Wednesday, August 21, 2019

Family And Gender Roles Changing Attitudes Sociology Essay

Family And Gender Roles Changing Attitudes Sociology Essay From this point of view, female employment can be seen as a necessary means of family income and support. But the question remains whether and in which way family life and children will be affected by employed womens temporary absence from the household. Gender has been an important principle of stratification throughout Japanese history, but the cultural elaboration of gender differences has varied over time and among different social classes. After World War II, the fixed image of the Japanese woman has been that of the young office lady, who becomes a housewife and a stay-at-home mother after marriage. But a new generation of educated women is emerging, who are seeking a career as a working woman with a family at home. They continue to have nearly total responsibility for home and children and often justify their employment as an extension of their responsibilities for the care of their families (Molony, 2000). But how are the views on that commitment compared to women in a much more liberal country, such as the Netherlands? The subject of gender equity and working women in the Netherlands is often viewed by known two-tier societies (such as Japan, Austria, Italy and Venezuela) as progressive, maybe even too liberated for the tastes of some governments. Japan generally constitutes a case of low gender equity and low female labour market participation while the Netherlands has moderate to high gender equity and high part-time female labour market participation. This bachelors thesis will therefore attempt to identify whether a link exists between asymmetrical gender division of household labour and low gender equity. Gender equity is the process of being fair to women and men. To ensure fairness, strategies and measures must often be available to compensate for womens historical and social disadvantages that prevent women and men from otherwise operating on a level playing field. Equity leads to equality. Japan and the Netherlands will be very interesting countries to compare, because inequality between men and women in the Netherlands is relatively small compared to the other countries. As you can see, the Netherlands has a relatively high score on the Gender Empowerment Measure (GEM):Arbeidsdeelname vrouwen in de EU, 2009 It comes fourth behind Sweden, Finland and Denmark. Despite anti-discrimination laws and a steadily growing number of employed women, Japan is falling behind the rest of the world on gender equality. Widespread discrimination persists, and has only grown more subtle over the past years. According to the United Nations Development Programme, Japan has consistently ranked as the most unequal of the worlds richest countries. Our means for above endeavor will be the Multidimensional Unfolding technique; can this technique be applied successfully to the large dataset of the ISSP 2002 survey programme: Family and Changing Gender Roles III? This large annual cross-national survey includes questions about attitudes towards gender role distribution, the attitudes towards employment of mothers and married women and management of marriage or partnership. I will also try to give a practical explanation of unfolding and the procedures that are used for this. The first phase of the project will involve an analysis of the ISSP data with SPSS PREFSCAL. With this programme we will try to find a common quantitative scale that allows us to visually examine the relationships between our two sets of objects/countries; Japan and the Netherlands. The main issue of this study is to show how attitudes towards marriage, motherhood, and the morality of family behavior differ across nations, both in the Netherlands and in Japan. What is the difference in attitudes about the changing gender roles in two different cultures? Theory 2.1 Attitudes towards gender equity in Japan and the Netherlands In the beginning, woman was the sun. An authentic person. Today, she is the moon. Living through others. Reflecting the brilliance of others (Sievers, 1983). It was Toshiko Kishida (1863-1901) who said: If it is true that men are better than women because they are stronger, then why arent our sumo wrestlers in the government? (Kishida, 2007) This famous Japanese woman used clever phrases such as this to attack the view that men by nature were superior to women. She became a talented and exciting public speaker and the first woman to travel all over Japan, addressing huge crowds. She was imprisoned for her beliefs, but continued to speak out. She said that a civilized country such as Japan should be ashamed to respect men and yet despise women. When after this the Meiji-Taisho era (1868-1926) began, Japanese leaders were open to new ideas; male nationalists argued that improving the status of women was essential if other technologically advanced nationals (such as the Netherlands) were to accept them. This opened the door for a small group of women who called for new rights and freedoms. The phrase good wife, wise mother was coined, meaning that in order to be good citizens, women had to become educated and take part in public affairs (Sievers, 1983). But even after this, the dominant male-breadwinner family model, accompanied by tax and state benefits for families that favor one-earner couples, and a lack of available or affordable institutionalized childcare make it difficult for Japanese women to combine work and parenthood. When they do attempt this, they take on not only the role of paid worker, but also remain the largest contributor to housework and child-caring tasks (Kreyenfeld Hank, 2000). I dont interfere with my husbands business, not with my mouth, hands or legs. (Jordan, 2002). This statement, made by Kumiko Hashimoto, the wife of former Japanese Prime Minister Ryutario