Poverty Risk in Taiwan

Transcription

1 The Fourth Annual East Asian Social Policy research network (EASP) International Conference Restructuring Care Responsibility: Dynamics of Welfare Mix in East Asia 20 th -21 st October 2007 In Hongo Campus, The University of Tokyo, Japan Poverty Risk in Taiwan Chin-Fen Chang Associate Professor Soochow University, Taiwan Shao-Hua Wang Postgraduate Student Soochow University, Taiwan Presented in The Fourth Annual East Asian Social Policy Research Network (EASP) International Conference Restructuring Care Responsibility: Dynamics of Welfare Mix in East Asia 20 th -21 st October 2007 Hongo Campus, The University of Tokyo, Japan 0

2 The Fourth Annual East Asian Social Policy research network (EASP) International Conference Restructuring Care Responsibility: Dynamics of Welfare Mix in East Asia 20 th -21 st October 2007 In Hongo Campus, The University of Tokyo, Japan Poverty Risk in Taiwan* Chin-Fen Chang Associate Professor Soochow University, Taiwan Shao-Hua Wang Postgraduate Student Soochow University, Taiwan Keywords: Poverty, Risk, Taiwan, Social Change, Social Policy, Low Viewing the trends of income distributions in Taiwan, the gap between Top 20 to Button 20 of income distribution is getting wider. In 2005, there are 6.04 times between Top 20 incomers and Button 20. Also, the phenomena of new poverty raised in recent years. To what extent the risk of poverty have been faced in Taiwan? The aim of this paper tries to use time-series method to find the phenomena of poverty risk in Taiwan. Taiwan Social Change Survey has been used in this study as a secondary data. In hence, this paper will analyse the Databases of The Taiwan Social Change Survey in 1995, 2000 and The SPSS(14.0) Statistic analysis Software will be used to analyse time-series data in this study. This paper tries to separate three dimensions, personal characters, family conditions and employment conditions, to find who faces the challenge of poverty risk. First, for personal characters, sex, age, education and living area are significant effects people whether to face poverty risk or not. Secondly, viewing family conditions, marriage indeed influence female to face the challenge of poverty risk or not. Also, the conditions of single families, no matter is single-mother family or single father family, is getting worse. Third, the employment situation will be analysed. This paper will make conclusion remarks for social policy to rethink the phenomena of poverty risk in Taiwan Acknowledge: This paper is based on research project (period: )sponsored by the National Science Council (NSC: H MY2). 1

3 The Fourth Annual East Asian Social Policy research network (EASP) International Conference Restructuring Care Responsibility: Dynamics of Welfare Mix in East Asia 20 th -21 st October 2007 In Hongo Campus, The University of Tokyo, Japan Issues of Poverty Risk The Concept The concept of poverty and low income has been defined by absolute poverty and relative poverty for setting poverty line. In EC of 1975, the poorer is defined as individuals or families whose resources are so small as to exclude then from the minimal acceptable way of life in which they live (Room, 1995). As Atkinson (2002) noted poverty cannot be defined unequivocally, as it is essentially a relative, graduated and multi-dimensional notion (Atkinson, 2002:78-79). It is on account and potential ambiguity that some prefer to use alternative expressions, such as low income, rather than poverty (ibid.). In this study, the in sufficiency of resources did not be included as poverty. Also, the useful way to discuss poverty is from collecting information about a household s total income or about its total expenditure. wadays, Taiwan has come to use income as the indicators of financial poverty in survey data. Nevertheless, expenditure has been raised to concern with standards of living. In this paper, income should be our concern for considering minimum rights. Although Atkinson (2002) reminded that household income need to be concerned for measuring risk of financial poverty. The poverty risk measurement should be based on household income adjusted for differences in household size and composition by the OECD-modified equivalence scale, each household being given a weight equal to the number of household measures, with the sensitivity of the results being assessed on the a regular basis (Atkinson, 2002:100). However the survey data we used in Taiwan Social Change Survey collected for general conditions not specific for this paper. In hence, the limitation of this database, we used in this study could not analyse poverty risk through house income. Considering the Social Assistant Act, the term of low income will be used as to measure the condition of facing poverty risk in this paper. Although we did concern to use 60% of the median as a income poverty level to segregate low income and non low income in the early stage of research design, the limitation of Database caused we choose the other way to do. wadays, the poverty line in Taipei city, Kaohsiung city and Taiwan Province considered for defining low income and non low income. 2

4 The Fourth Annual East Asian Social Policy research network (EASP) International Conference Restructuring Care Responsibility: Dynamics of Welfare Mix in East Asia 20 th -21 st October 2007 In Hongo Campus, The University of Tokyo, Japan Also, the minimum wage 1 of 15,840 was concerned by definition, owing to Database are from 1995 to Also, the range of income in the Database we used was Therefore, we defined above as non low income. The Situation Turning to the factors of causing poverty in Taiwan, there are three reasons: personal factor, social factor and events (Wan, 2000; Shu et al., 2000).The personal factors included personal conditions, such as disabled, low education etc. In social factors included social change, structural unemployment. Events causing people faced poverty was car accident and disasters etc. Viewing the factors of poverty, there are several risks to cause poverty, not only focused on older people, disabled people or no incomers, but also spread to non low incomers and working level. As Houng et al. (2003) mentioned that, since 1980s the structure of poor has been change. Viewing the acceptance of social assistance, the proportion of older people fallen to poverty was getting down; the young unemployed people are getting raised. Also, Li & Lai (2004) interviewed five disadvantaged groups which were easier to drop into poverty: middle-high aged unemployed, indigenous unemployed, single family, foreign bridge and the group which did not paid NHI contribution fees. Li & Lai s (2004:4) findings, the change of social structure made people struggled in poverty. Moreover, some people did not covered by Social Insurance faced higher risk of poverty (Guo, 2003). As Cho w (2003) mentioned that some employers are work as part-time job, temporary job without social insurance. Those marginal labourers are excluded by employment insurance when those groups lost their jobs. For those disadvantaged people, they could not get any allowance from unemployment allowance; they also could not qualify to get support from social assistance under means-test. As Li & Lai (2004); Guo (2003) & Shu (2005) all criticized social security system could not help people in need. The gap of rich and poor is getting larger. In hence, the factors of poverty are different to traditional poverty. Also, there is different groups faced poverty risk which is different to traditional poverty. The welfare system is one of the factors to affect people to drop into poverty. However, there is no study to view the changing situation of poverty risk. This study tries to do so. 1 Before 2007, the minimum wage was

