Basic Review Flashcards
Levels of measurement
- Nominal
- Ordinal
- Interval
- Ratio
Nominal
Categories are mutually exclusive, but not ordered. Ex: species (cat, dog, fish, hermit crab)
Ordinal
Order matters, but the difference between values is NOT informative. Ex: place in a race (1st, 2nd, 3rd,…)
Interval
Order matters and difference between values IS informative. Ex: temperature (80, 90, 100)
Ratio
Interval variable and true definition of zero. Ex: height, weight, pieces of candy eaten, NOT temp since there is still heat at 0 degrees.
Measures of central tendency
- Mean: The arithmetic average of scores in a dataset.
- Median: The midpoint of scores in a dataset (when arranged in numerical order).
- Mode: The most frequently occurring value(s) in a dataset.
Measures of variability
- Range
- Standard deviation
- Variance
Range
Spread of scores in a dataset. Range = largest score - smallest score
Standard deviation (SD)
How far the average score in the dataset is from the mean of that dataset. SD = √variance.
Variance
SD squared
Hypotheses
- Null hypothesis (H0)
2. Alternative hypothesis (H1)
Null hypothesis (H0)
Predicts no change, difference, or relationship. “The IV has no effect on the DV.”
Alternative hypothesis (H1)
Predicts a change, difference, or relationship. “The IV has an effect on the DV.”
Decision errors
Possibility of drawing incorrect conclusions from a study.
Type I error
Incorrectly reject the null hypothesis (false positive). Conclude that treatment has an effect when it really doesn’t - there isn’t actually an effect.
Type II error
Incorrectly retaining the null hypothesis (false negative). Conclude treatment doesn’t have an effect when it actually does.
p-value
Indicates the likelihood that H0 is correct/that it would be a mistake to reject the null hypothesis. Indicates the probability/likelihood of your sample data when H0 is true.
p < .05
Less than 5% chance that H0 is correct (low likelihood). Reject H0 if alpha = .05 (and retain if alpha is greater than .05).
p < .01
Less than 1% chance that H0 is correct (very low likelihood). Reject H0 if alpha = .01.
Effect size
Measures the size of treatment effect. “x% of the variability in the DV can be attributed to the IV”. Only compute if you reject H0/yield significant results!
Power
Probability that the test will correctly reject a false H0; probability that the test will correctly identify a treatment effect if it exists. Influenced by effect size, sample size, alpha level and if the test is one or two-tailed.
As effect size increases, power increases. As sample size increases, power increases. As alpha level decreases, power decreases. A one-tailed test has more power than a two-tailed test.
When to use a z-test
You have a nominal IV and want to compare the sample mean to the population mean when the population mean (σ) is known.
When to use a one-sample t-test
You have a nominal IV and want to compare the sample mean to the population mean when the population mean (σ) is unknown.
When to use an independent samples t-test
You have a nominal IV and want to compare two unrelated sample means to each other.
When to use a dependent samples t-test
You have a nominal IV and want to compare two related sample means to each other.
When to use correlation
You have an interval/ratio IV and want to examine the relationship/association between two variables.
When to use regression
You have an interval/ratio DV and want to predict the DV using the IV.
Z-test
Hypothesis test used to compare a sample mean to a population mean when the population SD is known.
One-sample t-test
Hypothesis test used to compare a sample mean to a population mean when the population SD is UNknown.
Independent-samples t-test
Hypothesis test used to compared two unrelated sample means to each other.
Dependent samples t-test
Hypothesis test used to compare two related sample means to each other.
ANOVAs
Hypothesis test used to compare 3+ unrelated sample means
Correlation
Hypothesis test used to examine the relationship between two variables.
Regression
Hypothesis test used to predict DV using IV.