Hashimoto, underlines the traditional role of women in Japan. After this we can conclude that traditional gender roles in Japan are characterized by a strong sense of patriarchy in their society, which accounts for the bifurcation of the productive and reproductive spheres, with a distinct separation of gender roles. In the family, this refers to the idea of the man as the primary breadwinner of the family, and the woman as the primary caregiver in the family (Iwao, 1994). But all this is completely different in the Netherlands. After being oppressed by men in the 19th century women started the first feminist wave here around the year 1870. Wilhelmina Drucker and Aletta Jacobs were the two main women of this wave and both meant a lot for the position of women nowadays. At the end of the first feminist wave women obtained the right to vote and were able to attend college and universities, and had the right to work outdoors. In the 1950s and early 1960s it seemed that the emancipation of women in the Netherlands was completed. Formally, women had the same rights and possibilities to work outdoors as men. Though in practice, married women did not work outdoors and the public opinion was that both men and women had both different competences; women were the ones to take care of the children and men had paid jobs outdoors. In 1980 the law for equal treatment for men and women was finally ratified. Nowadays, the amount of part time jobs has increased massively over the last twelve years. Though, women work much more in part time jobs as men. The government wants to stimulate more women to work more hours a week. Furthermore, men should have the chance to work in part time and take care of children for example as well (Van de Loo, 2005). Most women in the Netherlands today continue to work after the birth of their children, and compared with Japan the employment rate of mothers in the Netherlands is high. The majority of women in the Netherlands do however reduce their working hours after the birth of their first child. This fits in with views in the Netherlands on looking after children; the fact that a mother is working is no longer an issue, but a mother having a full-time job still goes too far for most Dutch people. The majority feel that it is best for children to be looked after exclusively by their own parents, and they are very reserved about formal childcare. In addition, a third of women and more than half of men believe that women are better suited than men to looking after small children. Working part-time offers mothers an opportunity both to participate in the labour market and to look after their children largely or entirely themselves (Versantvoort, 2008). Multi-dimensional Unfolding Technique Unfolding is a data analysis technique that was invented in 1950 by Clyde Hamilton Coombs and his students in one dimension and is later extended to multiple dimensions (Busing, 2010). The unfolding model is a geometric model for preference and choice. It locates individuals and alternatives as points in a joint space, and it predicts that an individual will pick the alternative in the choice set closest to its ideal point. It is perhaps the dominant model in both scaling of preferential choice and attitude scaling. By scaling we refer to the process following data collection, by which numbers are assigned to entities such as items or individuals. Coombs proposed a joint scale for preference data: his J scale positions both judges and items on a single continuum such that an item is located closer to the judge the more it is preferred. Unfolding  is the name he coined for the operation of deriving the joint scale from the individual preference rankings (Coombs, 1950). Coombs ideas were later extended by Bennett and Hays (1960) to the multidimensional case: multidimensional unfolding  becomes then the operation of representing both individuals and preference items as points in a low-dimensional space such that the distance orders re ¬Ã¢â‚¬Å¡ect the rankings. This multidimensional unfolding model, which relies on distances and that is also known as ideal point model, is a very attractive one: it gives a geometric representation of nonmetric data in a sparse way, and with a distance model that is easy to grasp. This statistical method is explorative in the sense that it can identify latent dimensions in a given dataset. The model will find coordinates in a low-dimensional joint space, in the particular case at hand both for respondents and statements about gender equity and inequity. The resulting configurations are very easy to interpret and give a quick first insight into the overall structure of the data and its particularities (Van Deun, Marchal, Heiser, Engelen, Van Mechelen, 2007). The multidimensional unfolding technique computes solutions to the equations of unfolding model. It can be defined as multidimensional scaling of off-diagonal matrices. This means the data are dissimilarities between n row objects and m column objects, collected in an n ÃÆ'- m matrix 1. An important example is preference data, where _i j indicates, for instance, how much individual i dislikes object j . In unfolding we have many of the same distinctions as in general multidimensional scaling: there is uni-dimensional and multidimensional unfolding, metric and nonmetric unfolding, and there are many possible choices of loss functions that can be minimized (De Leeuw, 2011). Unfolding also finds an optimal solution by minimizing what is called a stress function. (To be continued) 2.3 Data This research is a secondary analysis based on the study monitoring survey conducted by ZUMA for the ISSP on the 2002 Family and Changing Gender Roles module. Thirty-four member countries archived the 2002 Family and Changing Gender Roles module; all of them have completed the monitoring questionnaire. The ISSP 2002 module contains 362 variables and 60000 