THE QUANTITATIVE RESEARCH PAPER 143
c h
a
P t
In a quantitative paper, numerical data are collected to answer a sociological question. Because quantitative research depends on specific techniques of data collection and analysis, this chapter (unlike the previous three chapters) may be most useful to students who have taken or are taking an introductory research methods and an elementary statis^ixuuxie^^^^
Mostquantitative papers areja^edjjr^^ is, the in-
vestigator, starting with a theory or with previousTesearcrT, "expects a certain answer to her or his research question. The investigator develops one or more hypotheses with the aim of predicting the results. However, some quantitative papers are~Dasecl or'i|^^aive^reasonirig^ The investigator, unsure of the answer but with some idea^J whartoTook for, sets out to explore a particular topic. Here the purpose Prescription rather than prediction. No matter which approach is taken, the data collected can be represented numerically.
The sample quantitative research paper that appears at the end of this chapter illustrates both types of logic. The student writer, Shannon Prior, uses the deductive approach in examining; (1) the relationship between race/ ethnicity and gender stereotyping, and (2) the relative frequency of the gender stereotype, objeciification, vis-a-vis other forms. But she uses the inductive approach in examining which race/ethnic group will be stereotyped the most and in what way. Basing her expectation on previous research, Shannon anticipated that women from all race/ethnic groups would be portrayed stereo-typically in magazine advertisements. She also estimated that the gender stereotype, objectification, would be the most common form for all three groups of women. However, her review of the literature did not suggest which racial/ethnic group would be gender-stereotyped most or which gender stereotype, other than objectification, would occur most frequently. Therefore, she was unable to make an educated guess as to the rate of occurrence of the six forms of gender stereotyping for each of the three groups of women.
Like most quantitative research papers, Shannon's is written in journal rather than essay format (see the section "Developing an Argument: Logic and Structure" in Chapter 1 for more information on these two formats) and 142
is divided into four major sections, and within the Methods section, two to three minor subsections. Briefly, the issues that should be covered in each section and subsection of a quantitative research paper include:
1. Review of the Literature. After a review of the relevant theory and literature, what sociological question do you feet needs to be addressed? What, if any, expectations (hypotheses) do you have about the answer?
2. Methods.
Sample. How did you select your sample? Measures. What measures did you use?
Procedure. What method did you use in trying to answer your question?
3. Results. What patterns of numerical data did you find?
4. Discussion. Do the data support or refute your hypotheses? What do your data mean? How do they relate to theory and/or previous empirical research?
Although the length of the four major sections is about equal for published papers utilizing sophisticated methods and analyses, the introduction and discussion sections for student papers are generally slightly longer than the other two sections. However, the relative length of the sections will depend on the amount of detail required by your instructor for describing your methods and results. With the exception of the Review of the Literature, which requires no heading, you should identify the other major and minor sections with a heading. The major headings should be in all capital letters, left-justified, and bolded; the minor headings should be left-justified and italicized and have only the first letter capitalized. In addition to these major sections and subsections, other important, components of your paper include the title, abstract, references, and appendix (which are discussed later in the chapter).
Since journal styles vary, ask your instructor which professional or scholarly journal format you should use. For example, the advice given in this chapter is based on the American Sociological Association Style Guide (2007) <http://www.asanet.org/>; however, many journals used by sociologists follow the procedures described in the Publication Manual of the American Psychological Association (2001) <http://www.apa.org/>. Both of these publications are available in the reference section of your library as well as available for purchase through the Web sites of each professional organization. It may be sufficient to simply follow the style used by the journal your instructor recommends. Before starting your paper, examine recent articles from the recommended journal. (If none is recommended, use the American Sociological Review, the top-tier journal published by the American Sociological Association). Download to a storage device, email, or photocopy one or two sample journal articles (as discussed in Chapter 4) to use as a model of format and tone. Your paper should not only look professional but sound professional as well. Scientific communication uses a formal prose style.
We have arranged the topics covered in this chapter according to the steps you should follow in writing your quantitative research paper. For example,
144
WRITING FROM VARIOUS DATA SOURCES
THE QUANTITATIVE RESEARCH PAPER
145
although the title and the abstract go at the beginning of your paper, you should write them during the final stages so that they describe your entire study. Therefore, we cover these components toward the end of the chapter.
REVIEWING THE LITERATURE
Once you decide on a topic, use the reference sources listed in Chapter 4 (such as CSA Sociological Abstracts and Google Scholar) to search for similar empirical studies of the topic. Also gather books and articles on the theory you plan to use as a framework for your study. Read and take notes as you would for a textual analysis paper or an article critique (see Chapter 5). That is, the purpose of the literature review is not simply to discuss what has been done on your topic but to determine what has been done well. Rather than accept the existing research at face value, students writing a quantitative research paper are trying to develop a justification for doing additional research. To do this, a student needs to find a weakness in the logic and/or methods of previous studies. Alternatively, a student might look for gaps in what is known. After reviewing what has been done, it should become evident what has not been done or what has not been done well By finding an area of research that has been understudied or one that has not been adequately studied, you will be able to justify new data collection. You will also be able to make a contribution to the literature. Thus, the purpose of the literature review is to provide a context for the formulation of your hypotheses.
In writing your review of the literature, provide enough background material to place the hypotheses in their proper setting. Begin your review with a summary of the theory (or theories) from which your question is derived, specifying its major tenets and focusing on one or two aspects (or major concepts) that you would like to test. Next, discuss each relevant study, summarizing in a few sentences the theoretical approach, major hypotheses, operational definitions (measures), and conclusions drawn in each one. It is often helpful to arrange the studies in chronological order. How do these studies fit together? Do they form a pattern or are they inconsistent? Do they fail to account for an important variable? What direction do they suggest for future research?
STATING THE PROBLEM AND CHOOSING A QUESTION
The statement of the problem reveals the gaps or contradictory findings that you found after reviewing the chosen theory and relevant literature. Its purpose is to point out theoretical inconsistencies in need of resolution, methodological problems apparent in the empirical literature, and/or the logical next step that research in this area should take. For example, you may want to concentrate on a different interpretation of a theory not adequately tested, set up a critical test of two rival theories, extend the theory to a new
population or substantive area, use a new operational definition of a concept, correct the faulty methodology of a previous study, use a different design or method, or include more variables in order to look for possible interactions. For some class assignments a simple replication of a published study may be sufficient. Be sure to check with your instructor. However, unless your study is an exact replication of an earlier study, you must explain how your study differs from previous works, how your study will extend their findings, and what your study will contribute.
Following from the review of the literature, the statement of the problem should suggest questions that need to be answered (for example, "Is education always related to occupational attainment?"). These questions can be refined and developed into hypotheses (for example, "If ethnicity is held constant, an increase in education will be associated with an increase in occupational attainment"). Most quantitative papers examine more than one hypothesis.
STATING YOUR HYPOTHESES
A hypothesis in a quantitative paper is the counterpart of the thesis in a library research paper. Each is a formal statement expressing the relationship you expect to find between two or more of your variables. Your literature review and statement of the problem show the logic that led to the development of your hypotheses. They serve as a sort of preliminary evidence. If your reasoning is sound, the numerical data you collect will provide further support for each of your hypotheses.
Each hypothesis should be stated in such a way that it can be unambiguously confirmed or rejected by the results. Each should also be stated in such a way as to make clear the type (causal or correlational) and direction (positive or negative) of the expected relationships. That is, does any particular hypothesis postulate that one variable causes the other, or does it simply state that the two variables are correlated? If the relationship is believed to be correlational rather than causal, are the variables expected to be related positively (as X increases, Y increases) or negatively (as X increases, Y decreases)? Explain how the independent and dependent variables will be operationally defined, that is, the manner in which the variables will be measured. For example, occupational attainment may be operationally defined as whether the job involves the supervision of other workers. (In correlational studies where there is no assumption about the causal order of the variables, no distinction is made between the independent and dependent variables.) Be sure to specify your unit of analysis, for example, individuals, groups, institutions, or countries.
As we explained earlier, if your study is purely descriptive you will not have specific hypotheses. You may simply have questions that suggest themselves as interesting problems in need of further investigation. In this case, you should explain why you feel an exploration of these topics is important.
146
WRITING FROM VARIOUS DATA SOURCES
THE QUANTITATIVE RESEARCH PAPER
147
For some topics, a descriptive study is often an important first step toward the formulation of a good deductive study.
DEVELOPING A METHODS AND ANALYSIS PLAN
Once you have drafted the review of the literature and stated your hypotheses, you are ready to proceed with the development of a methods and analysis plan. Collecting and analyzing quantitative data can be time-consuming tasks. Unanticipated problems or events may interfere. Start early in the term and apportion enough time for these and other tasks. Conducting a sound investigation is crucial to writing a good quantitative research paper. Therefore, it is necessary to consider in advance all the decisions you must make in collecting your data. Drawing up a methods and analysis plan will greatly improve the quality of your paper and will make the writing process go more smoothly. Although it is beyond the scope of this book to discuss the multitude of methodological and statistical factors that need to be considered in conducting a good quantitative study, we do address those issues important to writing a good report of your study. These tips should be useful to both the novice and the more experienced student. However, if you are currently taking a research methods or statistics course, or have taken one or both in the past, you might want to consult your text(s) for further details.
You will be limited to certain methods depending on the hypothesis you are testing or the kind of research question you are trying to answer. For example, if you are interested in the mortality (death) rates of upper- versus lower-class men, you obviously would have to use archival sources rather than a survey. Your choice of method will also be influenced by your assignment and by time and cost constraints. However, the four most common methods are archival sources, structured observation, experiment, and survey.
Archival sources are records of preexisting data. Although some of these data are obtained from surveys—the census, for example—we include them as archival because the results are published in tables available in government documents and books. Most of these data consist of official records of "rates," such as birth, death, marriage, divorce, crime, suicide, and accident rates. For example, you may want to examine the change in divorce rates from 1977 to 2007, the crime rates in urban versus rural areas, or the suicide rates of males versus females. For statistics about the United States, two excellent government publications are the Annual Statistical Abstract of the United States (available online at <http://www.census.gov/prod/www/ statistical-abstract.html>) and the Historical Statistics of the United States: Colonial Times to 1970 (available online at <http://www2.census.gov/prod2/ statcomp/documents/CT1970pl-01.pdf>), which are also available in most college and university libraries.
Structured observation can be conducted in a laboratory or in the "field" {real-world settings). Structured field observation, unlike ethnographic field
research, is guided by set hypotheses or specific measurement objectives. Structured observation, whether it is done in a laboratory or in the field, can often involve simple counting, such as counting the occurrence of certain behaviors or counting the number of people in different situations. For instance, you may want to observe the frequency with which men as compared to women make supportive statements during group discussions, or count the number of students who attend political rallies on versus off campus.
There are many different experimental designs, but the basic model involves two groups—an experimental group and a control group. Both groups are treated exactly the same except for the independent variable(s), which is (are) manipulated. Although we usually think of experiments as conducted in a laboratory, experiments can also be conducted in everyday settings (called "field experiments"). For example, you may want to examine whether people in a shopping mall are more likely to come to the aid of a well-dressed victim or a shabbily dressed victim. You could manipulate the situation so that in half the cases your confederate (accomplice) comes to the mall wearing a suit and in the other halí wearing dirty jeans and a torn T-shirt.
The survey method includes both questionnaires and interviews. The logic of the survey is to replicate the experimental method artificially, although without the same degree of control by comparing two or more groups. The groups can be based on response scores (for example, those who score high or low on a particular attitude measure) or on demographic characteristics (for example, Catholics and Protestants, blacks and whites, young and old, or high and low socioeconomic status). If you are interested in the different responses of males and females to a series of questions, your independent variable would be gender of respondent. In a survey, unlike an experiment, the independent variable is not manipulated. Instead, the researcher focuses on response differences that result irom the naturally existing differences in the respondents.
Whichever method you choose, be sure that your proposed research is in line with the guidelines set forth by the Institutional Review Board or Office for the Protection of Human Subjects on your campus (ask your instructor for details). You may need to get approval for your project from this committee before you collect your data.
