I just read comments from 2 reviewers of a journal. The first reviewer gives a minor correction because of English editing, copy editing, and elaboration in the discussion section. The first reviewer gives good comments too, for instance, we have to have another robustness test, which is a very good idea to make our paper better. The second reviewer gives a rejection because two issues. First, he/she suspect we just wrote down the R-Square (R2) before running the factor analysis. I just stunned, and read it again and again. R2 from factor analysis? Second issue is about the data collection. He/she addresses our manuscript is too weak as it does not mention the philosophical data collection. I was confused that time thinking what the reviewer means by philosophical data collection. After reading it again, reviewer mentions that I have to write down, not only whether we adopt or adapt the items of questionnaire, but also the grounded theory of the data collection and the phenomenological of items construction. I was very surprised because actually grounded theory data collection and phenomenological are qualitative approach which is totally different view from our approach which is quantitative method. I just smile facing this type of reviewing process. But it makes me think what the root of this “evil” is. Why nowadays people are good in click-click the software interface but fail in understanding comprehensively the statistic.
The second reviewer misconception in methodology is not a new issue for me. I face some students who very insist that certain software does not need a normality test. Yet, when he/she got all the regressors contribute significantly to the dependent variable (DV), and I said maybe there is type 1 error, he is still with his stand, until I grabbed methodological book.
I have faced also students who have negative R2. Usually they are very happy with a relatively big R2, even though it is negative. When I told them the interpretation of negative R2, some yelled me back arguing that I just wanted to make them down. They never believe that negative R2 means your regressor has failed to explain your DV.
I have faced also students who insist that few samples were good enough for a research because the software that they used does not need huge samples. I have to explain the purpose of number of sample in quantitative, yet they did not believe me. Until they were very happy because of having superb R Square: 0.9! I told them that it is actually an indication that you do not have enough sample to generalize a conclusion in quantitative analysis. They said, they already asked their professors, and I was wrong; even though I showed already Gujarati and Porter (2012).
Sometimes they scold me back with high temper. Sometimes they listen to me (after I show the evidence). Sometimes they talk behind me saying that actually I envy them because they will graduate (yeah rite, I already graduated and I do not care with your life). But it is okay. To me, that’s a journey of life.
I know nowadays statistical software becomes very easy to use. You just click this and click that. The funny thing is that some claim their software much better than others software. But software is just software. I can drive a car, but it does not mean that I know how to repair the car if it is broken. If the car suddenly shuts down in the middle of nowhere and I do not know about mechanical, it is indeed I cannot drive the car again. Most probably I get panic, confuse, and have to call mechanics to fix it. Same goes to statistical software, you know how to click this and that, the professor even gives you a journal article to interpret the result, but it is not necessary you know about statistic. My suggestion is that you have to attend methodological class or statistical class first before starting click this and click that. Otherwise, you will keep losing in the journey of your research.