Near-infrared spectroscopy has been successfully used by the industry for chemical analysis due to its recognised advantages as a rapid and accurate tool (Bartholomew & Osuala, 1988; Mitsumoto, Maeda, Mitsuhashi, & Ozawa, 1991; Tøgersen, Isaksoon, Nielsen, Bakker, & Hildrum, 1999). The dairy industry uses infrared technology effectively to determine lactose, protein and fat contents in milk. Furthermore, this technology is widely used in the grain and livestock feed industry (Bjarno, 1982; Rotolo, 1979; Shenk, Westerhouse, & Hoover, 1979). Near-infrared reflectance (NIR) spectroscopy has also been used previously in meat analyses (Ben-Gera & Norris, 1968) and, recently, numerous studies and applications have been developed for the prediction of meat properties, particularly chemical composition (Denoyelle & Cartier, 1996; Hildrum, Nilsen, Tøgersen, Rødbotten, & Isaksoon, 1999; Lanza, 1983; Mitsumoto et al., 1991; Oh & Großklaus, 1995; Oliván, 1999; Oliván, de La Roza, Mocha, & Martínez, 2001; Tøgersen et al., 1999). Other characteristics of meat, such as tenderness, are studied by means of slow and destructive methods such as texturemeters equipped with Warner–Bratzler, Kramer or compression devices. Similarly, sensory analyses with trained panellists require a great amount of meat samples and this is a complex, expensive and time-consuming technique. Applications of NIR spectroscopy in the prediction of sensory meat quality are less developed than in other fields. Taking into account that covalent bonds that involve hydrogen are dominant in the near-infrared region (Davies & Grant, 1987), changes in meat that involve bond absorption energy could be measured by NIR spectroscopy. Such changes evidently also occur in meat during ageing (Hildrum et al., 1995) by oxidation or enzyme activity and are reflected in many characteristics such as tenderness, flavour and juiciness. Thus, both instrumental and sensory tenderness, and related variables such as sensory overall appraisal and juiciness could be predicted by NIR spectroscopy, although until now tenderness predictions have usually been inconsistent (Liu et al., 2003). The aim of this work is to evaluate NIR spectroscopy as a tool for determining instrumental texture variables and sensory properties, as well as chemical characteristics of beef.