The paper’s title is “Unnecessary Roughness? School Sports, Peer Networks, and Male Adolescent Violence”. It is written by Derek A. Kreager in 2007. This paper investigated the “extent to which participation in high school interscholastic sports contributes to male violence” from two aspects, which are the type of sport and peer atheletic participation. (Kreager, 2017) The data used by the author comes from the National Longitudinal Study of Adolescent Health. Add Health survey includes the longitudinal examination of violent outcomes, as well as a wealth of individual background variables and sociometric data. The paper used logistic regression model to predict the probability of a male conducting violence behavior.
The paper mentioned that if peer's bond in school is strong enough, the student is not likely to conduct violence behavior. However, through the model construction and analysis, the finding is that these two factors are positively related. The author explained that there might exist unobserved heterogeneity that can explain the association between peer athletic participation and violence. (Kreager, 2017) I think it is possible to fit the data into a hidden Markov model to eliminate this limitation and make the result clearer. We can think a person’s behavior as a stochastic process, which means the behavior of a person depends on the previous behavior of a person. Thus, give all the data parameters, we can take the longitude data as a time related data. Therefore, all the variable at a specific stage can be one state in a Markov model. In addition, we can add the hidden state, which is the unobserved heterogeneity. I think using hidden Markov model has a large chance to eliminate the limitation mentioned in the paper.
Moreover, since this paper published in 2007, there should be more data available to us. Thus, the longitude of the dataset becomes longer, which is possibly a chance to improve the accuracy of the model, since we have a larger sample size.
Reference:
Kreager, D. A. (2007). Unnecessary Roughness? School Sports, Peer Networks, and Male Adolescent Violence. American Sociological Review,72(6), 1019-1019. doi:10.1177/000312240707200612
The paper’s title is “Unnecessary Roughness? School Sports, Peer Networks, and Male Adolescent Violence”. It is written by Derek A. Kreager in 2007. This paper investigated the “extent to which participation in high school interscholastic sports contributes to male violence” from two aspects, which are the type of sport and peer atheletic participation. (Kreager, 2017) The data used by the author comes from the National Longitudinal Study of Adolescent Health. Add Health survey includes the longitudinal examination of violent outcomes, as well as a wealth of individual background variables and sociometric data. The paper used logistic regression model to predict the probability of a male conducting violence behavior.
The paper mentioned that if peer's bond in school is strong enough, the student is not likely to conduct violence behavior. However, through the model construction and analysis, the finding is that these two factors are positively related. The author explained that there might exist unobserved heterogeneity that can explain the association between peer athletic participation and violence. (Kreager, 2017) I think it is possible to fit the data into a hidden Markov model to eliminate this limitation and make the result clearer. We can think a person’s behavior as a stochastic process, which means the behavior of a person depends on the previous behavior of a person. Thus, give all the data parameters, we can take the longitude data as a time related data. Therefore, all the variable at a specific stage can be one state in a Markov model. In addition, we can add the hidden state, which is the unobserved heterogeneity. I think using hidden Markov model has a large chance to eliminate the limitation mentioned in the paper.
Moreover, since this paper published in 2007, there should be more data available to us. Thus, the longitude of the dataset becomes longer, which is possibly a chance to improve the accuracy of the model, since we have a larger sample size.
Reference: Kreager, D. A. (2007). Unnecessary Roughness? School Sports, Peer Networks, and Male Adolescent Violence. American Sociological Review,72(6), 1019-1019. doi:10.1177/000312240707200612