Once you have decided on a method, draw up a plan for data collection and analysis to show your instructor. A methods and analysis plan ensures, before you collect the data, that your study will actually provide a test of your hypotheses. Further, it guarantees that you will be able to make sense of your data and analyze them successfully. Many students waste time and effort coUecting large amounts of data only to discover later that the data do not provide a test of their hypotheses. Or, they find that they don't know how to go about analyzing the data. A methods and analysis plan can prevent these problems.
When you write your methods and analysis plan, address those issues relevant to your type of study:
1. What population will you sample? How wili you select your sample? If you are conducting a field study, either a structured field observation or field
148
WRITING FROM VARIOUS DATA SOURCES
THE QUANTITATIVE RESEARCH PAPER
149
experiment, what setting will you choose? Do you anticipate any problems in gaining access to the respondents or the field setting? How many observations will you need to make? Witli the exception of an archival study, no matter which type of method you decide to use to collect your data, you need to specify how many respondents (of each type) you will need to question or observe. Describe relevant characteristics of the sample, such as the number of respondents of each race/ethnicity, class, and gender.
If you are doing an archival study, documents, rather than people, constitute your sample. For example, in the sample student paper at the end of this chapter, the sample is magazine advertisements. Just as with human samples, you must describe the important characteristics of your archival sample. Shannon's sample included advertisements in women's magazines targeted to three different race/ethnic groups (African American, Hispanic, and white).
2. What measures will you use? That is, how will you operationalize your variables? (To operationalize your variables means to define the specific operations, methods, or procedures you will use to measure your variables.; If you are designing an interview or questionnaire, what question(s) will you ask to measure each concept? For example, if one of the concepts you are interested in is traditional gender role altitudes, you might operationally define this as an affirmative answer (either "strongly agree" or "agree") to the statement "It goes against nature to place women in positions of authority over men." On the other hand. liberal gender role attitudes would be operationally defined as a negative answer (either "strongly disagree" or "disagree") to the same statement.
How long will the questionnaire or survey take to answer? If you plan on conducting an interview or survey, will you use closed-ended questions (also known as "fixed-response") or open-ended questions? Closed-ended questions compare to open-ended questions as multiple-choice exams compare to short-answer essay exams. What will be the possible range of the response scale (for example, a five-point Likert Response Scale) for the closed-ended questions?
If you plan to do a structured observation, whether it is in a field or laboratory setting, what exactly will you look for? How long should each observation last? What things will you want to have on your observation checklist (the list of things that you intend to count or measure)? For example, if you want to observe differences in how near people of different cultures tend to stand to one another, you might want to have a checklist that includes several different races/ethnicities and distances.
If you plan to conduct an archival study, you will need to go to the library to find out the types of data that are available to you. Further, you will need to determine the form in. which these data are presented. For example, if you are interested in comparing the birthrates of different religious groups, you will need to find out if the birthrates presented in the census tables are broken down by religious affiliation. Librarians can very often help you find what you need.
If you plan to conduct an interview or survey, you will need to develop a questionnaire. Even when conducting an experiment in the laboratory or in the field, you generally will want to interview respondents or have them complete a questionnaire at the conclusion of the experiment. Although it is beyond the scope of this book to discuss all the details of creating a sound instrument, there are general guidelines that you should consider in order to facilitate the writing of your report. Instructors often expect you to include a copy of your instrument in the appendix to your paper. It is best to include your instrument in your methods and analysis plan and to have it approved by your instructor before you collect your data.
In developing your measures, it is best to begin by looking at those developed by other researchers. Scales exist that have already been shown to be valid and reliable. Many of these scales are reproduced in the appendixes of books or journal articles; others appear in Measures of Social Psychological Attitudes by John Robinson and Philip Shaver of the Survey Research Center at the University of Michigan's Institute for Survey Research. If you use an existing scale, be sure to refer to the name of the scale and its originator in the body of your paper (for example, "Rosenberg's Self-Esteem Scale was administered to respondents") and to include the source of the scale in your list of references.
However, you might want to construct some original questions to use in conjunction with an existing scale or to modify existing questions to better suit the purpose of your study. In constructing your own questions, try to avoid the following pitfalls:
♦ Avoid using ambiguous terms or slang. Define the terms you use. For example, in the questionnaire on student dating, the term "date" is defined in question #7 (see Figure 6-1). Since dating patterns have changed over the years, there might have been some confusion as to what was meant by this term had this point not been clarified.
♦ Avoid "double-barreled" questions. Questions that contain "and" or "or" (such as "Do you feel that physical attractiveness or attitude similarity are important characteristics in a dating partner?") make it impossible to know whether the respondent views one or bath characteristics as important.
♦ Avoid biased questions that lead the respondent to answer in a socially desirable way. For example, rather than asking, "Have you ever had a date?" You might ask the question, "Have you dated in the last year?" Respondents may be reluctant to say that they have never dated. However, they may feel comfortable saying they have not dated recently.
With regard to other aspects of your instrument, consider the following:
♦ If it is not necessary to know your respondent's name, do not ask for it. Anonymous questionnaires, that is, those that don't ask for a respondent's name, are more likely to yield honest answers.
150
WRITING FROM VARIOUS DATA SOURCES
Figure 6-1 EXAMPLE OF A QUESTIONNAIRE
Questionnaire # Student Dating Questionnaire
The following questionnaire is anonymous. Your answers will be held in the strictest confidence.
Please circle your answers to the following questions. 1.  What is your gender?
a. male
b. female
Using the five-point scale (where 1 = not at all important and 5 = extremely important), please rate how important each characteristic is to you in a dating partner. e
2. Physical Attractiveness
3. Personality
4. Sense of Humor
5. Potential Occupational Success
6. Attitude Similarity
Not at All Important
Extremely Important
5
5
5
5
5
7. Have you dated in the last year (that is, gone out with someone of the opposite sex, or of the same sex, for purely social purposes with the possibility of developing a romantic involvement)?
a. yes
b. no
8. What is your sexual orientation?
a. heterosexual
b. homosexual
c. bisexual
d. prefer noi to answer
Thank you for taking the time to complete this questionnaire. Your cooperation is greatly appreciated!
THE QUANTITATIVE RESEARCH PAPER
151
♦ Provide adequate instructions about how to answer the questions. For example, the instructions ("rate how important each date characteristic is to you") let respondents know they are being asked for their own opinion, not the opinion they believe to be held by their peer group.
♦ Number each question in the questionnaire. Space questions out on the page so that they are easy to read.
♦ Be careful about the order in which questions are listed. Put easy questions first and difficult or sensitive questions last.
♦ Show respect for your respondents. Thank them for their cooperation and retain the confidentiality of their responses, that is, do not reveal any information that could be used to identify them, including their names, if the questionnaire was not anonymous.
Alternatively, your assignment may allow you to use not only existing measures developed by other researchers, but their data as well. This is called secondary data analysis. Many large universities subscribe to the quantitative data library service provided by the Interuniversity Consortium for Political and Social Research {ICPSR). One of the most commonly used data sets provided by the ICPSR is the General Social Survey (GSS) collected by the National Opinion Research Center (NORC). Check with your instructor about this data library. If ICPSR serves your university, and if its use is acceptable for the purposes of your assignment, your task for the methods and analysis plan would involve the selection of an appropriate database, and, within that, the selection of specific questions to be used in your analysis.
3. How will you get the data into an analyzable form? For example, have you assigned an appropriate numerical equivalent (low = 1, high = 2) to each of the response categories of closed-ended questions? Have you developed a coding scheme for open-ended questions? For example, your coding scheme could involve counting the number of respondents who made some reference to social mobility in response to an open-ended question, or counting the number of times different types of respondents mentioned themes of alienation. Remember that in a quantitative paper, you must be able to represent all responses numerically.
If you are an advanced student planning on forming an attitude index or scale from a set of closed-ended questions, will the response scores of any of your questions need to be reversed? That is, before adding together the response scores of several questions to form a single scale, will the response scores of negatively worded questions be reversed so as to bring them in the same direction as positively worded questions? Will you leave the index as a continuous variable or will you divide it at the median so as to compare high and low scorers?
4. How will you analyze the data? Depending on your hypotheses and the level of statistical knowledge required for your assignment, there are different ways to do this. If you haven't taken a statistics course, the two simplest ways
152
WRITING PROM VARIOUS DATA SOURCES
THE QUANTITATIVE RESEARCH PAPER
153
to analyze your data would be to calculate percentages or averages on each variable, independently of other variables. Independent percentages or averages are adequate for reporting the results of a descriptive study.
However, in testing hypotheses it is usually necessary to look at the relationship between two variables. The complexity of calculating percentages or averages increases when you examine the relationship between two variables, because the variables must be examined jointly. Some common methods of doing this include constructing a cross-tabulation table (see Table 1 in the sample quantitative paper), a table of means (see Figure 6-2), and a correlation matrix {not shown). In each type of table, one variable is designated as the row variable and the other as the column variable.
Although constructing a correlation matrix is most feasible for advanced students, beginning students may be able to calculate a cross-tabulation table or table of means (averages). For example, in Table 1 of the sample paper Shannon uses race/ethnicity as the column variable and type of gender stereotype as the row variable. In preparing the table, Shannon first sorted the magazine advertisements into three piles based on the race/ethnicity of the women (African American, Hispanic, and white). Separately far each pile, she counted the number of times each of the six gender stereotypes was found. She repeated this procedure for each pile, producing a set of frequencies to be used in calculating the percentages. Notice at the top of each column in Table I that Shannon specifies the number of advertisements analyzed for each of the three racial/ethnic groups: African American (N -10), Hispanic (N - 10), and white (N = 10). This lets the reader know that the denominator used in calculating the percentages for each pile is 10. For example, Shannon found that out of the 10 magazine advertisements featuring African American women, 4 used the gender stereotype objectification (4/10 - 40%). At the bottom of each column. Shannon includes the total percentage so that the reader knows how to read the table (for example, 10% + 10% + 10% + 20% + 10% + 40% - 100%, indicating that the column variable is the independent variable).
In Figure 6-2, we present a table based on a calculation of means or averages instead of percentages to show the relationship between two variables (gender and desired date characteristics). We use date characteristic as the row variable and gender as the column variable. We designate in parentheses the number of male and female respondents. In preparing this table, we sorted the questionnaires into two piles: one for males and one for females. Then, separately for each pile, we added together the numerical scores given by every respondent for each date characteristic. We divided each sum by the number of students in the pile. For example, the responses of the 15 males in pile 1 for the date characteristic "physical attractiveness" summed to 69 (5 + 4 + 4+ 5 + 4+ 5 + 4+ 5 + 5 + 5+ 4+ 5 + 5 + 4+ 5 = 69). We divided 69 by the number of the respondents in the pile to obtain the average score for males (69 + 15 = 4.6). We repeated this procedure for each date characteristic. We
Figure 6-2 SAMPLE TABLE OF MEANS
[ Table 1. Mean Level of importance of Five Date Characteristics by Gender
	Males	Females
	(N - 15)	(N = 15)
Physical Attractiveness	4.6	3.1
Personality	3.0	3.9
Sense of Humor	2.5	3.2
Potential Occupational Success	1.4	4.5
Attitude Similarity	2.2	2.4
then did all the same calculations for females. The table reports the mean level of importance of each date characteristic for males and females separately. The results show that males (mean = 4.6) place a greater importance on physical attractiveness than do females (mean = 3.1), while females (mean = 4.5) place greater importance on potential occupational success than do males (mean - 1.4).
Try to make a mock table for analyzing and presenting your results. That is, try to specify which variable you will use as your row variable and which variable as your column variable. Which variables will you use to sort respondents into piles? Determine whether the numbers in the cells will be percentages or averages.
If you are required to carry out more sophisticated statistical analyses of your data, determine the level of measurement of your variables (nominal, ordinal, interval, or ratio). This will allow you to decide which statistical tests can be appropriately calculated. Computer software packages that calculate social science statistics, such as SPSS and SAS, are available. Check with your instructor.
