Many scientific publications on combat sports have dealt with physiological, biomechanical, psychological and training issues. Different physiological parameters have been studied in disparate combat sports such as silat (Aziz et al., 2002), judo (Borkowski et al., 2001; Callister et al., 1991; Kalayci, I., 2008, Taylor and Brassard, 1981; Thomas et al., 1989), boxing (Smith, 2006), wushu (Ribeiro et al., 2006), and wrestling (McGuigan et al., 2006).

As far as judo is concerned, research has been focused on training (Blais and Trilles, 2006; Azevedo et al., 2007), on the influence of psychological variables during the competition (Stevens et al., 2006) and in several biomechanical aspects (Imamura et al., 2006; 2007).

During the 1970's the study of the knowledge of performance of a sport and particularly the study of the technical errors made during said execution has turned out to be auseful tool to learn the basics at the initiation stage of sport techniques (Newell and Walter, 1981; Pieron, 1976; Schmidt, 1988; Schmidt and Gordon, 1977; Schmidt and Lee, 2005) and a valuable help for the later development of specific disciplines such as tennis (De Knop, 1983), gymnastics (Magil and Schoenfelder-Zohdi, 1996) and recently judo (Gutiérrez and Prieto, 2007). All these studies corroborate Gentile’s conclusions (1972), suggesting that it is far more useful to correct a movement or technical gesture from the error’s perspective if there is information about the nature of these mistakes than simply to point out the result of the execution.

The analysis of technical errors through observational methodology (Anguera et al., 2001) allows us, in the case of this particular study, to obtain a tool which helps to improve the teaching and learning process of the Uki Goshi judo technique in those subjects whose characteristics are similar to the ones studied in this piece of research. To this end, two fundamental goals are proposed: first, to obtain the frequency of the errors during the technique’s execution and second, to prove the existence of important relations among said errors, in order to be able to confirm if certain mistakes lead to others.

With the intention of analysing the different approaches to the Uki Goshi technique in depth we have observed that the existing publications on judo follow five different trends:

The reason why Uki Goshi is the object of this study is that the usage of this technique in the early stages of the initation to judo has been strongly recommended since its creation in 1882 until present day by the most prestigious authors of this sport in their teaching proposals. Thus, Jigoro Kano included this throw in 1895 in the first Gokio group, which can be considered as the first manual on the organisation of the process of teaching and learning the throws. From that moment on, new proposals have come into being, keeping this technique in the same group. These new plans are adapted to the pedagogical reality of the time and try to overcome the difficulties in the learning process that this classification entails. In this respect, Castarlenas and Calmet, 1999, Geesink, 1967, Kawaishi (1995), Sacripanti, 1989 and even the Federation Française de Judo et Ju-Jitsu et Disciplines Associés [FFJDA] (1967) have established new organizational systems for the teaching of this sport, where Uki Goshi is always present in the early stages of the initiation.

In view of the exposed antecedents, it can be concluded that this research comes into being as a consequence of the lack of scientific studies on judo aiming to determine technical errors in the throws and the relations among them, and also as a complement to new teaching trends created to facilitate the work of teachers and professionals of this sport.

Thus, the final purpose of the present study is to analyze the most frequent mistakes made in the execution of the Uki Goshi hip throw by adults who have no previous experience in judo techniques, showing how the observed errors relate in order to provide judo professionals with a useful technical support tool for the teaching-learning process of subjects of similar characteristics.

Observational methodology (Anguera, 1999) was used as it has the necessary rigour and flexibility to study the situations which present themselves during the teaching-learning processs of the judo technique. Specifically, this method can affirm that the kind of observation carried out was standardized, open and non-participant (Heinemann, 2003).

Likewise, the observational design (Anguera et al., 2001) is nomothetic (several participants execute the Uki Goshi throw at the same time) monitoring (a technical Uki Goshi gesture developed during five academic years) and multidimensional (the dimensions correspond to the criteria of the observation instrument). From the N/M/M approach (nomometic/ monitoring/ multidimensional) the following decisions derive: participants, observation instrument -OSJUDO-UKG (Uki Goshi Judo Technique Observation System)-, registering instrument and procedure.

The group of study was formed by 44 pupils of the Faculty of Education and Sports Sciences (21 men and 23 women) who belong to the 2003/2004, 2004/2005, 2005/2006, 2006/2007 and 2007/2008 academic years.

The execution of the Uki Goshi technique was filmed after a training period of approximately four months. The data were recorded with two digital video cameras using in both cases a full shot and a normal focus. The cameras were placed on two tripods, at a height of 1,50 metres, forming an angle one in front of the other and approximately 2m away from the tatami. Afterwards, the different filmed throws were edited with Pinnacle Studio software (version 8, 9, 11 and 12).

Appendix 1 describes the technical model used for the teaching of this throwing technique which was also used for the elaboration process of the system of categories with which the observation was carried out. For this purpose and due to the difficulty pointed out by different authors (Imamura et al., 2007; Kano, 1994; Koizumi, 1960) to divide the technique into the three traditional phases Kuzushi, Tusukuri and Kake, this division will not be strictly used, but rather a general description of it following the sequence of the different movements.

