Some of you apparently are a little worried because I said your themes aren't in "Shako style". But what is "Shako style"?
Remember that "社会工学" literally translates as "social engineering". The social part is easy enough to explain: study and research in Shako involves the relationships and interactions among people in groups. Sometimes these relationships are very concrete, as in organizational behavior, like "job loyalty" and "job satisfaction". Sometimes they are abstract, as in supply and demand in economics.
But what is engineering? Engineering is the application of scientific knowledge to the design of products and processes. That is, knowledge of cause and effect relationships. Scientific knowledge is not just facts and measurements. The more important aspects are in the cause and effect. There are two other important things to know about science: scientific knowledge is repeatable (if you act according to scientific knowledge twice, you get the same results) and trainable (often to computers as software!) At the other end of the spectrum from science is expert knowledge, which is not teachable and often not repeatable (i.e., very context-dependent so that often the expert never does exactly the same thing twice). Engineering is in the middle: guided by scientific knowledge, but adjusted by expertise.
Then social engineering is the design of social structures and processes based on scientific knowledge. It includes use of quantitative techniques in finance and supply chain management, experiments and statistics in labor relations and marketing.
What is "Shako style" in research? Research is the systematic creation of new scientific knowledge. Although deciding what to study and how to conduct your research is a matter of expertise (thus something you "learn by doing", with the help of an advisor), the techniques you use are scientific (quantitative and disciplined) and the knowledge you produce must be repeatable and teachable. It's not enough to discover and collect facts. You need to generate understanding ("cause and effect") and the methods you use should give the audience confidence that they would get the same results if they used the same methods.