Quantitative Research Design

Introduction

Area of Interest and Statement of the Issue.  My primary area of research interest is the influence of sexism in media propaganda upon the self-esteem of Black American teen girls and young women.  Queries into propaganda research and the influence on gender and racist stereotypes have revealed no direct studies of the subject or any related subjects.  Let me be clear here.  There is an abundance of quantitative and qualitative research into the influence of stereotypes, of sexism and racism embedded in the English language (Sunderland, 2006), and even of the influence of sexist media programming and advertising upon women and (Sunderland, 2006).  However, I am unable to find any research that explicitly identifies this as unintentional propaganda (Doob, 1966) or as “penetration of an ideology by means of its sociological context” (Ellul, 1969). There is a gap between media and propaganda and between sexism and propaganda that this research hopes to connect and fill.

Theoretical perspectives.  The quantitative theory that would be most aligned with this particular topic is correlational modeling research where path analysis maps the relationships between a number of variables and displays the degree to which any of them can be used to predict one or more variables (Locke, et al, 2010).  In this theory, lines are given a direction of influence and the number given for each line indicates the degree of influence that has been exerted.  Swim, et al. (2004) utilize modelling to measure the association between Modern Sexist beliefs and identifying and engaging in subtle sexist behavior.  Connelly and Heesacker (2012) utilized structural equation modeling to explore the extent to which benevolent sexism is positively associated with life satisfaction.  The scales included benevolent sexism, hostile sexism, system justification and life satisfaction.  Oehlhof (2011) use structural equation modeling that took into account objectifying experiences, internalized experiences, and psychosocial outcomes related to self-objectification of overweight women.

The study of sexism in media propaganda has not been researched specifically via the influence of unintentional propaganda but it has been studied via the influence of media programming (Lafky, et al, 1996; Plakoyiannaki, et al, 2008, Plakoyiannaki and Zotos, 2009).  The manner that influence has been described in past studies is described as propaganda elsewhere (Doob, 1966) as well as sociological, cultural, and societally embedded (Ellul, 1969).

Formulating the Quantitative Research Question

Research Question.  The research question that could be investigated with this approach, “How are black teen girls’ and women’s unintentional propaganda (Doob, 1966, pp. 370-371; Ellul, 1969, pp. 62-65) social media postings about cultural artifacts (e.g., clothing, music, fashion) affected by mass media that uses sexist themes and may be defined as propaganda of one form or another?”

Variables.  The dependent variable will be the unintentional propaganda that participants use to influence others and are in turn influenced by.  Independent variables will include the use of primary social networks (Facebook, Twitter, Tumblr, Instagram), YouTube (for fashion and music videos), television programming, fashion and music magazines, websites popular with this audience, as well as news articles.  The control group will be a group of participants who not exposed to any of the above or below media artifacts.  The question is, though; given the nature of propaganda and in this case, unintentional propaganda, it is everywhere, should they be isolated from all media (impossible) or allowed to read, watch, listen to, and read what they want?  Even a control group will be influenced by the media artifacts around them whether the research study is controlling the exposure or not.  Additionally, while the negative effects of media artifacts are to be studied here, if there’s room and time within the research, I will add another independent variable that is not controlled by the research study itself.  If a large enough sample can be found among the participants who are interested in alternative to mainstream media cultures (Goth, Punk, etc.) they will be included to study an outlier to determine the influence they have on their peers as well as the pressure they are subject to, to conform.  The participants in this outlier group will determine the media artifacts that will be used (through questionnaires and/or focus groups). This portion of the study may reveal nothing definitive, but it may open up ideas for future studies.

Hypotheses (Null & Alternative).  The null hypothesis here is that there is no statistically significant degree of [unintentional propaganda] influence upon the study participants.  The alternative hypothesis will predict that the study participants exert a statistically significant degree of [unintentional propaganda] influence upon their peers and in turn are influenced similarly.

Asking the Quantitative Research Question

Quantitative Design.  Given my research interest is one where I am measuring the effects of several independent variables, namely specified media artifacts and news articles upon one dependent variable, the unintentional propaganda used by participants (Doob, 1966, pp. 370-371; Ellul, 1969, pp. 62-65) that they unintentionally use to influence other participants, the most appropriate research design will be correlational.  I will use a correlational statistical design to describe and measure the relationship between these independent variables (Creswell, 2014; Locke, et al, 2010) and their effects upon the dependent variable.  Correlationally, lines will have a direction of influence and the number given for each line will indicate the degree of influence that has been exerted upon the dependent variable.

