I agree with Robert. Too little details here to work with. Are you trying to reproduce some already published spectra? Since you mentioned scaling, is your whole spectra off by a simple factor? What experimental technique are you using?
I agree with the Robert and Avishek. At first please read this link:
http://www.molecularinfo.com/MTM/UV.pdf
On other part you must have e.g. use pure compunds ( correct elemental analysis (+/- 0,2 % )for the values of (C,H,N...) , MS,NMR) for the concentration of the solution.
Please give us wether you use single beam or dualbeam or diode arrays as spectrometer.
Are you using quartz cuvette?
A (absorbance) is unitless but epsilon is the molar absorptivity of the compound and is given in solution as follows M-1 cm-1 (1 is exponent). You can use log epsilon too for the spectra.
The wavelength range
for glass is 380-780 nm
for plastic 380-780 nm
for fused quartz below 380 nm
The cutoff of the solvent (spectranal quality) is important too.
All these aspects are important for a comparison between your value and the experimenta value cited in reference.
I assume you mean infrared absorptions? And I assume you are talking about comparison with some ab initio or DFT calculation of the harmonic frequencies/intensities??
It really depends on the level of theory you are using, and even the wavelength range you are comparing... but yes, if you are comparing experimental data to theoretical you usually need to scale the results since the latter usually don't account for anharmonicity. Different levels of theory do better jobs than others, and different scaling factors are recommended for each, and the basis set.
Take a look at: http://pubs.acs.org/doi/abs/10.1021/jp052793n
Thank you so much for your kind reply to my question.
Yes i am trying to reproduce the UV-Visible spectra for different chlorophyll molecules theoretically not experimentally.
Here I attached a file in which i am trying to reproduce the same as they have done for different solvents and different levels of theory but i could not do it .
In this paper , they are talking about scaling factors to reach the experimental spectra.