Week 2 Flashcards
What does the mean measure ?
central tendency
* add K: New mean = old mean + k
* multiply k: New mean = old mean * k
What does standard deviation measure ?
measures the variability
* add K: New SD = old SD
* multiply k: New SD = old SD * k
What is the null hypothesis ?
predicts that no difference exists between the two groups being compared
What is the alternative hypothesis ?
is the research hypothesis which predicts that there is a significant difference existing b/w the 2 groups being compared
* what the researcher wants to support
What is a two-tailed hypothesis ?
It is a non-directional in which the alternative hypothesis is that a difference exist b/w groups, but it doesn’t predict the direction of the difference
What is a one-tailed hypothesis ?
It is directional in which the alternative hypothesis predicts the direction of the expected difference b/w the groups
What is Type I error ?
error in hypothesis testing in which we rejected a true null hypothesis
* ex: there was no actual difference b/w A and B, but we rejected the null and said that there was a difference
What is Type II error ?
error in hypothesis testing in which we failed to reject a null hypothesis
* ex: there was a difference b/w A and B, but the data failed to reject the null and we said there was no difference
What is statistical significance ?
when the observed difference b/w 2 descriptive statistics is unlikely to have occured due to chance
What is alpha level ?
an acceptable probability of observed difference occuring due to chance
What is p value ?
the actual probability of type I error in a statistical test
What is a z-test ?
It is a parametric inferential statistical rest of the null hypothesis for a single sample where the population variance is known
* compare sample against population
What is a sampling distribution ?
A distribution of sample means based on random samples of fixed size from the population
What is the central limit theorem ?
Regardless of the shape of the parent population, the random sampling distribution of the sample means will tend towards normal, and this tendency towards normal increases with the sample size (N)
What is the standard error of the mean ?
the SD of the sampling distribution
What are the assumptions of z tests ?
- The sample must be drawn randomly with replacement
- The sampling distribution is normal
- Z-tests require we know the SD of the population
- The sample size must be relatively large ( 30 or more )
What is a t-test ?
A parametric inferential statistical test of the null hypothesis for a single sample where the population variance is not known
What is a student’s t distribution ?
a set of distributions that, although symmetrical and bell-shaped, are not normally distributed
What are degrees of freedom ?
the number of scores in a sample that are free to vary
* any single distribution: df = n-1
What are the assumptions of t-tests ?
- The sample was drawn randomly with replacement
- The sampling distribution is symmetrical and bell-shaped
What are the benefits of t-tests over z-tests ?
- T-tests doesn’t require knowing the standard deviation of the population
- The n of the sample could be large or small
What is effect size ?
The proportion of variance in the dependent variable that is accounted for by the manipulation of the independent variable
What is Cohen’s d ?
An inferential statistic for measuring effect size when using a t-test
What occurs to the null and alternative hypothesis wen comparing two sample means ?
Now the hypotheses become whether the two sample are the same (null) or not the same (alternative)
What is a independent samples t-test ?
A parametric tests that compares the performance of two different sample of participnants
What is the standard error of the difference b/w means ?
The SD of the sampling distribution of differences b/w the means of the independent samples in a two sample experiment
What are the assumptions for independent samples t-test ?
- The samples were interval and ratio
- The samples were drawn randomly with replacement
- The distributions are normally distributed
- The samples are independent of each other
- Homogeneity of variance - the variance in the first sample should be similar to the variance in the second sample
What are the issues with independent sample t-tests?
- Results may be influenced by individual difference factors - a type of inherent variance
- To reduce this, we can have studies designed to have dependent-samples
What is a dependent samples t-test ?
A parametric test that compares the performance of two related or matched sample of participants
* two samples are connected (either same person at different times or pairs or matched subjects)
What is the standard error of the difference score ?
The SD of the sampling distribution of differences b/w the means of dependent samples in a two-sample study
What are the assumptions of dependent samples t-test ?
- The samples were interval or ratio
- The samples were drawn randomly with replacement
- The distributiond are normally distributed
- The samples are NOT independent of each other
- Homogeneity of variance
What are the pros and cons of dependent tests ?
Pros
* reduced inherent variability
Cons
* Fatigue effect
* Carry-over effect
* Practive effect