Saturday, November 12, 2011

Correlation: The Relationship Between Hazing and Team Cohesion


Why do people continue to haze?  Some find it to be fun while others believe that it makes the team better.  Is hazing a part of team building? Do teams who participate in hazing have better team chemistry? These are some of the questions that encouraged a study to be conducted on hazing, team building, and cohesion.
            There is an article I would like to discuss about hazing and how it correlates to team cohesion.  First though, I would like to talk to you about correlation when used for statistical analysis. The book, “Research Methods in Physical Activity” by Thomas, Nelson, and Silverman, explains correlation as, “A statistical technique used to determine the relationship between two variables” (Thomas, Nelson, and Silverman, 2011). Correlation is basically looking at two things and sees how they relate to one another. The book explains measuring cardiovascular through a distance run and a step test.  Correlation is also used to traits and behavior, personality, characteristics, and many other variables. Thomas explains, “Correlation may involve two variables, such as the relationship between height and weight. It may involve three or more variables, such as when a researcher investigates he relationship between a criterion (dependent variables) such as body weight, percentage of fat, speed, muscular endurance, and so on. This technique is called multiple correlation” (Thomas, Nelson, and Silverman, 2011).  The book also explains, “You can use correlation to identify relationships between variables, but you cannot use them to establish causation. The only way causation can be shown is with an experimental study in which an independent variable can be manipulated to bring about an effect” (Thomas, Nelson, and Silverman, 2011).

Below you will find the definition related to correlation from the book, “Research Methods in Physical Activity”:

  • Correlation: 
    • A statistical technique used to determine the relationship between two or more variables.
  • Positive Correlation: 
    • A relationship between two variables in which  a small value for one variable is associated with a small value for another variable, and a large value.
  • Negative Correlation: 
    • A relationship between two variables in which a small value for the first    variable is associated with a large value for the second variable.
  • Pearson r: 
    •  Pearson product moment coefficient of correlation: the most commonly used mouthed of computing correlation between two variables.
  • Significance: 
    • The reliability of or confidence in the likelihood of a statistic occurring again if the study were repeated. 
  • Multiple Regression: 
    • Model used for predicting a criterion from two or more independent, or predictor, variables. 
  • Factor Analysis:  
    • A statistical technique used to reduce a set of data by grouping similar variables into basic components. 
(Thomas, Nelson, and Silverman, 2011).
      Through correlation, the article, "The Relationship Between Hazing and Team Cohesion" found that hazing does not result in better team cohesion.  In fact, here is what they discovered, "Results indicated that the more appropriate team building behaviors that athletes were involved in, the more socially cohesive they perceived their team to be. The more hazing activities they reported doing or seeing, the less cohesive they perceived their team to be in sport-related tasks. The results of this study suggest that the argument that hazing builds team cohesion is flawed. Hazing is associated with less, not more, team cohesion" (Van Raalte, Cornelius, Linder, Brewer 2007).  

During their research, they used a questionnaire to ask 167 athletes (66 female and 98 male, 3 did not report gender) question about their hazing experiences and how they perceived them.  The respondents represented six colleges and universities in the United States, “participants were members of basketball (26% n=44), gymnastics (26%, n=43), track and field (22%, n-36), ice hockey (10%, n=17), and swimming and diving (16%, n=27). Thirty-five percent of the sample (n=59 were freshman, 31% (n=52) were sophomore, 23% (n=39) were juniors, 7% (n=12) were seniors, and 2% (n=3) were 5th year or graduate students (3 did not indicate their year in school)” (Van Raalte, Cornelius, Linder, Brewer 2007). The article goes on to express demographics of nationality as well. I liked that they had a variety of represented sports. However, I feel that they did not get many sports that are known for hazing such as soccer and football. This makes me wonder about their research and if they would have received different results had they included more sports.
            Most of the participants had not been a part of a fraternity or sorority. Nearly three fourths of the participants lived on campus while the other twenty-four percent lived off campus.  They used a Group Environment Questionnaire to gather their data on team cohesion. They used a 9-point scale for responses and they confirm adequate reliability and validity for the GEQ. The article states, “Coefficient alphas for the four subscales ranged from .57 to .70 for this sample” (Van Raalte, Cornelius, Linder, Brewer 2007).
            A questionnaire or team initiation was used to gather information as well. Activities were described and participants would explain how they perceived them. Some activities included acceptable behaviors while others were questionable, alcohol related, and unacceptable behaviors. “Respondents were presented with 24 activities and for each activity were asked whether they Did it or saw it, Heard about it, or suspected it, or Not done, seen, or heard about it. For those activities that they had done or head about, students were asked to indicate whether the activity was A tradition or requirement, Appropriate, Inappropriate, and Done when drinking alcohol.  Students were instructed to check all of these options that applied to the behavior” (Van Raalte, Cornelius, Linder, Brewer 2007). I would assume this would have give the research team some very valuable information on why someone was participating in hazing and what type of hazing is most common.  Another aspect to look at in this research is what participants perceived hazing to be.  This is an ongoing issue. Many think hazing is only defined as hazing when there has been an injury from the actions. If no harm is done, is it still hazing?
            Another questionnaire that was given was the Social Desirability Questionnaire. These were measured using the Marlowe-Crowne Social Desirability Scale. They explain that this information was both valid and reliable.
            The last questionnaire to be discussed is the Demographic Questionnaire.  Demographic questions included those of age, gender, year in school, race/ethnicity, location, participation in a fraternity or sorority, and sports participated in. (Van Raalte, Cornelius, Linder, Brewer 2007). Demographics are a necessity when trying to gain valuable information.  It also helps you see the age range you were able to get in your sample. I could see for this research if they only questioned freshman they may not get results on hazing from people who were administering the hazing. This would have skewed results. The procedure, “Athletes who gave their informed consent to participate in this research were given a packet of questionnaires in a counterbalanced order. All participants completed the demographic questionnaire, GEQ, TIQ, and social desirability questionnaires” (Van Raalte, Cornelius, Linder, Brewer 2007).