Don't proceed with your data collection or analysis until your instructor has approved your methods and analysis plan and answered your questions. Once you complete your data collection and analysis, you are ready to begin writing the other sections of your paper.
WRITING THE OTHER SECTIONS OF YOUR PAPER
THE METHODS SECTION
The methods section directly follows the review of the literature. It should contain three subparts: sample, measures, and procedure. Each subpart should be labeled with an italicized heading at the left margin; capitalize only
154
WRITING FROM VARIOUS DATA SOURCES
THE QUANTITATIVE RESEARCH PAPER
155
the first letter. (See pages 167 and 169 of Shannon's paper for examples of these headings.) Begin the methods section by describing your sample.
Describing Your Sample
Specify the population studied. Discuss in detail how you selected your sample from this population. Did you randomly select respondents—that is, give every member of the population an equal chance of being included in the sample—or did you select whomever you could get? If respondents were randomly selected, describe the steps you took to ensure randomness (for example, tossing a coin or systematically selecting every fifth respondentl. If you are a more advanced student, did you stratify your sample on any particular variable?
Describe all the relevant characteristics of your sample (for example, age, gender, race). If you had to eliminate any subjects because of incomplete data or for other reasons, state the number and the reason. Specify the final overall sample size and the size of each group.
Describing Your Measures
If you obtained your data irom secondary sources, describe where the original data came from and how they were measured. If you did a structured observation, describe the behaviors, types of people, situations, and so on, that you observed.
If you used a questionnaire or interview, state whether you used closed-ended or open-ended questions, questions newly developed by you, or questions adapted from previous research. If you used existing scales or indexes, include information about their validity and reliability, if available. "Validity" refers to the extent to which the questions actually measure what they are supposed to measure. "Reliability" refers to the stability of measurements taken at different times.
In the body of the paper, quote the actual question(s) used to operationally define each variable. If several questions were used, as in the construction of an index or a scale, give a sample of the questions and include the others in a table or an appendix. For example: "Gender role attitudes were measured by agreement-disagreement with 20 statements, such as 'The woman's place is in the home' and 1 would vote for a woman presidential candidate' (see Appendix A)." If the questions were closed-ended, state the range of the response scale and describe the anchor points. For example, you might state that you used "a five-point Likert Response Scale ranging from (1) not at all to (5) extremely." If you averaged several questions together to form an index, state what the high and low scores on the index signify. For example: "A high score on the gender role index indicates liberal gender role attitudes; a low score indicates conservative attitudes." If the questions were open-ended, describe the coding scheme (hat you used to code the data.
Describing Your Procedure
Identify the method you used. Describe when (time of day, day of week, date), where (the geographic location, type of institution, building), and under what circumstances the study took place. This information is especially important if the study was conducted in a field setting.
If you conducted an experiment, be sure to also specify the design. Discuss the procedure by which the independent variable^) was (were) manipulated and the instructions given to respondents in each group. Specify any additional precautions taken to control extraneous variables or to exclude bias from your sample. For example, did you randomly assign respondents to experimental conditions? (Random assignment to groups is different from random selection.) If you employed confederates (accomplices), describe who they were, what they did, and whether or not they were kept "blind" to (ignorant of) the hypotheses.
Whichever method you chose, summarize each step you took in collecting your data. A good rule of thumb is to describe your methods in enough detail that another researcher could replicate your study.
THE RESULTS SECTION
Discuss how you examined the relationship between your variables. Did you count the number of people who gave each type of response, or did you average the scores of several people? If you calculated percentages or took averages, state the number of people used as the denominator in your calculations.
If you have a large amount of data to report, consider displaying it in a table or figure. Put each table or figure on a separate page at the end of the paper, just after the list of references. Each table should be numbered consecutively. The word "Table" and the number should be bolded and end with a period and be followed by an indented descriptive title. The line consisting of the table number and descriptive title should be flush with the left margin, and it should be followed by a double rule, as shown in our Figure 6-3. A
Figure 6-3 SAMPLE TABLE OF PERCENTAGES
Table 1, Gender Role Attitudes by Gender of Respondent
	Gender of Respondent	
	Male	Female
Gender Role Attitudes	(N - 100)	(N = 100)
Liberal	55%	85%
Traditional	45	15
TOTAL	100%	100%
156
WRITING FROM VARIOUS DATA SOURCES
good title allows the reader to tell what is in the table without having to refer to the text. For figures, the number and descriptive title should also be flush with the left margin and bold, but they should be positioned below the graph or diagram, as shown in Figure 6-4. Figures should also be numbered consecutively. In the body of the paper, refer to each table or figure by number; then explain it. Remember that the numbers presented in the table never speak for themselves.
Another way to present your results is to use graphs or charts. You can use one of the many computer graphics programs that are available. The best known of these programs for the advanced student are SPSS (formerly known as the Statistical Package for the Social Sciences) and the SAS (Statistical Analysis System), which can also be used to perform statistical analyses. Programs such as Microsoft's Excel can be most useful for the less advanced student. Often visual aids can dramatically illustrate the relationships between variables. Whether you should use a bar chart, a line graph, or a causal model will depend on the type of data that you have and the analyses you perform. Ask your instructor for suggestions on the best way to present your results.
If your assignment required statistical analyses, state the statistical tests performed; their critical values, degrees oi freedom, and significance levels; and the direction of the results. For example: "The relationship between gender role attitudes and gender of respondent is reported in Table 1. The results of a chi-square test indicate that a significantly greater proportion of females (85%) than males (5 5%) hold liberal gender role attitudes (x2 - 24.24, d.f. = 100, p < .01)." In the discussion section, you will take this explanation a step further.
THE DISCUSSION SECTION
In the discussion section you should lie your results back in to your hypotheses. Did the data support any of your hypotheses? Remember that the statistical significance of your findings does not indicate the theoretical, substantive, or practical significance of your findings. The latter is a judgment you must make in the discussion section. Whai does a relationship between X and Y mean in the larger theoretical context? How do your findings corn-
Figure 6-4 SAMPLE FIGURE
Income-
:Gender role attitudes
Education
Figure 1.   Measurement Model oi Gender Role Attitudes
THE QUANTITATIVE RESEARCH PAPER
157
pare with previous research? That is, are your findings consistent or inconsistent with those found by other researchers? What has the study contributed to the existing body of literature on this topic? What are the practical implications of your findings, if any? What ethical issues were raised?
What is the internal and external validity of your study? That is, to what extent does your study provide an adequate test of your hypotheses? To what other populations can your findings be generalized? Discuss any methodological or design flaws, particularly if your hypotheses are not supported. Make suggestions for improving future research. If the study is methodologically sound, how can you account for your unexpected findings? Do the data support an alternative theory?
What conclusions can you draw? What direction should further research
on this topic take? THE TITLE
Now that you've completed the main sections of your paper, you will be able to come up with a good descriptive title. It should be short (rarely over 12 words) and include the theoretical perspective taken and/or the major variables examined (both independent and dependent, where appropriate). (See Part 3 for guidelines on formatting a title page.)
THE ABSTRACT
The abstract, usually about 100 to 200 words in length, is a very brief summary of your paper. It describes the problem, methods, sample, results, and conclusions of your study and should contain only ideas or information already discussed in the body of the paper. The abstract goes on a separate (labeled) page right after the title page. For a heading, type ABSTRACT (in capital letters) and left-justify it. Triple-space between the heading and the body of the abstract. Indent the first line. (See page 163 of the sample paper for an example of an abstract.) Although almost always included in a journal article, an abstract may not be required by your instructor.
THE LIST OF REFERENCES
If no specific journal style is required, follow the guidelines given in Chapter 3 for formatting your list of references. Include in your list only those sources you actually cite in the body of your paper. The list of references should appear on a separate page at the end of your paper (see page 179 for an example).
THE APPENDIX
The appendixes (also called appendices) are optional. Some instructors may want you to include your questionnaire, observation checklist, instructions to
158
WRITING FROM VARIOUS DATA SOURCES
THE QUANTITATIVE RESEARCH PAPER
159
respondents, raw data, statistical calculations, or other items in an appendix Each appendix should be numbered or lettered (see Part 3) and given a de' scriptive title. The appendixes, when included, go after the list of references
Babbie, Earl R 2007. The Practice of Social Research. 11th ed. Belmont, CA- Thomson Wadsworth Publishing Co.
Babbie, Earl R^ 2008. The Basics of Social Research. 4th ed. Belmont, CA: Thomson-Wadsworth Publishing Co.
Babbie Earl R Fred Halky, and Jeanne S. Zaino. 2007. Adventures in Social Research 6th ed. Thousand Oaks, CA: Sage Publications, Inc.
The following sample quantitative research paper, a study of gender stereotyping in magazine advertisements, was written by Shannon Prior for an undergraduate course in writing for sociology. Shannon's study reveals that women are portrayed stereotypically in the advertisements of women's magazines, regardless oi their race or ethnicity.
Shannon's assignment required her to review the literature on a topic, design a study, and construct measures; however, she only had to simulate rather than collect the data. (She included this simulated "raw data" in Appendix B [not shown], so that her instructor could verify that she tabulated
the results correctly). Because Shannon wrote this paper for a class in writing rather than in research methods, the purpose of the assignment was to demonstrate critical thinking about a sociological topic all the way from the formulation of a research question and/or hypothesis to the analysis, interpretation, and discussion of findings. Thus, there was less emphasis on the mechanics of collecting data than on the logic behind the entire research and writing process. If your assignment entails the collection of data, it would be prudent to share your methods and analysis plan with your instructor beforehand, to ensure you're on the right track.
Shannon reviewed the relevant literature on stereotyping in magazine advertisements and developed her main hypotheses to address questions raised by this review. Shannon went beyond other studies by looking at gender stereotypes of African American and Hispanic women, in addition to white women, in magazine advertisements. Thus, this work is located within an ongoing discussion but also contains something original, both of which are important considerations in designing sociological research.
The paper follows a journal article format and is based on simulated quantitative results from an archival study. Notice that it contains the major and minor sections discussed in this chapter; title page, abstract, review of the literature, methods (including subsections on the sample, measures, and procedure), results, and discussion. Shannon also provides a table that presents important results in an easily read format. Her list of references appears toward the end of the paper, but it should have been placed just before her table and appendixes. Our comments on the facing pages detail other important features of the paper as well as ways in which it could be improved.
THE QUANTITATIVE RESEARCH PAPER
Because of the length of her paper, Shannon includes a title page (as suggested in Part 3). Shannon correctly begins numbering the pages of her paper after the title page.
Shannon produces a descriptive title that appropriately identifies the dependent variable (stereotypes) and the independent variable (race/ethnicity), and hints at the method (examining magazines). (Note that this reference to magazine advertisements implies that she uses archival data to examine the relationship between the independent and dependent variables). However, this title is ambiguous. It does not identify the kinds of stereotypes investigated, whether these stereotypes are racial/ethnic stereotypes, class stereotypes, or gender stereotypes. Since Shannon's paper is on gender stereotypes, this topic of gender stereotypes should be specified in the title.
STEREOTYPES OF AFRICAN AMERICAN, HISPANIC, AND WHITE WOMEN IN MAGAZINE ADVERTISEMENTS
Shannon Prior Sociology 301 Dr. Giarrusso December IS, 2006
THE QUANTITATIVE RESEARCH PAPER 163
OUR COMMENTS
Shannon includes an abstract or brief summary of her paper. In less than 200 words she describes the purpose of the paper, the methods, and the results. Because Shannon follows the ASA Style Guide (2007), she left-justifies, bolds, and capitalizes all the letters of the heading. However, she begins the fourth sentence with a number; instead, she should write it out, "One hundred and twenty advertisements. . . " Also, she could strengthen her abstract with the addition of a concluding sentence, such as "Two explanations that could account for die present findings are 'the beauty myth' and old fashioned economics."
