Ilovepdf Merged

Independent Samples t-
test
PSYC2098
DİLAN ÇABUK, PH.D.(C)
Independent Samples t-test
Between Subjects Design
-2 groups
-Not more
Are my groups coming from different populations?

Null hypothesis
- The means of the groups are coming from the same population.
- There is no difference between group means.
Formula…
T calculation for one-sample t-test
T calculation for independent samples t-test

Estimated Standard Error in Independent
Samples t-test
The standard error in the denominator measures how accurately the
sample statistic represents the population parameter.
In the single-sample t formula, the standard error measures the
amount of error expected for a sample mean.
For the independent-measures t formula, the standard error
measures the amount of error that is expected when you use a
sample mean difference to represent a population mean difference.
Symbol for Standard Error
 For one sample t-test
For independent samples t-test

How to Calculate Estimated Standard
Error in Independent Samples t-test?
Each of the two sample means represents it own population mean, but in each case there is
some error.
M1 approximates its population with some error.
M2 approximates its population with some error.
-We have 2 sourves of error
This formula is only applicable when sample sizes are equal!!!

When Sample Sizes are not Equal…
Law of large numbers: Statistics obtained from large samples tend to
be better (more accurate) estimates of population parameters than
statistics obtained from small samples.
The variance obtained from a large sample is also a more accurate
estimate of the population variance than the variance obtained from
a small sample.
Pooled Variance
A procedure that allows the bigger sample to carry more weight in determining
the final value
Pooled Variance
The pooled variance is actually an average of the two sample variances, but the average is
computed so that the larger sample carries more weight in determining the final value
We will learn more detail in lab session

Estimated Standard Error in Independent
Samples t-test
Remember the Formula for Independent
Samples t-test
Cohen’s d for Independent Samples t-
test
Assumptions of Independent Samples t-
test
1. The observations within each sample must be independent.
2. The two populations from which the samples are selected must be normal.
3. The two populations from which the samples are selected must have equal
variances. (homogeneity of variance)
LEVENE’S TEST OF FOR VARIANCE
EQUALITY
VARIANCE HOMOGENITY TESTS:
Null hypothesis =>

SAMPLE 1 SAMPLE 2
VARIANCE VARIANCE
Alternative hypothesis =>
SAMPLE 1 SAMPLE 2
VARIANCE VARIANCE
Introduction to
t-statistics
Chapter 9
Dilan Çabuk, Ph.D. (c)
Remembering z-statistic
Remembering z-statistic
• You need population standard deviation

• Most of the time, this is unknown
• Estimating standard error using sample values –
estimated standard error (sM)
T-statistic • Take square root for standard deviation
Calculating estimated
standard error
T-statistic
- used to test hypotheses about an
unknown population mean
- Denominator is estimated standard
error (the only difference between z-
statistics and t-statistics)
Degrees of freedom describe the number of
scores in a sample that are independent and
free to vary.
Because the sample mean places a

Degrees of restriction on the value of one score in the
sample, there are n – 1 degrees of freedom
Freedom for a sample with n scores
Greater df, better sample variance
representing population variance, better t-
statistic.
A t distribution is the complete set
of t values computed for every
possible random sample for a
specific sample size (n) or a specific
T- degrees of freedom (df).
disribution The t distribution approximates the

shape of a normal distribution.
T-distribution
The Shape of T-
distribution
• t distribution has more variability than a normal z
distribution, especially when df values are small
• The t distribution tends to be flatter and more spread
out
• Because, for z-scores, bottom of the formula does not
vary, but it varies for t-scores
T-
distribution
Table
T-
distribution
table
Hypothesis testing • The estimated standard error in the denominator
measures how much difference is reasonable to expect
with t-statistic between a sample mean and the population mean.
Assumptions 1) The values in the sample must consist of
independent observations.
of the t-test 2) The population that is sampled must be normal.
The Role of • Large variance  Large error, small t-value
• Large sample  Small error, large t-value
Sample Size
Effect size • A hypothesis test simply determines
whether the treatment effect is greater
for t- than chance

• It does not really evaluate the size of the
statistic treatment effect

• It is possible for a very small treatment
effect to be “statistically significant,”
especially when the sample size is very
large.
Effect size for t-statistic
• Cohen’s d
• Estimated d
• an estimated d of 1.00 indicates that the size of the treatment effect

is equivalent to one standard deviation.
Percentage of the Variance Explained
• How much of the variation can be explained by treatment?
- r2
CONFIDENCE • a sample mean tends to provide a
INTERVALS reasonably accurate estimate of the
population mean.
FOR ESTIMATING • Sample mean tends to be near to the
POPULATION MEAN population mean
• A confidence interval consists of an
interval of values around a sample mean,
and we can be reasonably be confident
that the unknown population mean is
located somewhere in the interval
One-
tailed t-
test
For directional hypotheses

Ilovepdf Merged

Uploaded by

Copyright:

Available Formats

Ilovepdf Merged

Uploaded by

Document Information

Original Title

Copyright

Available Formats

Share this document

Share or Embed Document

Sharing Options

Did you find this document useful?

Is this content inappropriate?

Copyright:

Available Formats

Ilovepdf Merged

Uploaded by

Copyright:

Available Formats

Independent Samples t-

Are my groups coming from different populations?

T calculation for independent samples t-test

For independent samples t-test

This formula is only applicable when sample sizes are equal!!!

We will learn more detail in lab session

Null hypothesis =>

Alternative hypothesis =>

• You need population standard deviation

Because the sample mean places a

disribution The t distribution approximates the

for t- than chance

statistic treatment effect

• an estimated d of 1.00 indicates that the size of the treatment effect

You might also like