We use 101 galaxies selected from the Nearby Field Galaxy Survey ( NFGS ) to investigate the effect of aperture size on the star formation rate , metallicity and reddening determinations for galaxies .
Our sample includes galaxies of all Hubble types except ellipticals with global SFRs ranging from 0.01 to 100 { M } _ { \odot } yr ^ { -1 } , metallicities between 7.9 \lesssim \log ( { O / H } ) +12 \lesssim 9.0 , and reddening between 0 \lesssim A ( V ) \lesssim 3.3 .
We compare the star formation rate , metallicity and reddening derived from nuclear spectra to those derived from integrated spectra .
For apertures capturing < 20 % of the B _ { 26 } light , the differences between nuclear and global metallicity , extinction and star formation rate are substantial .
Late-type spiral galaxies show the largest systematic difference of \sim 0.14 dex in the sense that nuclear metallicities are greater than the global metallicities .
Sdm , Im , and Peculiar types have the largest scatter in nuclear/integrated metallicities , indicating a large range in metallicity gradients for these galaxy types or clumpy metallicity distributions .
We find little evidence for systematic differences between nuclear and global extinction estimates for any galaxy type .
However , there is significant scatter between the nuclear and integrated extinction estimates for nuclear apertures containing < 20 % of the B _ { 26 } flux .
We calculate an ‘ expected ’ star formation rate using our nuclear spectra and apply the commonly-used aperture correction method .
The expected star formation rate overestimates the global value for early type spirals , with large scatter for all Hubble types , particularly late types .
The differences between the expected and global star formation rates probably result from the assumption that the distributions of the emission-line gas and the continuum are identical .
The largest scatter ( error ) in the estimated SFR occurs when the aperture captures < 20 % of the B _ { 26 } emission .
We discuss the implications of these results for metallicity-luminosity relations and star-formation history studies based on fiber spectra .
To reduce systematic and random errors from aperture effects , we recommend selecting samples with fibers that capture > 20 % of the galaxy light .
For the Sloan Digital Sky Survey and the 2dFGRS , redshifts z > 0.04 and z > 0.06 are required , respectively , to ensure a covering fraction > 20 % for galaxy sizes similar to the average size , type , and luminosity observed in our sample .
Higher-luminosity samples and samples containing many late-type galaxies require a larger minimum redshift to ensure that > 20 % of the galaxy light is enclosed by the fiber .