ABSTRACT
Research by Lindner (2004) indicates that women have been consistently stereotyped in magazine advertisements since the 1950s. However, her research does not address how women of different races are stereotyped. A content analysis of African American, Hispanic, and White women's magazine advertisements is conducted to compare the stereotypes of each group. 120 advertisements are analyzed using the coding categories relative size, function ranking, licensed withdrawal, body display, location, and objectification. The results show that all groups of women are stereotyped. Advertisements of white women are slightly more likely than advertisements of other women to contain licensed withdrawal and body display. Objectification occurs most frequently in all groups.
THE QUANTITATIVE RESEARCH PAPER 165
j OUR COMMENTS j _
Appropriately, Shannon begins the review of the literature section on a separate page without a heading. At the outset of the paper, she provides a state- : ment of the problem: advertisers often use stereotypes to sell products. This paragraph creates a context for the rest of the paper, explaining why it is important to study the use of stereotyping by advertisers. However, at this point it is still unclear whether Shannon will be focusing on gender stereotypes or racial/ethnic stereotypes.
Shannon provides a concise description of a relevant empirical journal article that includes key information about the: (1) independent variable (race/ethnicity), (2) dependent variable (types of roles in which minorities were cast), (3) method (archival), (4) sample (four types of magazines— business, women's, general interest, and technical), and (5) results. Importantly, she ends the paragraph with a critique of the article, setting the stage for her study to make a contribution to the literature. However, her description of the study would be stronger if she indicated the gender of the individuals studied in the advertisements—it is unclear whether the study by Taylor, Lee, and Stern is based on males and/or females. Also, this is the first time Shannon suggests she is interested in studying gender role stereotypes rather than racial/ethnic stereotypes. The paper should have prepared the reader beforehand to expect this emphasis. Moreover, the reader still doesn't know at this point whether Shannon studied the stereotyping of one or both genders.
Again, Shannon provides a concise description of an empirical journal article. Without using the actual terms, she specifies the independent variable (time), dependent variable (gender role stereotypes), method (archival), sample (mainstream magazines), and results (stereotypes remain stable over time). Once more, her critique of the study (the findings cannot be generalized to minority women) shows critical thinking skills. Further, she is careful to give the operational definitions of the gender role stereotypes that are studied (for example, objectification is "when the woman's sole purpose is to be looked at"). However, because Shannon uses the ASA publication style, she should not include Lindner's first name in her citation.
The power of advertising is clear. The fact that companies annually expend over $200 billion on advertising is a testament to its power and prevalence in our society (Kilbourne 1999). Advertising not only sells products, it also sells images (Baker 2005). Too often these images are stereotypes of women and minorities. These portrayals, good or bad, influence our perceptions of the people in our society. For these reasons, the study of minorities and women in advertising is the focus of this research.
Taylor, Lee, and Stern (1995) explore the portrayal of minorities in magazines by conducting a content analysis of four types of magazines (business, women's, general interest, and technical). The researchers analyze the types of roles (major, minor, background, etc.) minorities are cast in and the types of products they are selling. The authors conclude that some stereotypes still persist. African Americans are over-represented in minor or background roles. Asian Americans are consistently represented as tech-savvy workaholics while Hispanic Americans are either non-existent or somewhere in the background. Although the study includes women's magazines in their content analysis, they do not provide a specific analysis of gender role stereotypes.
To gain a better understanding of gender portrayals in advertising, I consulted Katharina Lindner's (2004) longitudinal study of women in Time and Vogue magazine advertisements. The study utilizes Goffman's frame analysis and coding scheme, which is designed to detect subtle messages in advertising. Lindner finds that 78% of the advertisements portray women stereotypical^. The study finds decreases over time in certain categories of stereotypes. For instance, the feminine touch category in which the woman depicted is touching her hair, face, or clothing in an unnatural manner or stroking an object declines throughout the years. However, Lindner was surprised to find that the other types of stereotypes measured remain relatively stable over time. Overall, objectification, when the woman's sole purpose is to be looked at, occurs most frequently. Although this study provides insight into gender portrayals, it only includes publications directed toward a mainstream, predominantly White audience. Therefore, the results do not apply to minority women.
l
THE QUANTITATIVE RESEARCH PAPER
167
OUR COMMENTS
Shannon provides another succinct description of an empirical journal article, clarifying how the dependent variable in this article differs from the other articles she reviews. It appears that the Baker article is not as relevant to her topic as the previous article by Lindner; however, there may not have been another article that more closely related to her topic. Notably, Shannon shows the limitations of the article in addressing her research question, thereby adding to her argument for the need to study gender stereotypes among women from different race/ethnic groups. Because Shannon uses the ASA publication style, she should not have included Baker's first name or middle initial in her citation.
This reference to the source being cited is correct according to ASA Style Guide.
This paragraph ties all the studies reviewed together leading to the conclusion that a study of gender stereotypes among African American, Hispanic, and white women in corresponding women's magazines will make a contribution to the literature. A common error is capitalizing the word "white." According to the ASA Style Guide, the words "black" or "white" should not be capitalized; however, it is appropriate to capitalize the words "African American," "Asian," and "Hispanic" because these names represent groups with a common geographic location or language.
The final paragraph in the review of the literature section is where most researchers propose their hypotheses. Appropriately, Shannon offers two hypotheses, each about the frequency with which she expects various gender stereotypes will be found; she is able to provide "educated guesses" about the likelihood of these findings because there is adequate literature on this aspect of her research in the literature. These hypotheses constitute the deductive component of her research. Conversely, the inductive component of her research deals with "which race/ethnic group will be stereotyped most frequently" and "what ways." Because there is not enough literature on these topics for her to render hypotheses, she poses these as research questions.
Shannon divides her paper into four major sections. She indicates the beginning of the second section with the heading, METHODS, all in capitals, bolded, and left-justified. She then goes on to divide this major section into several subsections: sample, measures, and procedure.
In contrast to Lindner's study, Christina N. Baker (2005) studies the portrayal of African American and White women's sexuality in advertising. She conducted a content analysis of African American and White-oriented men's and women's magazines to compare the portrayals of African American and White women's sexuality. Included in her study is an analysis of the role of women and the authority of the women in relation to the men. In women's magazines, Baker finds that African American women are more likely to be portrayed as independent compared to White women. White women are more likely to be portrayed as "partners" or have their faces hidden, signifying subordination. However, the results are not generalizable because Baker only includes advertisements that display women in a sexual manner.
It is clear from the previous research that White and minority women are stereotyped in magazine advertisements. These stereotypes are not simply images, they are powerful messages that "depict for us not necessarily how we actually behave as men and women but how we think men and women behave" (Gormck 1979:7). Therefore, I wish to further explore the types and frequencies of gender stereotypes in magazine advertisements. I contribute to the literature by examining and comparing the stereotypes of African American, Hispanic, and White women in African American, Hispanic, and White oriented women's magazines.
Based on the literature reviewed, I hypothesize that all groups of women will be stereotyped. Based specifically on Lindner's (2004) study, I hypothesize that objectification will occur most frequently out of all the stereotype categories. Due to the lack of research on the topic, I explore rather than hypothesize which racial group will be stereotyped most frequently and in what ways.
METHODS
Sample
A content analysis is conducted to examine and compare the portrayals of African American, Hispanic, and White women in magazine advertisements. Since minority women are underrepresented in mainstream magazine advertising, I analyze African American and Hispanic-oriented
OUR COMMENTS
The word "amount" is used incorrectly here because "women" is a count noun. The correct term is "number."
Although for most research studies, the sample refers to people, in an archival study, the sample refers to existing documents. Importantly, Shannon specifies how she selected her sample (magazines): she purposefully chose magazines that would provide her with a close-to-equal representation of women from each race/ethnic group. Moreover, she shows her critical thinking skills by selecting the January and June issues to control for possible seasonal biases in the advertisements.
Again showing her critical thinking skills, Shannon integrates the coding schemes of different researchers in examining her hypotheses and answering her research questions. She also provides an operational definition of each type of stereotype to be investigated in an appendix at the end of her paper. While many students believe they will impress their instructor by developing new measures, it is best to use existing measures unless your instructor tells you otherwise. Better to use "tried and true" measures, especially if they are based on classic studies like those of Goffman, than risk using inadequate or inappropriate measures. Because Shannon includes more than one appendix, she assigns a letter to each one (as suggested in Part 3), designating the first one as Appendix A.
Shannon provides further specification about how the advertisements are selected and coded. To her credit, she trained two coders to examine the data and made sure that there was high interrater reliability before having the coders analyze the rest of the advertisements. The fourth sentence of the paragraph begins incorrectly with a number. Shannon could have eliminated this error by putting a semi-colon at the end of the previous sentence and joining them to make one sentence or by writing out the number in words: "Twenty-five percent/'
Shannon uses the major heading, RESULTS, to delineate another major section. She refers the reader to Table 1 for the results and then goes on to explain the content of the table. Although Shannon uses the word "significant," she does not mean the term "statistically significant/' which means that the differences found could not be attributable to chance. To avoid any possible ambiguity here, Shannon should have substituted the word "noteworthy" or "noticeable" when she was revising her final draft.
THE QUANTITATIVE RESEARCH PAPER
169
magazines in addition to a mainstream or White-oriented magazines, i do so in the hope of obtaining a close to equal amount of Black, Hispanic, and White women. I chose Redbook for White women, Latina for Hispanic women and Essence for Black women. Aii of the magazines contain similar topic matter including sections on beauty, fashion, sex, and health. The sample of advertisements is taken from the January and June issues published in 2005. 1 chose a winter and summer month to control for any bias that one season might create. For example, summer months would most likely contain more advertisements with body display.
Measures
Goffman's (1997) coding scheme categories relative size, function ranking, and licensed withdrawal are used to analyze the advertisements. In addition, Umiker-Sebeok's (1996) category location, Lindner's (2004) category objectification, and Kang's (1997) category body display are used. See Appendix A for the full list of the categories including their operational definitions.
Procedure
Only full-page advertisements containing at least one woman with or without one or more men were coded. At least one woman in the advertisement had to appear to be Black, Hispanic, or White. Other minority women were excluded from this study due to the lack of a comparable magazine that targets them. Two coders were trained to use the coding scheme and were not informed of the exact purpose of the study. 25% of the sample was analyzed by each coder. The results were compared and inconsistencies were resolved. Each coder then analyzed half of the sample.
RESULTS
Table 1 displays the percentage of each gender stereotype by race/ethnicity. I found that African American, Hispanic, and White women are all consistently stereotyped. There were significant differences in the frequencies of stereotypes based on race with the exception of the
3
THE QUANTITATIVE RESEARCH PAPER
171
OUR COMMENTS
Shannon explains how she derived the percentages presented in the table. She also provides the raw data from which she constructed the percentages in an appendix at the end of the paper (not shown); she distinguishes between the two appendixes by assigning the second appendix a different letter ("Appendix B").
Here Shannon provides further description about how she obtained the percentages presented in the last column of Table 1, and she talks about how the gender stereotypes differ in their frequency. Shannon could have avoided beginning the fourth sentence with a number (13.3%) by combining it with the previous sentence. In that case, she would need to be very careful to avoid a common grammar error of run-on or comma splice sentence. A correct version would be "Location occurred in 13.3% of the advertisements, while 13,3% of the advertisements contained the stereotype licensed withdrawal"
Students need to be careful about the use of the word "significant" in quantitative papers. Although Shannon does not use the word "statistically significant," the reader may not be sure whether she understands the difference. A better way to note the differences in occurrence among the types of gender stereotypes when statistical tests were not calculated would be to say that "large differences" were found.
Shannon should avoid beginning the sentence with a number (0%).
Shannon uses the major heading, DISCUSSION, lo divide her paper into the last major section.
To avoid the confusion of the reference to "we," Shannon should have omitted it here by revising the statement to "Whether acknowledged or not, advertising. ..."
Be careful about using the pronoun "this" ("This was consistent with Lindner's . . .") when it is not clear what it refers to: Hypothesis? Category? Women? Note as effective examples that Shannon writes "these reasons" and "This hypothesis" in sentences before the ambiguous "This" here.
categories location and objectification. 1 also found particular categories of stereotypes to be more prevalent than others among all women tested.
