You are on page 1of 5

Using Excel to do Basic Statistical Tests

Part 1: Reviewing Basic Statistics and Ways of


Representing Data.
Mean
Mean is also called the average. Mean used to determine the trend when numbers
are fairly similar to each other.
Use Excel to calculate the mean of the class grades above.

Title the second column on the Excel Spreadsheet Grades on Project.


Enter the data into that column with one grade per row.

Highlight the entire column plus one extra row below the numbers.

Click on the formulas tab at the top of the spreadsheet.

Click on the down arrow to the right of autosum. Now some options will appear.

Click on "average".

The number produced in the last row is the mean.

In the 1st column next to the value of the mean, type mean to label the calculation.

1. What is the mean of the project grade? ______________


2. According to the mean, did most of the students understand 80% of the concepts?
________
3. Why is the mean so different from the median?
______________________________________________________________________________________
______________________________________________________________________________________
______________________________________________________________________________________
______________________________________________________________________________________

Standard Deviation
Standard deviation is a number that reflects how close the data numbers are to each other
and, thus, to the mean. If the standard deviation is small, then the numbers are very close to
each other. If the standard deviation is large, then the numbers vary greatly. Standard
deviation can be used only if there are 5 or more numbers in the data set.
A student wanted to see if leaves grow longer if the plant is placed in sunlight. She
hypothesized that leaves in the sun will grow longer than those in the shade. Here are her
results.

Length of Leaf (mm)


Leaves in Shade
24, 26, 27, 30, 31, 32, 32, 33, 37, 38, 43

Leaves in Sunlight
33, 34, 35, 35, 36, 36, 36, 37, 41, 42,

Using Excel to do Basic Statistical Tests


45

On your Excel Spreadsheet, label two columns- Shade and Sunlight


Enter the data into the columns.

To calculate Standard Deviation, click on the next empty cell in the shade column.

Choose more functions under autosum. Click on "STDEV" the OK. (You may
need to search for standard deviation if it is not visible on the list.)

Highlight your column of numbers for shade. Click OK

The number produced is your standard deviation.

Repeat for the numbers in your sunlight column.

Label the columns and the standard deviations.

4. What is the standard deviation of the leaves in shade? _________


5. What is the standard deviation of the leaves in sunlight? _________

Standard Error of the Mean


Standard error of the mean tells you how close the mean of your sample is to the true mean
of that group as a whole. If two groups have overlapping SEMs then they are very similar.

Place the cursor in the cell where you wish the standard error of the mean to appear,
and click on the fx symbol in the toolbar at the top.
A menu will appear that says Insert Function. Select Statistical from the top of
the menu, if necessary. Scroll down on the bottom of the menu and select STDEV;
then click OK.
Click on the picture of the spreadsheet, and highlight the numbers you averaged
earlier. Hit enter, and OK to calculate the standard deviation.
With the cursor still on the same cell, now click in the formula bar at the top of the
spreadsheet (the white box next to the = sign) to put the cursor in that bar so you
can edit the formula.
Put a ( in front of STDEV and a ) at the end of the formula. Ex. (STDEV (A1:A7)) Add
a / sign to indicated you are dividing this standard deviation. Put 2 sets of
parentheses (()) after the division symbol. Ex. (STDEV (A1:A7))/(()) Put the cursor in
the middle of the inner set of parentheses.
Now click on the fx symbol again. Choose Statistical on the upper menu, and then
COUNT on the lower menu.
Click on the spreadsheet picture in the pop-up box, and then highlight the list of
numbers you averaged. Hit enter and OK as before.
Move the cursor to be between the 2 sets of parentheses, and type SQRT. Hit enter.
The standard error of the mean should now show in the cell. Your formula in the
formula bar should look something like this, =(STDEV(A1:A7))/
(SQRT(COUNT(A1:A7))).
(This formula would calculate the standard error of the mean for numbers in cells A1
to A7.)