5 The Fourth Annual East Asian Social Policy research network (EASP) International Conference Restructuring Care Responsibility: Dynamics of Welfare Mix in East Asia 20 th -21 st October 2007 In Hongo Campus, The University of Tokyo, Japan Methodology This study used Taiwan Social Change Survey Database (TSSS) which is sponsored by National Science Council and collected by Survey Research Data Archive, Academia Sinica. We choose the original data in 1995 (the First General Questionnaire in the Third Period); in 2000 (the First General Questionnaire in the Fourth Period) and in 2005 (the First General Questionnaire I & II, the Fifth Period). The reason of choosing two questionnaires in 2005 is TSSS tried to adopt International Social Survey Program (ISSP) and reduced items of questions and caused some questions which this study tried to analyse are separated to different questionnaires. Sampling Viewing the TSSS, the whole questionnaires used face to face interview. The sample frame was Household Database in Taiwan Area. The TSSS used probability proportional to size (PPS) to get survey samples. The conditions of samples are Nationality and age. The Nationality was Republic of China; the age was different in different survey year. In 1995, the range of ages is 20 to 70 years old; in 2000 was above 20 years old and in 2005 was above 18 years old. The original samples are 2093 cases in 1995; 1960 cases in 2000 and 2146 cases in Questionnaire I and 2171 cases in Questionnaire II in Concerning the need s of comparison, this study set the age for each was above 20 years old. In hence, there are 43 cases in Questionnaire I and 29 cases Questionnaire II in 2005 excluded. Answering the question of income is required for the purpose of this study to analyse poverty, we excluded those samples answer the item of income: don t know or don t want to answer. The numbers of excluded samples were 162 cases in 1995; 81 cases in 2000 and 158 cases in Questionnaire I and 291 cases in Questionnaire II in 2005 (see Table 1). Table 1 Samples Samples\Survey Time QI 2005 QII Original Samples Excluded Samples Aanlysed Samples Framework This study tries to analyse personal characters, family conditions and 4

6 The Fourth Annual East Asian Social Policy research network (EASP) International Conference Restructuring Care Responsibility: Dynamics of Welfare Mix in East Asia 20 th -21 st October 2007 In Hongo Campus, The University of Tokyo, Japan employment situation to examine the trends and groups to face poverty risk (see Figure 1). The variable description as followed: Personal Characters Sex Age Education Living Area Family Conditions Marriage Partner Relations Poverty Risk employment situation Industry Position Figure 1 Research Framework 1. Independent variables 1.1. Personal condition A. Sex: Male and Female. B. Age: it divided into four groups, 20~35, 36~50, 51~64 and Above 65. C. Level of Education: it divided into four groups, n-education, low level of education (elementary school and junior high school), middle level of education (senior high school and vocational school) and high level of education (above university). D. Living Area: it divided into six groups, Taipei City, Kaohsiung City, rthern Taiwan (including Keelung, Taipei County, Taoyuan, Hsinchu and Miaoli), Midland of Taiwan (including Taichung, Changhua, Yunlin, Chiayi and Nantou), Southern Taiwan (including Tainan, Kaohsiung County and 5

7 The Fourth Annual East Asian Social Policy research network (EASP) International Conference Restructuring Care Responsibility: Dynamics of Welfare Mix in East Asia 20 th -21 st October 2007 In Hongo Campus, The University of Tokyo, Japan Pingtung) and Eastern Taiwan (including Yilan, Hualien and Taitung) Family condition A. Marriage: single and married. B. Relation of spouse: it divided into three groups, live together, not live together with marriage relations (because of job) and single parent (divorced or partner death) Labor market A. Industry: According to the Standard Industrial Classification of Taiwan, it is classed as ten category, Agriculture, Forestry, Fishery, Husbandry and Hunting, Mining and Quarrying, Manufacturing, Electricity, Gas and Water, Construction, Commerce, Transportation, Storage and Communication, Finance, Insurance, Real Estate and Business Services, Public Administration, Social, Personal and Related Community Services and Others. B. Position: According to the Position Classification of Taiwan, it is classed as twelve category, administrator, professional (including Engineer), the assistant and technical fields, the clerical field, servicing and sales fields, agriculturist, fisher, herdsman and woodcutter, skilled worker, mechanical operator and assembler, unskilled worker, military, housewife and others (such as student). 2. Dependent variables According to the monthly income (including salary, overtime pay, bonus, pension, interest, rent etc.) of samples, it divided into four groups,,, and and n-low (above ). The Limitation There are several limitations when the secondary data we tried to use in this study. First, in 1995v & 2000, the item of income overlapped. For example, the item 3 and 4 have overlapped place (3) /per month; (4) /per month. Some samples might set their income range at lower level. Also, the income range is too large. Second, although in questionnaires there were one item asked who live together and who did not live together, there are difficult to justify the dependent population in household. Third, the item of living area is not the same as working area in the real situation. Fourth, The TSSS did not ask health conditions and caused the limitation of this study could not put health issue into personal condition. Finally, the TSSS only ask income, the expenditure and 6

8 The Fourth Annual East Asian Social Policy research network (EASP) International Conference Restructuring Care Responsibility: Dynamics of Welfare Mix in East Asia 20 th -21 st October 2007 In Hongo Campus, The University of Tokyo, Japan income distribution did not be included into questionnaires. Also, this study might be Type I and Type II errors; i.e., the count of those in poverty risk will miss some people who are indeed in poverty risk ( false negatives ) and will incorrectly identify as in poverty risk some people who are not ( false positives ). Analysis This Section tries to separate three parts, personal characters, family conditions and employment situations, to analyse the way in which poverty risk has been faced by different income groups. Personal Characters In Table 2 showed that sex and income were significant relations (χ 2 = Cramer's V=0.378 P=0.000 *** in 1995; χ 2 = Cramer's V=0.315 P =0.000 *** in 2000 andχ 2 = Cramer's V=0.252 P=0.000 *** in 2005). Comparing sex, most male samples were non-low incomers in 1995, 2000 and Viewing low income condition, female group dropped to low income category more than male group. the consideration of income, female faced disadvantaged than male. Moreover, the proportion of non low income in male group was decreased; female group was increased 5.9% from 1995 to The gap of income between male and female was getting decreased from 36.6% in 1995 to 23.8% in Concerning age and income relations, Table 3 showed that both age and income were significant relations (χ 2 = Cramer's V=0.210 P=0.000 *** in 1995; χ 2 = Cramer's V=0.262 P=0.000 *** in 2000; χ 2 = Cramer's V=0.248 P=0.000 *** in 2005). Segregated to different categories, the non low incomers were focused on the category of age between 36 and 50. Viewing the changing situation through year, this group increases 8.6% into non incomers. Relatively, more and more age group on 51 to 64 years old stayed at whether no income or non low income dramatic different conditions. It showed that middle-high aged group faced two types income structure: no income and non low income. Viewing older people who is over 65 years old, the proportion of no income was decreased from 51.3% to 40.1%. The proportion of income under increased 13.0%. Returning to view old aged policy, there were Old-Age Farmers' Welfare Allowance and Living Allowance for Elderly People implemented. This might impact the old age economic security when there was still no national pension 7

