RD Week 4 ANOVA Study

RDWeek 4 ANOVA Study

RDWeek 4 ANOVA Study

Researchquestion: “Does the type of exercise affect the weight loss of anindividual”. This research involves an experiment where people aresubjected to three types of exercises and their weight loss recordedafter one month then the comparison is done to determine if there isa difference in the average weight loss between the three types ofexercises. The statistical test that applies to this researchquestion is the one-way ANOVA given that the study seeks to comparethe average weight loss of participants between the three exercises(Heiberger &amp Neuwirth, 2009).

Thenull hypothesis (H0): there is no difference in the weight lossbetween the three types of exercise and the alternative hypothesis(H1): there is a difference in the weight loss between the threetypes of exercises. The errors that can occur are the type 1 errorand the type 2 error. The type 1 error will occur if we conclude thatthere is a difference in the weight loss between the three types ofexercise yet the evidence we have indicates that we should haveconcluded that there is no significant difference in weight lossbetween the three groups. Type 2 error will occur if the researchconcludes that there is no significant difference in the weight lossbetween the three groups, yet the evidence that is available,indicate that the research should have concluded that there is adifference in the weight loss between the three types of exercises.


Thestudy will have 10 participants per exercise that will make 30participants in the study. The participants will be 15 males and 15females, and they will be from different age groups so that we can beable to make good conclusions. The participants will be selectedthrough simple random sampling among overweight people where theconfidence interval, significance level and probability of beingincluded in the sample and probability of not being included in thesample will be taken into account. That is to come up with the samplesize. The 30 participants were recruited and then randomly split intothree groups of 10 people each. Each group was assigned to a specifictype of exercise.


Thevariables in the study will be weight loss and the exercise type. Theweight loss variable will be continuous, quantitative with aninterval scale of measurement. This is the dependent variable in thestudy. The exercise type variable is discrete, categorical with anominal scale of measurement. This variable has three categories,which are exercise A, Exercise B and Exercise C. The exercise typevariable is the independent variable in this study.

Weightloss is the amount of weight that an individual loses over time. Inthe study, the weight loss variable is measured by subtracting thefinal weight of a participant at the end of the one month of takingpart in a particular exercise from the initial weight the participanthad before enrolling in a particular exercise. The exercise typevariable is the kind of exercise that a participant undertakes withan aim of losing weight. In this case, there are three types ofexercise, which are exercise A, Exercise B, and Exercise C.


Inthis study, one-way ANOVA is the statistical test that will becarried out. This test is relevant for this research given that itinvolves comparing of the averages for more than two groups withregard to a continuous quantitative dependent variable (Heiberger &ampNeuwirth, 2009). In this research, there is the continuousquantitative dependent variable, which is the weight loss and thecategorical dependent variable, which is exercise type, which hasthree categories. Upon the determination that there is a significantdifference between the exercise types’ weight loss means then thatthe research does a post hoc analysis, which simply helps the studyto identify the combinations of exercise types that could be causingthe significant difference in weight loss between the exercise types.Since there is an assumption of equal variance, the study will employthe tukey post hoc test.

Fromthe ANOVA table, there is the p-value that is obtained, which helpsin reaching a conclusion. From this, it is deduced that if thep-value is higher than the level of significance, then the studyfails to reject the null hypothesis. On the other hand, if it is lessthan the level of significance the study then rejects the nullhypothesis. In the case of post hoc we look at the p-value of eachgroup with the other and for those combinations that the p-value isless than the level of significance (0.05), a conclusion is reachedthat it is statistically significantly different (Brown, 2005).


Theassumptions taken up by this study takes into account that thesamples are of equal variance and have homogeneity of means. There isan assumption that the weight loss is usually distributed.Additionally, there is bias during data collection or the choosing ofthe participants. From the study, it can be concluded that the typeof exercise affects the weight that the participant loses but wecannot conclude that only exercise type leads to the difference inthe weight loss. The results can be used to recommend the mosteffective type of exercise to the people who want to lose weight.


Brown,A. M. (2005). A new software for carrying out one-way ANOVA post hoctests.&nbspComputermethods and programs in biomedicine,&nbsp79(1),89-95.

Heiberger,R. M., &amp Neuwirth, E. (2009). One-way anova. In&nbspRThrough Excel(pp.165-191). Springer New York.