Q: How does random assignment help eliminate confounds in experiments?
Any differences between the participants in the conditions, both known and unknown, tend to average out
Quasi-experiment
Can’t use random assignment because participants need to be split into groups based on certain characteristics
Population
Everyone in the group the experimenter is interested in
Sample
A subset of the population
Culturally Sensitive Research
Studies that take into account the role that culture plays in determining thoughts, feelings, and actions
LO: Identify ethical issues associated with conducting psychological research on human participants
LO: Identify the key issues regarding the humane treatment of animal subjects
LO: Identify three characteristics that reflect the quality of data
LO: Describe measures of central tendency and variability
LO: Discuss the rationale for inferential statistics
Validity
degree to which an instrument or procedure measures what it claims to measure (is it valid?)
Construct Validity
How well does it measure the concept
External Validity
- degree that effects observed can be generalized to other people, settings, situations
- does it relate to the real world?
Internal Validity
- degree to which the effects observed are due to the independent variable and not to confounds
Reliability
- degree to which a measure is stable and consistent over time
Accuracy
- degree to which an experimental measure is free from error
Q: You want to know whether the results of your study generalize to other groups. What kind of validity are you most concerned about?
External Validity
What are the two types of statistics?
1) Descriptive
2) Inferential
1) Descriptive Statistics
Statistics that summarize the data collected in a study
Central Tendency
A measure that represents the typical response or behaviour of a group as a whole
Mean
- measure of central tendency
- (average)
- sum of scores/number of scores
Median
- measure of central tendency
- middle number
Mode
- measure of central tendency
- most common number
Variability
In a set of numbers: how widely dispersed the numbers are from each other and from the mean
Standard Deviation
Statistical measure of how far away each value is, on average, from the mean
Range
difference between highest and lowest score
2) Inferential Statistics
Set of procedures that enable researchers to decide whether differences between two or more groups are just chance or whether they reflect true differences in the populations being compared (ex. placebo)
Q: Why might you prefer the median to the mean in certain circumstances?
Because the mean can be distorted by outliers, but the median is simply the middle value
