ESCI 340: Biostatistical Analysis Nomenclature
Populations and samples: N (usually) is number of individuals in population
n is number of individuals in sample
ni is number of individuals in ith sample
Summation:
∑
=
+++=
n
ini XXXX
121 L e.g.,
4321
4
1
XXXXX
ii+++=
∑
=
∑
=
++=
n
jniiiji XXXX
1,2,1,, L e.g.,
4321
4
1iiii
jij XXXXX +++=
∑
=
()(
LLL ++++++=
∑∑ n
A
a
B
bnba XXXXXX ,21,2,12,11,1,
)
Means: population mean:
N
X
N
ii
∑
=
=1
µ
sample mean: n
X
X
n
ii
∑
=
=1
Median: middle measurement in ordered set of data (central data point)
if N even, average of 2 MX
n
=+()/12
Geometric mean (GM): XXXXX
Gn
ni
i
n
n
==
=
∏
123 1
LX
Variance: population variance: ∑−
=
N
Xi2
2)(
µ
σ
sample variance: ∑
−
−
=1
)( 2
2
n
XX
si
Standard Deviation (SD): population SD: 2
σσ
=∑−
=N
Xi2
)(
µ
sample SD: 2
ss ==∑−
−
1
)( 2
n
XXi
Coefficient of Variation (CV)
X
s
CV ×= 100
Standard Error (SE): SE of the mean: n
s2
X
s=
Note: different formulae for SE of difference between means 21 XX
s−, etc
1−
α
Confidence Interval: SECI ,×±=
να
tX
Null hypothesis: H0 Alternative hypothesis: HA
Significance level:
α
= probability of (incorrectly) rejecting a true null hypothesis
P-value: Given a true H0 , P-value is the probability of obtaining a test statistic at least as
extreme as the one obtained.
Degrees of Freedom:
ν
l_symbol.pdf 1 McLaughlin