Experiments are essential for scientific research. After getting the original experimental data, we need to analyze it to judge whether the obtained results are statistically significant and support the research ideas, and then interpret the data results and finally draw a conclusion.

Whether it is a journal paper or a dissertation, the purpose/purpose and hypothesis of this study are put forward in the introduction or introduction. After a series of experiments are completed, data analysis should be carried out in the materials and methods section.

Through data analysis, it is found that the conclusions are relevant, thus verifying your research hypothesis and realizing your research purpose.

However, it is also possible that the correlation between the experimental results is not significant, and the results obtained are inconsistent with the research assumptions, or even the opposite. Your first reaction may be to ignore these data and even think about eliminating them. This is the wrong thing to do.

A scientific researcher should have scientific research quality, respect science and be rigorous in his studies. In fact, the correlation is not significant, which is the scientific conclusion of your experiment, but it does not support your research ideas. Your experimental results prove that your hypothesis is not valid, thus denying this hypothesis, which is a conclusion in itself.

In general, if the correlation of experimental results is not significant, the author should also analyze the reasons, such as the sample size is not large enough, variables are not easy to control, human factors and so on.

Let's take an SCI article as an example to see how to handle "imperfect" data.

? We have achieved mixed success in our goal. ? We believe that our results show that trust is best described as a concept with two different dimensions. ? On the contrary, we found the overall trust dimension that best describes the characteristics of data. ? There are at least two plausible reasons for this difference, each of which provides a rich field for further research. ? To some extent, some inconsistencies may exist because of cross-cultural differences. ? In addition, due to differences in methods, there may be some differences in the results.

The first sentence directly points out that some results are different from the hypothesis, and the second and third sentences respectively explain the original hypothesis and the actual experimental results. There are two reasons for writing the fourth sentence, and the fifth and sixth sentences are analyzed in detail.