Population and Sample.  The population in question is Black American girls and young women from the age of 12 to 19 (pre-teen to legal adult) who are actively savvy and participate in pop culture, whether perceived to as mainstream or alternative in one or more ways.  Ideally, my sample will include a cross section of socioeconomic backgrounds of this population through the selection of participants from specified school districts.  Accessing the tax rolls will be too time consuming and not as effective as other means.  And since affluence is still determined by neighborhoods and how those neighborhoods effect the tax revenue of the local school districts, this will be one of the deciding factor that will be used to determine the population and sample.  Recruiting that sample will be occur via social network postings (and perhaps, ironically, advertising propaganda, if budget allows), and contacting local school administrators formally.  What is in question, still, is how much information regarding the study should be revealed to the participants in order to convince them to participate without compromising the integrity of the research.  The selection of participants will be randomized to provide the largest cross-section within the selected sample.  The size of the sample will be determined by consulting a survey book to determine the appropriate margin of error that I will accept for this study (Creswell, 2010).

Instrumentation.  The instrumentation that I would use for the quantitative portion of this study would be a correlational analysis (Locke, Silverman & Spirduso, 2010) where I can examine the nature of the relationships between the selected variables.  Given the tentative nature of influence and the concepts we are discussing, the most realistic instrument for this analysis will be structural equation modeling in order to determine the degree and relationship among and between each of the variables.

Data Collection.  The data would be collected via electronic survey tallies and structured interviews for a small portion of participants and interviewer assistants who would help with tallying the results that would be plugged into software such as SPSS to analyze the results using a variety of statistical tests.

Answering the Quantitative Research Question

Quantitative data analysis.  The tests and measures that I propose for the quantitative study will be a combination of questionnaires and structured interviews that will be designed through a series of open-ended interview questions with a select group of pre-participants to determine the overall shape and form of quantitative instrumentation.  The data would be collected via electronic survey tallies and structured interviews for a small portion of participants and interviewer assistants who would help with tallying the results that would be plugged into software such as SPSS to analyze the results through structural equation modeling to make the results more tactile, visual, and visible as a valuable tool for future research. However, the sample I will select will not be large, probably under one hundred participants. And due to the nature of the study, I.e. unintentional propaganda in one on one and small-group situations (Doob, 1966, pp. 370-371; Ellul, 1969, pp. 62-65), it will be effective for the participants of this study to be located in one or two communities where the effects can be easily measured.  The steps involved in the data analysis Creswell (2014) would include, describing the number of respondents and nonrespondents, a bias tally of nonrespondents vs respondents, a descriptive analysis of the data, the usage of SPSS, the statistical test and its rationale, and presenting the results in the form of tables along with an interpretation of those results.  The discussion following would explain the data analysis in a manner that everyone could understand and other researchers could extract ideas for further future research.

Limitations.  There is a myriad of limitations of quantitative research associated with this study (Locke, et al, 2010).  Those limitations include replicating the results without stringent guideline in this original study or errors in how the sample was generated and even how the original questions were determined as well as the creation of the survey and the structured questions.  What will be key is keeping the language of the interviews and the surveys as clear, concise, and as neutral as possible.  To the extent that the results will reflect the experiences of the Black American community, the results may not be exclusive.  There may be elements of this research that are relevant to other women and girls of color, and there may be more than one area of the results that are relevant to women and girls in the larger population.  To what degree that will be is difficult to commit to at this point until the study is conducted and the results are tallied.

Ethical Issues.  Ethical issues for this study will include seeking university approval through the Institutional Review Board (IRB) as well as permission from several local site schools and participants, remaining sensitive to the needs of vulnerable populations (children and teens) and scrutinizing the study for any areas that may be inappropriate or damaging to said population.  To facilitate some and all of these ethical issues, I will also enlist a teen who would have made an ideal participant and explain the study and its goals so she may be able to analyze the study elements from a perspective that I am not able to and be able to determine that all participants are treated equally and fairly (Creswell, 2014).

 

Qualitative Research Design

Introduction

Area of Interest & Statement of the Issue.  My primary area of research interest is, the influence of sexism in media propaganda upon the self-esteem of Black American teen girls and young women.  Queries into propaganda research and the influence on gender and racist stereotypes have revealed no direct studies of the subject or any related subjects.  Let me be clear here.  There is an abundance of quantitative and qualitative research into the influence of stereotypes, of sexism and racism embedded in the English language (Sunderland, 2006), and even of the influence of sexist media programming and advertising upon women and (Sunderland, 2006).  However, I am unable to find any research that explicitly identifies this as unintentional propaganda (Doob, 1966) or as “penetration of an ideology by means of its sociological context” (Ellul, 1969). There is a gap between media and propaganda and between sexism and propaganda that this research hopes to connect.  The key here is the concept of unintentional propaganda as an influence upon girls and women via their peers.