To obtain percentages of the stereotypes for each group of women, 1 took the total number of times the stereotype occurred within that race and divided it by the total number of women in that group. To calculate the average percentage of stereotype categories, 1 added up the percentages of each category and divided it by three. See Appendix B for the raw data.
I averaged the percentages for the three race groups to determine which categories occurred most frequently overall. The categories relative size and function ranking occurred the least among all of the categories at 6.7% Location occurred in 10% of the advertisements. 13.3% of the advertisements contained the stereotype licensed withdrawal. Body display occurred in 23.3% of the sample. Objectif/cal/on was the most frequent stereotype, occurring in 40% of the advertisements.
There were significant differences among the women in the categories relative size, function ranking, licensed withdrawal, and body display.
0% of the advertisements of White women contained relative size or function ranking. In contrast, 10% of advertisements depicting African American women and 10% depicting Hispanic women contained relative size and function ranking. Advertisements depicting White women contained more instances of licensed withdrawal (20%) and body display (30%) compared to African American (10% and 20%, respectively) and Hispanic (10% and 20%, respectively) women.
DISCUSSION
Whether we acknowledge it or not, advertising permeates almost every aspect of our lives. The images in advertisements define and solidify gender roles (Baker 2005). For these reasons I studied the stereotypes of African American, Hispanic, and White women in magazine advertisements. I hypothesized that each group of women would be stereotyped. This hypothesis was supported by my findings. African American, Hispanic, and White women were stereotyped in almost every category tested. This was consistent with Lindner's (2004) study, which found that 78% of the advertisements examined in 7mieand Vogue contained stereotypes of
4
THE QUANTITATIVE RESEARCH PAPER
173
OUR COMMENTS
As with most sociologists. Shannon begins the discussion section by reiterating the reason for her study and reminding the reader of the hypotheses she set out to examine. She goes on to conclude that the data supported her first hypothesis. Moreover, she ties her first finding back to the literature she reviewed at the beginning of her paper by indicating whether or not her results were consistent with those of other researchers. She is careful to state that the study by Taylor et al. was not similar enough to make a meaningful comparison.
Shannon follows the same strategy in discussing the results for her second hypothesis. Because she already cited all the authors of the Taylor article, she now refers to this article using the "et al." convention.
These two paragraphs are very strong because they provide an interpretation of the results. The purpose of the discussion section is to give meaning to the numbers that were presented in the results section, rather than simply repeat the findings that were reported. Shannon explains how her findings support the idea of a beauty myth.
women. This finding was also supported by Baker's (2005) study which found that African American and White women were stereotyped in White and African American men's and women's magazine advertisements. Although Taylor et al. (1995) found minorities to be stereotyped in magazine advertisements, they did not specifically examine stereotypes of women. Therefore, our findings could not be confirmed or denied by their study.
My second hypothesis, that objecSification would be the most prevalent stereotype, was supported by my data. Objectification occurred in 40% of the advertisements analyzed, more than any other category. This finding was consistent with Lindner's (2004) study, which also found objectification to be the most common stereotype in 7/meand Vogue magazine advertisements. Since Baker (2005) and Taylor et al. (1995) did not use objectification in their content analysis, their studies could not substantiate or deny my second hypothesis.
The prevalence of objectification can be interpreted as a reaction against the economic and social freedoms enjoyed by the modern woman. In Naomi Wolf's (1991) book, The Beauty Myth, she argues that the beauty myth, the unattainable image of women perpetuated by the mass media, is a tool of patriarchy. As women have become stronger materially they have been weakened psychologically by the beauty myth (Wolf 1991). This unrealistic ideal of beauty keeps women mentally and physically preoccupied with their bodies (Wolf 1991). This theory was supported by the frequent occurrence of objectification in advertisements. Objectification, which occurs when a woman is depicted for the sole purpose of being looked at, fully acquaints women with the ideal they must strive to attain.
Wolf's theory was also supported by the low occurrence of function ranking and relative size. Function ranking occurs when the man in the advertisement plays a dominant role as a boss or instructor. Relative size occurs when the man in the advertisement is larger or takes up more space than the woman. Both of these stereotypes display more traditional gender roles in which the woman is subordinate to the man. Since the women's movement, these gender roles are considered to be outdated and unacceptable to today's woman. Hence the emergence of the modern beauty myth and the prevalence of more subtle forms of stereotypes like objectification. However, the requirement of a man and woman to be
OUR COMMENTS
THE QUANTITATIVE RESEARCH PAPER
This is another strong paragraph because Shannon postulates yet another possible explanation of her findings. The two explanations are mutually exclusive, that is, only one of them can be correct. Also, telling the reader about Lasn's credentials as founder of Adbusters effectively supports the credibility of the borrowed explanation cited here.
A comma is necessary after "profitable" in order to clarify the meaning of this sentence. Without the comma, a reader might think "profitable" referred to "advertisers."
Here Shannon explains the implications of her findings and gives some concrete ways to counter the pernicious effect of gender stereotyping in advertising.
Shannon proposes several interesting ways to extend her work in future research Any one of these ideas would produce an interesting quantitative research paper.
______i
depicted together could also account for the low occurrence of function ranking and relative size.
The high occurrence of objectification can also be explained by economics. Katie Lasn, the founder of Adbusters magazine, argues that advertising starts a vicious cycle. It bolsters our insecurities by bombarding us with unattainable images of beauty and then it "offers us a variety of ways to buy our way back to security" (Lasn 1999:17). For Lasn, the beauty myth does not discriminate based on gender. He argues that because it is more profitable advertisers launch their psychological assault on both men and women,
Although my study was not perfect, the results suggest that women should be aware of the images they are absorbing. Women can resist the psychological assault of the beauty myth by being conscious of the subtle messages conveyed in advertisements. We must also teach our children to be media literate and aware of damaging stereotypes perpetrated by the media. In addition, we can take action by writing letters to advertisers and magazine editors expressing our disapproval of the stereotypes conveyed in their advertisements and magazines.
For the future, it would be interesting to focus on the racial stereotypes present in women's magazines. It would also be interesting to study the stereotypes of men in men's magazines. If stereotypes like objectification are also high, it would mean the beauty myth is less about gender and more about money. I would also like to study stereotypes in teen magazines. It would also be interesting to expose teens to popular advertisements and .then test their perceptions of themselves and the opposite sex.
THE QUANTITATIVE RESEARCH PAPER
177
OUR COMMENTS
According to the ASA Style Guide, Shannon puts the table on a separate page at the end of her paper; however, she should have put it after the References. By giving the table a descriptive title, the reader can understand what's in the table without having to read the text. Shannon includes the sample size of each race/ethnic group and gives the total percent for each column; this indicates that race/ethnicity is the independent variable and that the type of gender stereotype is the dependent variable.
Table 1. Type of Gender Stereotype by Race/Ethnicity of Women in Magazine
Advertisements___
Race/Ethnicity _
In line with the ASA Style Guide, Shannon should have put the appendix on a separate page after the References. However, at least she provides the appendix with a descriptive title.
Type of Gender Stereotype	African American (N - 10)	Hispanic (N =10)	White (N -10)	Average of Three Groups
Relative Size	10%	10%	0%	6.7%
Function Rank	10%	10%	0%	6.7%
Licensed Withdrawal	10%	10%	20%	13.3%
Body Display	20%	20%	30%	23.3%
Location	10%	10%	10%	10%
Objectification	40%	40%	40%	40%
Total	100%	100%	100%	
APPENDIX A
Operational Definitions of Coding Categories
1. Relative Size. In advertisements with both men and women present, the man is larger than the woman and more space in the ad is used by him.
2. Function Ranking. When both men and women are depicted, the man plays a dominant part as a superior or instructor.
3. Licensed Withdrawal. The woman is mentally not present. Her gaze is off in the distance or she is smiling or laughing, covering her face or mouth.
4. Body Display. The woman depicted is wearing revealing clothing or no clothing at all.
5. Location. The woman is depicted in a domestic or unidentifiable environment. Examples would be a kitchen, bedroom, bathroom, or a setting that does not lend itself to any purposeful activities.
6. Objectification. Occurs when being looked at is the major purpose of the woman depicted.
7
THE QUANTITATIVE RESEARCH PAPER
179
OUR COMMENTS
Shannon uses the term "references" rather than "bibliography" because she includes only those articles and books that she actually cites (see Chapter 3). a. bibliography is rarely, if ever, appropriate for a quantitative journal article, unless indicated by your instructor. Shannon follows the ASA Style Guide in formatting the bibliographic information for the articles and books she cites.
Consistent with the ASA Style Guide, Shannon includes both the volume number and the issue number of the journal articles in her References.
1 TT'? ^e Styk GUide' WUh the excePtion * New York Shannon in eludes both the city and state for all book references.
references
Baker, Christina N. 2005. "Images of Women's Sexuality in Advertisements: A Content Analysis of Black- and White-Oriented Women's and Men's Magazines." Sex Roles 52(l-2):13-27,
Goffman, E. 1979. Gender Advertisements. Cambridge, MA; Harvard University Press.
Gornick, V. 1979. "Introduction." Pp. vii-ix in Gender Advertisements by E.
Goffman. Cambridge, MA: Harvard University Press. Kang, M.E. 1997. "The Portrayal of Women's Images in Magazine
Advertisements: Goffman's Gender Analysis Revisited." Sex Roles
37(ll-12):979-97. Kilbourne, Jean. 1999. Can't Buy My Love: How Advertising Changes the
Way We Think and Feel. New York: Touchstone. Lasn, Kalie. 1999. Culture Jam.- How to Reverse America's Suicidal
Consumer Binge-And Why We Must. New York: Harper Coilins
Publishers.
Lindner, Katharina. 2004. "Images of Women in General Interest and Fashion Magazine Advertisements from 1955 to 2002." Sex Roles 51(7-8):409-21.
Taylor, Charles R., Yu Yung Lee, and Barbara B. Stern. 1995. "Portrayals of African, Hispanic, and Asian Americans in Magazine Advertising." American Behavioral Scientist 38(4):608-21.
Umiker-Sebeok, J. 1996. "Power and Construction of Gendered Spaces." International Review of Sociology 6(3):389-404.
Wolf, Naomi. [1991] 2003. "The Beauty Myth " Pp.515-524 in Signs of Life in the USA: Readings on Popular Culture for Writers, edited by Sonia Maasik and Jack Solomon. Boston, MA: Bedford/St.Martin's Press.
8
THE ETHNOGRAPHIC FIELD RESEARCH PAPER
181
h   a   p t
v
In an ethnographic field research project, your data come from observing or inter-acting with people in everyday social settings, which are known as "the field." The data are gathered when a researcher visits the setting (allowing him or her to conduct observational research), takes part in the setting's activities (called Vart^c^^observation), and sometimes interviews participants in the setting.
Ethnographic research is one type of qualitative method used by sociologists, as well as by their colleagues in fields such as anthropology, education, social work, journalism, nursing, and management. Other qualitative methods include narrative analysis, comparative historical methods, ethnomethod-ology, and conversational analysis.' These are likely to be assigned with very specific instructions from your instructor.
What makes a research method "qualitative" is a subject of considerable livS^XJ^b3^ among sociologists. Some suggest that qualitative methods are defined by what they are not: they do not involve counting or measuring the social phenomenon of interest. Others argue that qualitative methods take a distinctive approach to exploring the social world, setting out to represent as accurately as possible the process of social life from the point of view of the participants (or "members") in the setting (or "field") being investigated. Since a scientific hypothesis is an explanation of social processes proposed by someone outside the research setting, the qualitative researcher who adopts an ethnographic approach generally does not engage in the testing of hypotheses. (The kind of fieldwork in which sociologists do use field work to test hypotheses about what happens in social settings or why it happens—which we refer to as "structured field observation" or as a "field experiment"—is discussed in Chapter 6.) The following list of goals and methods defines the specific type of qualitative research—ethnographic field research—that we describe in this chapter.
