Week 8 Flashcards
Descriptive Statistics
Statistics that describe the results as they are - “get to know” the data.
Central Tendency
Different ways of measuring the average.
Mean (average),
Median (middle),
Mode (most frequent).
Variability
Range (difference between highest and lowest value). Standard Deviation (the average distance between the scores and the mean).
Effect Size
A measure of the strength/magnitude of a statistical relation.
Interpretation Guidelines (d)
Small: 0.2
Medium: 0.5
Strong: 0.8.
Interpretation Guidelines (r)
Small: .1
Medium: .3
Strong: .5.
Cohen’s d
The difference between two means in standard deviation units. Make comparisons between different studies by standardizing.
Inferential Statistics
Making educated guesses based on acceptable mistakes.
Null Hypothesis
Hypothesis that population means are equal.
Relationship in sample is sampling error.
No relationship in the population.
Research Hypothesis
Hypothesis that population means are different.
There is a relationship in the population.
Relationship in the sample reflects the relationship in the population.
P-Value
Probability that the data would happen if the null were true.
Type I Error
False positive. Saying there is an effect in the population when there is not. Rejecting null when null is true. Alpha.
Type II Error
False negative. Saying there is not an effect in the population when there really is. Fail to reject null when null is false. Beta.
Statistical Power
The probability of rejecting the null given the sample size and expected relationship strength.
The complement of the probability of committing a type II error.
Four Criticisms of NHST
- Most people don’t understand p-values.
- Oversimplifies as “significant/not-significant”: incentivizes cheating/questionable practice.
- Pointless - the null is almost never true.
- We’re actually interested in effect sizes.
Distributions
The way the scores are distributed across levels of that variable.
Negatively Skewed
Peak to upper range - skewed to the left.
Positively Skewed
Peak to lower range - skewed to the right.
Restriction of Range
When one or both variables have a limited range in the sample relative to the population.
Ex: age as a variable, but only having access to 18-25 year olds.
Parameters
Corresponding values in the population
Sampling Error
Random variability in a statistic from sample to sample.
Alpha Level
How low the p-value must be for the sample result to be considered unlikely.
Rejecting null when null is true.
Practical Significance
The importance or usefulness in a real-world context.
Correlation Matrix
Presenting correlations (r) among several variables.