            The results discuss the correlation between many different aspects in their research. They also categorized many of their findings, such as did the respondents find the activity to be appropriate or inappropriate?  “Eleven activities were categorized as hazing and the other 13 activities were categorized as appropriate team building behaviors. Hazing consists of being the passive victim of physical and psychological abuse, being coerced into self-abuse, or being coerced into abuse other. Acceptable team building included required skill development or assessment, being coerced to engage in deviant behavior, required team socialization activities, and required positive behaviors. Coerced deviant behaviors under acceptable team building included tattooing, piercing, head shaving, or branding, and engaging in or simulating sex acts. Theses activities may appear unacceptable to many segments of our society, and they were intended to be perceived as questionable or unacceptable in the original Hoover (1999) study, but they were rated as acceptable by the participation in this study” (Van Raalte, Cornelius, Linder, Brewer 2007).

          The results showed, “The hazing index and the appropriate Team Building Activity Index were correlated with the four subscales of the GEQ. Three significant correlations were found. The Appropriate Team Building Activity Index was positively correlated with the ATGS subscale of GEQ, indicting the more appropriate activities the participants did or saw, the more positive feelings they had toward the group. The Hazing Index was significantly negatively correlated with the ATGT subscale and the GIT subscales of the GEQ indicating that the more hazing activities the participants did or saw, the less they were attracted to the group’s task and the less bonding and closeness they felt about the group’s task. Separate correlations were run for males and females, and there were no significant gender differences in the magnitude of any of the correlations (all ps> .10)” (Van Raalte, Cornelius, Linder, Brewer 2007).

“The pattern of correlations shown in Table 3 suggests that hazing was negatively related to task cohesiveness, whereas appropriate team building was positively related to social cohesiveness. To explore this possibility two composite indices were constructed. A Task Cohesiveness Index was constructed by summing ATGT and GIT scores. A Social Cohesiveness Index was constructed by summing ATGS and GIS. The conceptual and empirical justification for combining subscales of the GEQ in this manner may be found in the article describing the development of the instrument. The Hazing Index was negatively related ot the Task Cohesiveness Index (r=-.22, p<.005), but not to the Social Cohesiveness Index (r=.07, p=.40). The Appropriate team Building Activity Index was positively related to the Social Cohesiveness Index (r=-.06, p=.48)” (Van Raalte, Cornelius, Linder, Brewer 2007).

            My hope is that through the beginning of this post where I explained correlation with the help of the Thomas text and the article I just discussed that your knowledge of correlation has been enhanced!


Thomas, J., Nelson, J., Silverman, S. (2011). Research Methods in Physical Education (6th Edition). Champaign, IL: Human Kinetics/Thomson-Shore, Inc.

Van Raalte, J. L., Cornelius, A. E., Linder, D. E., & Brewer, B. W. (2007). The Relationship Between Hazing and Team Cohesion. Journal Of Sport Behavior, 30(4), 491-507.

No comments:

Post a Comment