Summaries Flashcards
What do you need to think of when you design a study?
Designs
Variables
Hypotheses
What are IVs?
Variables we intentionally manipulate
What are DVs?
Variable we measure
What do you do after you generate research hypotheses?
Turn these into null and alternative statistical predictions
What are designs, variables and hypotheses important for?
Helping to choose, conduct and interpret inferential tests
What might you want to use to represent data?
Frequency distributions
Histograms
What are the measures of central tendency?
Statistics based around central data point of distribution
What measure of central tendency is used the most?
Mean
What are measures of spread?
How dispersed data is
What measure of spread is used the most?
Standard deviation
What does probability theory assume?
Data should be normally distributed and use a Gaussian curve
What are Z scores useful for?
Reducing skew and comparing individual data to the mean
How are sampling errors accounted for?
Calculating standard error of the mean or confidence intervals
Are mean differences in the sample equal to the mean difference in the population?
No
What does a t-test tell us?
If two groups have different means
Whether the sample means are different enough to be confident that there is a difference in the population
What kind of t-tests used for to determine whether there is a significant different between two means?
Independent samples
Paired samples
What is a t-statistic?
Difference between means in standard errors
How do you know if there is a statistically significant difference between means?
p < 0.05
What is an independent sample t-test?
Compares two different groups of people
What kind of design is an independent-samples t-test used for?
Between-subjects
What kind of design is a paired-samples t-test used for?
Within subjects
What is a one-sample t-test?
Determines whether a mean is significantly different from a given value
How do t-tests work?
Calculate t-statistic
- size of effect in standard errors
Calculate degrees of freedom
- usually df = n-1
Calculate p-value
- probability of seeing the effect in the sample, if there was no effect in the population
What are measures of covariance used for?
Measuring the degree to which two variables vary with each other
What can be used to measure the degree to which two variables are related to each other?
Pearson’s r
Spearman’s Rho
Does correlation imply causation?
No
What does regression tell us?
Statistical test that tells us about relationships between variables (correlation)