Theoretical perspective.  The theoretical approach that I would take with this would be grounded theory where I would derive a general, abstract theory of the interaction grounded in the views of participants in the study (Creswell, 2014) to avoid categorizing the participants in this analysis before allowing the results reveal themselves and therefore selecting a theory before one actually manifests in the results.  Several studies on the subject of sexist and media influence on the self-esteem of women (Goffman, 1976; Holmes and Meyerhoff, 2008; Thiyagarajan, et al, 2012).

Formulating the Qualitative Research Question

Research Question.  My research question, in order to accommodate a grounded theory approach could be revised from, “How are black teen girls’ and women’s unintentional propaganda within social media postings about cultural artifacts (e.g., clothing, music, fashion) effected by mass media that uses sexist themes and may be defined as propaganda of one form or another?,” to one where I conduct prior very open-ended interviews with select individuals about their social media activity, what they respond to, what they post, and what they forward to their social networking groups.  From these interview results I can form a research question that is more in line with a grounded theory approach.  Given that I will be conducting open-ended interviews with the participants to determine the direction this study should take via grounded theory, the participants themselves will ultimately determine the variables involved in this study.

Formulating the Qualitative Research Question

Concepts and Constructs.  Recruiting the sample will occur via social network postings (and perhaps, ironically, advertising propaganda, if budget allows), and contacting local school administrators formally.  In this case, there will be no question regarding how much information of the study should be revealed to the participants in order to recruit them to participate since this is a grounded theory-participatory action research study.  The selection of participants will be randomized to provide the largest cross-section within the selected sample.  The size of the sample will be determined by consulting a survey book to determine the appropriate margin of error that I will accept for this study (Creswell, 2010).  The constructs involved in my research question will be developed via initial focus groups of Black American teen girls and young women who may participate as co-researchers in the impending study.  While speculative at this point, I expect to see the influence of media postings and cultural artifacts upon the participants.  I expect that influence to manifest itself in the forms of dress, makeup, and attitudes towards so-called appropriate behavior for a Black American girl and young woman.  This will be through “feminine” behavior, clothing, magazines, television programming, and advertising.  What will be less evident without further analysis with the participants will be the extent of the influence of White European Beauty Standards and White Patriarchy upon the self-esteem of the participants and how they really feel about that influence.

Asking the Qualitative Research Question

Population and Sample.  I prefer to enlist participants into my study to fully immerse my research into their experience.  For this study, the most effect means to do that is grounded theory utilizing participatory action research.  For this, I would derive a general, abstract theory of the interaction of the study participants’ views measured against a method that allows the participants to fully participate in the study research to fully understand what I am attempting to understand about their experience and to help them help themselves (Creswell, 2014).

While there are major obvious differences between quantitative and qualitative research, I don’t see a difference in the ways the population is determine or the sample is selected.  Since this is qualitative, however, it will be possible to decrease the size of the sample without destroying the integrity of the study or the sample since there is more total immersion in a shared outcome that the researcher and co-researchers participate in to effect social change in their immediate surroundings (Brydon-Miller et al 2003).  The population involved in the study will be the same, Black American teen girls and women from 12 to 19 years of age.  The sample will include groups from three general socioeconomic groups in or near three economically different school districts.  Those three will be separate from each to determine influence and results; however, there will be a fourth, the independent one as described above where alternatives to mainstream media are considered from the participants from across the three schools regardless of socioeconomic background. Participants will be recruited via Facebook internet ads, social network postings, and school presentations.

Data Collection and the Role of the Researcher.  I see the role of researcher as one where I am not directing and controlling the study (to avoid influencing the results thereby destroying the integrity of the study) but keeping it on course and guiding it towards its logical conclusion that allows the participant researchers to understand the subject a little better as well as exercising some degree of agency over an understanding of the subject (Kidd and Kral, 2005). Data would be collected in the form of focus groups, one initial, and then smaller ones that teams of two participant researchers conducted following some form of IRB training during the recruitment and initial focus group phase.  Additionally, open-ended interviews would be conducted and a form of data collection via social networks through participant researchers (also following some form of IRB training).

