Here is a really good example of a scholary research critique written by a student in EDRS 6301. The student who submitted this paper last semester earned a 100 on his critique.  The content of the paper is right on track.  A succint summary is provided in the first paragraph.  This paper would have been even better if the student had added a sentence or two about the results of the study.  That way, after reading the first paragraph, the reader would know the purpose, hypotheses, and findings.

Next, the student described specifics about the research design, including the sample, instrumentation, and data analysis.  Ecological and population generalizability were discussed.  The student spoke at length on threats to internal validity.  Following the information on threats to internal validity, the student provided suggestions regarding how these threats could have been dealt with.  This shows a high level of understanding.  Not only does the student know what the weakness of the study are, he provides ways the study could have been improved.

One thing that was not discussed in this paper is the literature review.  In previous classes we spent more time talking about statistics than the literature review.  That's why you'll see some fairly complex explanations in this paper on the data analysis but no information on the literature review.

Your paper will contain information on the literature review and less specific information on statistics.
 
 
 

Chris K.
Research Critique 1

       Jamber, E. A., & Zhang, J .J. (1997).  Investigating leadership, gender, and coaching level using the Revised Leadership for Sport Scale. Journal of Sport Behavior, 20, 313-322.
 

       The purpose of the study was to determine possible differences in leadership behaviors,

using the Revised Leadership for Sport Scale (RLSS), between male and female coaches 

and among different coaching levels.  The researchers submitted two hypotheses.  The first 

hypothesis was that male and female coaches would respond differently to the RLSS in 

overall leadership behaviors.  The second hypothesis was that differences on the RLSS 

would occur among coaching levels: junior high, high school, and college.

     The sample was nonrandom, including 162 coaches that were chosen on a volunteer 

basis.  Within the sample, 118 (0.73) of the coaches were male, while 44 (0.27) were 

female.  With regard to coaching level, 25 (0.15) were junior high coaches, 99 (0.61) high 

school, and 38 (0.24) at the college level.   While this is a good sample size, the problem lies 

with the distribution of the sample.  The sample number for junior high coaches, in particular, 

is rather low.  A larger sample with regard to all categories would have aided in the data 

analysis, particularly when looking for possible interactions between gender and coaching 

level. 



     The instrument utilized was the Revised Leadership for Sport Scale (RLSS) developed 

by Zhang, Jensen, and Mann in 1996.  This scale is used to measure six leadership 

behaviors:  training and instruction, democratic, autocratic, social support, positive feedback, 

and situational consideration.  The scale uses 60 statements, which were preceded by “In 

coaching, I:” A Likert scale was then given for each statement: 1 = never; 2 = seldom; 3 = 

occasionally; 4 = often; and 5 = always.  This produced an ordinal level data set.  Scales 

were administered in a number of environmental settings: classrooms, gymnasiums, practice 

fields, and offices.  The internal consistency for each section was calculated: 0.84 for training 

and instruction; 0.66 for democratic; 0.70 for autocratic; 0.52 for social support; 0.78 for 

positive feedback; and 0.69 for situational consideration.  There was no information, 

however, regarding the validity of the RLSS. 

     A MANOVA was used to analyze the data for differences between male and female 

coaches with regard to leadership behaviors.  This is not consistent with the type of data 

collected.  The RLSS used a Likert scale (ordinal), yet a MANOVA would be most 

applicable for normally distributed, quantitative data.  The analysis showed there were no 

significant differences between male and female coaches in overall leadership behaviors. 

When the six leadership styles were examined separately, there was a significant difference 

in social support between males and females.  In general, females scored much higher than 

did the male coaches.



       A MANOVA was also used to examine the data for differences between the three 

levels of coaching (junior high, high school, and college) with regard to leadership behavior 

in general.  There were significant differences between the three levels.  When breaking 

down the six behaviors and examining them individually, an ANOVA was used to analyze 

the data.  Again, because the data for the RLSS is ordinal, an ANOVA is not the best 

analysis tool.   The three coaching levels scored differently on three of the six behaviors: 

democratic behaviors, training and instruction, and social support. High school coaches 

scored much higher than college level coaches in democratic behavior.  Junior high coaches 

were significantly lower in training and instruction than either high school or college coaches. 

Junior high coaches also demonstrated a lesser degree of social support than either the high 

school or college coaches.

     A MANOVA was again used to analyze the data for any interaction between gender and 

coaching level with regard to overall leadership behavior.  Once again, a better analysis 

method could have been chosen based on the nature of the data collected. The results 

indicated no significant interactions. 

      The ecological generaliziability for the study is fairly high.  The surveys were mailed out, 

and returned on a volunteer basis.  However,  due to the nonrandom nature of the sample, 

the results would not generalizable beyond the 162 participants in the study. There was no 

effect size is listed for the study. 

      In order to reduce threats to internal validity, the participants were asked to respond 

honestly and confidentiality was stressed so that the “coaches might feel more at ease in 

responding.”   No other efforts were indicated. 



      The researchers mention that the scales were given in a variety of settings.  This could 

present a threat to the internal validity in that participants might not have been entirely 

focused on completing the scale, but instead on coordinating practice, completing 

paperwork, etc.  There are a number of other factors that could effect the internal validity of 

the study, yet were not addressed by the researchers. Coaching experience would greatly 

effect the responses of the participants, yet this was not considered in the study.  The gender 

of the athletes may be a contributing factor to the coaches’ responses.  It is not unreasonable 

to suppose that coaches of female athletes, particularly at the junior high and high school 

levels, will demonstrate more social support than those of male athletes. The nature of the 

sport could also be critical.  Certain coaching styles are more applicable for individual sports 

(wrestling, track, and tennis) than for team sports (football, soccer, and basketball).  The 

socioeconomics and population of the school itself could play a factor.  Certain schools have 

better athletes and programs in a particular sport, while others may not be able to field a 

winning team.  In addition, at the high school level, coaches are occasionally asked/forced to 

work with a program they have no knowledge of or desire to coach due to staffing 

shortages.  This could dramatically influence a coach’s response to the scale questions. The 

history of the program as well as the individual coach’s personal coaching history could 

greatly influence responses.  If the program has had several losing seasons in a row, perhaps 

the attitude of the coach could be different than that of a coach who has recently won a state 

title. 



     An additional set of questions regarding the personal history of the coach in question 

could have helped reduce many of these threats.  With additional information, the 

researchers may have been able to use a modified matching system when analyzing the 

results.  By increasing the number of independent variables to include things such as 

coaching experience and gender of the athletes, the researchers could have reduced some of 

the potential threats to internal validity.  In addition, bringing coaches together to a common 

setting could have reduced location threat.  Coaches meet seasonally for clinics.  Perhaps 

obtaining permission to administer the survey during these meetings would have been 

possible. It would have also been possible to actually go to individual schools and meet with 

the coaches as a group to administer surveys.  This method would have given a good 

cross-section of gender and coaching experience for a variety of sports. 

     While the study has merit, the methods need to be re-evaluated.  The power of the study 

needs to be increased by obtaining a larger sample size.  The numerous potential threats to 

internal validity need to be addressed and minimized where possible.  It would also be 

helpful to be given data regarding the validity of the RLSS.  Without these, it is impossible to 

evaluate the potential meaningfulness of this study.