Some individuals indicated that they had not complied with the dietary advice and were omitted from the final analysis but it was impossible to determine how accurate the exact dietary intake of each subject or even whether they had consumed high- or low-cholesterol foods.
Certainly the precise consumption of cholesterol was not monitored and could have varied from week to week.
Whilst your results may not be enough to reject the null hypothesis, they may show a trend that later researchers may wish to explore, perhaps by refining the experiment.
For this purpose, you should criticize the experiment, and be honest about whether your design was good enough.
Similarly, over the same period, consumption of a low cholesterol diet did not significantly reduce plasma cholesterol concentration.
This result should be interpreted with caution because each of the 3 subjects used in the analysis for the high-cholesterol group finished the study with an identical plasma cholesterol concentration of 8 m M (probably coincidentally).
In this study, the sample size was small (n=3-4 for each group) and was heterogeneous with respect to age, weight, starting cholesterol concentration and other lifestyle factors.
In addition, each subject only consumed either the high- or low-cholesterol diet The greatest problem with this study, however, was compliance.
Use the questions in the Content section and the stages mentioned here to help you to develop your argument in a logical way.
Remember that not all of the questions or the stages will be relevant to every experiment.