A continuous development of the observation instruments will allow us to deepen our knowledge of the teaching process of the judo technique. The chosen observation instrument for this study is OSJUDO-UKG, which combines field format and category systems (Jonsson et al., 2006). OSJUDO-UKG includes in its criteria the objectives of our study: different technical errors and their interrelation.

OSJUDO-UKG fits the proposed observational design, is multidimentional and consists of the following structure of criteria, dimensions and categories:

Fixed criteria (FC): course and sex.

Variable criteria: grip, unbalancing, right foot positions, right arm position, hip position, right arm action, left foot position, leg action, blocking action, throwing stage, control stage, rebalance and globality. Each of these criteria gives rise to their respective category systems which comply with the conditions of thoroughness and mutual exclusivity (T/ME).

The dimensions considered in this study have given rise to the following system of categories (thorough and mutually exclusive in every dimension), which is described in Appendix 2 of this article.

A continuous registering process was carried out during the observation of all the filmed sessions, using the software Match Vision Studio Premium v.1.0. (Castellano, Perea, Alday and Hernández, 2008). This is an interactive multimedia computer program which allows watching and registering in the computer’s screen itself the digitalized recording of the videos in avi, mpg or wmv format.

This program is highly flexible, and enables the input of all of the corresponding codes for each one of the variable dimensions criteria of the observation instrument OSJUDO-UKG (Table 1) in order to register their sequence. It must be pointed out also that when one of those

In addition, the quality of the data (Blanco-Villaseñor and Anguera, 2000) was calculated through the kappa coefficient -obtained using the program SDIS-GSEQ, version 4.2 (Bakeman and Quera, 1992; 2001)- getting a value of 0.91.

Following the registering of the 44 Uki Goshi throws performed by the participants, a first descriptive analysis of the frequencies and percentages of the occurrence of the technical mistakes was carried out. Afterwards, time patterns -sequential in this case-(Magnusson, 1996; 2000) were analyzed using the THÈME software, obtaining the corresponding dendogrames, which show the occurrence of mistakes in the execution of the studied technique.

Table 1 shows that the participants made a large number of different mistakes. It is apparent how some of these mistakes occurred more often than others. Out of 59 mistakes observed by the researchers, more than half (37) had a very low frequency of occurrence (only 1 to 5 participants). Therefore, it can be stated a priori that there exist a few mistakes which could be considered as common, due to the high frequency with which they were observed, and many others which cannot be considered as significant or important because they were scarcely registered in the videographic analysis. In fact, as it can be observed in Figure 1, only 8 of them were made by more than 10 participants out of the total 44 people who participated in the study.

In the analysis of the T-Patterns (temporary behaviour patterns) a series of important links related to the occurrence of chained mistakes, were observed which enable us to deepen our knowledge of the Uki Goshi technique.

As a result of this T-Pattern represented in Figure 2, it can be deduced that a possible chained sequence of Uki Goshi errors could be as follows: At the beginning of the throw, the Tori does not unbalance his opponent correctly (NU) and afterwards, he places his right arm (skilled arm) under his opponent’s left armpit surrounding the dorsal region of the back instead of the lumbar region (AB). These last two mistakes bring about another two new ones, specifiaclly, the Tori places his left foot after the Tai Sabaki on the inner part of the Uke’s same foot (ILF) and his iliac crests are parallel to those of his adversary (PCR) instead of occupying the right position described by the ideal technical model.

These mistakes lead firstly to an insufficient pull with the skilled arm (NT) probably due to its incorrect initial position (AB), and secondly to an inefficient block, thus the Uke’s body is not throwed down (by blocking it) over the Tori’s hip’s right side, but through the attacker’s body’s right side (STH), not existing a close contact with the Tori’s right iliac crest.

According to what has been observed in the presented T-Pattern, it can be concluded that this last error is originated by the Tori following his opponent’s body while it falls instead of blocking it (FOFA) and by the absence of a spin movement of his body in the Tsukuri stage (ITRT), a fundamental task in this hip throw.

The following dendrograme, Figure 3, shows the close relation that exists, as we pointed out before, between the incorrect position of the right arm during the initial stage of the throw (AB) and its lack of pull afterwards (NT)

Lastly, it must be emphasized that another important relation, not observed in the previous dendrogrames, is probably fundamental in order to determine the chain of erros and incorrect actions that lead to mistakes during the throw. As it can be observed in the next dendrogram, Figure 4, the existing relation between NBLC and STH, is high, since, every time the first error happens, the second one follows. Not blocking the opponent’s hip (NBLC), which is the primary movement in this technique, leads to throw the Uke down through the Tori’s side, following an irregular trajectory (STH) instead of throwing the Uke down over the Tori’s hip perpendicularly.

A Mann Whitney U Test found significant differences between males and females in the amount of observed mistakes in the Uki Goshi technique (p < 0.05). In the AB mistake (Tori puts his right hand under the Uke’s body’s left armpit, holding the dorsal region of his opponent’s back in order to throw him down). Specifically, the highest percentage of mistakes was observed in the male participants.

