Theoretical value of a simple pendulum in motion

[S00_SUNNiE]

1. The problem statement, all variables and given/known data
I'm new to this forum and i'm not very sure with how i'm suppose to state my problems, thus if it's written wrong please tell me so.

I'm currently doing a motion lab report for my physics class and we have to analyse the motion of a pendulum.
The material used for the lab are: Aparatus, string, 100g bob, and stopwatch (to measure time it takes to complete 30 cycles).
I'm asked to find the relationship between the length and the frequency of the simple pendulum.
i have finhsed mostly everything that's asked for in the lab but for the conclusion i'm asked to add the percent error, but my problem is what is a theoretical value?
2. Relevant equations
This is the percent Error Formula:
(|theoretical value - experimental value|/theoretical vaule) x 100%

I've asked my teacher what is a theoretical value but she told me to look it up.
3. The attempt at a solution
i've tried googling for the theoretical value and one site has stated that the theoretical value is 9.80m/s2. It too is a lab report about the motion of a simple pendulum but i'm not quite sure if the value is correct.
1. The problem statement, all variables and given/known data



2. Relevant equations



3. The attempt at a solution

[LowlyPion]

Look at this entry at Wikipedia.
http://en.wikipedia.org/wiki/Pendulum_(mathematics)#Rule_of_thumb_for_pendulum_ length

Plug in the nominal values that you have for length and gravity and determine the calculated result.

Then consider the error that may have been introduced in measuring your string and timing and see if your observed values are close to the theoretical values you calculated.

[Delphi51]

I am thinking that you graphed your data, something like Period vs squareRoot(L), to get a straight line. The accepted formula is T = 2(pi)*squareRoot(L/g) so on that graph theory predicts a slope of 2(pi)/squareRoot(g).
The % error would then be the % difference between your slope and the slope theory predicts.

Personally, I never liked that approach. The scientist usually doesn't know the "correct" value and must estimate the accuracy of measurement. Hey, maybe you can earn a bonus mark! What you do is run your data through a calculator and get the variation or deviation of the data from the line of best fit. (Calculators usually show it when you use the line of best fit feature.) Then you say, "the slope is ___ plus or minus ___".
Then you could say "this is (or isn't) equal to the accepted value to within the experimental error."