Objectives To measure and compare rectal temperature measurement with and without thermometer protective sheaths. To survey veterinary practices regarding thermometer use and disinfection. Materials and Methods Thermometers were validated with and without protective sheaths for accuracy and repeatability in a water bath with a high precision thermometer as reference. Then, the rectal temperature of 500 dogs was measured with and without protective sheaths in randomised order. The difference in temperature measured by the two methods was analysed with a generalised linear model with order of temperature measurement, life stage, dog size, body condition, level of consciousness, class of temperature and operator as explanatory variables. Practical thermometer use and disinfection was surveyed by online questionnaire. Results In the pre‐clinical phase, study thermometers recorded average temperature differences of −0.05 ± 0.07°C (mean ± SD) and −0.03 ± 0.08°C with and without sheath, respectively and the coefficients of variation ranged from 0.1 to 0.5% with and without sheath. In the clinical study, temperatures measured with and without sheaths differed by less than ±0.2°C for 92% of the measurements. None of the explanatory variables were associated with the observed differences. Twenty‐five percent of survey respondents were aware of the guidelines for disinfection of thermometers. Clinical Significance Thermometer protective sheaths do not affect the measurement of canine rectal temperature in the clinic.