Topic 8 Biostats

what is a normal distribution

a continuous probability distribution describing a bell-shaped curve

what is the shape in a normal distribution determined by

the mean and variance

where is the normal distribution symmetric around

mean

how many modes are in a normal distribution

single mode

what is highest at the mean in a normal distribution

the probability density

what is a t distribution

a sampling distribution used for testing for differences between two population means

where are t distribution symmetric around

the mean of 0

where is uncertainty of the true value of the standard deviation and the standard in a t-distribution

visible in the tails of the distribution

what a sampling distribution used to describe

all possible values we might obtain from a population

what are degrees of freedom for continuous numerical variables related to

sample size

what is the purpose of the t-test

compare the means of a numerical variable between two groups

what do the two types of t-tests depend on

how an experiment is designed

in a two sample t-test what is the null hypothesis

that the mean of group a is the same as the mean of group b

what are the assumptions for a two sample t-test

the numerical variable is approximately normally distributed in each population, the variances and standard deviations of the populations to be compared are approximately equal and each of the two samples is a random sample from its population

to calculate the standard error difference between the two means what do you use

the pooled variance

if the null hypothesis is true in a two sample t-test then the mean of treatment 1 is what

equal to the mean of treatment 2 and thus test statistic is 0

your chosen alpha determines what

the confidence level

when alpha is 0.05, it is a what

95% confidence interval

when two sets of confidence intervals don’t overlap then it supports that

the inference that the true population means are different

in a two sample t-test there is no what

direct relationship between observations in treatment and the second treatment

in a paired one sample design treatment is what

applied to a pair of samples (dependent)