Effects of Social Class Identification on Government

Effects of Social Class Identification on Government Walsh, K. C., Jennings, M., Stoker, L. (2004). The Effects of Social Class Identification on Participatory Orientations Towards Government.  British Journal of Political Science,Vol.34, No.03, pp.469-495. Brief Summary There is common claim that class identity does not matter for American political behaviour. The researchers tried to find out the influence of social class identification on participatory orientations towards government. The researchers focus on class identity perspective which people use to interpret and interact with the political world. The researchers used panel-study data spanning thirty-two years and two generations. Research Description This study’s independent variable is social class identification. There are two approaches, the first respondents subjectively identified themselves, and the second one is objective assessment. The researchers distinguish two classes (working-class and middle-class) by levels of income, education, union membership, and occupation. This study’s dependent variable is participatory orientations towards government. The researchers employed four measures. All of these measures concern political involvement and engagement (political interest, political efficacy, conception of good citizen, and political participation). Because there were many variables in this research, the researchers used multivariate analysis. Two variables are control variables, they are gender and race. â€Å"In a multivariate analysis, a variable is said to controlled when its influence is removed. We control variable by holding its value constant† (Agresti, 1999, p.305). The methodology in that research was a quantitative approach. Creswell (2003, p.18) explained quantitative as â€Å"an approach which the investigator primarily uses post positivist claims for developing knowledge, employs strategies of inquiry such as experiments and survey, and collects data on predetermined instruments that yield statistical data†. Because the purpose of this research is to make predictions about relation or effects of social class on political world, researchers used inferential statistics. â€Å"Inferential statistical analysis can predict characteristics of entire populations quite well by selecting samples that are small relative to the population size. In the past quarter-century, social scientists have increasingly recognized the power of inferential statistical methods†, (Agresti, 1999, p.5). This research used longitudinal research, from 1965-1997 with panel-study type. The sampling technique is proportionate stratified random sampling. The sample is 1,669 seniors from 97 secondary high schools in America in 1965 and also their parents. Furthermore, the instrument which researchers used is a questionnaire. Research Critique Clarity This research methods use inferential statistics to predict population characteristics by using sample, so the concept of population and sample must be clear. According to Babbie (1990, p.72), researchers must begin with a careful specification of population and should not mislead or deceive the readers. He gave example â€Å"Adult New Yorkers† is clearer than â€Å"American† because there is specification of the age and boundaries. In this study, the population all senior high school students in USA, and the unit analysis is an individual senior high school student in 1965. The sample is 1,669 students in 1965 and their parents. The researchers describe the different attitude between two generations. Besides that, the researchers also show the trend or the changing of respondents attitude using four wave panels for student generation in 1965 (G2) and three waves panel for their parents (G1). This article published in July 2004 and at the beginning of the article (p.469) clearly stated that it is about American political behavior. In addition, the brief result or answer for research question is â€Å"class may be particularly important in the present political context†. However the researchers used data survey in 1965, 1973, 1982, and 1997. The focus of research is one of issues here. Longitudinal research usually compares the changing of the subject time after time. One of types of longitudinal research is panel-study. â€Å"Panel studies involve the collection of data over time from the same sample respondents. The sample for such a study is called the panel†(Babbie, 1990, p.58). Researchers make claim the result particularly important for present political context instead of stated trend of last five decades or comparison of two generations. 2.2 Theory and Hypotheses â€Å"Theory has an important role in research and is an essential ally for the researcher. Researchers interweave a story about the operation of the social world (the theory) with what they observe when examine it systematically† (Neuman, 2007, p.24). Dennis Gilbert (2010, p.11) define social classes as â€Å"groups of families, more or less equal in rank and differentiated from other families above or below them with regard to characteristics such as occupation, income, wealth, and prestige†. Furthermore he figure out American social class structure become six parts: underclass (12%), working poor (13%), working class (30%), middle class (30%), upper-middle class (14%), and capitalist class (1%). Each of class has characteristics in terms of typical occupation and education. In this research, researchers just use two social classes. Justification to do that because based on previous research by Jackman and Jackman in 1975, the vast majority respondents identified as working-class (37 %) or middle-class (44 %). In subjective social class identification, respondents have to choose two options. As a result there are several respondents did not give answer. Instead of beginning with theories and concepts related to engagement in governmental affairs such as political participation and civic engagement, the researchers directly gave their measurement. They used four measurements to measure participatory orientations