Tests of

Significance
 Lesson Objectives

 Learn how to use test statistics to

examine hypothesis about population
mean, proportion
Inference

Use a random sample

to learn something
about a larger
population
 Inference
 Two ways to make inference
 Estimation of parameters
* Point Estimation (X or p)
* Intervals Estimation
 Hypothesis Testing
 Statistic Parameter

Mean: X estimates ____

Standard
deviation: s estimates ____

Proportion: p estimates ____

from entire
from sample
population
 Sampling Distribution

X or P X or P X or P
Population Point estimate Interval estimate
I am 95%
Mean confident that 
Mean, , is is between 40 &
X = 50
unknown 60

Sample
 Parameter
= Statistic ± Its Error

Sampling error versus Standard error

 Standard Error

S
Quantitative Variable SE (Mean) = n

p(1-p)
Qualitative Variable SE (p) = n
 Confidence Interval

Range of likely values for the (unknown)

population mean, based upon the observed
sample mean & its Standard error
So, population mean (u = x̅ + 1.96 x s.e.) or
95% CI (µ)= x̅ + 1.96 x s/√n
90% CI (µ) = x̅ + 1.64 x s/√n
99% CI (µ) = x̅ + 2.58 x s/√n
 Confidence Interval

α/2 α/2
1-α
_
X
SE  SE Z-axis

95% Samples

X - 1.96 SE X + 1.96 SE

95% sample means will lie within limits of 2 SE on

 Confidence Interval

α/2 α/2
1-α

SE  SE Z-axis p

95% Samples

p - 1.96 SE p + 1.96 SE

Critical Values: 1.96 at α=0.05

2.58 at α=0.01
 Level of Significance, a
 Defines unlikely values of sample statistic
if null hypothesis is true. Called rejection
region of sampling distribution
 Typical values are 0.01, 0.05
 Selected by the Researcher at the Start
 Provides the Critical Value(s) of the Test
Level of Significance, a and the Rejection Region

a Critical
Value(s)
Rejection
Regions
0
 Interpretation of
CI

Probabilistic Practical

In repeated sampling 100(1-

We are 100(1-a)%
a)% of all intervals around
confident that the single
sample means will in the
long run include  computed CI contains 
 Tests of significance

SE (p) =
n
Example:
Assume, we have obtained a random sample of 25
males, age 20-24years whose mean temp. was 98.140
F with a standard deviation of 0.6.
What can we say of the true mean of the universe
from which the sample was drawn?
Using standard error as the yard stick,
S.E. (x̅)= S / √n = 0.6 / √25 = 0.6 /5 = 0.12
95% CI (µ)= x̅ + 1.96 x s.e. or + 2x s.e.
= 98.14 + 2x0.12 = 97.900F – 98.380F
The chances will be 1 in 20 (p=0.5) that µ will be
outside these confidence limits
 Example (Proportion)

In a survey of 140 asthmatics, 35% had

allergy to house dust. Construct the 95% CI
for the population proportion.
 = p + Z P(1-p) SE = 0.35(1-0.35) = 0.04
n 140
0.35 – 1.96  0.04   ≥ 0.35 + 1.96  0.04
0.27   ≥ 0.43
27%   ≥ 43%
 p Value Test
 Probability of Obtaining a Test Statistic
More Extreme  or ) than Actual Sample
Value Given H0 Is True
 Called Observed Level of Significance
 Used to Make Rejection Decision
 If p value a, Do Not Reject H0
 If p value < a, Reject H0
 Example Normal Body Temperature
What is normal body temperature? Is it actually
37.6oC (on average)?
 Example Normal Body Temp (cont)
Data: random sample of n = 18 normal body temps

37.2 36.8 38.0 37.6 37.2 36.8 37.4 38.7 37.2

36.4 36.6 37.4 37.0 38.2 37.6 36.1 36.2 37.5

Summarize data with a test statistic

Variable n Mean SD SE t P
Temperature 18 37.22 0.68 0.161 2.38 0.029

 95%CI=36.89-37.54
 Observed difference=37.6-37.22=0.38 which is
more than 0.32
 EFFECT OF A DRUG ON KIDNEY
WEIGHT OF 24 MICE

CONTROL
N12
MEAN
318
SD
10.2
SEM
EXPERIMENTAL 12 370 24.1
 Trial of 2 whooping cough vaccines
 N CASES ATTACK RATE
A 2400 22 24.4%

B 2300 14 16.2%