Logistic regression in psychiatric literature: evaluation of articles published between and in a prominent journal. Medicina basada en evidencia. Assessments of articles published in other medical specialties have shown that the report of logistic regression is incomplete in many cases and this leads to difficulties in interpretation. OBJECTIVE: Assess the quality related to logistic regression analysis and accomplishment of internal validity criteria of articles published in a high impact journal specialized in psychiatry.
|Published (Last):||19 October 2015|
|PDF File Size:||15.99 Mb|
|ePub File Size:||17.23 Mb|
|Price:||Free* [*Free Regsitration Required]|
Application of logistic regression models in observational methodology: game formats in grassroots football in initiation into football. Daniel Lapresa 1 , Javier Arana 2 , M.
Teresa Anguera 3 , J. This study shows how simple and multiple logistic regression can be used in observational methodology and more specifically, in the fields of physical activity and sport. We demonstrate this in a study designed to determine whether three-a-side futsal or five-a-side futsal is more suited to the needs and potential of children aged 6-to-8 years. We constructed a multiple logistic regression model to analyze use of space depth of play and three simple logistic regression models to determine which game format is more likely to potentiate effective technical and tactical performance.
Key words: observational methodology; logistic regression; football; grassroots. They can be used for estimation or prediction purposes. A range of regression models exist that vary in numerous aspects, including number of predictor variables simple vs multiple regression and the nature of the variables continuous or dichotomous.
The present study had two aims: to show how logistic regression analysis could be applied to an observational study in the field of physical activity and sport and to determine how futsal competition models could be optimized for children aged years old. In this study, we compared F-5 with an alternative format 3-sided futsal, or F-3 to see if the latter might be more suited to the needs and potential of children aged years old.
We specifically compared the use of space depth of play and ball handling skills effectiveness and suitability for age during offensive play in F-3 and F Observation was nonparticipatory and governed by scientific criteria, and the level of perceptibility was complete. To collect the data for the study, we organized a three-way tournament for each of the game formats. To ensure between-session consistency, we used the same pitch, the same game duration, the same referee, the same ball no.
This combined approach is feasible when the field format criteria, used to build the category system, are atemporal and supported by a theoretical framework. Table 1 summarizes the criteria applied in the observation instrument. The sessions were recorded using the software program ThemeCoder, with consideration of the work of Jonsson and Jonsson et al.
Video recordings of each of the matches were used to record the study data. Each match consisted of a given number of moves, which, in turn, consisted of a given number of actions. To study the use of space, we analyzed moves for F-3 and moves for F-5, and to analyze the technical performance of players, we analyzed contacts for F-3 and for F We designed a multiple logistic regression model to determine the likelihood of a move being successful in F-3 and F-5 according to the use of space depth of play.
The criterion variable Move Conclusion was dichotomous; the two possibilities were Success if the move concluded in Zone 80, which contained the rival team's goalposts or Failure if the move concluded outside this zone.
Because the criterion variable was dichotomous and there was more than one predictor, we used a multiple logistic regression model. Because the predictor variable Move Initiation Zone had more than two categories, we generated dichotomous dummy variables, maintaining the information provided by the original variable. We designed three simple logistic models to analyze the children's technical performance in F-3 and F To assess the reliability of the data collected using the observation tool, we analyzed the level of agreement between the different data sets.
According to the criteria of Landis and Koch , p. We next analyzed the possibility of confounding and interaction. It was therefore decided to maintain the three dummy categories for this predictor in the multiple logistic regression model. To choose which variables to include in the predictive regression model and to estimate the strength of the association between the criterion variable and the predictor variables, we used three SPSS procedures: Enter, Forward Selection Wald , and Backward Elimination Wald.
The model accurately predicted Furthermore, it yielded a sensitivity of We then performed the estimation process based on the resulting logistic regression model. In the first case, the Exp B value corresponding to Game Format was 2. In the second case, the predicted probability of success for F-3 was.
The respective results for F-5 were. We therefore established Game Format as the criterion variable in the three simple logistic regression models, which we constructed using the Enter selection method in SPSS.
