Measuring behaviour, especially oral behaviour, has always been a debated issue: therefore the aim of this paper is to closely examine the study of oral behaviour in calves and the approaching methodology. Behavioural observations were conducted by two media (direct observations by check sheets and indirect observations by videotapes recorded by cameras connected to a digital field switcher and a time-lapse video recorder) in order to compare data and assess the reliability and validity of the two methods in identifying some oral behavioural patterns in calves.The study was carried out on 54 Polish Friesian calves housed in group pens and in individual crates. The behaviour of the calves was investigated during the fattening period on the 2nd, 7th, 13th, 18th and 23rd week, one hour before and one hour after each of the two meals. Two experienced observers checked the behaviour of the calves, including oral behaviours on structures and buckets and oral stereotypies, by direct observations using a scan sampling every 2 minutes. The calves' behaviour was also video recorded on the same days in which the direct observations were carried out and analysed by the same two observers. Percentages of time spent on each type of behaviour were calculated and analysed by Chi-square test. Regardless of the housing system, the comparison between direct and indirect observations revealed significant differences in almost every behavioural category. Licking, biting and nibbling structures, nibbling and sucking the bucket, playing with the bucket and the teat, chewing and oral stereotypies were significantly higher in direct observations compared to indirect (P < 0.001), while inactivity was higher in video recorded observations (P < 0.001). In conclusion, regardless of the type of housing, our results revealed an objective difficulty in analysing videotapes with very detailed behavioural categories, like oral behaviours. Although video recording can certainly represent a useful and practical alternative to direct observations in many situations, the video recording system used in this study for investigating calves’ oral behaviour, in spite of the use of multiple cameras, could not replace direct observations, probably due to the restricted field of view, the low depth of focus, the black and white vision, the lack of audio and the time-lapse feature.