"Reverse" from frequency table or histogram to (pseudo)data and then do stats.

In case you don't have the data but only a classed/binned/grouped frequency table 
or histogram how can you calculate the mean?

class    Freq.
-----    -----
 0         6
10         2
20         8
30        12
40         4
50         7
        ____
        n=______
The midpoint of the classes: 5 15 25 35 45 55
Use the midpoint of each class as the average data value of the data in that class
and multiply by that class's frequency, and then sum them all:
(6 5s + 2 15s + 8 25s + 12 35s + 4 45s + 7 55s)
 5*6 + 15*2 + 25*8 + 35*12 + 45*4 + 55*7    
= 30 +  30  + 200  +  420  + 180  +  385
= 1245
Tedious! Error-prone. 

(pseudo)mean = 1245 / n  = _______   
Inaccurate because the midpoint is probably not the average value of the data in a class
but won't be too far off. Well, it's all you have in this case.

OR

grouped freq dist→data

Type the class/group/bin midpoints into "class/bin midpoints:" 
 In this example: 5 15 25 35 45 55
Paste the frequencies into "their Frequencies:"
 In this example: 6 2 8 12 4 7

It generates that many (i.e. its frequency) of each midpoint values.
 In this example:
5 5 5 5 5 5 15 15 25 25 25 25 25 25 25 25 35 35 35 35 35 35 35 35 35 35 35 35 45 45 45 45 55 55 55 55 55 55 55
(i.e. 6 5s, 2 15s, 8 25s, 12 35s, 4 45s, 7 55s)

Copy and paste into
freq dist, histogram, stats ***

and let that do the tedious and error-prone arithmetic.
mean=_______  And all the other stats...

*******************************************************************

class    Freq.
-----    -----
 100       37
 130       18
 160        5
 190       15
 220       18
 250       22
 280       31

I recommend using the software. 
n=_____
What is the (pseudo)mean:______