It is not easy to justify the reason for this, but after a close observation of the videos, a possible explanation could be that males who make this mistake performed the technique with an Uke who was shorter than them. The difference in the partner’s height makes it more difficult for the Tori to surround the Uke’s lumbar region with his hand. Female participants performed the technique with a partner who was either similar in height or taller than them, thus reducing the chances of making the same mistake.

As it was previously pointed out different researchers (De Knop, 1983; Gutiérrez and Prieto, 2007; Magil and Schoenfelder-Zohdi, 1996; Newell and Walter, 1981; Pieron, 1976; Schmidt, 1988; Schmidt and Gordon, 1977; Schmidt and Lee, 2005) have dealt with the study of knowledge of performance (Schmidt, 1988) of the technique and particularly the technical mistakes made in it. However, there are very few scientific studies which have used observational methodology as a tool to obtain the best results in the early stages of the teaching learning process of judo (Gutiérrez and Prieto, 2006; 2007). In fact, after a long research in specific databases and specialized magazines, we did not found any article which focused on the Uki Goshi technique nor any scientific studies linked to the sequential analysis of judo techniques.

Several of the most prestigious judo experts do reflect in their works on the most essential points to describe the Uki Goshi technique and offer a specific section on technical errors. Their conclusions are probably based on their own personal and professional experience, and the typical mistakes observed in the current study coincide in many cases. For example, many of these authors point out as a fundamental aspect or frequent mistake the incorrect performance of the initial unbalancing movement with both arms forward after the kumikata (NU error) (Daigo, 2005; Inman, 1997; Kobayashi and Sharp, 1995; Kolychkine, 1989; Mifune, 2004). In a similar vein, some authors pay attention to the position of the Tori’s right arm on the Uke’s body (AB, AW, AWGR, SBGR and ALAP errors), and claim that the Tori must never hold the Uke’s belt or judogi, he must firmly embrace his opponent’s waist (Daigo, 2005; Kolychkine, 1989; Kudo, 1967; Ohlenkamp, 2006; Taira et al., 1992). In this particular case, we observed that it is an error which manifests itself frequently, especially referring to the position of Tori’s right arm on the Uke’s back’s middle dorsal region, in more than 1 out of 3 of the analyzed participants (AB error).

Another group of authors consider the pulling of the Tori’s free body inwards very important in order to attract the Uke towards his own hip’s right side (NT) and block him with it (NBLC) (Kudo, 1967; Mifune, 2004; Uzawa, 1981). One of the most frequently observed mistakes in this technique’s analysis is the lack of a left foot pivot during the final stage of the throw (used in the initiation in order to favour the body’s spinning movement -FLFAT, PLFAT and ILFL errors-). Many of the most relevant consulted international authors agree that this spinning action is vital for the succesful performance of Uki Goshi technique (Daigo, 2005; Kawaishi, 1955; Kobayashi and Sharp, 1995; Koizumi, 1960; Ohlenkamp, 2006; Taira, Heguedas and Román, 1992; Uzawa, 1981; Watanabe and Avakain, 2001).

On the other hand, there are two key points in the technique which are conspicuously reflected, in the form of mistakes, in the videographic analysis. The first of them is the lack of a blocking movement performed with the Tori’s hip on the opponent’s body (NBLC) and the second one is related to the trajectory that Uke’s body follows during the throw (STH). In accordance with the technical model’s recommendations and following the premises of the world’s most prestigious authors, the Uke’s body must be blocked by the Tori’s right iliac crest, and afterwards, he must be thrown down forwards over the hip, following a perpendicular trajectory (Daigo, 2005; Kolychkine, 1989, Ohlenkamp, 2006; Taira et al., 1992; Watanabe and Avakain, 2001).

One of the original aspects of this study which will provide information concerning chained technical mistakes in the Uki Goshi throw (that is, the existing relations among the different technical errors) is the analysis of the sequential patterns in the occurrence of mistakes in this technique.

Undoubtedly, one of the most important sequencies found in this piece of research has been the existing relation between the lack of the Uke’s body’s block (NBLC) and the incorrect trajectory which the body follows during the throw (STH). The consulted bibliography does not mention this behaviour specifically, although most authors do point out the importance of the block in order to perform a successful throw (Daigo, 2005; Kolychkine, 1989; Ohlenkamp, 2006; Taira et al., 1992; Watanabe and Avakain, 2001).

It also seems relevant the existing relation between the incorrect positions of the Tori’s right hand on the Uke’s back (AB) and the deficient subsequent pulling inwards of the Tori’s free body (NT). Due to biomechanical reasons, pulling the Uke’s body towards the Tori’s hip is more complicated from a higher position (the back) than from a lower position (the hip). Both Kudo, 1967 and Taira et al., 1992 in their section on most important points refer to this mistake specifically. Likewise, the other previously mentioned authors also explain the importance of embracing the dorsal region during this technique (Daigo, 2005; Kolychkine, 1989; Ohlenkamp, 2006).

After a statistic-descriptive-sequential analysis of the Uki Goshi technical errors and after a process of interpretation and discussion, this piece of research came to the following conclusions: