You'll have to do it for your long list of data.
This one stops at 19, but yours goes to 62.
I'm just going to show you how you would do it
if there were only the first 4 values in your
list. So don't use the sums below.  Do it for
you whole list, which will take a long time. 

Make this chart.  The frequencies f are how many of that number
are in the list. You have 2 16s, 4 17's, 5 18's and 6 19's.
Then multiply the f's by the x's. And add up the f's and the
f·x's.

x   f  f·x   x-x  (x-x)²  f·(x-x)²
----------------------------------
16  2   32   
17  4   68 
18  5   90
19  6  114
---------------------------------
   17  304


Now divide the ∑f·x = 304 by the ∑f = n = 17 and get x = 17.88

Subtract x = 17.88 from each x and put it in the next column.

x   f  f·x   x-x  (x-x)²  f·(x-x)²
----------------------------------
16  2   32 -1.88   
17  4   68  -.88 
18  5   90   .12
19  6  114  1.12 
---------------------------------
   17  304  

We square each number in that column
and put it in the next column. 

x   f  f·x   x-x  (x-x)²  f·(x-x)²
----------------------------------
16  2   32 -1.88   3.53   
17  4   68  -.88    .77
18  5   90   .12   1.44  
19  6  114  1.12   1.25 
---------------------------------
   17  304    

Now we multiply the number in column f times
the numbers in column (x-x)² and put those
values in the last column.  Add that last column

x   f  f·x   x-x  (x-x)²  f·(x-x)²
----------------------------------
16  2   32 -1.88   3.54     7.09   
17  4   68  -.88    .77     3.11
18  5   90   .12    .01      .07     
19  6  114  1.12   1.25     7.49  
---------------------------------
   17  304                 17.76

Now divide that ∑f·(x-x)² by n-1 which is 17-1 or 16,

and you get the variance = s² = 1.11

You'll have to do that will all your data.  It's
going to be a lot of work.  And maybe you'll need
to take the decimals out further to get the
accuracy you need.

Edwin