Dear Alperen, apparently you are using a spectroradiometer in reflection mode, so I will assume that in this answer. In reflection measurements ideally only photons are detected that come back from the sample to be measured. Hence, a totally black truly spectrallt neutral background would be best. To set your exposure time, gain, slit width and aperture, preferably use a spectrally neutral known grey-tone that returns light just a little above the spectral peak value of the sample to be tested. This provides you with optimum dynamic range.

Also recommend to download and read the free Light Measurement Handbook by Alex Ryer.

If your question is about dental composite resins, which are semi-transparent, and you will get different reading using different background

The color of the background / substrate will always be included in the measurement. But you can take a measurement of the background alone so you can see the CIELAB values for it. Typically it will have a suit (a * and b * values other than 0).

It is only if you want to measure Density with the spectrophotometer that you will be able to avoid "paper white" (the color of the substrate