Instrumented spinal stiffness measurements have shown high test-retest reliability. However, factors that may affect reliability have yet to be investigated. The objective of this study was to compare the: 1) within- and between-day reliability of a mechanical indentation device (MID) in measuring spinal stiffness, 2) measurement precision of averaging multiple measurements, and 3) reliability of stiffness measurements between individuals with and without low back pain (LBP). The spinal stiffness of 26 volunteers with and without LBP was measured 3 times by MID in each of two visits 1-4 days apart. Two stiffness measures were calculated from the resulting force-displacement data: global stiffness and terminal stiffness. Intraclass correlation coefficients (ICCs) were used to estimate reliability. Measurement precision was measured by minimal detectable changes, bias and 95% limits of agreement. Using the mean of three spinal stiffness measurements, the measurement precision was improved by 33.7% over a single measurement. Averaging three measurements, the within- and between-day reliability point estimates of both global and terminal stiffness were 0.99 and 0.98, respectively. The reliability estimates of spinal stiffness measurement using MID were not significantly altered by the participants' LBP status across all circumstances (95% confidence intervals overlapped). With our experimental protocol, averaging three spinal stiffness measurements using MID produces reliable stiffness measurements regardless of individuals' LBP status. Copyright © 2013 Elsevier Ltd. All rights reserved.