T Tests for the Terrified v1.0
TWO SAMPLE T TEST: This test is used to compare the means
between two separate samples of individuals. It assumes the data are normally distributed.
PAIRED T TEST: A test on the change that occurs in a measurement
carried out on the same subject under two different conditions, eg. before and after a therapeutic treatment.
Glossary
T tests are generally the first tests encountered by undergraduate students, and are very
commonly used in the wider world. 'T Tests for the Terrified' covers basic statistics
as it applies to the t tests, from the mean and median through to hypothesis testing,
in a user-friendly, commonsense, non-mathematical and highly interactive way. Using the same
approach as its big brother 'Statistics for the Terrified', the new tutorial will concentrate
specifically on this one area and its underlying techniques.
Purchasing
Because of its more focussed content, T Tests for the Terrified is available for under
half the price of Statistics for the Terrified.
Buy now
Content
Each chapter begins with an outline of its aims and objectives and finishes with a summary
of the key points covered. Click on the
chapter headings below to see a fuller outline of the course contents:
Basic numerical data description:
- • data types
- • average values - the mean and median introduced
- • the effect of outliers on the mean and median
- • the advantages and disadvantages of the mean and median
- • range and variance
- • standard deviation
- • datagame - how each point contributes to variance and sd
- • what is the coefficient of variation?
Graphical data description:
- • introducing the normal curve
- • the normal curve shows relative likelihood
- • the normal curve is always centred above its mean
- • how sd affects normal curve shape
- • normal distribution and calculating probabilities
- • dealing with data that isn't normally distributed.
Standard error and confidence intervals:
- • introduction to standard error
- • standard error measures accuracy
- • standard error decreases as standard deviation increases
- • accuracy increases as standard error decreases
- • how this looks with normal curves
- • how to calculate standard error in theory
- • how to calculate standard error in practice
- • isn't it cheating to use estimates in the calculation?
- • what is a confidence interval?
- • confidence intervals are based on standard error
- • a 95% ci contains the population mean 95% of the time.
What does p<0.05 really mean?
- • what is hypothesis testing?
- • posing the hypothesis
- • introducing the null hypothesis
- • random sampling
- • what do you take into account?
- • the steps involved in hypothesis testing
- • the relationship between the test statistic and p values
- • how difficult is it to reject the null hypothesis?
- • the relationship between power and type II errors
- • power worsens as the alternative gets closer
- • making significance tests more powerful.
The paired t test:
- • what are repeated measures?
- • collapse the data
- • types of repeat measure studies
- • before/after tests
- • the paired t test
- • the paired t test datagame
The two-sample t test:
- • overview - tests for differences between group means
- • two-sample t test overview
- • two-sample t test datagame
- • extending the t test: introducing analysis of variance
Analyse your own data:
- • describe your data
- • paired t test
- • two-sample t test
