What are the features of a good, reliable study/ investigation?

To produce a good reliable study, you would first need a sufficient number of participants to represent the actual population. Therefore a large sample size (number of participants in the study) would produce reliable results. This is usually worked out through statistics and is dependent on the estimated effect size of the variable you are looking at. There should be a control group to compare the effects of your particular intervention so your control group and the group taking the intervention should be equally matched. A suitable measure of the outcome should be used, this should be objective. Finally, use a statistical test such as a t test or compare the means of the two different groups.