Discussion
The previous conclusions did not support the hypothesis. The following factors could have effected the results:
Problems in subject selection
Subjects were not randomly selected. Weight training subjects were those who already were weight training at the school gym. These subjects were chosen to minimize any additional risk of injury. Subjects throwing were also selected. They were asked if they were able to meet the criteria (practice twice a week for 15-20 min. at each practice). Few subjects responded to requests to participate.
Variability
Each of the subjects was at a different experience level. Some of which had no experience at all. Ideally they would have all been quarterbacks or at least all football players. This may have reduced the MSWG so that the trends may have been more significant.
Sample size
Too few subjects were involved. This reduced the power of the Two-Way ANOVA test.
Repetitions
The number of throws at each distance may also have been too small to observe a difference.
Throwing
Training may not have been long or frequent enough to show a significant difference.
Weight training
Weight training may not have been long or frequent enough to show a significant difference.
Training interruptions
The March break interfered with weight training and throwing practices between the 3 and 6 week intervals for some subjects because the school was not available during that time. This may have affected their performance at 6 weeks.
Normal distribution
The “distance from the centre” variable may not be normally distributed. There were not enough subjects to take averages to normalize the data.