Statistics 2 Flashcards
What are Inferential statistics?
- allow to generalize from the sample to the greater population, which the sample represents
- are crucial because the effects that researchers find in a study may be due to random variability caused by sampling error
Difference between Null and Alternative Hypothesis?
Null Hypothesis
there is no difference in the population (H0)
Alternative Hypothesis
there is a difference in the population (Ha)
Analogy: criminal trial
Difference between Directional and Non-directional alternative hypothesis?
Directional alternative hypothesis
* the direction of the effect is determined
* delayed reward leads to slower learning
Non-directional alternative hypothesis
* the direction of the effect is not determined
* delayed reward leads to either slower or faster learning
Level of significance refers to a criterion of judgment upon which a decision is made regarding the value stated in a null hypothesis.
What’s the significance threshold in percent (in psychology)?
significance level is usually 5% or 1%.
What is a Type I Error? What is a Type II Error?
Can you provide an example?
Type I error (false positive): the test result says you have coronavirus, but you actually don’t.
Type II error (false negative): the test result says you don’t have coronavirus, but you actually do
What is a t-test?
it tells you how significant the differences between two group means are
it is usually used when data sets follow a normal distribution but the population variance is unknown
What is ANOVA?
Analysis of Variance (ANOVA)
* An ANOVA is a statistical test used to compare variances across the means of different groups
* It is used when you have more than two groups.
* Furthermore, you can have more than one independent variable (factor).
Difference between one-way and two-way ANOVA?
One-way ANOVA has only one idependent variable (factor) with two or more levels
EX: What is the effect of three different dosages of an antidepressant on depression?
Two-way ANOVA has two or more idependent variables (factors) with two or more levels
EX: What is the effect of three different dosages of an antidepressant (factor 1, three levels) on depression in two different age groups (factor 2, 2 levels)?
What are the 3 types of t-tests and what are they each used for?
Independent Sample t-test: means of two DIFFERENT groups
Paired Sample t-test: compare means of SAME group at 2 different times
One Sample t-test: compare sample mean with a known (population) mean
How can Correlation be statistically computed?
Pearson’s Correlation. Correlation coefficient = degree of linear correlation between two variables. Ranges from -1 to 1.
What is Statistical Power? How is it computed?
The statistical Power of a test is the likelyhood of rejecting a wrong null hypothesis. It is 1-beta (type 2 error).
What does Cohens d express. And how is it computed?
Cohens d is a measure of the effect size. It is comupted as the difference of the means divided by the Standard Deviations.
What is the effect size?
Computed by Cohens d. it expresses how LARGE an effect is. Important because signifanct differences must not be large. (and vice versa)
How is an a priori power analysis conducted? Why conduct one?
Ingredients:
1. wanted statistical power (normally more than 80%)
2. signifance level alpha
3. expected effect size (Cohens d)
——-
Its important because it tells you how many participants are needed to reach the wanted statistical power.
How does
1. a more liberal alpha
2. a smaller sample size
3. a larger effect size
each impact the statistical power?
More liberal alpha -> higher power
smaller sample -> less power
larger effect size -> more power
(think: more power means more likehood of rejecting a false H0. That increases if your test is better able to capture the difference that exists. And if that difference is larger).
