Biostatistics 3

Question 1

Accept Null and it is false =

Answer 1

Type II error

Think beta and power

Question 2

Reject Null and it is true =

Answer 2

Type 1 error

Think alpha and pvalue

Question 3

Alpha and multiple comparisons?

Answer 3

Divide the alpha by the number of comparisons

Question 4

Multiplicity of comparisons

Answer 4

A typical large scale clinical trial may describe as many as 500 individual terms descrfibing adverse events
If P value were calculated for wach pairwise, the one would by chance alone expect –> 25 events to have a P <0.01

Question 5

Correction for MC

Answer 5

Bonferroni correction

Hochberg sequential procedure

Question 6

Bonferroni correction

Answer 6

Take 0.05/the number of comparisons

Result is the threshold your p value must meet

Question 7

Hochber sequential procedure

Answer 7

Less rigid

Question 8

Number needed to treat

Answer 8

Determine % of patients in treatment and control group that demonstrates outcomeCalculate ARR and take the reciprocal

Question 9

ARR

Answer 9

Absolute Risk Reduction

% outcome from intervention group - % outcome from control group