'For more information on the range and nature oi research methods considered "qualitative," see Dcnzin, Norman K. and Yvonna S. Lincoln. 2005. The Sage Handbook of Qualitative Research. 3rd ed. Thousand Oaks, CA: Sage Publications; or Merriam, Sharon B. 2002. Qualitative Research in Practice: Examples for Discussion and Analysis. San Francisco, CA: Jossey-Bass.
180
The ethnographic researcher usually conducts research by closely observing what people are doing, by talking with them informally, and often by participating in activities with them. If interviews are conducted, the ethnographer uses open-ended questions that encourage respondents to answer in their own ways and with their own words. The choice of methods used in ethnographic research depends on the characteristics of the setting and its inhabitants and on the personal style of the researcher.
Unlike most deductive researchers, then, the ethnographic field researcher does not use a predesigned research instrument, such as a written questionnaire. And unlike the structured fieldwork and field experiments described in Chapter 6, ethnographic field research rarely involves quantitative measurement. While predesigned and quantitative methods are useful for measuring some aspects of the social world, they do not convey the intricate and subtle transactions that the ethnographer seeks to understand.
Reports based on ethnographic field research—called ethnographies— often produce new theoretical insights, but they are most distinctive for their vivid descriptions of actual social scenes and transactions. In other words, even after collecting data, the ethnographer typically does not attempt to propose a hypothesis about why something happens in the social world. Instead, ethnographic research attempts to uncover what happens in a social setting, how social relationships are conducted, and what those events and relationships mean to those involved. Assignments involving these methods may require you to connect these descriptions to specific concepts covered in the class.
In doing ethnographic research, your sociological imagination is exercised by the opportunity to see society's institutions, such as the police, the judicial system, and the health care system, as they are enacted by specific individuals in everyday settings. Because it takes sociology out of the classroom and into the "real world," and because it allows you to view the world through the eyes of people often very different from yourself, an ethnographic field research project can be especially challenging and exciting.
Often the goal of your research project—whether to explore unfamiliar settings or to see a familiar environment through a sociological lens—will be specified by your instructor, who may ask you to do one of the following:
1. Look at social interaction in your everyday life—among family members, friends, fellow students, or coworkers, for example—in new ways. The goals are to describe patterns and processes that often pass unnoticed in your daily interactions and to use your sociological imagination to relate these
182 WRITING FROM VARIOUS DATA SOURCES
personal patterns and processes to specific course concepts. This kind of project might ask you, for instance, to talk to fellow students about their relationships with friends; to observe how those in your dorm, apartment, or family deal with odd behavior; or to watch how individuals attempt to present a certain impression of themselves to others.
2. Visit a setting selected by your instructor, in which social activities of special concern in your course occur, and investigate how those present carry out routine activities and make decisions. Examples of this kind of assignment are going on a police ride-along, observing small-group interactions, or interviewing a mental health professional.
In some classes, however, you may have to develop your own question to address through ethnographic research, or you may simply be assigned to visit a setting of your choice and describe what it is like. If so, remember that, unlike much other sociological research, the goal of ethnographic field research is not to determine what causes some social event or relationship. Therefore, avoid devising a research question thai asks why something happens in your research setting. Instead, concentrate on asking what {for example, "What does a police officer do during his or her time on the job?") or how (such as, "How do those sharing an elevator ride deal with one another in the limited space available?"). In the sample student paper at the end of this chapter, the author addresses this question: "How do students act in academic settings?"
REVIEWING THE LITERATURE
In a deductive research process, a review of relevant research done on the same topic is used to develop a hypothesis for testing through data collection. However, because the kind of heldwork we are describing here does not involve hypothesis testing, instructors assigning ethnographic research projects often do not require that you use a summary of relevant research literature for that purpose. Nevertheless, your instructor, in order to encourage you to become familiar with work already done on the question you are investigating, may prefer that you conduct an overview of relevant research on your subject. Or you may find a literature review useful in getting a feel for ethnographic research, perhaps as you choose a setting or a question for your research or understand the issues of concern to those you will be observing in the field. In this case, use the guidelines for library research in Chapter 4 to get an overview of the sociological literature relevant to your project.
THE ETHNOGRAPHIC FIELD RESEARCH PAPER 183
UNDERSTAND THE ASSIGNMENT
It's important that you understand the instructions you've been given for completing the assignment. Where are you supposed to go? What are you to look for? Is there a specific question you should address?
The most common mistake that students make in conducting an ethnographic research project is to focus so intently on describing a setting that they neglect to discuss it from a sociological perspective. Be sure you are clear on whether the assignment requires you to provide a detailed account of interaction in the observed setting, to demonstrate your ability to apply course concepts to what you see, or both. Ask the instructor for any necessary clarification.
PLAN AHEAD
1. Begin early in the quarter or semester. Field data cannot always be collected predictably or on short notice. Furthermore, you may have to return to your field setting several times to get the additional information or understanding that you need.
2. Make arrangements to interview and/or observe. While the prospect of getting permission may make you nervous at first, you will find that most people are receptive to showing or telling you about their lives. You can assure them that, if they prefer, their identities will not be known to anyone besides you and your instructor. Be sure to follow the procedures established by your college's Office for the Protection of Research Subjects or Institutional Review Board, which might require you to submit your research plan for approval or to obtain written permission from those you observe or interview. Consult your instructor for details.
When scheduling your observation or interview, allow plenty of time. Unanticipated events may occur, your subject may begin to talk at length about some particularly interesting topic, or you may think of additional questions on the spot. Also, you will need to allow time to record, transcribe, or elaborate on notes immediately after the contact.
3. Plan how you will record your data (a summary of recording options follows later in this chapter). In interview situations, it is best to electronically record or to make notes during the interview. Likewise, notes made while observing are more reliable than those made after you've left your field setting. The data collection methods you choose will depend on trie situation,
184
WRITING FROM VARIOUS DATA SOURCES
THE ETHNOGRAPHIC FIELD RESEARCH PAPER
185
your personal style, and the ethical constraints of the situation. But what ever approach you take, be prepared ahead of time with adequate supplies such as blank recording tape, batteries, paper, and pencils, as appropriate '
Although you may know a lot about the setting and the interactions you observe, it is crucial that you leave behind your previous assumptions and even your knowledge about them in order to learn something new. Adopt the attitude of a naive newcomer {as though you are a visitor from another planet!) so that you can begin to look in a new way at events and experiences you used to take for granted, fn other words, don't try to figure out beforehand what conclusions you should come to or how you will use the information you are collecting. Just be as attentive to detail as you can in order to get as much valuable information as possible.
When observing, don't presume you know which events or interactions matter most. Keep your eyes and ears open to everything that is going on around you. Notice your surroundings, all the people who are present, the time taken by events, and so on. Attempt, above all, to look at the setting or situation through the eyes of the participants.
When interviewing, consider your place within the interview situation. How might the person you are interviewing see you or think about the questions you are asking? How might he or she relate to someone of your age, race/ethnicity, gender, and/or educational level? Your sensitivity to these social dynamics will be helpful in managing the interaction successfully. In terms of the interviewing process, follow these guidelines;
1. Don't talk more than you have to. Listen carefully to the respondent's comments.
2. Ask open-ended questions. Avoid leading questions that define the respondent's answer, and avoid questions that point to "yes" or "no" answers.
3. Rather than asking why something happened, concentrate on asking what transpired and how it occurred. "Why" questions often put people on the defensive, making them feel forced to justify their actions or lifestyle. Also, respondents' answers to "how" questions are usually more specific about real events, providing you with the concrete examples you need to describe in detail what goes on in the setting.
4. Don't overwhelm your interviewee with multiple questions. If you are a new interviewer, you may be uncomfortable with silence, but don't rush in with comments, requests for clarification, or further questions if the respondent pauses. Allow the respondent time to think and to complete his or her responses fully.
5. Encourage the respondent to be fairly specific about the details of events or experiences: Exactly who was involved? What happened? When did it take place? Remember, however, that probing should be gentle (for example, "Could you tell me more about that?"), not an interrogation.
6. Relax, allow your natural curiosity about your subject to direct you, and listen.
Since the final paper you produce will be only as good as your recorded data, it is crucial that you record observations or interview responses accurately, in detail, as soon as possible after the event. Otherwise, you will inevitably forget or distort what was said or done.
In observational research, take notes on what you see or hear as it happens. II that is impossible or bothers those you are observing, then write notes on what you observed as soon as possible afterward. You may even want to take periodic note-taking breaks away from the setting during your observation to jot down a few words or phrases that will trigger your memory later. A camera, or even your cell phone, makes it possible to easily create a video record of the setting; however, your obligation as a researcher to respect the privacy of all subjects would require you to get the written agreement of everyone involved in order to use their video images. Check with your instructor on any questions regarding these ethical issues.
If you are interviewing, it is best to electronically record the conversation (with the interviewee's permission). The most effective way to record is to use an audiotape recorder with a microphone. In a pinch, though, it is also possible to record it as a voice message by using a cell phone to call yourself. Don't be shy about asking permission to tape or take notes during the interview. People are often agreeable once they understand your interest in accurately representing what they say.
If subjects seem reluctant to let you tape, don't force the issue. Just listen to their responses and reconstruct the interview in writing as soon as you can afterward. Don't editorialize in your reporting of what was said. Likewise, don't edit your interview to make responses seem more sensible or because something seems "inconsequential." If you edit or editorialize, you may leave out something significant. Report all the respondent's comments, keeping them in their original order. And be sure to include all your questions, as well
as the answers to them.
If you do tape-record an interview, it will be helpful to transcribe it into written form. There are special dictation machines rhat make transcription easier. Or, at the least, you should listen carefully to the interview and take notes on both your questions and the responses.
In all cases, make your notes specific. Describe in detail whar you observed, did, and/or heard. Like a good reporter or a novelist, give the specifics of who, what, when, and where. Include concrete details about the physical setting, what went on, and your reactions: How did you feel about the people with whom you were involved? Remember that in ethnographic field research, you are the research instrument; it is through your own thoughts and feelings, your interactions, and your relations that you learn about the people
186 WRITING FROM VARIOUS DATA SOURCES
and sellings you are studying. For that reason, your personal reactions are especially important.
You may be required to submit the originals, or a typed version, of your field notes as an appendix to your paper (see page 21S of Part 3). Or you might choose to include your notes in an appendix, in order to give your instructor a better appreciation of what happened in your setting or interview (and of how hard you worked to gather your data!). Even if you will not be submitting your notes, keep them legible and organized. Be sure, for example, to date every entry. The quality of your paper relies on the quality of your field notes and interview notes or transcript; the more clear data you have available, the stronger your paper will be.
THE ETHNOGRAPHIC FIELD RESEARCH PAPER 187
■ '■i
EXAMPLE OF OBSERVATIONAL FIELD NOTES
- The following sample observational field notes will give you an idea of their
\ general format and content. The example is excerpted from the notes written
up by Gloria Fong, an undergraduate student in a class on the sociology of | student and campus culture at a large university. The assignment asked par-
ticipants to investigate aspects of student life. The field notes were in prepa-l ration for Gloria's paper, which is presented later in this chapter. Her
| observations focused on the non-academic activities in which students en-
"i gage while in lecture and seminar classes.
OUR COMMENTS
Gloria is careful to identify where and when her observations took place.
Unlike your paper, which should be double-spaced, it is usually acceptable to single-space your field notes. Be sure to leave wide margins for making notes, as described later in the chapter.
It's perfectly normal to be nervous about beginning observations, even in a familiar environment. Generally, though, people being observed are friendly and willing to talk about their experiences.
Gloria describes the setting. Elsewhere in her field notes she includes descriptions of all students vn the seminar room. The set-up turns out to be significant: by noting the seating arrangement, she eventually realizes that students sitting a certain distance from the instructor are more likely to engage in non-academic activities.
Remember that your notes are not a formal treatise. Describe gestures, sounds, even smells, through whatever means necessary, to bring the setting or interactions alive. This will help you to remember and to analyze what happened. It will also bring the field notes alive for your instructor.
THE ETHNOGRAPHIC FIELD RESEARCH PAPER
Be careful not to make assumptions. It's obvious that the two students look up at the instructor. But we don't know why: Is it to keep from getting caught, out of interest in what the instructor is saying, or some other reason?