toward government (political interest, active citizen, political efficacy, and political participation). Based on the purpose of research to get to know relationship social class identification toward participatory orientations toward government, the simply hypothesis would be: H0 = There is no significant correlation/effects between social class identification and participatory orientations towards government. Ha= There is significant correlation/effects between social class identification and participatory orientations towards government. In analytical surveys usually are driven by theoretical questions. The purpose is to collect evidence which supports or contradict some hypothesis about the cause of people’s behavior. Buckingham and Saunders (2004, p.14) define hypotheses as â€Å"statements about what our theoretical propositions lead us to expect to find. They enable theories to be tested by predicting patterns of observations that should occur. Hypotheses therefore predict patterns of association in observed data as a means for testing causal theories†. 2.3 Research Design and Methodology â€Å"Research design is a strategy for collecting and analyzing data. It must be appropriate for answering the questions which the project is seeking to address, and it must take into account the practical constraints which the project is likely to encounter† (Buckingham and Saunders, 2004, p.294). The strategy which researchers used to collect data here is a survey. Sampling In order to get the view and attitude of population, the researchers used random sampling. This is the selection of a sample that should be representative of the population. The researchers choose from sampling frame available, and each individual has same opportunity. In this study, type of probability sampling which researchers used is proportionate stratified random sampling. The researchers said (p.473) â€Å"the students were distributed across a stratified sample of ninety-seven secondary schools, with the probability of schools selection being proportionate to size†. As described above, the sample took in 1965 from 97 schools with target subjects 1,669 students and one of their parents as well. We can ask questions, why the researchers used stratified random sampling? Why make a senior high school student as a unit analysis? Why there are four wave panels? Actually it is a series of data survey about political attitudes and behaviors by obtaining data on the same individuals as they aged from approximately 18 years of age in 1965 to 50 years of age in 1997. Based on that series of data survey, in 2004 the researchers make study about the effects of social identification on participatory towards government affairs. According to this information, we can assume the stratum here is the generation (G1 and G2). These four wave panels yielded 935 respondents for G2 (students) and 898 respondents for three wave panels G1 (parents). However, the surveys in 1973 and 1982 did not include social class questions. Because of geographical problem to make personal interview, the base panel survey in 1973 and 1982 are 728 for G1 and 659 for G2. One problem here, can we compare every wave panel, if the sample compositions are different? Instrumentation â€Å"A research instrument is a survey questionnaire or interview schedule that researcher uses to measure variables† (Neuman, 2007, p.168). The researchers used a questionnaire which conducted by several ways (face-to-face interview, computer-assisted personal interview, computer-assisted telephone interview, and self-enumerated questionnaire). This research used Logistic regression model for binary response variables which only has two possible outcomes. Education level: This index was scaled from 0 (less than high school) to 1 (PhD). Union Member: 0 = no; 1 = yes. Income: all of responses about household income were coded to the midpoint of the category and then rescaled within each wave and each generation from 0 (least) to 1 (highest). Occupation Skill Level Self-Employment and Homemaker For the better education level, income, and participate in union member would take into account in middle class status. However, for occupation indicators, the researchers just gave information about the consideration without description where those characteristics were placed. There are several questions in the questionnaire which measure political attitudes of American people. Researchers employed four measures, and it is categorized as a dichotomous dependent dummy variable. The value is most often a representation for a measured variable (Hagle, 2004). Political interest, 0 = hardly at all, 1 = most of the time Being an active citizen, 0 = if they did not volunteer such a response, 1 = respondents volunteered a response of this nature in any one of the responses. Political Efficacy, 0 = least efficacy, 1 = most efficacy Political participation, 0 = least active, 1 = most active The measurement scale which researchers used is an ordinal scale. Neuman (2007, p.375) defines ordinal level measurement as â€Å"a level of measurement that identifies a difference among categories of a variable and allows the categories to be rank order†. Even though just there are two categories (0 and 1), we can order it, which is 1 is more active than 0 in terms of political participation variable for example. Ordinal scale can have more than two options, for example (very interested, interested, uninterested, very uninterested). 