Answering the Qualitative Research Question

Qualitative Data Analysis.  The tests and measures that I propose for the qualitative study will be a combination of grounded theory and participatory action research where the researcher and co-researchers participate equally to effect social change in their immediate surroundings (Brydon-Miller et al 2003).  The population involved in the study will be the same, Black American teen girls and women from 12 to 19 years of age.  The sample will include groups from three general socioeconomic groups in or near three economically different school districts.  Ideally, I will attempt to find schools that are not far from each other to allow for some cross-pollinating of participants who at least are aware of each other from social events.  However, those three groups will be separate from each to determine influence and results; however, there will be a fourth, an independent one where alternatives to mainstream media are considered from the participants from across the three schools regardless of socioeconomic background. Participants will be recruited via Facebook internet ads, social network postings, and school presentations.

The steps involved in the data analysis (Creswell, 2014) will include a natural setting for the research to take place, researcher as key instrument where I, as the researcher, collect data through examining documents, observation, and interviews, use multiple sources of data, employ inductive and deductive data analysis, learn participants’ meanings of the issues at hand, allow the research process to emerge from the data, reflect on how my role and experiences shape my views of the research, and maintain a holistic account of the larger picture that emerges from the research.  The data that would emerge from the study would determine the research design as well as how it is to be used.  In explaining the design, it would be necessary to describe its origin, applications, a definition of its use in the current study, and how it shapes the overall study’s data collection, analysis, problem, and research questions.  The discussion that follows would explain the qualitative analysis in a manner that everyone could understand and other researchers could extract ideas for further future research.

Limitations.  Limitations in a qualitative study of this kind may include scenarios where very little data will be collected and where so much data is collected that I will have more than I can use in this study.  In the former instance, that will be avoided through extremely open ended, but guided, questions, in the focus groups and within the collaboration with participant researchers. In the latter instance, if there is an abundance of research there will be opportunities for writing several papers and perhaps opportunities for few more studies that give me an opportunity to apply for additional funding to continue the study.

Ethical Issues.  Some ethical issues I foresee are those of informed consent, where the parents of the participants decline to allow their children to participate.  However, I may plan to schedule meetings with the schools and the parents (especially the mothers or female guardians) of the girls whom I would like to participate in the study to explain the reasoning behind and the goals of the study.  I plan on a completely transparent line of communication between the participant researchers, the schools, the guardians and the study.  Additionally, as I indicated in the ethical considerations within the quantitative portion of the study, above, those specific considerations may have to be addressed here, too.

 

References

Brydon-Miller, M., Greenwood, D., & Maguire, P. (2003). Why action research? Action research, 1(1), 9-28.

Connelly, K., & Heesacker, M. (2012). Why is benevolent sexism appealing? Associations with system justification and life satisfaction. Psychology of Women Quarterly, 36(4), 432-443.

Creswell, J. W. (2014). Quantitative Methods. In Research design: Qualitative, quantitative, and mixed methods approaches (4th ed.). Thousand Oaks, CA: Sage Publications.

Doob, L. W. (1966). Public opinion and propaganda (2nd ed.). Hamden, CT: Archon Books.

Ellul, J. (1969). Propaganda; the formation of men’s attitudes. New York: Knopf.

Goffman, E. (1976). Gender advertisements. New York, NY: Harper & Row.

Holmes, J., & Meyerhoff, M. (Eds.). (2008). The handbook of language and gender (Vol. 25). John Wiley & Sons.

Kidd, S. A., & Kral, M. J. (2005). Practicing participatory action research. Journal of Counseling Psychology, 52(2), 187.

Lafky, S., Duffy, M., Steinmaus, M., & Berkowitz, D. (1996). Looking through gendered lenses: Female stereotyping in advertisements and gender role expectations. Journalism & Mass communication quarterly, 73(2), 379-388.

Locke, L. F., Silverman, S. J., & Spirduso, W. W. (2010). Chapters 6, 7, 8, 9. In Reading and understanding research (3rd ed.). Thousand Oaks, CA: Sage Publications.

O’Donohoe, S. (1994). Advertising uses and gratifications. European Journal of Marketing, 28(8/9), 52-75.

Oehlhof, M. E. W. (2011). Self-objectification among overweight and obese women: An application of structural equation modeling (Doctoral dissertation, Bowling Green State University).

Plakoyiannaki, E., Mathioudaki, K., Dimitratos, P., & Zotos, Y. (2008). Images of women in online advertisements of global products: does sexism exist?. Journal of business ethics, 83(1), 101-112.

Swim, J. K., Mallett, R., & Stangor, C. (2004). Understanding subtle sexism: Detection and use of sexist language. Sex roles, 51(3-4), 117-128.

Sunderland, J. (2006). Language and gender. Routledge.

Thiyagarajan, S., Shanthi, P., & Naresh, G. (2012). Viewers’ Perception of TV Ads: The Role of Sex Appeal. Psychology Research, 2(8), 452.