9 The Fourth Annual East Asian Social Policy research network (EASP) International Conference Restructuring Care Responsibility: Dynamics of Welfare Mix in East Asia 20 th -21 st October 2007 In Hongo Campus, The University of Tokyo, Japan for all senior citizens. Combined sex, age and income variables, there were significant relations (see Table 4). Viewing female group, no income group focused on aged over 65; non low income group focused on 36 to 50 aged groups. Viewing time-series from 1995 to 2005, there were 80% of male aged group between 20 to 50 were non low incomers, but the percentage dropped to about sixty percent in However, there was more male group on aged 51 to 64 stayed at no income group. Concerning female group, the large proportion of non low incomers was the group of aged 20 to 35. Concern time-series situation, the proportion increased from 46.2% in 1995 to 58.6% in 2000 and slightly decreased to 57.9% in For female group on aged 36 to 50, they increased to stay at non low income group, from 36.2% in 1995 to 52.5%. The economic ability was increased in female group on aged 36 to 50. Relatively, over forty percent of the aged group on 51 to 64 dropped to no income category. For over 65 years old female group, 77.3% of this group used to stay at no income in 1995 and dropped to 57.3% in It was significantly decreased 30%. Turning to Education, Table 5 showed that education and income did have significant relations (χ 2 = Cramer's V=0.261 P=0.000 *** in 1995; χ 2 = Cramer's V=0.292 P=0.000 *** in 2000; χ 2 = Cramer's V= P=0.000 *** in 2005). Most of no income category was no educators, 52.3% in 1995; 58.0% in 2000 and 46.8% in % of no educators under in 1995, 20.7% in 2000 and 33.5% in Almost no educators stayed at low income condition. The higher the education they had, the more proportion on non low income they dropped. However, there were gradually increasing of proportion on higher education stayed at no income group from 11.7% to 16.3%. Finally, living Area should be concerned on Table 6. The relations of living area and income was significant (χ 2 = Cramer's V=0.148 P=0.000 *** in 1995; χ 2 = Cramer's V = P = *** in 2000;χ 2 = Cramer's V=0.145 P=0.000 *** in 2005). Nearly 70% of people lived at Taipei Area were non low incomers. Comparing different areas, Taipei city was more no incomers whilst no income category and income below were focused on East Taiwan Area. Family Conditions Viewing marriage situation, Table 7 stated that marriage conditions, sex and income were significant related. For unmarried person, sex and income did not have significant relations. Relatively, for married person, sex and income did have 8

10 The Fourth Annual East Asian Social Policy research network (EASP) International Conference Restructuring Care Responsibility: Dynamics of Welfare Mix in East Asia 20 th -21 st October 2007 In Hongo Campus, The University of Tokyo, Japan significant relations through time series analysis (χ2= Cramer's V=0.433 P=0.000 *** in 1995; χ2= Cramer's V=0.365 P=0.000 *** in 2000 andχ 2 = Cramer's V=0.344 P=0.000 *** in 2005). Viewing male condition, 74.9% of married male was on non low incomers in 1995; 70.4% in 2000 and 70.5% in Concerning female situation, 41.9% of married female was on no income group in 1995; 38.2% in 2000 and 34.6% in The condition of married female was changed over ten years. More and more married female stayed at non low income group and raised nearly 4%. Viewing spouse relations, spouse relations and income cross-tabulations showed significant relations (see Table 8). For no income group, there was more single parent dropped into no income group. If we put sex into consideration, there was different result between male and female group(see Table 9). There were more single mother stayed at no income group, the percentage were 43.7% in 1995; 45.0% in 2000 and 35.2% in When single mother stayed at worse income condition from 1995 to 2005, single father also faced income insecurity. For single father, 22.0 % of income group in 1995; 29.0% in 2000 and 29.2% in 2005 dropped into no income group. Single father suffered disadvantaged conditions also need to take into account. Employment Conditions Working in different industry did show different income condition (χ 2 = Cramer's V=0.472 P=0.000 *** in 1995; χ 2 = Cramer's V =0.449 P=0.000 *** in 2000; χ 2 = Cramer's V=0.413 P=0.000 *** in 2005)(see Table 10). The best income condition was working in Finance, Insurance, Real Estate and Business Services, there were over 90% at non low income group. However, the worst condition of industry was Agriculture, Forestry, Fishery, Husbandry and Hunting category, their incomes were below per month. Seeing position and income situation (see Table 11), Professionals (including engineer) was at the best income situation, 91.1% at non low income level in 1995; 88.5% in 2000 and 94.6% in The second one was manager, 72.6%of this group stayed at non low income group in 1995; 92.7% in 2000 and 93.2% in Relatively, the worst condition still dropped at the category of agriculturist, fisher, herdsman and woodcutter. 9

11 The Fourth Annual East Asian Social Policy research network (EASP) International Conference Restructuring Care Responsibility: Dynamics of Welfare Mix in East Asia 20 th -21 st October 2007 In Hongo Campus, The University of Tokyo, Japan Conclusion Remarks Following the above analysis, there were some findings help us to answer the question of who faced the challenges of poverty risk. Also, we might turn to think does policy matters to overcome poverty risk. Who Faced the Challenges of Poverty Risk? Viewing sex issue, although the general situation of income and education on female is getting better, female group still faced disadvantaged than male group in general. Female group faced higher risk to drop into poverty risk. Also, combined sex, age and income variables, the economic ability was increased in female group on aged 36 to 50. Relatively, over forty percent of the aged group on 51 to 64 dropped to no income category. Moreover, there were more single mother stayed at no income group, they faced higher poverty risk than single father. In hence, female group at different range of age and life circle will impact them to face poverty risk. Relatively, middle-high aged group faced two types income structure: no income and non low income. Moreover, there was more male group on aged 51 to 64 stayed at no income group. This condition reminded as to concern middle-high aged group, especially male group, faced higher risk of poverty. Also, more and more single father dropped into no income group from 1995 to Single father suffered higher risk of poverty also need to take into account. Education did affect whether poverty risk faced or not. Almost no educators stayed at low income condition. As the above analysis found that the higher the education they had, the more proportion on non low income they dropped. Living area need to take into consideration of examining poverty risk when Taipei city was more no incomers whilst no income category and income below were focused on East Taiwan Area. Finally, for working class and worked at Agriculture, Forestry, Fishery, Husbandry and Hunting faced higher risk to face poverty. This situation might relate to wage structure and working period. Does Policy Matters to Overcome Poverty Risk? As literature told us, if older people retirement from labour market without pension or assistant, they are easier to drop into poverty. The situation of old age poverty also happened in Taiwan; older people still stayed at income insecurity and had higher risk to face poverty. As result shown that o the proportion of no income 10