Wherever possible, make a note of direct quotations right away. Otherwise, paraphrase as accurately as possible.
EXCERPTS FROM GLORIA'S FIELD NOTES
SEMINAR FIELD NOTES, May 11, 1-3 p.m.
I entered the seminar room through the one and only door. I was nervous about observing, even though I usually attend this class anyway. In the middle of the small room is a large rectangular (able with sixteen chairs around it, seven on each long side of the table and two individual chairs at each end of the table. The room is cramped, to say the least. There are fifteen students and one professor. The students sit around the table, and the professor sits at the head of the table, the seat facing opposite the door. A wall of windows is behind the professor, and the other three walls are blackboards.
Two females are each sitting seven seats away from the professor, directly across the table from each other, one on the left and one on the right side of the tabte. The female on the right side of the table {female A) is a Hispanic female, wearing a white t-shirt and black pants. She has black hair pulled back with a black elastic into a ponytail. The female on the left side of the table (female B) is a Caucasian female, wearing a tan sweatshirt and blue jeans. She has long, dark brown hair, and it hangs loosely around her face.
During the seminar meeting, these two females pass back and forth female A's copy of a newspaper clipping and female B's open notebook. I can see that the two students are doing the campus newspaper's daily crossword puzzle together, even though they are across the table from each other. They signal to each other when they need help, and take turns filling in the blanks of the crossword puzzle. They both look up at the professor periodically to see where he is in the room so they do not get caught. Neither of them seems to notice that they are being very obvious with their passing the crossword puzzle and notebooks back and forth across the table. When I asked them about it later, female A said, "We're not trying to disrespect the professor, but we just get bored and it's a challenge to try to finish the puzzle. We help each other out."
As with the seminar setting, Gloria carefully describes the academic setting in which her observations took place. We really get a sense of the large size of the room and the distance between students and between students and instructor.
LECTURE FIELD NOTES, 5/13, 2-3:30 p.m.
I walked into the large lecture hall through the left set of doors. There are three main sections of seats, a left, right, and middle section. The right
1
OUR COMMENTS
THE ETHNOGRAPHIC FIELD RESEARCH PAPER
191
It is important to separate out what is obvious (the exact words or movements or gestures that anyone in the same setting would be able to see or hear) and what « assumed (what we conclude based on those observations). Gloria 1 t0Ad0Cmnent Ihe exact **nss she observed that led her to conclude
that the student was intermittently sleeping and that he was nonetheless attempting to take notes.
and left sections each consist of fourteen rows of seats with three seats across each row. The middle section consists of thirteen rows of seats with eight seats across each row. There are no windows in the room, and the ceilings are much higher than in the seminar room. The professor stands at the front corner of the lecture hall, behind a lectern. Behind the professor is a large movie screen with a picture of lecture notes projected upon it. The class is not full at all, and only the first five rows and the last two rows of the room have students in the seats.
One student I observed was an Asian male in a black t-shirt and black pants, wearing a red baseball cap backwards, sitting alone in the last row of the lecture hall. The student is hunched over his desk, with his elbows on the desk, his arms supporting his head, and his eyes are closed. His notebook is open on the desk, supporting his arms and head, and he has a black pen uncapped in his right hand. He opens his eyes wide and shakes his head from side to side. He looks to the front of the room when his eyes are open and then looks down at his notebook and begins to write. He writes in his notebook for ten minutes and then rests his head on his arms and closes his eyes again. Five minutes pass and his whole body hunches over the desk. He drops his pen on the floor and does not pick it up. His body moves up and down in rhythm with his breathing. My assumption is that this male student is attempting to take notes, but falls asleep in lecture instead. He is trying to pay attention, but just keeps falling asleep in between his frantic note taking sessions.
2
192 WRITING FROM VARIOUS DATA SOURCES
EXAMPLE OF INTERVIEW NOTES
Following is a brief excerpt from the interview notes that Gloria Fong (GF) took while researching her paper on college student and campus culture (see pp. 199-213). Although she could have asked for permission to make an electronic recording of the interview, this is an example of the type of notes jotted down during or immediately after speaking with one of the students (SI) she had observed.
GF: What do you think of attending class? 51:   It's okay. Sometimes kind of boring. GF: How do you deal with it?
51:   I try really hard to take notes but then I just space out. But I don't want to skip lecture because then I might miss something that is on the final, I stop by and get a newspaper every time I come to this class. I rip out the croseword and put it in my notebook before i walk in. ! just can't take notes the whole time so 1 need something else to do to pass the time.
The interview excerpt illustrates how Gloria's attempt to speak directly to a participant in the process she observed paid off. Being able to convey the voices of actual actors in a social scene makes her paper much richer and more graphic.
Gloria's' inquiries are phrased as "what" and "how" questions, which typically elicit more useful information from interview subjects than "why" questions do. If she had asked, "Why don't you pay attention in class?" the student might have felt too defensive to share her thoughts about classroom behaviors. Also, her questions are open-ended rather than phrased, so they cannot be answered with a "yes" or "no." She might have gotten more specific responses, though, if she had been able to focus her questions. For example, "What do you think of attending class?" is a broad inquiry, to which the reply was very general. She might have learned even more if she had asked specifically what the subject thought of a particular aspect of attending a college class, such as "What do you do during class besides listening and taking notes on what the instructor says?"
The observations and answers you collect in your fieldwork are the data on which your paper will be based. In this step of your research process, you use the material you have collected to analyze the setting and/or to answer the question that your instructor assigned or that you formulated. In order to do
THE ETHNOGRAPHIC FIELD RESEARCH PAPER 193
that, you will need to identify, sort, and order and re-order segments of your notes, either electronically or on hard copy. This is an exciting process; as you work with your notes, you will notice that the setting you have learned about in a personal way reveals interesting information about the nature of social life.
ANSWERING AN ASSIGNED QUESTION
If your instructor asked a specific question in your paper assignment, now is the time to consider how what you saw, heard, and experienced addresses that question. Here are some guidelines:*
1. Go through your notes and make a mark by every comment, observation, or response that seems relevant to the question being asked. Some researchers prefer to do this kind of review in the margins of a printout; others do it on screen (as described in Warren and Karner 2005). Either way, don't be too discriminating at this point. Better to include too much at this stage than too little. Use a relevant word (or "code") in the margin to highlight what is important about that section of the data.
2. Extract these relevant pieces of data so that you can review them together and put them in categories. You can do this by copying them by hand onto separate note cards; by photocopying your original notes and cutting and pasting the relevant excerpts onto the cards; or by electronically blocking and copying the identified sections into a separate file. Whatever approach you take, you should be able to see the bits of data side by side, much as you would for materials in a library research paper (see Chapter 4).
3. Now consider what the information on each card or segment says in response to the question asked in the assignment. What does it tell you about the setting you observed and/or the people you interviewed?
4. Look for patterns among your data excerpts. Move them around (on cards or electronically) to illustrate to yourself how the information fits together. For instance, you might stack together cards or put together blocks of data to which you assigned the same code because they contain examples of the same kind of behavior. Or you might arrange data sections to reflect stages in a process.
Course materials and the paper assignment itself may be useful in helping you notice the patterns in your data. Recall concepts covered in the class that are relevant to your project. Review carefully just what the assignment directs you to look at. Then consider how your data illustrate those concepts or teach you something about the social relations you observed.
Gloria Fong, the student author of the sample ethnographic paper that appears at the end of this chapter, was asked by her instructor to observe and write about some aspect of college student life. Several class readings,
* These guidelines assume you will he doing this sorting on hard copy. If you have the time and resources, the coding and sorting process can also be done electronically, with a program such as The Elhnograph <http://www.qualisresearch.com/>.
194
WRITING FROM VARIOUS DATA SOURCES
THE ETHNOGRAPHIC FIELD RESEARCH PAPER
195
including the classic work by Horowitz (1987) that she cites, focused on how college students do and don't feel engaged in their studies. In organizing her data, she might have listed the different roles that Horowitz noticed that students take, then stacked in separate piles (or moved to separate electronic files) the excerpts from her data that illustrated them. Or, looking for stages, she could have arranged her description of students in the "back of the class" as they selected their seats, began their non-academic activities, and moved from one activity to another.
Perhaps you were simply assigned (or chose as your project) to participate in and describe a social setting. It will still be useful to sort out excerpts from your notes as described in the preceding section, but you will probably have more freedom to establish the categories in which you will organize your observations.
Begin by carefully rereading your field notes to refresh your memory about the events. Then start to look for patterns in your notes. As in the case of an assigned question, you might use course concepts to organize this search. Better yet, you might try to find the categories and terms used by people you observed, asking yourself how they understand and describe their activities.
For example, if you were taking a course in deviant behavior, your text would probably spend considerable space defining "deviance" in terms of breaking social norms. But in ethnographic research, you would find that the people you observe don't talk about "deviance" or "norms." Thus, rather than looking in your notes for examples of what your text would define as "deviance," it would be more enlightening to pinpoint what specific behaviors your subjects perceive as odd or disruptive and to note the ways in which they categorize and describe those who exhibit disliked behavior (such as referring to them as "weird" or as "different from us"). Similarly, a course in stratification may present sophisticated ways of measuring socioeconomic status; but, since the most interesting and valuable findings from your ethnographic field research concern the ways in which the people you observe perceive their own position in relation to society, you might look through your notes for all the ways they compare themselves to other groups.
Some of the most common themes ethnographers look for include the ways members characterize their group; the ways they distinguish between insiders and outsiders; the special language they develop to describe their shared activities and values; their patterns of interaction; the ways they teach new members the ropes; the ways they identify and respond to behavior they don't like; and the ways in which members experience their setting through the course of an event, a workday, or any other unit of experience. You may find some or all of these reflected in your notes, and you may find interesting themes not listed here. You may choose to focus on one area or on several related themes.
As you begin to identify themes that run through your field notes, you can proceed to sort the excerpts, either electronically or on hard copy, as described in the preceding section.
Because ethnographic field research does not involve hypothesis testing, the essay format discussed in Chapter 1 is more appropriate than the journal format for this type of paper. Simply modify the essay format slightly: in place of the three (or more) "claims" or "points" relevant to a paper that proposes and supports a thesis, substitute the themes or concepts that you identified in your field notes. These will serve as the body of your paper. Remember that you are responsible for demonstrating, through effective use of your data, why your description of the setting is believable. Be sure to describe your research methods—where you went, how long you stayed, with whom you spoke, and so on—and to include as evidence the most illustrative excerpts from your field notes.
If the question you are addressing was assigned, you can use one of two approaches to present a written report on your observations:
One approach is to describe what happened or what was said, chronologically, and comment on how course concepts apply to the things you describe, as you report them in the order in which they occurred. This strategy requires you to be especially careful to avoid spending too much time on describing "what happened next."
Another approach is to organize your paper around concepts, defining and indicating the importance of each and using your data to illustrate them. In this case, you can follow the essay format, taking for each of your main points a selected concept or group of concepts.
Unless your assignment specifically requires one approach, either can be successful. If you are uncertain about which approach to take or which may be preferred, discuss your plans with your instructor. In either case, return to course concepts and themes frequently. Ask yourself how the events or comments you are describing reflect or illustrate sociological ideas. This will help you avoid the common mistakes of focusing too much on description or making overly psychological interpretations of those whom you observe or interview.
If your assignment doesn't specify a particular question for you to answer or a specific setting for you to analyze according to course concepts, then you can simply organize the themes you discovered in your notes in the essay format, You might choose three points to make about one of the themes that you found most interesting or revealing. Or, you might develop your paper around three different themes.
Whichever format you use, it is important for your paper to incorporate the reactions you experienced in your research and recorded in your notes.
196
WRITING PROM VARIOUS DATA SOURCES
THE ETHNOGRAPHIC FIELD RESEARCH PAPER
197
Inevitably, those engaged in ethnographic field research encounter people, events, and experiences that fascinate, surprise, confuse, or even upset them. It is a challenge to make effective use of such reactions without getting sidetracked into self-analysis. A good way to make your personal reactions relevant is to ask yourself what they illuminate about the setting. Describe in your paper how your own feelings and thoughts helped you better understand the people you studied and their interactions.
