Scribd
Upload
26 views28 pages

T Test One Sample

The document discusses statistical significance testing using the Student's t-test. It defines key terms like independent and dependent variables, explains the assumptions and types of t-tests, and provides examples of performing one sample and paired t-tests to test hypotheses about population means.

Uploaded by

imayan.2700
Copyright
© © All Rights Reserved
We take content rights seriously. If you suspect this is your content, claim it here.
Available Formats
Download as PPTX, PDF, TXT or read online on Scribd
26 views28 pages

T Test One Sample

The document discusses statistical significance testing using the Student's t-test. It defines key terms like independent and dependent variables, explains the assumptions and types of t-tests, and provides examples of performing one sample and paired t-tests to test hypotheses about population means.

Uploaded by

imayan.2700
Copyright
© © All Rights Reserved
We take content rights seriously. If you suspect this is your content, claim it here.
Available Formats
Download as PPTX, PDF, TXT or read online on Scribd
You are on page 1/ 28

TESTS OF STATISTICAL

SIGNIFICANCE (t – test)
• The independent variable (traditionally plotted on the x axis) is what
can freely change.
• The dependent variable (traditionally plotted on the y axis) is what is
affected by the independent variable.

• Examples:

Independent - outdoor temperature; dependent - refreshment sales


Independent - time of the day; dependent - electricity consumption
Independent - price of aviation fuel; dependent - price of airline tickets
between specific destinations
STUDENT‘s t-TEST
• This test was invented by a statistician working for the brewer
Guinness. He was called WS Gosset (1867-1937), but preferred to
keep himself anonymous so wrote under the name “Student” It is used
for CONTINUOUS DATA and comparison of means
The t-test distribution
• lived from 1876 to 1937
• Gosset invented the t -test to handle small samples for quality control
• In brewing. He wrote under the name "Student
THE t-DISTRIBUTION
• Also known as the "Student's T Distribution"
• A type of probability distribution that is theoretical and resembles a
NORMAL DISTRIBUTION
• It is UNI- - MODAL
• BELL SHAPED
• It is symmetrical around the mean and extends indefinitely on both
sides
• It has many properties which differentiate it from the standard normal
or z-distribution
THE t-DISTRIBUTION
• It is actually a family of distribution, i.e. different distribution for
each sample value of n-1
• It deviates form normality with small sample sizes i.e. it is more flat
with long tails & variance is more than 1 but approaches 1 when
sample size becomes large
• 95 % t value is 2.36 ( depends on DF)
• 95 % z value is 1.96
THE t-DISTRIBUTION
• T distribution differs from the normal distribution by its degrees of
freedom. The HIGHER THE DEGREES OF FREEDOM, THE CLOSER
that distribution will resemble a STANDARD NORMAL DISTRIBUTION
with a mean of 0, and a standard deviation of 1
• Critical values used to decide significance obtained from t table
• With any level of significance, critical value to achieve statistical significance
varies with the size of sample
• At indefinite degree of freedom critical value for “t” is same as “z”
• It is a “probability distribution that is used to estimate population parameters
when the SAMPLE SIZE IS SMALL And/or when the POPULATION
VARIANCE IS UNKNOWN
t DISTRIBUTION
TABLE
• Using the t distribution table, critical values are determined given the
form of the alternative hypothesis the level of confidence, and the
degree of freedom (n-1)
• Example:
• Find the critical values of t given that:
• a) Ha: ≠ , α = 0.05 df =12
t (two tailed,α,df,) = +/- 2.179
• b) Ha: > , α = 0.01 df =29
t (one tailed,α,df,) = + 2.462
z-TEST OR t TEST ??
n>30 n<30
σ known (pop SD) Z test Z test
σ Unknown (pop SD) Z test T test

σ = Population standard deviation


n = sample size
Degrees of Freedom
• Refers the number of values that are free to vary after certain
restrictions have been placed on the data.
• Number of elements in a set that can be arbitrarily defined before the
rest of the elements in the set are determined.
t – test – when to use it?
• t distribution is used for statistical testing when the population
parameters are unknown and / or when the sample size is small
• Comparison of means
• Does that mean that we can not use the t – test if population
parameters are known and / or the sample size is sufficiently large?
t-test – Assumptions / Conditions
• Samples are drawn randomly and independently from their respective
populations
• Sample size is small
• Normal distribution
In these contexts t-test is similar to a z-test
Types of t - test
T test