12 The Fourth Annual East Asian Social Policy research network (EASP) International Conference Restructuring Care Responsibility: Dynamics of Welfare Mix in East Asia 20 th -21 st October 2007 In Hongo Campus, The University of Tokyo, Japan was decreased from 51.3% to 40.1% in old age group. The proportion of income under increased 13.0%. Viewing old aged policy, there were Old-Age Farmers' Welfare Allowance and Living Allowance for Elderly People implemented to support older people over past ten years. Especially, there were old aged women advantaged from this policy, as the above already presented that, 77.3% of over 65 years old female group used to stay at no income in 1995 and dropped to 57.3% in It was significantly decreased 30%. This might impact the old age economic security when there is still no national pension for all senior citizens. Although There were more single mother stayed at no income group, the percentage were 43.7% in 1995; 45.0 % in 2000 and 35.2% in The decreasing of single mother stayed at no income group can trace to view the implementation of Act of Assistance for Woman in Hardship (promulgated on 2000). Policy indeed helps some disadvantaged group reduced the risk to drop into poverty. Nevertheless, there still a far way to go to reexamine the phenomena of poverty risk and rethink who has been excluded by the contemporary policy. Reference Atkinson, Anthony B (2002) Social Indicators: the EU and SocialIinclusion, Oxford: Oxford University Press. Room, G. (ed.) (1995) Beyond the Threshold, Bristol: Policy Press. Wan, Y-W (2000) Social Welfare Services Theory and Implementation, Taipei: San-Min Press. Shou, C. et al. (2000) Social Problem, Taipei: Shiou-Fu Press. Houng S. C. et al. (2003) New Poverty and Social Welfare Policy, Journal of National Policy, Vol.2. 4, pp Guo, Y. L. (2003) 109 Million New Poor, Journal of Business, Li, S. R. & Lai, L. L. (2004) Survey Analysis and Strategies on Disadvantaged Family, Research Report, Ministry of Interior Chow, W. C. (2005) The Impact of Labour Policy under Unemployment, New Poverty and distribution, paper presented at Labour Policy: Contents, Impact and Prospect, Council of Labour Affairs. Shou, C. T. (2005) Low and Policy in Taiwan, website: time, 25, December,

13 Table 2 Sex & Cross-tabulation: 1995, 2000 & 2005 Year Different Different Different Variable Item and n-low and n-low and n-low Number Male % of Sex % 12.50% 5.10% 9.80% 72.50% % 13.00% 7.00% 9.80% 70.20% % 14.30% 10.20% 9.90% 65.60% Sex % of 51.70% 6.50% 2.60% 5.10% 37.50% 51.10% 6.70% 3.60% 5.00% 35.90% 50.60% 7.20% 5.10% 5.00% 33.20% Number Female % of Sex % 38.50% 7.20% 18.40% 35.90% % 35.40% 9.20% 15.00% 40.40% % 31.40% 12.70% 14.20% 41.80% % of 48.30% 18.60% 3.50% 8.90% 17.30% 48.90% 17.30% 4.50% 7.30% 19.70% 49.40% 15.50% 6.30% 7.00% 20.60% Number % of % 25.10% 6.10% 14.00% 54.80% % 23.90% 8.00% 12.30% 55.70% % 22.70% 11.40% 12.00% 53.80% Chi-Square Test χ 2 = Cramer's V=0.378 P=0.000 *** χ 2 = Cramer's V=0.315 P=0.000 *** χ 2 = Cramer's V=0.252 P=0.000 ***

14 Table 3 Age & Cross-tabulation: 1995, 2000 & 2005 Year Variable Item Different Different Different and n-low and n-low and n-low Number ~35 % of Age % 20.00% 4.20% 13.20% 62.60% % 14.80% 4.40% 10.90% 70.00% 100.0% 17.9% 10.0% 12.1% 60.0% % of 34.60% 6.90% 1.50% 4.60% 21.70% 31.70% 4.70% 1.40% 3.50% 22.20% 32.2% 5.8% 3.2% 3.9% 19.3% Number ~50 % of Age % 19.10% 3.80% 13.50% 63.50% % 16.80% 4.20% 12.30% 66.80% 100.0% 14.6% 3.9% 9.4% 72.1% Age % of 40.90% 7.80% 1.60% 5.50% 26.00% 40.30% 6.80% 1.70% 4.90% 26.90% 32.8% 4.8% 1.3% 3.1% 23.7% Number ~64 % of Age % 35.60% 7.60% 14.20% 42.50% % 37.50% 15.20% 13.40% 33.90% 100.0% 31.3% 10.8% 15.8% 42.1% % of 14.20% 5.10% 1.10% 2.00% 6.10% 14.70% 5.50% 2.20% 2.00% 5.00% 20.9% 6.5% 2.3% 3.3% 8.8% Above 65 Number % of Age % 51.30% 19.80% 18.30% 10.70% % 52.80% 20.60% 14.90% 11.70% 100.0% 40.1% 32.8% 12.4% 14.6% % of 10.20% 5.20% 2.00% 1.90% 1.10% 13.20% 7.00% 2.70% 2.00% 1.50% 14.1% 5.7% 4.6% 1.7% 2.1% Number % of % 25.10% 6.10% 14.00% 54.80% % 23.90% 8.00% 12.30% 55.70% 100.0% 22.7% 11.4% 12.0% 53.8% Chi-Square Test χ 2 = Cramer's V=0.210 P=0.000 *** χ 2 = Cramer's V=0.262 P=0.000 *** χ 2 = Cramer's V=0.248 P=0.000 *** 13

