2. In Excel, click on ‘Tools” at the top of the screen, select ‘Data Analysis’ on the
drop-down, and select either ‘t-Test: Two-Sample Assuming Unequal
Variances’, ‘t-Test: Two-Sample Assuming Equal Variances’ (read
‘Equal/unequal variance above to decide which one to use) and click ‘ok’.
3. In the small screen that pops up, you will need to add your data in the ‘Input’. For
the ‘Variable 1 range’, click on the small box with the red arrow at the right end of
the entry area. On the main screen, click and drag your mouse over the entirety of
one of your data columns. Click “return”. Do the same for the ‘Variable 2 range’
with your 2
nd
data column. Enter the value of your hypothesized mean difference
between two groups if you have one in “Hypothesized Mean Difference”. Then,
the ‘Output Options’ is where you’ll decide to show the result output, it can either
be the cell you selected or create a new worksheet. Click ‘ok’.
4. In your t-test results, focus on three parts: ‘t Stat’, ‘P(T<=t) two-tail’, and ‘t
Critical two-tail’. If the ‘t Stat’ value is larger than the ‘t Critical’, then you know
your H
a
is supported, and if the reverse is true, then you know your H
o
is supported.
Paired two-sample t-test: compare two means of paired samples
Assumptions:
▪ Independence within groups: each observation in the group is not associated with
any observation from the same group.
▪ Pairs of observations: each subject is measured twice, and two measurements
constitute a pair; in other words, we are comparing the mean measurement between
two groups in which each observation in one sample can be paired with an
observation in the other sample.
o Examples include repeated measurements of the same individual
before and after a treatment, or the effect of two different
treatments apply on the individual at the same time.
▪ Dependent variable must be from a normal distribution: refer to section 3.2 for
testing of normality (if the normality assumption is violated, then you might
consider using a non-parametric test).
Conduct Two-sample t-test in Excel:
1. Make sure that the two columns of data you are comparing are side-by-side.
2. In Excel, click on ‘Tools” at the top of the screen, select ‘Data Analysis’ on the
drop-down, and select either ‘t-Test: Paired Two Sample for Means’, and click
‘ok’.
3. In the small screen that pops up, you will need to add your data in the ‘Input’. For
the ‘Variable 1 range’, click on the small box with the red arrow at the right end of
the entry area. On the main screen, click and drag your mouse over the entirety of