t - DISTRIBUTION

‹#›
 Student t Distribution
( t-dist )
When σ is unknown, we must use the
Student t distribution instead of the
normal distribution.
Requires new parameter
df = Degrees of Freedom
‹#›
 t-Distribution – Probability Density Function

A random variable T is said to have the t-distribution with

parameter  , called degrees of freedom, if its probability
density function is given by:

  1  / 2
 (  1) / 2   t 2

h (t )   1   , -<t<
  / 2      
where is a positive integer

‹#›
 Student t Distribution…[don’t really need to know
formula]
1) Here the letter t is used to represent the random variable,
hence the name. The density function for the Student t
distribution is as follows…

2) (nu) is called the degrees of freedom, and

3) (Gamma function) is (k)=(k-1)(k-2)…(2)(1)
‹#›

4
 Student t Distribution…[1 parameter]

1) In much the same way that and define the normal

distribution [2 parameters], , the degrees of freedom,
defines the Student [will use df]
2) t Distribution:

‹#›
Figure 8.24
1) As the number of degrees of freedom increases, the t
distribution approaches the standard normal distribution.

5
 Definition
The number of degrees of freedom (df) for
a collection of sample data is defined as:
“The number of sample values that can
vary after certain restrictions have been
imposed on all data values.”

In this section: df = n – 1

Basically, since σ is unknown, a data point has to

be “sacrificed” to make s. So all further
calculations use n – 1 data points instead of n.
‹#›
 Using the Student t Distribution

The t-score is similar to the z-score but applies for the

t-dist instead of the z-dist. The same is true for
probabilities and critical values.

α (area)

-1 0 0
P(t < -1) tα
(Area under curve) (Critical value)

NOTE: The values depend on df

7
 Important Properties of the Student t
Distribution

1. Has a symmetric bell shape similar to the z-dist

2. Has a wider distribution than that the z-dist
3. Mean μ = 0
4. S.D. σ > 1 (Note: σ varies with df)
5. As df gets larger, the t-dist approaches the z-dist

8
 Properties of a Student’s t
Distribution
1. The distribution is symmetric about the mean 0.
2. The distribution depends on the degrees of
freedom, d.f. (d.f. = n – 1 for μ confidence
intervals).
3. The distribution is bell-shaped, but has thicker
tails than the standard normal distribution.
4. As the degrees of freedom increase, the t
distribution approaches the standard normal
distribution.

9
Student’s t Variable

x
t
s
n
10
 The shape of the t distribution depends only
the sample size, n, if the basic variable x has a
normal distribution.

When using the t distribution, we will assume that the x

distribution is normal.

11
Student t Distributions for
n = 3 and n = 12

12
z-Distribution and t-Distribution

df = 2 df = 100

Wider Spread Almost the same

As df increases,
the t-dist approaches the z-dist
13
Progression of t-dist with df
df = 2 df = 3 df = 4

df = 6 df = 7 df = 8

df = 20 df = 50 df = 100
df = 5

14
CHOOSING THE APPROPRIATE
DISTRIBUTION

15
Choosing the Appropriate Distribution

 known and normally

Use the normal (Z) distributed population
distribution
or
 known and n > 30
 not known and normally
distributed population
Use t distribution
or
 not known and n > 30
Methods of Ch. 7 Population is not normally
do not apply distributed and n ≤ 30

16
 t-Distribution – Table of Probabilities

Remark: The distribution of T is usually called the Student-t

or the t-distribution. It is customary to let tp represent the t
value above which we find an area equal to p.

p
0 tp t
Values of T, tp,ν for which P(T > tp,ν) = p

17
t-distribution - Probability Density Function for
various values of 

 
 5
 2

-3 -2 -1 0 1 2 3

18
 Table of t-Distribution

t-table gives values of tp for various values of p and ν. The

areas, p, are the column headings; the degrees of freedom, ν,
are given in the left column, and the table entries are the t
values.

19
 t-Distribution - Example

If T~t10,
find:

(a) P(0.542 < T < 2.359)

(b) P(T < -1.812)

(c) t′ for which P(T>t′) = 0.05 .

20
 Example Solution

(a) P(0.542 < T < 2.359)

= 0.3-0.02 =0.28
t
0 0.542 2.359

(b) P(T < -1.812)=F(-1.812)

=P(T > 1.812)=0.05

-1.812 1.812
t
0

(c) t′ for which P(T>t′) = 1-F(t′ ) =0.05 .

t′ = 1.812 0.05

0 t′ t

21
 HAWKES LEARNING Continuous Random
SYSTEMS Variables
math courseware specialists 6.5 Finding t-Values Using
the Student t-Distribution

Comparison of the Normal and Student t-Distributions:

22
 Sec.
10.1
Critical Values of t

Degrees of
freedom

Use column 3 for a 90%

confidence level for a two-
tailed test (or confidence
Use column 2 for a 97.5%
interval)
confidence level for a one-
tailed test