15 Table 4 Age, Sex & Cross-tabulation: 1995, 2000 & 2005 Year Variable Different Different Different Age 20~35 36~50 51~64 Sex Item and n-low and n-low and n-low Male Number % of Sex % 9.40% 2.40% 5.10% 83.20% % 10.20% 2.60% 6.30% 80.90% 100.0% 15.4% 13.2% 9.3% 62.1% % of 44.40% 4.20% 1.00% 2.20% 36.90% 51.00% 5.20% 1.30% 3.20% 41.30% 49.6% 7.7% 6.5% 4.6% 30.8% Female Number % of Sex % 28.50% 5.60% 19.60% 46.20% % 19.50% 6.20% 15.80% 58.60% 100.0% 20.3% 7.0% 14.9% 57.9% % of 55.60% 15.80% 3.10% 10.90% 25.70% 49.00% 9.60% 3.00% 7.70% 28.70% 50.4% 10.2% 3.5% 7.5% 29.2% Number % of % 20.00% 4.20% 13.20% 62.60% % 14.80% 4.40% 10.90% 70.00% 100.0% 17.9% 10.0% 12.1% 60.0% Male Number % of Sex % 5.40% 1.40% 5.90% 87.20% % 3.50% 2.70% 7.20% 86.50% 100.0% 4.5% 1.5% 4.2% 89.9% % of 53.50% 2.90% 0.80% 3.20% 46.70% 52.90% 1.80% 1.50% 3.80% 45.80% 52.5% 2.4%.8% 2.2% 47.2% Female Number % of Sex % 34.90% 6.50% 22.30% 36.20% % 31.70% 5.90% 17.90% 44.50% 100.0% 25.7% 6.6% 15.2% 52.5% % of 46.50% 16.20% 3.00% 10.40% 16.80% 47.10% 14.90% 2.80% 8.40% 21.00% 47.5% 12.2% 3.1% 7.2% 24.9% Number % of % 19.10% 3.80% 13.50% 63.50% % 16.80% 4.20% 12.30% 66.80% 100.0% 14.6% 3.9% 9.4% 72.1% Male Number % of Sex % 19.90% 7.70% 14.70% 57.70% % 25.80% 14.40% 15.20% 44.70% 100.0% 15.3% 8.2% 15.8% 60.7% 14

16 Year Variable Different Different Different Age Above 65 Sex Item and n-low and n-low and n-low % of 56.70% 11.30% 4.40% 8.40% 32.70% 47.70% 12.30% 6.90% 7.20% 21.30% 48.3% 7.4% 3.9% 7.6% 29.3% Female Number % of Sex % 56.30% 7.60% 13.40% 22.70% % 48.30% 15.90% 11.70% 24.10% 100.0% 46.2% 13.3% 15.7% 24.8% % of 43.30% 24.40% 3.30% 5.80% 9.80% 52.30% 25.30% 8.30% 6.10% 12.60% 51.7% 23.9% 6.9% 8.1% 12.8% Number % of % 35.60% 7.60% 14.20% 42.50% % 37.50% 15.20% 13.40% 33.90% 100.0% 31.3% 10.8% 15.8% 42.1% Male Number % of Sex % 35.20% 21.30% 28.70% 14.80% % 37.10% 23.40% 21.00% 18.50% 100.0% 33.6% 26.6% 16.8% 23.1% % of 61.90% 21.80% 13.20% 17.80% 9.10% 50.00% 18.50% 11.70% 10.50% 9.30% 52.2% 17.5% 13.9% 8.8% 12.0% Female Number % of Sex % 77.30% 17.30% 1.30% 4.00% % 68.50% 17.70% 8.90% 4.80% 100.0% 47.3% 39.7% 7.6% 5.3% % of 38.10% 29.40% 6.60% 0.50% 1.50% 50.00% 34.30% 8.90% 4.40% 2.40% 47.8% 22.6% 19.0% 3.6% 2.6% Number % of % 51.30% 19.80% 18.30% 10.70% % 52.80% 20.60% 14.90% 11.70% 100.0% 40.1% 32.8% 12.4% 14.6% 20~35 χ2= Cramer's V=0.381 P=0.000 *** χ2= Cramer's V=0.246 P=0.000 *** χ 2 = Cramer's V=0.141 P=0.006 ** Chi-Square Test 36~50 χ2= Cramer's V=0.530 P=0.000 *** χ2= Cramer's V=0.461 P=0.000 *** χ 2 = Cramer's V=0.418 P=0.000 *** 51~64 χ2= Cramer's V=0.403 P=0.000 *** χ2= Cramer's V=0.259 P=0.000 *** χ 2 = Cramer's V=0.399 P=0.000 *** Above 65 χ2= Cramer's V=0.453 P=0.000 *** χ2= Cramer's V=0.340 P=0.000 *** χ 2 = Cramer's V=0.309 P=0.000 *** 15

17 Level of Education Table 5 Education & Cross-tabulation: 1995, 2000 & 2005 Year Variable Item n-education Junior school below Senior school Vocational school University and above Different Different Different and n-low and n-low and n-low Number % of Education % 52.30% 24.60% 16.90% 6.20% % 58.00% 20.70% 15.50% 5.70% % 46.80% 33.50% 13.90% 5.70% % of 6.70% 3.50% 1.70% 1.10% 0.40% 10.30% 6.00% 2.10% 1.60% 0.60% 8.10% 3.80% 2.70% 1.10%.50% Number high % of and % 31.90% 7.60% 20.90% 39.60% % 28.80% 13.20% 18.10% 39.90% % 30.40% 17.60% 18.50% 33.60% Education % of 40.00% 12.80% 3.10% 8.40% 15.90% 36.20% 10.40% 4.80% 6.60% 14.40% 29.30% 8.90% 5.10% 5.40% 9.80% Number High and % of % 17.80% 2.70% 9.70% 69.70% % 13.10% 1.70% 9.00% 76.20% % 15.90% 4.00% 10.30% 69.90% Education % of 42.20% 7.50% 1.10% 4.10% 29.40% 40.80% 5.30% 0.70% 3.70% 31.10% 41.50% 6.60% 1.60% 4.30% 29.00% Number % of Education % 11.70% 2.30% 2.30% 83.60% % 17.30% 3.40% 4.20% 75.10% % 16.30% 9.00% 5.90% 68.80% % of 11.00% 1.30% 0.30% 0.30% 9.20% 12.60% 2.20% 0.40% 0.50% 9.50% 21.10% 3.40% 1.90% 1.20% 14.50% Number % of % 25.10% 6.10% 13.80% 54.90% % 23.90% 8.00% 12.40% 55.70% % 22.70% 11.40% 12.00% 53.80% Chi-Square Test χ 2 = Cramer's V=0.261 P=0.000 *** χ 2 = Cramer's V=0.292 P=0.000 *** χ 2 = Cramer's V=0.265 P=0.000 *** 16

