The researchers obtained informed consent from the participants.  This is important since they will be asked to carry out some moderate intensity aerobic exercise.  They would also need to be told that they have the right to withdraw at any time.  If they felt that they were overly exhausted, unwell or were simply bored, they have the right to pull out of the study.  The researchers also attempted to control for undue stress or harm.  The participants were asked to fill in a health survey to make sure that they were fit enough for the experiment.  In addition, the participants heart rates were monitored throughout the experiment. If anything was problematic, the researchers should stop the study. All data that is obtained must remain anonymous.  This can be done by coding the participants so that names are not used in the study. Although there is no outright deception in the study, it is not clear whether the participants understood the hypothesis that was being tested.  This could be done to avoid expectancy effect, where the participants try their best to satisfy the goals of the research. Finally, the participants would have to be debriefed as to the true aim of the study as well as the findings.  They would also have the right to withdraw their data at that point in the study.

The researchers could avoid bias in several ways. The most basic way is to have peers review the study before publication. If there are any biases that may influence the findings of the study, peers should be able to point it out. As the study is quantitative in nature, the conclusions are drawn based on statistical analysis. In this case, the statistics show that one condition is significant compared to the control - but the other is not.  It is important that the researchers do not attempt to present their conclusions as supporting the hypothesis, and that they actually discuss that the hypothesis is only partially supported.  As part of reflexivity, the researchers may want to reflect on why this may be the case, but it is important that this is discussed within the report.

The researchers may also want to use a blind control.  For example, they would have the experiment run by assistants who are not clear of the goals of the experiment.  As they are evaluating how exactly the participants recall the text, it is important that they do not do this too subjectively.  A blind control would mean that the researchers would assess the level of recall without knowing which group/condition the data is coming from.  In the actual study, this was not done.  Instead, they only counted recall if it is was 100% verbatim.  Any variation meant that it was thrown out.  Although this lowers the level of bias, it may also mean that the results are skewed. This could also be avoided by having more than one researcher evaluate the level of recall in order to establish inter-rater reliability and to lower the chance that subjective biases influenced the evaluation.

This study also suffers from a sampling bias. Students are used - and this may bias the findings as they are used to memorizing.  In addition, it was a volunteer sample, which means that only those that would be interested in exercising would take part in the study.  A random sample from the population would lower the level of sampling bias.