Case Study of Par
Case Study of Par
Case Study of Par
Par, Inc., is a manufacturer of golf equipment. The company wants to check whether this new
type of golf ball is comparable to the old ones. Therefore they did a test with 40 samples of both
new balls and old ones. The testing was performed with a mechanical hitting machine so that
any difference between the mean distances for the two models could be measure to evaluate
the difference in the two models, presented with the sample sie of 40 and confidence level of
!"#. The table below depicts a descriptive statistical summary for the case study.
$ %urrent &odel$ 'ew &odel$
(ample &ean $ )*+.+!$ ),).!+$
(tandard -rror$ ../0$ ..""$
(tandard 0eviation $ +.)0$ !.+.$
(tandard variances$ *,.)+$ !*.)+$
(ample (ie$ 40$ 40$
%onfidence 1evel 2!"#3$ ).*)$ /..4$
The hypothesis for Par, Inc. to compare the driving distance of the current and new golf
balls is 40 5 6. 7 6) 8 0 2they are the same3
4a 5 6. 7 6) 9 0 2the are not the same3.
P7:alue for this two tailed test is 0.0*, which is greater than level of significance ; 20.0"3.
4ence, 40 will not be re<ected which shows that Par, Inc. should take a new ball in
production as the P value indicates that there is no significant difference between
estimated population mean of current as well new sample model. The !"# confidence
interval for the population mean of the current model is )**.)* to ),*... and of the new
model is )*!.+4 to ),*.... It means that the estimated population mean for Par, Inc.
should lie within this range for consistent result. 4owever, the !"# confidence interval
for the difference between the means of the two populations is 7+../ to 70.0". The
calculated test statistic value is far from the re<ection area. The larger the sample sie
the smaller the standard deviations which means point estimator of mean will become
more precise. 4ence there is no need to take larger sample sie.
%onclusion57