Use column 2 for a 95%

Use column 3 for a 95% confidence level for a two-
confidence level for a one- tailed test (or confidence
tailed test interval)

23
 HAWKES LEARNING Continuous Random
SYSTEMS Variables
math courseware specialists 6.5 Finding t-Values Using
Student t-Distribution Table: the Student t-Distribution

Student t-Distribution Table

d.f. 0.100 0.050 0.025 0.010 0.005
1 3.078 6.314 12.706 31.821 63.657
2 1.886 2.920 4.303 6.965 9.925
3 1.638 2.353 3.182 4.541 5.841
4 1.533 2.132 2.776 3.747 4.604
5 1.476 2.015 2.571 3.365 4.032

24
 HAWKES LEARNING Continuous Random
SYSTEMS Variables
math courseware specialists 6.5 Finding t-Values Using
Student t-Distribution Table (continued): the Student t-Distribution

1. The numbers across the top row represent an area to

the right of t, known as .
2. The numbers down the first column represent the
degrees of freedom, d.f.  n – 1.
3. Where the appropriate row and column intersect,
we find the t-value associated with the particular
area and degrees of freedom.

When calculating the t-values, round your answers to three

decimal places.

25
 HAWKES LEARNING Continuous Random
SYSTEMS Variables
math courseware specialists 6.5 Finding t-Values Using
the Student t-Distribution
Find the value of t0.025 with 25 degrees of freedom.

Student t-Distribution Table

d.f. 0.100 0.050 0.025 0.010 0.005
23 1.319 1.714 2.069 2.500 2.807
24 1.318 1.711 2.064 2.492 2.797
25 1.316 1.708 2.060 2.485 2.787
26 1.315 1.706 2.056 2.479 2.779

t0.025  2.060

26
 HAWKES LEARNING Continuous Random
SYSTEMS Variables
math courseware specialists 6.5 Finding t-Values Using
the Student t-Distribution
How many degrees of freedom make t0.005 4.604?
Student t-Distribution Table
d.f. 0.100 0.050 0.025 0.010 0.005
1 3.078 6.314 12.706 31.821 63.657
2 1.886 2.920 4.303 6.965 9.925
3 1.638 2.353 3.182 4.541 5.841
4 1.533 2.132 2.776 3.747 4.604
5 1.476 2.015 2.571 3.365 4.032

d.f.  4

27
 HAWKES LEARNING Continuous Random
SYSTEMS Variables
math courseware specialists 6.5 Finding t-Values Using
the Student t-Distribution
Find the value of t such that the shaded area to the
right is 0.1 for 17 degrees of freedom.
Student t-Distribution Table
d.f. 0.100 0.050 0.025 0.010 0.005
15 1.341 1.753 2.131 2.602 2.947
16 1.337 1.746 2.120 2.583 2.921
17 1.333 1.740 2.110 2.567 2.898
18 1.330 1.734 2.101 2.552 2.878
19 1.328 1.729 2.093 2.539 2.861

t0.100  1.333

28
 HAWKES LEARNING Continuous Random
SYSTEMS Variables
math courseware specialists 6.5 Finding t-Values Using
the Student t-Distribution
Find the value of t such that the shaded area to the left
is 0.05 for 11 degrees of freedom.
Student t-Distribution Table
d.f. 0.100 0.050 0.025 0.010 0.005
9 1.383 1.833 2.262 2.821 3.250
10 1.372 1.812 2.228 2.764 3.169
11 1.363 1.796 2.201 2.718 3.106
12 1.356 1.782 2.179 2.681 3.055
13 1.350 1.771 2.160 2.650 3.012

t0.050  1.796, however the table assumes that the area is to the right of t.
Since the t-curve is symmetric at t  0, we can simply change the sign of the
t-value to obtain the correct answer.
t0.050  1.796
29
 HAWKES LEARNING Continuous Random
SYSTEMS Variables
math courseware specialists 6.5 Finding t-Values Using
the Student t-Distribution
Find the value of t such that the shaded area in the
tails is 0.02. Assume there are 7 degrees of freedom.
This type of problem is called two-tailed.
If the area in both tails is 0.02, then the area in one tail
would be 0.01.