18 Living Area Table 6 Living Area & Cross-tabulation: 1995, 2000 & 2005 Year Variable Item Taipei City Kaohsiung City Different Different Different and n-low and n-low and n-low Number % of Living Area % 25.00% 2.50% 4.20% 68.30% % 20.40% 5.20% 6.40% 68.00% 100.0% 13.8% 4.8% 12.4% 69.0% % of 14.70% 3.70% 0.40% 0.60% 10.00% 13.30% 2.70% 0.70% 0.90% 9.00% 7.5% 1.0%.4%.9% 5.1% Number % of Living Area % 28.20% 3.10% 16.00% 52.70% % 17.70% 5.30% 15.00% 61.90% 100.0% 26.6% 14.0% 8.4% 51.0% % of 6.80% 1.90% 0.20% 1.10% 3.60% 6.00% 1.10% 0.30% 0.90% 3.70% 7.4% 2.0% 1.0%.6% 3.8% rthern Taiwan Number (including Keelung, Taipei % of Living % 28.30% 2.70% 13.10% 55.90% % 21.80% 4.30% 9.80% 64.30% 100.0% 22.4% 7.4% 8.4% 61.8% County, Area Taoyuan, Hsinchu and % of 34.70% 9.80% 0.90% 4.60% 19.40% 21.30% 4.60% 0.90% 2.10% 13.70% 35.4% 7.9% 2.6% 3.0% 21.9% Miaoli) Midland of Number Taiwan (including % of Living % 21.10% 8.60% 18.00% 52.40% % 25.20% 9.90% 12.70% 52.20% 100.0% 28.5% 12.1% 13.8% 45.6% Taichung, Area Changhua, Yunlin, Chiayi % of 25.30% 5.30% 2.20% 4.60% 13.30% 26.40% 6.70% 2.60% 3.40% 13.80% 22.0% 6.3% 2.7% 3.0% 10.0% and Nantou) Southern Taiwan Number (including Tainan, % of Living % 27.50% 3.30% 20.90% 48.40% % 26.30% 9.20% 15.00% 49.40% 100.0% 24.9% 12.8% 14.2% 48.1% Area 17

19 Year Variable Item Different Different Different and n-low and n-low and n-low Kaohsiung County and % of 4.70% 1.30% 0.20% 1.00% 2.30% 28.30% 7.50% 2.60% 4.30% 14.00% 19.2% 4.8% 2.5% 2.7% 9.3% Pingtung) Number Eastern Taiwan (including Yilan, % of Living % 21.90% 16.60% 15.80% 45.70% % 30.70% 19.30% 19.30% 30.70% 100.0% 9.0% 26.3% 20.4% 44.3% Hualien and Area Taitung) % of 13.70% 3.00% 2.30% 2.20% 6.30% 4.70% 1.40% 0.90% 0.90% 1.40% 8.6%.8% 2.3% 1.7% 3.8% Number % of % 25.10% 6.10% 14.00% 54.80% % 23.90% 8.00% 12.30% 55.70% % 22.7% 11.4% 12.0% 53.8% Chi-Square Test χ 2 = Cramer's V=0.148 P=0.000 *** χ 2 = Cramer's V=0.118 P=0.000 *** χ 2 = Cramer's V=0.145 P=0.000 *** 18

20 Table 7 Marriage, Sex & Cross-tabulation: 1995, 2000 & 2005 Year Variable Different Different Different Marriag e Sex Item and n-low Incom e and n-low and n-low Male Number % of Sex % 16.30% 7.60% 13.60% 62.50% % % 6.00% 9.30% 69.80% 100.0% 18.6% 16.8% 11.7% 52.9% % of 57.90% 9.40% 4.40% 7.90% 36.20% 59.90% 8.90% 3.60% 5.60% 41.80% 58.2% 10.8% 9.8% 6.8% 30.8% single Female Number % of Sex % 17.90% 3.00% 23.10% 56.00% % % 6.30% 11.80% 61.80% 100.0% 18.8% 8.1% 12.2% 60.9% % of 42.10% 7.50% 1.30% 9.70% 23.60% 40.10% 8.10% 2.50% 4.70% 24.80% 41.8% 7.9% 3.4% 5.1% 25.5% Number % of % 17.00% 5.70% 17.60% 59.70% % % 6.10% 10.30% 66.60% 100.0% 18.7% 13.2% 11.9% 56.3% Male Number % of Sex % 11.70% 4.40% 9.00% 74.90% % % 7.20% 9.90% 70.40% 100.0% 12.7% 7.6% 9.3% 70.5% % of 50.40% 5.90% 2.20% 4.50% 37.80% 49.10% 6.10% 3.60% 4.90% 34.60% 48.2% 6.1% 3.7% 4.5% 34.0% married Female Number % of Sex % 41.90% 7.90% 17.70% 32.60% % % 9.70% 15.70% 36.50% 100.0% 34.6% 13.9% 14.7% 36.8% % of 49.60% 20.70% 3.90% 8.80% 16.10% 50.90% % 4.90% 8.00% 18.60% 51.8% 17.9% 7.2% 7.6% 19.1% Number % of % 26.60% 6.10% 13.30% 53.90% % % 8.50% 12.80% 53.10% 100.0% 24.0% 10.9% 12.1% 53.1% Chi-Square Test single χ2=7.613 Cramer's V=0.155 P=0.055 χ2=2.753 Cramer's V=0.088 P=0.431 χ 2 =7.867 Cramer's V=0.129 P=0.049 * 19

21 Year Variable Different Different Different Marriag e Sex Item and n-low Incom e and n-low and married χ2= Cramer's V=0.433 P=0.000 *** χ2= Cramer's V=0.365 P=0.000 *** χ 2 = Cramer's V=0.344 P=0.000 *** n-low 20