cases. The cases are a multi-stage stratified sample of the (adult) population of thirty-four countries worldwide. The data was essentially collected through face to face interviews, self-completion (with interviewer involvement) or, in some cases, telephone interviews. But these telephone interviews were later on not permitted in the ISSP, so they are not included in the data. Nine countries had advance letters, including Japan; while the Netherlands had a telephone pre-contact. The Netherlands also made use of an incentive (a gift token of 15 euros), while Japan did not use any incentive. With the exception of Japan, countries using interviewer-administered modes back-checked interviews (proportions ranging between 3%-95%). Japan and the Netherlands both had an age cut-off point at 16 years. The Study Monitoring Questionnaire (SMQ) has been modified from year to year. Questions on fieldwork, translation, and sampling have, for example changed and questions on documentation been added. Some countries used old versions of the SMQ, which they had kept. This means that some information for these countries is missing in the report. While the statements in the ISSP questionnaire are about different aspects of family and work life of women, they generally imply that an unequal burden of homemaking activities should lie with the female and/or that a females activities in the labour market are of secondary importance to her role as a homemaker and her husbands role in the labour market. Method The multidimensional unfolding technique usually runs with small datasets with a maximum of 100 subjects. Not much research is done yet on how well PREFSCAL handles larger datasets like the ISSP set with 2341 subjects (only Japan and the Netherlands) Im about to use, so this will be an interesting and instructive endeavor. To investigate whether or not unfolding can be used to make sense of the relations between the different gender-related items (i.e., statements) a first analysis will be done on a restricted sample from the dataset with respondents from the Netherlands and Japan only. Choosing only two countries turned out to be a wise decision, because PREFSCAL cant run smoothly with all thirty-four countries in one data-set, it is simply too much data for SPSS to handle; you will get an error about having insufficient memory to complete the procedure. After that the execution of the command will come to a halt. The next step was making a choice selection from the many variables; I chose seven variables which can best represent the construct of gender equality, these variables are given in table 1. The variables are comparable because they all measure subjects attitudes towards gender equality. V4 A mens job is work, a womans job is the household. V6 When a woman works, the family life suffers. V7 What women really want is a home and kids. V8 Work is the best thing for a womans independence. V9 A working woman should get paid maternity leave V11 Working in the household satisfies just as much as a paid job. V27 If a mom works she can still have a warm relationship with her children. The above questions were presented to each of the subjects, who were asked to indicate their degree of agreement on a 9-point rating scale. This scale goes from Strongly agree to Strongly disagree with an added Cant choose and No answer refused. I chose a (wide) variety of questions from my data; some are very positive about women in the workplace and household and some are very negative about women in the workplace and household. Entries in the following models indicate average similarities and dissimilarities between the Japanese and Dutch people across seven different points of view. The term similarity is used to indicate the degree of agreement between two objects, while dissimilarity indicates the degree of disagreement. I filtered the thirty-four countries with Select-Cases and using the option If condition is satisfied. My variable for country is called v3, with Japan having code 24 and the Netherlands having code 11; so I put in v3=24 OR v3=11. When I wanted to write the cases to a whole new data-set I chose the option copy selected cases to a new data-set in the first tab of Select Cases. Having done this, I had a whole new data-set with only the data from Japan and the Netherlands to work with. When I tried to make some models with PREFSCAL I got a missing data error: Row with only missing data found. This is not allowed. This means that at least one person in my data-set hasnt given a valid answer to the variables/questions Im using for this project. That is why I chose to use only persons that have given a valid answer to at least 50 percent of the questions in my analysis. I did this by going to Compute Variable and creating a filter-variable with Include if case satisfies condition: NVALID (v4, v6, v7, v8, v9, v11, v27)>=4. I simply took the amount of variables I had (7), divided them by two and rounded off upwards to four. Four being the minimal amount of valid answers I wanted in my analysis. After this I filtered my Japan-Netherlands data-set on this new variable (Valid Answers) via Select Cases. SPSS will then filter all the cases that have too little valid answers. When I had finally solved all the problems with my data-set I got the following error: Invariant part of the data found, check, depending on conditionality chosen, your data for constant parts. To solve this problem I made the decision to analyze my data with the option matrix-conditional and transforming the input data row conditionally, whereby tied observations were untied for each row separately. This means that the model was allowed to transform like-wise item scores into different values, as long as the overall order of the item scores was not altered. To better distinguish the difference between males and females on other variables, such as education and religion, I