Week 2: Research Methods Flashcards
Statistical Thinking, Research Designs, Conducting Psychology Research in the Real World
Key Components to a statistical investigation…
PLANNING THE STUDY - asking a testable research question & deciding how to collect data
EXAMINING THE DATA - determine appropriate ways to examine data; relevant graphs, descriptive statistics, patterns in data, individual observations that deviate from overall patterns - what does it reveal? - is there evidence for reliability/validity
INFERRING FROM THE DATA - what are the valid statistical methods for drawing inferences “beyond” data you collected
DRAWING CONCLUSIONS - based on what you’ve learned from data, what conclusions can you draw and who do they apply to; can you draw a cause-and-effect conclusion about your treatments?
Reliability
Refers to the consistency of a measure
Validity
The degree to which a measure is assessing what it is intended to measure
Cause-and-Effect
Related to whether we say one variable is causing changes in the other variable, versus other variables that may be related to these two variables
Distribution
The pattern of variation in data
2 Fundamental aspects of statistical thinking
DATA VARY - values of a variable vary
Analyzing the pattern of variation (DISTRIBUTION of the variable) often reveals insights
Level of Significance
A result is statistically significant if it is unlikely to arise by chance alone
Compare the p-value to the cut-off value
If p-value is smaller than cut-off value, we REJECT the hypothesis that only random chance is at play
P-value
Probability of observing a particular outcome in a sample, or more extreme, under a large conjecture abt the larger population/process
The LOWER the P-VALUE the HIGHER the STATISTICAL SIGNIFICANCE
Sample
Collection of individuals on which we collect data
Population
Larger collection of individuals that we would like to generalize our results to
Generalized
Related to whether the results from the sample can be generalized to a larger population
Random Sample
Using a probability-based method to select a subset of individuals for the sample from the population
Margin of Error
Expected amount of random variation in a statistic; often defined for 95% confidence
Key to the margin of error - When we use a probability sampling method, we can make claims abt how often the sample result would fall within a certain distance from the unknown population value by chance alone
Random Assignment
Using a probability-based method to divide a sample into treatment groups
Tends to balance out all the variables
Critical to experimentation bc if the only difference between two groups is the independent variable, we can infer the independent variable is the cause of any observable difference
Random Sampling/Assignment
Sampling - paramount to generalizing results from our sample to a larger population
Assignment - key to drawing cause-and-effect conclusions
Probability models help us assess how much random variation we can expect in our results in order to determine whether our results could happen by chance alone & to estimate a margin of error
Operational Definitions
How researchers specifically measure a concept
Psychologists measure many abstract concepts, such as happiness/intelligence, by beginning w operational definitions of the concepts
Independent Variable
Variable the researcher manipulates and controls in an experiment
Cause
Dependent Variable
Variable the researcher measures but does not manipulate in an experiment
Effect
Confounds
Things that could undermine your ability to draw casual inferences
Expectations that can influence you in a study
Ex. placebo effect, participant demand (participants try to behave in a way they think the experimenter wants them to), experimenter observation
Placebo Effect
sometimes a person just knowing that he or she is receiving special treatment or something new is enough to actually cause changes in behaviour or perception
Double-Blind Procedure
Neither the participant nor the experimenter knows which condition the participant is in - don’t know what’s going on
Because both parties are “blind” to the condition, neither willff be able to behave in a way that introduces a confound
Correlational Research
When scientists passively observe and measure phenomena - we do not intervene/change behaviour; we identify patterns of relationships but typically can’t infer what causes what; can only examine TWO variables at a time
Correlation
Measures the association between two variables, or how they go together
CORRELATION DOES NOT MEAN CAUSATION
Correlation Coefficient
Association between two variables can be summarized statistically using the CC (r); provides information abt the direction & strength of the association between two variables
Positive correlation - two variables go up/down together in a scatterplot - dots form a pattern that extends from bottom left to upper right - r value indicated by a POS number
Negative correlation - two variables move in opposite directions (one up one down) - r value indicated by a negative number
Strength of a correlation: WEAK - has to do w how well the two variables align - many exceptions; low absolute value close to 0 STRONG - few exceptions; tighter dots; high absolute value