Transcript No Slide Title

Lab 1
Sample calculation of statistics
Use Microsoft Excel Program
Save data from either Food Chemistry or Experimental Foods
•go to Food Chemistry or Experimental Foods Lab Data pages
•follow the save instructions at the top for the data you wish to
save and then run
•save data for your day on your disk
Run/Open data in Microsoft Excel
In Microsoft Excel, Select ‘Tools’ and then ‘Data Analysis’
Note: If you do not have ‘Data Analysis’, you may have to load this
through ‘Add-Ins’ - Select ‘Add-Ins’ then select ‘Analysis Toolpak’
From ‘Data Analysis’
Select ‘ANOVA: Two-Factor Without Replication’
Fill in the information for ‘Input Range’, and
‘Output Range’ - Remember if you select your data
with labels, the labels box must be checked.
Highlight
the Data
Make sure
Highlighted Data is
in the Input Range
Box
Make sure if Labels from the
data were Highlighted that
the Labels box is checked Alpha should always be
0.05.
Now you are ready to select
the Output Range - This is
where the ANOVA table will
be generated
Make sure the cursor is in the
Output data box - If it is not then
you will loose the information in
the Input Range box
Select the cell
Make sure the cell also
gets registered in the
Output Range box - If
nothing is in this box an
error will occur.
The ANOVA begins at the cell selected for the Output Range
First, look at the P-value
- If it is greater than
0.05 it is not significant
- if it is less than 0.05 it
is significant
Focus in on columns
(treatments) P-values
because these represent
whether treatments were
significant - In this case,
columns are
significantly different,
thus treatments are
different. What next?
Multiple Comparison or Least Significant Difference (lsd)
•place treatment means (averages) from high to low
Treatments
Sample 693
Sample 054
Sample 893
Average
7.3
7.3
4.6
Averages for
the 3 treatments
Now you must calculate the lsd to determine which mean or
means are significantly different from each other
Where:
(2)EMS
lsd = t
n
t is the critical t value from the t-table
with df as the error df from ANOVA
table
EMS is error mean square from
ANOVA table
n is the number of samples for each
mean (i.e. # panelists)
n
df for t-table
EMS
THE lsd CALCULATION
•You need to use the t-table for helping do this calculation. In the
table, you use the 0.05 for 2-tails probability
•For 12 df from df for EMS you get a t value of 2.179
•EMS from ANOVA chart is 1.5
•the value for n is 7 from the ANOVA chart.
THUS:
(2)1.5
lsd = 2.179
= 1.43
7
Now to determine which treatments are significantly different you
subtract each treatment from all the others. Thus, 7.3-4.6 = 2.7;
7.3-4.6 = 2.7; 7.3-7.3 = 0. This difference if less than 1.43 is not
significantly different, however, if greater than 1.43 it is significantly
different. You then assign letters or lines demonstrating significant
differences. Thus:
Treatments
Sample 693
Sample 054
Sample 893
Average
7.3 b
7.3 b
4.6 a
You conclude that Sample
893 is significantly
different from Samples 693
and 054, but that the latter
Samples (693 and 054) are
not different from each
other