22 Table 8 Spouse Relation & Cross-tabulation: 1995, 2000 & 2005 Year Variable Item Different Different Different and n-low and n-low n-low and Number Live together % of relation % 25.70% 5.40% 12.40% 56.50% % 23.70% 7.60% 11.80% 57.00% 100.0% 22.7% 9.0% 11.7% 56.6% % of 88.40% 22.70% 4.80% 10.90% 50.00% 85.90% 20.30% 6.50% 10.10% 48.90% 85.9% 19.5% 7.7% 10.0% 48.7% Number Spouse t live together(with Relation % of relation % 24.20% 9.10% 18.20% 48.50% % 21.90% 9.40% 15.60% 53.10% 100.0% 25.8% 9.7% 9.7% 54.8% marriage relations) % of 2.10% 0.50% 0.20% 0.40% 1.00% 2.10% 0.50% 0.20% 0.30% 1.10% 2.1%.5%.2%.2% 1.2% Single Parent Family (divorced or death) Number % of relation % 36.60% 12.40% 20.30% 30.70% % 39.60% 14.30% 19.80% 26.40% 100.0% 33.5% 24.4% 15.3% 26.7% % of 9.50% 3.50% 1.20% 1.90% 2.90% 12.00% 4.70% 1.70% 2.40% 3.20% 11.9% 4.0% 2.9% 1.8% 3.2% Number % of % 26.70% 6.20% 13.20% 53.90% % 25.50% 8.40% 12.80% 53.20% % 24.0% 10.9% 12.1% 53.0% Chi-Square Test χ 2 = Cramer's V=0.114 P=0.000 *** χ 2 = Cramer's V=0.142 P=0.000 *** χ 2 = Cramer's V=0.154 P=0.000 *** 21

23 Table 9 Spouse relation, Sex & Crosstabulation: 1995, 2000 & 2005 Year Variable Different Different Different Spouse Relation Live together Single Parent Family (divorced or death) Sex Item and n-lo w and n-low and n-low Male Number % of Sex % 11.00% 3.90% 7.80% 77.30% % 11.00% 7.00% 8.10% 73.90% 100.0% 11.1% 7.4% 9.1% 72.3% % of 52.40% 5.80% 2.00% 4.10% 40.50% 51.50% 5.70% 3.60% 4.10% 38.00% 51.1% 5.7% 3.8% 4.7% 37.0% Female Number % of Sex % 41.80% 7.10% 17.40% 33.70% % 37.00% 8.20% 15.70% 39.10% 100.0% 34.7% 10.7% 14.4% 40.2% % of 47.60% 19.90% 3.40% 8.30% 16.00% 48.50% 18.00% 4.00% 7.60% 19.00% 48.9% 17.0% 5.2% 7.0% 19.7% Number % of % 25.70% 5.40% 12.40% 56.50% % 23.70% 7.60% 11.80% 57.00% 100.0% 22.7% 9.0% 11.7% 56.6% Male Number % of Sex % 22.00% 10.00% 22.00% 46.00% % 29.00% 8.10% 27.40% 35.50% 100.0% 29.2% 12.5% 12.5% 45.8% % of 32.70% 7.20% 3.30% 7.20% 15.00% 34.10% 9.90% 2.70% 9.30% 12.10% 27.3% 8.0% 3.4% 3.4% 12.5% Female Number % of Sex % 43.70% 13.60% 19.40% 23.30% % 45.00% 17.50% 15.80% 21.70% 100.0% 35.2% 28.9% 16.4% 19.5% % of 67.30% 29.40% 9.20% 13.10% 15.70% 65.90% 29.70% 11.50% 10.40% 14.30% 72.7% 25.6% 21.0% 11.9% 14.2% Number % of % 36.60% 12.40% 20.30% 30.70% % 39.60% 14.30% 19.80% 26.40% % 33.5% 24.4% 15.3% 26.7% Chi-Square Test live together χ2= Cramer's V=0.447 P=0.000 *** χ2= Cramer's V=0.370 P=0.000 *** χ 2 = Cramer's V=0.340 P=0.000 *** single parent family χ2= Cramer's V=0.261 P=0.015 * χ2= Cramer's V=0.245 P=0.012 * χ 2 = Cramer's V=0.278 P=0.003 ** 22

24 Industr y Table 10 Industry & Cross-tabulation: 1995, 2000 & 2005 Year Variable Item Different Different Different and n-low and n-low Incom e and Agriculture, Number Forestry, Fishery, % of industry % 34.30% 22.40% 22.90% 20.40% % 32.70% 34.60% 16.30% 16.30% 100.0% 33.8% 38.3% 13.5% 14.3% Husbandry and Hunting % of 10.50% 3.60% 2.30% 2.40% 2.10% 8.20% 2.70% 2.80% 1.30% 1.30% 6.9% 2.3% 2.6%.9% 1.0% Mining and Quarrying Manufacturing Electricity, Gas and Water Construction Commerce n-low Number % of industry % 0.00% 0.00% 0.00% % % 75.00% 0.00% 0.00% 25.00% 100.0% 25.0%.0%.0% 75.0% % of 0.40% 0.00% 0.00% 0.00% 0.40% 0.20% 0.20% 0.00% 0.00% 0.10%.4%.1%.0%.0%.3% Number % of industry % 3.50% 2.80% 20.00% 73.70% % 6.80% 2.10% 16.80% 74.40% 100.0% 7.5% 5.2% 15.2% 72.2% % of 20.60% 0.70% 0.60% 4.10% 15.10% 22.90% 1.50% 0.50% 3.80% 17.00% 20.0% 1.5% 1.0% 3.0% 14.4% Number % of industry % 7.10% 0.00% 0.00% 92.90% % 14.30% 0.00% 0.00% 85.70% 100.0% 11.1%.0% 11.1% 77.8% % of 0.70% 0.10% 0.00% 0.00% 0.70% 0.40% 0.10% 0.00% 0.00% 0.30%.5%.1%.0%.1%.4% Number % of industry % 2.30% 5.50% 10.20% 82.00% % 8.70% 3.90% 10.70% 76.70% 100.0% 15.0% 6.2% 12.4% 66.4% % of 6.70% 0.20% 0.40% 0.70% 5.50% 5.50% 0.50% 0.20% 0.60% 4.20% 5.8%.9%.4%.7% 3.9% Number % of industry % 7.00% 2.90% 18.00% 72.10% % 5.70% 5.20% 14.30% 74.80% 100.0% 8.6% 9.9% 16.8% 64.7% % of 12.70% 0.90% 0.40% 2.30% 9.20% 12.30% 0.70% 0.60% 1.80% 9.20% 15.0% 1.3% 1.5% 2.5% 9.7% Number % of industry % 7.90% 4.80% 6.30% 81.00% % 2.70% 2.70% 13.50% 81.10% 100.0% 12.0% 4.8% 9.6% 73.5% Transportation, Storage and Communication % of 3.30% 0.30% 0.20% 0.20% 2.70% 3.90% 0.10% 0.10% 0.50% 3.20% 4.3%.5%.2%.4% 3.1% Finance, Number Insurance, Real % of industry % 6.90% 0.00% 3.40% 89.70% % 5.50% 2.20% 1.10% 91.20% 100.0% 1.0% 2.9% 4.9% 91.3% 23