Paired t test
(dependent ) Unpaired t test
Unmatched (independent)
within matched
Repeated measures
ONE SAMPLE T - TEST
One Sample t – test
(Sample Vs The Population)

• Finding out if the difference between sample mean and population


mean (may be hypothesized) is statistically significant or not. Or to
find out if the given sample is from a particular population.
• t = x - µ / s /√n
• Value of t calculated here is t calc.
This value is to be compared to t crit
• The formula is essentially the same as that for z test so the difference
henceforth would be ?
t-Test for single mean
• The following are the weight (mg) of each of 20 rats drawn at random
from a large stock. Is it likely that the mean weight for the whole stock
could be 24 g, a value observed in some previous work?

9 18 21 26
14 18 22 27
15 19 22 29
15 19 24 30
16 20 24 32
Steps for test for single mean
1.Question to be answered (Research Question)
Is the Mean weight of the sample of 20 rats is 24 mg?
N=20, x=21.0 mg, SD=5.91 , μ=24.0 mg
2 . Null & Alternate Hypothesis
The mean weight of rats is 24 g / not 24 g .That is, The sample mean is
equal/ not equal to population mean.
3. Assumption
• Sampled population is normal
• Small random sample (n < 30)
• σ is unknown
4. Level of Significance – 0.05
Steps for test for single mean
• 1.Apply test of significance
• 2.Justification: Small sample size
Degree of Freedom - 19
• 3.Find critical value from t table – t crit = t .05, 19 = 2.093
• 4.Compute test statistics t =x-μ /s /√n
• t(n-1) df
• 5. Test statistics : n=20, x=21.0 mg, SD=5.91, μ=24.0 mg
• t=21-24/5.91/√20=-2.27
Steps for test for single mean
6. Comparison with theoretical value
If t crit (n- 1)< t cal(n-1) reject Ho
If t crit (n-1)> t cal(n-1)fail to reject Ho
7. Reject or fail to reject null hypothesis - t crit = t .05, 19 = 2.093
Reject H 0 as t calc 2.30 > t crit 2.093
8. Inference : Answering the research question
THERE IS NO STATISTICALLY SIGNIFICANT EVIDENCE
THAT THE SAMPLE IS TAKEN FROM THE POPULATION
WITH MEAN WEIGHT OF 24 Gm
9. Calculate CI for population mean:
CI 95 = Mean± t 0.05 (SEM) = 21 ±2.093(SEM)
Exercise
• A medical investigation claims that the average number of infections
per week at a hospital in a central city is 16.3. A random sample of 10
weeks had a mean number of 17.7 infections. The sample standard
deviation is 1.8. Is there enough evidence to reject the investigators
claim at alpha = 0.05?
Solution
• H o : x = 16.3 (claim)
• H A : x ≠16.3 (claim)
• N = 10 , sd = 1.8, alpha = 0.05
• Df = n – 1 = 9
• Critical values = ±2.262 at df = 9
• t calc = 2.46
• Reject the null hypothesis since 2.46>2.262
• There is enough evidence to reject the claim that the average number
of infections is 16.3
Exercise
• A physicians claim that joggers maximal volume oxygen uptake is
greater than the average of all adults. A sample of 15 joggers has a
mean of 40.6 ml/kg and SD of 6ml/kg. if the average of all the adults
is 36.7 ml/kg, is there enough evidence to support the physicians claim
at alpha = 0.05
Exercise
• The following are the oxygen uptake during incubation of a random
sample of 15 cell suspensions:
14.0, 14.1, 14.5, 13.2, 11.2, 14.0, 14.1, 12.2, 11.1, 13.7, 13.2,
16.0, 12.8, 14.4, 12.9
• Do these data provide sufficient evidence at the 0.05 level of
significance that population mean is not 12 ml? what assumptions are
necessary?
Exercise
• Suppose it is known that the IQ score of a certain population of adults
are approximately normally distributed with a S.D of 15. A simple
random sample of 25 adults drawn from this population had a mean
IQ of 105. On the basis of these data can we conclude that the mean
IQ score for the population is not 100? Let the probability of
committing type I error is 0.05

You might also like