Topic 2 - Research Methods Flashcards
What are the limits of intuition and experience?
-Risk mistakes
-Don’t show all possibilities
-Overconfidence (bias blind spot)
What is the scientific method?
Theory data cycle: does theory match collected data? Need replication!
What kinds need variables are there?
Manipulated (independent) and measured (dependent)
How do you operationalize a variable?
Create operational definition (ex. Self-report on a scale)
What are the three types of surveys?
Descriptive, correlational, and experimental
What is random sampling?
Sample selected from population of interest, non-biased, that can generalize
What are naturalistic observations?
Observe everyday worlds without interfering (can use technology)
What is a case study?
An in depth study of one person with a rare condition - can provide new insights
What is a correlational study?
Understand relationship between two or more variables
What does a scatter plot show?
The strength and association between two variables (range from -1 to 1)
How to know if a relationship is causal?
Need to know which variable came first and to have no other explanation
What is experimental research?
Supports causal statements by manipulating causal variable and seeing the effect on measured variable
What are the three conditions of causality?
Covariance (correlation), temporal precedence, and internal validity
What is the difference between random sampling and random assignment?
Sampling: equal chance of being selected for study
Assignment: level of independent variable assigned
What questions are asked to confirm construct validity?
- Variable operationalization?
- Participants representative?
- Rule out other explanations?
How to assess external validity?
- Who was included/ left out?
- Can results from P.O.I generalize to another?
- Replication in non WEIRD populations?
What does WEIRD stand for?
Western, educated, industrialized, rich, democratic
What is a confound?
When experimental groups vary in more than the independent variable
What questions do we ask to evaluate claims in the media?
- What am I being asked to believe?
- What evidence is there?
- How strong is the result (graphs)?
- Causal claim?
- Replication?
What is the frequency distribution?
All possible scores (bar graph)
What is central tendency?
The centre of score batch (mean, median, mode)
What is standard deviation?
How much a batch of scores varies around the mean
What is effect size, how is it observed?
Magnitude of relationship between variables (tight scatterplot = stronger). Quantified with correlation coefficient r (-1 to 1)
What is d?
Effect size for experiments based on means. Strong effect size when not much variability
What is statistical significance testing?
Estimating whether results came from a particular population
What is the null hypothesis?
How likely is it that there is no relationship? Low p-value means significant result and data unlikely to occur under null hypothesis
What are some reasons for replication failure?
- false positive
- small samples
- underreporting non significant effects
What is HARKing?
Hypothesizing after results are known
What is p-hacking?
Removing extreme scores (increases bias) or over analyzing
What is open science and preregistration?
Available to other scientists (helps self- correct and progress). Statement of expected outcome before experiment
What three ethical standards does the institutional review board check for?
Autonomy (informed consent), beneficence (risks/ benefits), justice (representation)
What guidelines does animal welfare act set for ethical experiments with animals?
- Replacement if possible
- Refinement (modify to reduce stress)
- Reduction (few subjects)
What is a theory?
Idea designed to organize and explain existing facts and make predictions
What is a journal?
A periodical containing peer-reviewed articles on academic discipline for a scholarly audience
What are nominal, ordinal, and interval variables?
- Categorical
- Categorical with meaningful relative ranks
- Numbers
What is the third variable problem?
For a given relationship, there might be an additional variable that is associated with both
Within- subjects vs. Between subjects
- Same group does multiple conditions
- Different groups do each condition
What is inter-rarer reliability?
Do the research assistants agree on operational definitions?
Hawthorne effect?
Telling subjects what researchers are doing/ measuring which changes behaviour
Demand characteristics?
Cues from experimenter that tell participants how to behave
What is included in informed consent?
Enough info, can opt out anytime with no penalty, can withhold responses, debriefed at the end