Although friction variation is speculated to be a significant contributor to slip and fall incidents, it has not been related to a measurement of slipperiness in the literature. This field study investigated the relationship among multiple friction variations, friction levels and the perception ratings of slipperiness in six major working areas of 10 fast-food restaurants in the USA. The mean perception rating score for each working area was correlated with various friction reduction variables across all the restaurants in comparison with its correlation with the mean friction coefficient of each working area. The results indicated that the absolute and relative reductions in friction over the whole working area, among 12 friction reduction variables evaluated, could have a slightly better correlation with the perception rating score (r=0.34 and 0.37, respectively) than the mean friction coefficient of each working area (0.33). However, in friction measurements, more effort and time are needed to quantify friction variations than to obtain the mean friction coefficient. The results of the multiple regression model on the perception rating indicated that adding friction reduction variables into the regression model, in addition to the mean friction coefficient, did not make a significant impact on the outcomes. The results further indicated a statistically significant correlation between the mean friction coefficient and the maximum relative friction reduction over the whole area in each working area across all the restaurants evaluated (r=0.80). Despite a slightly lower correlation with perception rating than the friction variation, the mean friction coefficient of an area is still a reasonably good indicator of slipperiness.