2.4. Analysis of Data and Findings The researchers provide regression tables to describe their research result. In this study, researchers stated (p.479) that they used bivariate (type I) and multivariate regression models (type II). Bivariate analysis is an analysis of association between two variables. On the other hand, â€Å"multivariate analysis is statistical procedures which attempt to distinguish and measure the relative strength or significance of association between several independent variables and a dependent variables† (Buckingham and Sunders, p.292). The regression model formula: y= a + b1x1 + b2x2 + †¦ + e. Relationship between Class Identity and Political Interest The positive and significant coefficients on class identity in the bivariate models (I) indicate that middle-class identifiers display higher levels of interest. Generally, the stable middle-class identifiers are consistently and statistically significantly more interested in politics than are the stable working-class identifiers with p-value Relationship between Class Identity and Conception of Good Citizen as an Active Political Participant Middle-class identifiers are significantly more likely to emphasize the importance of being an active player in politics. In generation 2, the working-class mean drops from 0.53 to 0.41, whereas middle-class mean stays at 0.62. In model II, stable middle-class always more likely to mention active engagement than were stable working-class, 0.10 more in 1965 and 0.20 more in 1997. Relationship between Class Identity and Political Efficacy The bivariate models show the gaps between middle-class and working-class identifiers range from 0.14 to 0.21 for G1, and from 0.16 to 0.23 for G2 on the 0-1 efficacy scale. The dramatic decline of political efficacy happened from 1973 to 1992 (0.61 to 0.49) for G1 middle-class. On the other hand, for G2 slightly increase from 0.65 in 1982 become 0.67 in 1997. Relationship between Class Identity and Political Participation The constants and coefficients in the type I models demonstrate the expected life cycle: decreasing participation among parents as they become older and increasing participation among the offspring generation as they middle age. The type II models show that subjective identification becomes a less impressive predictor for G1 members as the coefficients on stable middle-class identity changed from 0.15 in 1965 to 0.09 in 1982. At confidence interval 95% and p-value (*p Limitation and Recommendation 3.1 Reliability and Outdated Neuman (2007, p.115) explain reliability as â€Å"dependability or consistency. It suggests that the same thing is repeated or recurs under the identical or very similar conditions†. This research seems like not reliable enough due to limited information given by the researchers. It may be difficult for the other researchers to conduct similar type of research in the future. The data which the researchers used derived from 1965-1997. Meanwhile, the research launched in 2004, and now 2014. Thus, this research could be less important to be used because the condition and preferences of people to participate on governmental affairs might be has been changed. 3.2 Sampling The representativeness is one of issues here. The researchers use data from their previous research. We can question, whether 1,669 students (with his parent) at 97 schools are representative? How about the distribution? Response rate from this study is just slightly more than a half of the total respondents, 935 of 1,669 or 56%. Then from 935 the researchers could only processed 728 data. Some of the reasons data not completed is because some respondents refuse to state their social class identity and geographical factor. It is easier for the researchers to focus on particular area or state. Besides that, the researchers can make shorter the range of research duration, only one decade for example. It is because not easy to keep the same and large respondents in long time period. 3.3 Data Analyzing The researchers not much explain about dependent variables in type II model, but from four those regression tables we know that, education and income are outstanding variables who make significant differences between working-class and middle-class. Research Good Practice and Contribution This study not only including subjective claim but also use objective indicator to determine of the social class of the respondents. Researchers effort to use panel study spanning two generations G1 and G2 for thirty-two years, allows the reader to check the trend of participation. The researcher concluded there is relevance of social class identification and participatory toward government affairs. It showed middle class with better income, education level, and occupation more contributive and active in political activity and civic engagement than working class. This research could be consideration for government to make strategies how to increase political participation like to vote in general election. Besides that, media, academician, and politicians could make effort to raise public awareness about how relevance and important of politics in their life. Conclusion To conclude this study, I would like to highlight back some points. Overall, it is a good research and useful in social political discipline. The researchers offered something new from previous research about social class and political participation. They used longitudinal research, compare two generations, and used objective measurement to determine social class. However, there are still few things need to be improved towards achieve a better result. Bibliography Agresti, A., Finlay, B. (1999) Statistical Methods for the Social Sciences. New Jersey: Prentice Hall. Babbie, E. (1990). Survey research methods . Belmont, CA: Wadsworth. Buckingham, A., Saunders, P. (2004)The survey methods workbook: From design to analysis. Cambridge: Polity Press Creswell, J.W. (2009) Research Design – Qualitative, Quantitative, and Mixed Methods Approaches 3rd edition, Thousand Oaks, California: Sage Publications Hagle, T. M. (2004) Dichotomous Variablesin The Sage Encyclopedia of Social Science Research Methods. Ed. Michael Lewis-Beck, Alan Bryman and Tim Futing Liao. Thousand Oaks, California: Sage Publications Jackman, M. R., and Jackman R. W. (1983) Class Awareness in the United States. Barkeley: University of California Press. Neuman, W. L. (2007) Basic of Social research methods: Qualitative and quantitative approaches. 2nd ed. Boston: Pearson Education. Walsh, K. C., Jennings, M., Stoker, L. (2004). The Effects of Social Class Identification on Participatory Orientations Towards Government.British Journal of Political Science,Vol.34, No.03, pp.469-495.