We have made an experimental investigation of how ion velocity and angle of incidence affect the magnitude and uniformity of secondary electron yield from thin amorphous oxide films. Flat target surfaces of Al2O3 and BeO in a high vacuum of 10−8–10−9 Torr are bombarded with singly-charged positive ions of Li, Na, K, Rb, Cs, and Ni isotopes. Each of these elements has a different yield versus velocity response. Our data are consistent with Parilis and Kishinevskii’s prediction of a common velocity threshold. Our estimate of the threshold velocity is 5.5×106 cm/sec. For a given element we find, within our experimental error, that isotopes of different masses but with the same impact velocity produce identical average secondary electron yields. Furthermore, the statistical variances of their secondary electron frequency distributions also are identical. These results demonstrate the absence of an isotope effect on secondary electron emission. In general, secondary electron yield is not linear with ion velocity. Additional results are obtained for MgO and Ta2O5 surfaces. We find that secondary electron yield depends strongly on the electronic shell structure of a projectile ion, that is, on its atomic number Z. Our data can be interpreted qualitatively using the Fano-Lichten-Barat electron promotion model for ion-atom collisions when the Z’s are unequal. We describe a statistical model of secondary electron emission in which each ion-atom collision below a surface may result in the release of one or more electrons. In this model the random sequence or order in which target atoms are struck, the unequal amount of electronic excitation generated during collisions of a projecticle ion with different kinds of target atoms, and the electronic properties of an oxide surface are all factors which shape the frequency distribution of observed secondary electrons. Each ion impact produces a pulse of secondary electrons. A pulse is just the sum of many localized showers of electrons released along the path of a projectile ion, reduced by electron absorption in the target during diffusion of the internal secondary electrons to the surface and escape. The statistics of ion-induced secondary electron emission are described remarkably well by the Polya (negative binomial) distribution.