25 Year Variable Item Estate and Business Services Public Administration, Social, Personal and Related Community Services Other Different Different Different and n-low and n-low Incom e and n-low % of 4.50% 0.30% 0.00% 0.20% 4.10% 4.90% 0.30% 0.10% 0.10% 4.40% 5.3%.1%.2%.3% 4.8% Number % of industry % 5.60% 3.60% 15.80% 75.00% % 4.60% 3.70% 14.50% 77.20% 100.0% 4.6% 7.4% 12.8% 75.2% % of 18.70% 1.00% 0.70% 3.00% 14.00% 17.30% 0.80% 0.60% 2.50% 13.40% 18.9%.9% 1.4% 2.4% 14.2% Number % of industry % 82.00% 7.30% 5.40% 5.20% % 70.40% 12.40% 7.20% 10.00% 100.0% 66.3% 18.0% 7.4% 8.3% % of 22.00% 18.10% 1.60% 1.20% 1.10% 24.50% 17.20% 3.00% 1.80% 2.50% 22.9% 15.2% 4.1% 1.7% 1.9% Number % of % 25.10% 6.10% 14.00% 54.80% % 24.00% 8.10% 12.40% 55.60% % 11.4% 12.1% 53.8% % Chi-Square Test χ 2 = Cramer's V=0.472 P=0.000 *** χ 2 χ 2 = Cramer's V=0.413 P=0.000 = Cramer's V=0.449 P=0.000 *** *** 24

26 Position Table 11 Position & Cross-tabulation: 1995, 2000 & 2005 Year Variable Item Administrator Professional(including Engineer) Assistant and technical fields Clerical field Different Different Different and n-low and n-low and n-low Number % of % 8.40% 7.40% 11.60% 72.60% % 3.60% 0.00% 3.60% 92.70% 100.0% 3.4%.0% 3.4% 93.2% position % of 4.90% 0.40% 0.40% 0.60% 3.60% 7.30% 0.30% 0.00% 0.30% 6.80% 4.5%.2%.0%.2% 4.2% Number % of % 4.40% 0.70% 3.70% 91.10% % 5.30% 0.80% 5.30% 88.50% 100.0% 2.0% 2.0% 1.3% 94.6% position % of 7.00% 0.30% 0.10% 0.30% 6.40% 7.00% 0.40% 0.10% 0.40% 6.20% 7.7%.2%.2%.1% 7.3% Number % of position % 1.50% 1.50% 3.90% 93.20% % 6.80% 0.50% 5.70% 87.00% 100.0% 4.2% 2.5% 5.5% 87.7% % of 10.70% 0.20% 0.20% 0.40% 10.00% 10.20% 0.70% 0.10% 0.60% 8.90% 12.2%.5%.3%.7% 10.7% Number % of % 5.30% 1.50% 15.30% 77.90% % 2.70% 1.40% 9.60% 86.30% 100.0% 7.5% 1.1% 12.1% 79.3% position % of 6.80% 0.40% 0.10% 1.00% 5.30% 7.80% 0.20% 0.10% 0.70% 6.70% 9.0%.7%.1% 1.1% 7.1% Number Servicing and sales fields % of % 7.10% 3.50% 22.00% 67.50% % 5.00% 6.10% 18.30% 70.60% 100.0% 10.3% 10.3% 19.8% 59.5% position % of 13.30% 0.90% 0.50% 2.90% 8.90% 9.60% 0.50% 0.60% 1.80% 6.80% 12.0% 1.2% 1.2% 2.4% 7.1% Agriculturist, fisher, Number herdsman and woodcutter % of % 34.00% 22.20% 22.70% 21.10% % 35.30% 34.70% 16.00% 14.00% 100.0% 31.7% 40.7% 13.8% 13.8% position 25

27 Year Variable Item Skill worker Mechanical operator and assembler Unskilled worker Military Housewife Others Different Different Different and n-low and n-low and n-low % of 10.10% 3.40% 2.20% 2.30% 2.10% 8.00% 2.80% 2.80% 1.30% 1.10% 6.4% 2.0% 2.6%.9%.9% Number % of position % 6.30% 3.80% 17.50% 72.40% % 8.90% 4.20% 15.20% 71.70% 100.0% 11.7% 6.3% 17.5% 64.6% % of 14.90% 0.90% 0.60% 2.60% 10.80% 10.20% 0.90% 0.40% 1.50% 7.30% 10.6% 1.2%.7% 1.9% 6.9% Number % of position % 3.40% 3.40% 22.10% 71.00% % 6.10% 3.00% 19.70% 71.20% 100.0% 11.8% 7.6% 20.1% 60.4% % of 7.50% 0.30% 0.30% 1.70% 5.40% 7.00% 0.40% 0.20% 1.40% 5.00% 7.4%.9%.6% 1.5% 4.5% Number % of % 8.50% 15.30% 35.60% 40.70% % 7.20% 10.10% 32.60% 50.00% 100.0% 14.4% 23.1% 25.0% 37.5% position % of 3.10% 0.30% 0.50% 1.10% 1.20% 7.40% 0.50% 0.70% 2.40% 3.70% 5.4%.8% 1.2% 1.3% 2.0% Number % of % 91.50% 4.30% 4.30% 0.00% % 4.20% 12.50% 25.00% 58.30% 100.0%.0% 24.3% 18.9% 56.8% position % of 2.40% 2.20% 0.10% 0.10% 0.00% 1.30% 0.10% 0.20% 0.30% 0.70% 1.9%.0%.5%.4% 1.1% Number % of % 83.30% 6.80% 5.10% 4.80% % 72.00% 12.50% 6.60% 9.00% 100.0% 68.9% 16.0% 6.2% 8.9% position % of 17.50% 14.60% 1.20% 0.90% 0.80% 15.40% 11.10% 1.90% 1.00% 1.40% 13.3% 9.1% 2.1%.8% 1.2% Number % of % 69.70% 6.10% 9.10% 15.20% % 69.70% 11.50% 7.90% 10.90% 100.0% 62.6% 20.9% 9.1% 7.5% position 26

28 Year Variable Item Different Different Different and n-low and n-low and n-low % of 1.70% 1.20% 0.10% 0.20% 0.30% 8.80% 6.10% 1.00% 0.70% 1.00% 9.7% 6.0% 2.0%.9%.7% Number % of % 25.10% 6.10% 14.00% 54.80% % 24.00% 8.10% 12.40% 55.60% % 22.8% 11.5% 12.0% 53.7% Chi-Square Test χ 2 = Cramer's V=0.486 P=0.000 *** χ 2 = Cramer's V=0.472 P=0.000 *** χ2= Cramer's V=0.444 P=0.000 *** 27