Research Exam 2

Nominal

— data: categorical data; objects fall into mutually exclusive unordered categories

Ex: insurance type, ethnicity, gender

ordinal

— data: categorical data; order is important

ex: likert-type scales, military rank

interval

— data: continuous data; legitimate mathematical values

Ex: temperature

ratio

— data: continuous data; equal intervals between values

ex: height, weight, time, length

mean

average of values

median

puts values in rank order distribution and selects the middle value

mode

number that occurs most frequently in a distribution

normal distribution

Bell-shaped data, median and mean will be about the same

negatively skewed

tail on the left

positively skewed

tail on the right

tail

the skew refers to the — not the bulk of the data

standard deviation

variablility or dispersion of scores around the mean of the sample, square root of the variance

range

difference between the smallest and largest values in data set

independent

the variable hypothesized to explain an observed clinical phenomenon

dependent

shows the effect of manipulating or introducing the independent variable

Null

hypothesis states that there is no difference or relationship between 2 variables

alternate

hypothesis states that there is a difference between variables

p

probability that an observed difference occurred by chance

p<a

reject the null hypothesis

p≥a

fail to reject the null hypothesis

hypothesis

it is a relationship that is being evaluated between an intervention/exposure and an outcome, or between 2 or more variables

equivalence

— study: you are trying to prove that the therapies are equivalent

non-inferiority

— study: you are trying to prove that the new therapy is not inferior

type 1 error

rejecting the H0 when it is true, you are stating that the observed differences may be due to chance, and not due to treatment

type 2 error

failing to reject H0 when it is false, you are stating that the observed differences are not due to chance but due to treatment

alpha

the maximum acceptable level of risk for a type 1 error, α is the probability of observed difference being due to chance/error and not detecting it

beta

the maximum acceptable level of risk for a Type 2 error, β is the probability of making a type 2 error

power

the probability that a test will reject the null hypothesis correctly

90% probability that the conclusion is correct, statistically significant

say you are given an alpha 0.10, a CI of 90% and a p-value ≤ 0.10; what does this mean?

not statistically significant

say you are given an alpha 0.05, a CI of 95% and a p-value > 0.05; what does this mean?

95% probability that the conclusion is correct, statistically significant

say you are given an alpha 0.05, a CI of 95% and a p-value ≤ 0.05; what does this mean?

confidence interval

a given range of values in a study which is likely to contain the true value of an unknown statistical parameter (ex. true population value)

wider

as your confidence level increases the confidence interval becomes —

not

forest plots: if there is a horizontal line that passes the black vertical line it is — statistically significant

0

if the forest plot is showing a difference comparison the black vertical line is at #

1

if the forest plot is a rates comparison the vertical black line is at #