The first model had a predictive accuracy of The Exp B value for Game Format was 1. The predicted probability of a contact being successful was. The second logistic regression model had a predictive accuracy of The predicted probability of a contact involving dribbling was. The third simple logistic regression model had a predictive accuracy of The Exp B value for Game Format was 2.
Finally, the predicted probability of a contact involving dribbling plus a shot at goal or continuation of attack was. The high kappa values obtained in our analysis of agreement confirm the reliability of the data used for the current study. We constructed a multiple logistic regression model to investigate the probability of a move being successful ball reaching the goal area according to the sector in which the move was initiated and the game format used F-3 vs F-5 in children aged years.
In both F-3 and F-5, the probability of the ball reaching the goal area decreased the further away the move was initiated. This observation is consistent with results reported by Castelo , Castellano and Perea for soccer and by Arda and Arana for grassroots soccer. The likelihood of success was higher for F-3 than for F-5 in all cases analyzed. Furthermore, the chances of the ball reaching the goal area was 2. A greater use of space observed in F-3 in our study is associated with better quality of play, as indicated by Vales in a study of tactical indicators of performance in soccer.
We also constructed three simple logistic regression models to analyze the quality of technical actions executed by children aged in F-3 and F In the first model, we saw that the likelihood of a contact being effective was 1. In the second model, we categorized the criterion variable as Adapted when the contact involved dribbling and, as Not Adapted, when it did not. We found that the smaller game format was better suited to the needs of the children in this age group as players were 1.
In the third model, we also saw that F-3 was more suited to the needs of children as the likelihood of a contact involving both dribbling and a shot at goal or continuation of attack was 2. Our findings show that compared with F-5, F-3 played by children aged according to the rules described by Lapresa et al.
We have shown how multiple and simple logistic regression models can be used in observational methodology, and more specifically in studies analyzing how football and sport in general can be adapted to the needs of children using multiple dichotomous variables in addition to those used in our study. Anguera, M. Observational Typology. European-American Journal of Methodology, 13 6 , Araguayo, E. Un modelo predictivo para el logro del rendimiento motor. Arana, J. Adapting football to the child: an application of the logistic regression model in observational methodology.
Tesis Doctoral. Ato, M. Bagley, S. Logistic regression in the medical literature: standards for use and reporting, with particular attention to one medical domain. Journal of Clinical Epidemiology, 54, Bakeman, R.
Cambridge: University Press. Benedek, E. Barcelona: Paidotribo. Buenos Aires: Efdeportes www. Broc, M. Camerino, O. Dynamics of the game in soccer: Detection of T-patterns. European Journal of Sport Science, 12 3 , Carvalho, J. Ensino do futebol: futebol de 11 ou futebol de 7. Revista Horizonte, 5 25 , Casal, C. Castellano, J. Castelo, J. Barcelona: Inde. Cohen, J. A coefficient of agreement for nominal scales. Educational and Psychological Measurement, 20, De Irala, J.
Pamplona: Newbook Ediciones. De la Vega, R. Escudero, J. Hellevik, O. Linear ersus logistic regression when the dependent variable is a dichotomy. Herrera, A. Influence of equal or unequal comparison group sample sizes on the detection of differential item functioning using the Mantel-Haenszel and logistic regression techniques.
This can be extended to model several classes of events such as determining whether an image contains a cat, dog, lion, etc. Each object being detected in the image would be assigned a probability between 0 and 1 and the sum adding to one. Logistic regression is a statistical model that in its basic form uses a logistic function to model a binary dependent variable , although many more complex extensions exist. In regression analysis , logistic regression  or logit regression is estimating the parameters of a logistic model a form of binary regression. In the logistic model, the log-odds the logarithm of the odds for the value labeled "1" is a linear combination of one or more independent variables "predictors" ; the independent variables can each be a binary variable two classes, coded by an indicator variable or a continuous variable any real value.
Please cite us if you use the software. Note that regularization is applied by default. It can handle both dense and sparse input. Use C-ordered arrays or CSR matrices containing bit floats for optimal performance; any other input format will be converted and copied. Read more in the User Guide. Used to specify the norm used in the penalization.