Student t-Distribution Table

d.f. 0.100 0.050 0.025 0.010 0.005
7 1.415 1.895 2.365 2.998 3.499
8 1.397 1.860 2.306 2.896 3.355
9 1.383 1.833 2.262 2.821 3.250
10 1.372 1.812 2.228 2.764 3.169
11 1.363 1.796 2.201 2.718 3.106

t0.010  2.998 30
 HAWKES LEARNING Continuous Random
SYSTEMS Variables
math courseware specialists 6.5 Finding t-Values Using
the Student t-Distribution
Find the value of t such that the shaded area between –
t and t is 99%. Assume 24 degrees of freedom.
Since 99% of the area of the curve is in the middle, that
leaves 1%, or 0.01 of the area on the outside.
Because of symmetry each tail will only have half of 0.01
in its area, 0.005.
Student t-Distribution Table
d.f. 0.100 0.050 0.025 0.010 0.005
23 1.319 1.714 2.069 2.500 2.807
24 1.318 1.711 2.064 2.492 2.797
25 1.316 1.708 2.060 2.485 2.787
26 1.315 1.706 2.056 2.479 2.779

t  2.797.
31
 Using the t table (Table 4) for values…

1) For example, if we want the value of t with 10 degrees of

freedom such that the area under the Student t curve is
.05: Area under the curve value (t) :
COLUMN

t.05,10
t.05,10=1.812

‹#›

Degrees of Freedom : ROW

 Student t Probabilities and Values
1) Excel can calculate Student distribution probabilities and
values. Warning: Excel will give you the value for “t” where
 is the area in “BOTH” tails

=TINV(0.1,10) "=" 1.812

‹#›

33
 t-Distribution

The student t distribution was first derived by William S.

Gosset in 1908.
1) t is used to represent random variable.
2) t-distribution is very commonly used in statistical
inference.
3) Like normal distribution, t-distribution is symmetrical
about 0.
4) For larger degree of freedom, the t-distribution
approaches standard normal distribution.

34
 t-distribution for
various degree of freedoms

35
 EXCEL FUNCTION
for t-Distribution

Given t, to find cumulative probability

TDIST(t, df, tails)
t: Random variable, t can not be negative
df: degree of freedoms
Tails: 1 for one tail, 2 for two tails

TDIST returns the probability for random variable >t

Example: TDIST(1.5,50,1) = 0.07

36
 EXCEL FUNCTION
for t-Distribution

Given cumulative probability, to find

random variable t for two-tail test
TINV(p, df)
p: probability
df: degree of freedoms
TINV is the reverse of TDIST. TINV returns the t-value of the t-
distribution as a function of the probability and the degrees of
freedom.
Example: TINV(0.05, 30) = 2.0423

37
 Right Tail Probability
One Tail
Probabilities from a to 

P(t>a)

a
38
 RIGHT TAIL PROBABILITIES
One Tail
P(t>a) = area between a and ∞
– Probability to the RIGHT of a

EXCEL:
=TDIST(a,df,1)

P(t100>0.56) = TDIST(0.56,100,1)
= 0.2884
39
RIGHT TAIL PROBABILITIES
One Tail
negative value

P(t>a)

P(t<a)

40
 RIGHT TAIL PROBABILITIES
One Tail
negative value

– Excel does not work for negative vales of t.

– But the t-distribution is symmetric. Thus,
• TDIST(-a,df,1) gives the area to the left of a
negative value of a.
• 1-TDIST(-a,f,1) gives the area to the right of a
negative value of a.
P(t100>-0.56) = 1-TDIST(0.56,100,1)
= 0.7116

41
 Left Tail Probability
One Tail
Probabilities from - to a

P(t>a)
P(t<a)

a
42
 LEFT TAIL PROBABILITIES
One Tail
P(t>a) = area between -∞ and a
– Probability to the LEFT of a

EXCEL:
=1-TDIST(t,df,1)

P(t100<0.56) = 1-TDIST(0.56,100,1)
= 1-0.2884
= 0.7116

43
 PROBABILITIES
Two Tails

EXCEL:
44
=TDIST(a,df,2)
 PROBABILITIES
Two Tails
1) TDIST(a,df,2) gives twice the area to the right of a
positive value of t.
P(t>|a|) = P(t<-a or t>a) =P(t<-a) + P(t>a)
= area between –a and -∞
and area between a and ∞

EXCEL: =TDIST(a,df,2)
P(t100>0.56) = TDIST(0.56,100,2)
= 0.5767
45
 Given Two-Tail Probability to find t value

1) TINV(p,df) returns the value a, such that

P(|t| > a) = probability or
P(t < -a or t > a) = probability

EXCEL:
=TINV(P,df)

t0.05,100 = TINV(0.05,100)
= 1.984

46
 Given One-Tail Probability to find t value

1) If one-tail probability is given, to find the t value, you

need to multiply the probability by 2

EXCEL:
=TINV(2*P,df)

t0.05,100 = TINV(2*0.05,100)
= 1.66

47
 REVIEW

1) If tails = 1, TDIST (a,df,1) is calculated to determine

P(t>a).
2) Excel does not work on negative a.
3) If tails = 2, TDIST (a,df,2) is calculated to determine P(|t| >
a) = P(t > a or t < -a).
4) If test is two-tail, TINV(,df) returns that value a, such that
P(|t| > a) = P(t < -a or t > a) = .
5) If test is one-tail, TINV(2*,df) returns that value a, such
that P(t > a).

48