Using Excel to do Basic Statistical Tests


6. What is the SEM of the leaves in shade? _________
7. What is the SEM of the leaves in sunlight? _________
+or -2 Standard Errors of the Mean represents range of values that gives you 95%
confidence that your mean represents the true mean. The larger your SEM is, the more
variable your sample was.
Use either Excel or a calculator to determine 2 SEM.
8. What is 2 SEM of the leaves in shade? _________
9. What is 2 SEM of the leaves in sunlight? _________

Error Bars
Error bars are used on graph to indicate how variable the data is. They can be used to
determine if two sets of data are significantly different from each other. For example, we can
use it to determine if leaf length is affected by the amount of light by comparing the error
bars of leaves grown in sunlight to those in shade.
Error bars can represent 2 Standard Error of the Mean or range of data (minimum and
maximum data values). In AP Biology, we will use SEM when possible (5 or more data
points). After graphing the data with error bars, look at the error bars. If the error bars
overlap, the data of the compared groups is not significantly different. If the error bars do
not overlap, the groups are significantly different. This means that the independent variable
did have an effect on the results. For example, if the error bars of the leaves in shade and
the leaves in sunlight do not overlap, sunlight causes leaves to grow faster.

On graph paper, make a bar graph of the mean length of leaves in shade and
sunlight. Leaf length will be on the Y axis. You will have two bars-one for shade and
one for sunlight.2SEM. For example, if your mean is 10 and your SEM is 1, error bars
are drawn from 12 (10+2) to 8 (10-2).

12
10
8

10. For leaves in shade, what is your mean? _______________


a. What is your SEM?_______________
b. What is the upper line of the error bar (mean + 2 SEM)? ___________
c. What is the lower line of the error bar (mean 2 SEM)? ____________

For the Leaves in Shade bar, draw a line at your upper and lower 2x SEM as shown
above. Connect the lines.

Using Excel to do Basic Statistical Tests


11. For Leaves in Sunlight, what is your mean? _______________
a. What is your standard deviation?_______________
b. What is the upper line of the error bar___________
c. What is the lower line of the error bar ____________

Draw the error bars for the sunlight data using 2 SEM.

Make a key that indicates that the error bars are for 2 SEM.

12. Look at the two error bars. Do they overlap the many of the same numbers? ________
13. Does this mean that the data is significantly different or not?
_____________________________

PART 2: Using T tests to Determine Significance


T-Tests are another way to determine if two sets of data are significantly different. In other
words, it determines if your independent variable really does have an effect on your results.
For example, does compost really cause plants to grow taller? Do males really do better in
math than females? Do mice gain weight significantly faster if they have more fat in their
diet?
Complete the T Test on the Leaf Length data from above.

Click on an empty cell where you would like to record your T Test result.
Look under the formula bar at the top of the spreadsheet.

Click "more functions"; choose "statistical" then "TTEST".

A dialog box will appear. Click in the box next to "Array 1".

Drag the dialog box out of the way, then highlight your first column of numbers.

Click in the box next to "Array 2" and highlight your second column of numbers.

Next answer the tails question with a 1 or 2.


o

To answer the "tails" question, think about your hypothesis. If you said that
the variable would have an effect, your tails is 1. If you had no hypothesis or
a null hypothesis, your tails is 2. (Null hypotheses state that the
independent variable will not have an effect on the results.

Next, answer the type of test with a 1, 2 or 3.


o

Experimental Investigations- "type 1"

Comparative Investigation- "type 2

Use type 3 if you get an error message.

Now hit "OK" and see your P-value.

Using Excel to do Basic Statistical Tests


A P-value below 0.05 is generally considered statistically significant, while one of 0.05 or
greater indicates no difference between the groups. In other words, if your value is less than
0.05, your two groups are different from each other and the independent variable appears to
effect the results. If your number looks like this: 2.03188E-7, Excel is giving you the number
in its version of scientific notation. This number is actually 2.03 X 10 -7, or 0.000000203.

14. What is your p value? _________________________


15. Is the difference between the groups significant? (i.e. does sunlight cause a
significant increase in growth of the leaf?) Explain how you know this from your p
value. ____________________________________
___________________________________________________________________________________

You might also like