Preferences of employers have critical roles to determine job-matching processes between the university graduates and the employers. Some researches show that the employers expect self-directedness to the graduates, but these results are only based on the sample of very large companies. This study investigates what types of ability employers demand from graduates, with a special focus on the differences among sizes of companies. Our data are collected from job-search website for new graduates in 2016, and consist of 20,859 companies including large and small companies. We analyze the descriptions obtained from the companies using topic modeling. The results show that large companies and small ones expect different abilities to the graduates. Large companies tend to expect self-directedness and entrepreneurial spirit to graduates. This is consistent with previous studies. By contrast, small companies tend to demand discipline and aggressiveness. These findings suggest that previous studies have overlooked the trends of small companies.

The purpose of this paper is to examine how law as social phenomenon can be described in mathematical models, especially in comparison with social norms. As a beginning we try to untangle the concepts of “law” used in the literature on social norms, and classify them into a few categories. We then point out the problems with assumptions made when analyzing the law in the usual mathematical models, including economic models. Particularly, we argue that public institutions are not always necessary to establish legal order and that law cannot affect human behavior without coordinating many more actors than usual models assume. In the latter part of the paper, we offer a simple model that describes the enforcement of social norms through collective and decentralized sanctions. On the basis of this model, we discuss the problem of how to capture the distinction between social norms and law, and conclude that the law can have an impact on players’ behavior only by changing people’s beliefs concerning other people’s action.

In this paper we formally discuss the mechanism of solidarity from the viewpoint of tolerance and norm. Defining solidarity as <we-relation> via net-base symbols, we formalize believes that specify what symbols are required for <we-relation> in terms of Boolean equations. Tolerant solidarity is confronted with contradictory relationship between universalism and relativism. In this case “less conflictual solutions” (LCS: Murakami,1994) will be an excellent guide for tolerance when we define it as restriction of norm activation. In addition, when we consider meta-believes that represent ideas like liberal nationalism (liberal universalism), LCS works as the standard of tolerance in the acceptance process of meta-believes. Through the above-mentioned analysis, we will discuss two issues. At first, believes show different level of tolerance from the viewpoint of LCS, and the most tolerant belief is different according to positioning of actors. Then, belief-switching should have rich implication to realize tolerant solidarity. Secondly, LCS does not guarantee that liberal universalism is the most tolerant meta-belief; however, it might best fit with tolerant solidarity because it holds moderate tolerance and a consistent ideal goal at the same time. Thus, we will suggest relation-epistemological micro-macro link in order to explain the mechanism of solidarity and norms.

Self-sanction in the repeated norm enforcing game is analyzed in order to analyze norm internalization of rational actors. Self-sanction is costly punishment which is carried out to self and includes psychological (e.g. conscience and regret), economic (resigning a job), and physical (hurting self) processes. The norm with self-punishment strategy equilibrium is characterized, and both self-sanction and external-sanction which is carried out by other players are considered. The equilibrium is sustained even when the social norm brings inefficient outcome. We obtained following results. Firstly, the equilibrium is sustained in wider setting of parameters when self-sanction is planned. Secondly, social norm in N-prisoners’ dilemma setting can be sustained without external-sanction. Lastly, key mechanism sustaining inefficient norms is not self-sanction but external one.

This paper presents a useful analysis of the evolution of preferences when investigating social norms. First, this research proposes the framework of analysis. Second, this research analyzes gender roles and the gendered division of labor with the evolution of preferences. Results show that improvements in the wage gap between men and women, and the low cost of housework and child bearing, promote a change in men's attitudes toward gender roles. This leads some couples to share costs and work full time. This analysis indicates that the evolution of preferences contributes to social norm research in two ways: (i) the evolution of preferences can treat the decision-making and replicating preferences separately, and (ii) the evolution of preferences can analyze the dynamics of preferences.

Japanese statistics act was revised in 2007 for the first time since 1947. The new statistics act stipulates further use of statistical microdata and confidentiality. Every five years, concreate measures concerning the further use of statistical microdata has been included in the “master plan concerning the development of official statistics”, which is designated by the law. Thus, Japanese Government has been trying to promote the use of statistical microdata. After ten years since the revision of the statistics act, “fundamental policy on statistical reform”, determined by the council on economic and fiscal policy, proposed to establish a statistical reform promotion council. The council was established in January 2017, and summarized a broad direction on statistical reform in May. This paper introduces a situation on microdata access for official statistics in Japan after the revision of the statistics act, how to tackle challenges, discussion on statistical reform, revision of the master plan and future direction.

The aim of this paper is to illustrate the process of accessing micro-datasets of official statistics. In Japan, according to the revision of the Statistics Acts in 2007, micro-datasets of official statistics can be accessed to conduct academic research with public interest. After the usage application is approved, researchers can obtain CDRs, including text files of micro-datasets, data layout tables, and code tables. In addition, they can not only generate and recode various variables but also analyze the interaction among variables using their preferred data management techniques. These analyses are expected to reveal more precise association of patterns among standard variables. Furthermore, any remaining questions can be addressed using theme-specific survey datasets. Thus, this paper suggests that a combination of mixed analytical strategy, micro-datasets of official statistics, and survey datasets can facilitate excellent quantitative research in sociology.