used Select Cases again to create two different data-sets, one with 1106 males and one with 1235 females after correcting for invalid answers again. The first two-dimensional unfolding model of Japan and the Netherlands was created. Its Normalized Stress level was a fair 0,1027924, which is an excellent stress-level for an unfolding model. This stress is intended to be a measure of how well the configuration fits the data. Stress is defined as a Standardized Residual Sum of Squares which should always be positive, and the smaller the better. Kruskal himself suggested the following benchmarks for measuring stress: .20 = poor, .10 = fair, .05 = good, .025 = excellent, and .00 = perfect. These benchmarks are based on experience with experimental and synthetic data (Kruskal, 1964). By adding more variables into my model I wanted to get a better grasp of the cultural differences between Japan and the Netherlands on the subject of gender equity, family life and marriage. I did this by adding these five questions to the seven I already had: V10 Both men and women should attribute to the household income. V12 Men should do a larger share of child care. V13 Men should do a larger share of household work. V19 A bad marriage is better than no marriage at all. V26 People without kids lead empty lives. These questions are aimed more at a mens responsibility in the household and add some more cultural information about attitude about marriage and having children. Before using the Prefscal method in SPSS, first the Correlation Matrix was analyzed. As can be seen in the above matrix table there are quite a few significant correlations between the twelve variables. The highest correlations can be found between questions about working mothers and questions about mens household duties, and also between questions about having children and questions about family life. So there are strong correlations between the variables. This is as expected, because the variables represent unique characteristics of two cultures about gender equity. Since Prefscal itself doesnt give a three-dimensional graphical representation in the SPSS output automatically, I wanted to see if I could find three different dimensions in my data, instead of two (see Figure under construction). Results I will hereby present the results of an attempt to classify 2341 Japanese and Dutch citizens using the unfolding model. The result of the SPSS PREFSCAL unfolding model for the sample from the Dutch and Japanese citizenship values data from the 2002 ISSP Citizenship Programme is a two-dimensional joint plot based on a classical initial scaling configuration, which shows points for 1241 Dutch respondents, 1100 Japanese respondents and twelve statements about gender-equity and inequity. This solution resulted in a great two-dimensional graphical representation which looked very interpretable in terms of the possibilities of interpreting the differences between the two countries in the data set. As you can see, the gender equity positive variables are all on the bottom of the Column Objects model. The negative variables are all at the top, and the one neutral variable is in the middle. The Joint Plot shows the separately derived two-dimensional spaces for the red (Japanese respondents) and blue (Dutch respondents) dots. The twelve statements are represented by black dots. The axes represent the primary and secondary dimensions, shown in normalized units. The interpretation of this unfolding solution is done intuitively. It will be investigated whether the dimensions can be given meaning. A good modeling solution will locate a subjects opinion about gender equity according to the most dominant cross-reference proximities in their answers. By first inspection, it can be readily seen that Prefscal indeed located points corresponding to similar objects close together, while those corresponding to dissimilar objects far apart. This is consistent with our intuition that the countries within the groups havent got much in common. To understand these different contexts, we can again turn to nation-specific contextual explanations. (Insert interpretation of the above model) The highest level of education for the respondent is represented by five comparative categories, not included No answer, dont know. Lowest formal qualification and above lowest qualification represents those who have not completed primary school to those who have completed this level or the first stage of basic education. Higher secondary completed refers to those who have completed higher secondary school or technical training. The above higher level of secondary education group includes those who completed higher secondary school theoretical training up to the last and highest level of tertiary education, University degree completed (i.e., PhD). Van Wel Knijn (2006) maintain that the part-time labour market participation of Dutch mothers is primarily caused by cultural factors and not economic or institutional constraints. They contend that a culture of care dominates, as does the one-and-a-half earner model where the man works full-time and the woman part-time. This model is particularly dominant among people with a lower education. For those with higher education, the tendency is for both partners to attempt to work part-time, although this is only achieved within a very limited group. Based on these institutional and cultural differences, we anticipate that women in Japan will face higher institutional and family constraints than in the Netherlands. As outlined in the main hypothesis, we expect that these constraints will be particularly poignant for Japanese working women who engage in both substantial paid labour combined with a heavy load of household duties. (Insert education model with only females). Conclusion Discussion

No comments:

Post a Comment

Note: Only a member of this blog may post a comment.