When writing your paper, you may quote your field notes directly. When you do, punctuate and cite them as you would any other source. Or, you may choose to summarize an incident or a response in an anecdotal way to illustrate a point. As long as they are relevant to your assignment, use your collected data in as many ways as you can; they make up the empirical basis for your discussion.
SUGGESTED READINGS
Denzin, Norman K. and Yvonna S. Lincoln. 2005. Handbook of Qualitative Research. 3rd ed. Thousand Oaks, CA: Sage Publications.
Emerson, Robert M., Rachel I. Fretz, and Linda L. Shaw. 1995. Writing Fieldnotes, Chicago, IL: University of Chicago Press.
Merriam, Sharon B. 2002. Qualitative Research in Practice: Examples for Discussion and Analysis. San Francisco, CA: Jossey-Bass.
Warren, Carol A.B. and Tracy X. Karner. 2005. Discovering Qualitative Methods: Field Research, Interviews, and Analysis. Los Angeles, CA: Roxbury Publishing Company.
A SAMPLE STUDENT PAPER
The following sample ethnographic field research paper was written by Gloria Fong for a class on the sociology of student and campus culture. She describes in her paper how she came to focus her research on the non-academic activities in which students engage in classroom settings. She observed students' activities both in a large lecture hall and in a small seminar room, noting where the students sat and what they did besides listening to the lecture.
Gloria used her original data to answer three research questions. Her findings are illustrated by specific observations, which she summarized in her field notes. Sometimes her field notes and interview notes are quoted directly, and sometimes they are incorporated into the text of her paper.
Gloria's paper demonstrates that she is a capable and diligent student. Her creative brainstorming resulted in an attempt to fill a hole in the research literature on higher education. Although research has been done on students' academic activities and on their out-of-class activities, little is known about their non-academic activities in the classroom settings that Gloria explored. She is a good example of a student whose basically good writing could be made even better by applying a closer eye to such details as grammar and punctuation.
THE ETHNOGRAPHIC FIELD RESEARCH PAPER
199
OUR COMMENTS
Because of the length of Gloria's paper, she has included a title page (as recommended in Part 3).
NON-ACADEMIC ACTIVITIES IN ACADEMIC SETTINGS
Gloria Fong
Sociology 197-A: Sociology of
Student and Campus Culture Professor Richlin-Klonsky June 2004
THE ETHNOGRAPHIC FIELD RESEARCH PAPER
201
OUR COMMENTS
Gloria's first paragraph introduces her research question: In what non-academic activities do students engage in a college classroom? Then she shows how it fits into existing research literature in higher education, such as the studies on student involvement by Astin and by Horowitz, and indicates its significance. Despite the length of time it's been since these studies were published, Gloria's use of them is acceptable because they are considered classics and because they were key readings for the class. You should be sure to look for recent relevant research whenever possible.
Gloria's writing would be more straightforward if she would use fewer descriptors, such as adverbs like "really," "actually," and "certainly."
INTRODUCTION
Researchers in higher education have been interested in the ways that students either are or aren't involved in their college experience (Astin 1993; Horowitz 1987). There seems to be quite a lot of research about the academic activities of students in the classroom and about the non-academic activities of students outside of the classroom. But there is not much about the non-academic activities of students in academic settings. No one seems to have really researched in depth what students actually do in class, even though it is certainly a part of student and campus culture. Such data could certainly help improve the quality of the college experience, particularly by helping faculty understand how things look from the students' point of view.
Gloria includes a description of how her research question was selected, because she feels that the roundabout way it occurred to her says something about the question she's studying. It also illustrates how the research question may emerge from personal experience.
Gloria's informal narrative style {telling about her walk home and how the question came to her "in a flash") was acceptable for this assignment, which was meant to lead to a preliminary research report. It might not be appropriate for more formal research papers.
RESEARCH DESIGN AND METHODS
I originally was interested in studying fashion on campus, but because there were already so many students in the class studying that, I was assigned to study something about academics. I spent a while trying to figure out what aspect of academics I wanted to look at most. Then in class the professor suggested that somebody look at where students sit in classrooms. Then another student in the class suggested that I look at how students do the daily crossword puzzle all the time during classes. As I was walking home after class, I started thinking back to all of the lectures and seminars I have been in during my four years in college. And I realized that the back of the class is usually more interesting than the front of the room because of the non-academic activities that take place in the back of the room. And so it came to me in a flash that I should just combine the two topics suggested to me in class and explore where people sit in classrooms and what people do depending on where they sit in class. The amount of time that it took me to string the suggested topics together to make my own topic shows how strongly people tend to separate the two realities of academics and non-academics.
f became an ethnographer to investigate, (1) whether there is a front-of-the room culture and a back-of-the-room culture, (2) whether students from one racial or economic background are more likely than others to sit in
1
THE ETHNOGRAPHIC FIELD RESEARCH PAPER
203
OUR COMMENTS
Although making only one visit to each kind of academic setting (lecture and seminar) does not produce enough data to draw any conclusions about them, it was sufficient for this preliminary study.
Gloria included her interview notes and field notes as an appendix to her paper (as suggested in Part 3), which will allow her instructor and any other readers to review the detailed data she collected. (However, due to space limitations, Appendix A is not included at the end of her paper.)
the front or the back of the classroom; (3) whether there is a difference between what students do in the back of the room in large lectures and what students do who sit in the back of the roam in small seminars. 1 made various observations and inscriptions during one three-hour seminar I attended and one two-hour lecture I attended. I sat in the back of the classrooms and took notes on what students were doing and 1 also asked a few of the students I had observed some quick questions after each class was over. I recorded small notes to myself while in the classes I chose to attend. I then took the inscriptions I made in the classroom back home and turned them into formal field notes, which I later reviewed for themes and patterns (see Appendix A).
In this introduction to this section, Gloria clearly identifies her major findings and organizes them around her research questions. This will serve as a road map for someone reading her findings.
Be careful to use the correct form of pronouns. The object of a preposition ("to") should be "whom" rather than "who."
It is not unusual that in the course of qualitative research, new insights lead to additional questions.
FINDINGS
In reviewing my field notes, I found answers to my three questions. First, there are definitely separate areas in the front and back of the lecture hall which can be seen either from where people are sitting or from who is talking to who. Second, social factors such as race, ethnicity, gender, and class standing did not seem to affect who sat in the back of a class versus who sat in the front of a class. Third, non-academic activities take place in both overall settings (lecture hall or seminar room).
Even though it wasn't part of my original questions, I also learned something about why students bother going to class when they are doing so many other things besides listening to the lecture. This was a question that came to my mind while I was doing my research.
Gloria quotes her field notes directly to illustrate the distribution of students in the classroom.
is there a back of the room culture?
The "front" and "back" areas in the lecture hall were easy to see because they were physically separated.
There are three main sections of seats, a left, right, and middle section. The right and left sections each consist of fourteen rows of seats with three seats across each row. The middle section consists of thirteen rows of seats with eight seats across each row. . . .The class is not full at all, and only the first five rows and the last two rows of the room have students in the seats.
THE ETHNOGRAPHIC FIELD RESEARCH PAPER
OUR COMMENTS
Watch out for accurate capitalization. All proper names ("Game Boys" and "Palm Pilots") need to be capitalized.
Gloria's notes do a good job of setting the scene for the reader. We can easily visualize the setting she was observing.
The middle of the room was basically a no-man's land. When it comes to interactions, it seemed that only those who sat in the front of the class spoke with and/or interacted with the students in the front of the class. And furthermore, only those who sat in the back of the class spoke with and/or interacted with the students in the back. The non-academic activities that i observed in the back of the class were: sleeping, eating, talking, doing the crossword puzzle from the school newspaper, doodling, and listening to music. And with the exception of listening to music, all of the non-academic activities were low-tech. I did not observe anyone text messaging on their cell phones, playing with gameboys or other portable video game devices, playing with their palm pilots, talking on their cell phones, or any other types of high-tech, non-academic activities in the back of the room.
In the seminar room, there was no space for the "front" and "back" to be physically separated.
In the middle of the room is a large rectangular table with sixteen chairs around it, seven on each long side of the table and two individual chairs at each end of the table. The room is cramped, to say the least. There are fifteen students and one professor. The students sit around the table, and the professor sits at the head of the table.
Before I observed the classes, I thought for sure that there would be less of a "back of the room" culture in a seminar room because of its size. I was surprised to find so many people doing the same types of non-academic activities in both academic settings. The back of the room in the seminar was just as active with non-academic activities as the back of the lecture hall. It is interesting to note that even though there were only 15 people in the seminar, there was still definitely a "back of the class" area. All the people sitting four seats or closer to the professor sat quietly and wrote down notes on their notebooks. They did not look at other people in the class and only looked up at the professor or the board when he was writing things down. All the people sitting five seats or more away from the professor felt that they were far enough away from him to be able to do other things besides take notes. This was the "back of the class" area for the seminar room.
THE ETHNOGRAPHIC FIELD RESEARCH PAPER 207
Are students of certain social characteristics more likely to be in the.&ack of the room?
There is no one type of person who sits in the back of the room and engages in non-academic activities. All races, ethnicities, genders, and class standings are represented in the back of the room. Students' social characteristics did not appear to piay a role in how involved they are in the academic part of what goes on in the classrooms.
Do non-academic activities take place in both lecture halls and seminar rooms?
Before I observed the classes, I thought for sure that there would be less of a "back of the room" culture in a seminar room because of its size. I was surprised to find so many people doing the same types of non-academic activities in both academic settings. The back of the room in the seminar was just as active with non-academic activities as the back of the lecture hall.
Two females are each sitting seven seats away from the professor, directly across the table from each other, one on the left and one on the right side of the table. During the seminar meeting, these two females pass back and forth female A's copy of a newspaper clipping and femafe B's open notebook. I can see that the two students are doing the campus newspaper's daily crossword puzzle together, even though they are across the table from each other. They signal to each other when they need help, and take turns filling in blanks of the crossword puzzle. They both look up at the professor periodically to see where he is in the room so they do not get caught. Neither of them seems to notice that they are being very obvious with their passing the crossword puzzle and notebooks back and forth across the table.
But because the professor was very close to them, students in the seminar setting did check on where he was looking and what he was doing far more often than the people in large lecture halls checked.
Why do students engage in non-academic activities while in class?
This wasn't one of my original research questions, but after observing students engaged in non-academic activities during class, I became curious about why they bothered to attend class at all. Two of my interview subjects
A
OUR COMMENTS
Sociologists are always interested in the role of social factors in the type of experience under study. This conclusion was made possible by Gloria's detailed field notes, which included descriptions of social characteristics of the students she observed, and by her interviews, in which she asked students about their class standings (first year, sophomore, and so on).
This kind of unexpected finding is not unusual for qualitative research and makes it especially interesting.
THE ETHNOGRAPHIC FIELD RESEARCH PAPER
209
iOUR COMMENTS
indicated that they wanted to be sure that they don't miss something that may be on a test. As one student put it:
I try really hard to take notes but then 1 just space out. But I don't want to skip lecture because then I might miss something that is on the final.
There also seems to be a strong norm among students that one has to go to class to feel like, and appear as, a "good student." Being a "good student" means you attend all of your classes. One of the students I interviewed said, "i feel like I have to come to lecture, so I come."
DISCUSSION AND IMPLICATIONS
Gloria tries to highlight the different types of students in Horowitz's typology. However, capitalizing an entire word is the written equivalent of yelling.
Gloria neglects to point out one difference between her observations and Horowitz's analysis, which suggests that "outsiders" and "college men and women" were of different racial and ethnic groups and different social classes.
In my study, there were distinct groups in the front and in the back of the classroom.This is like the "OUTSIDERS" and the "COLLEGE MEN AND WOMEN" that Horowitz (1987) described in higher education. She describes college men and women as those students for whom "classes and books exist [only] as a price one has to pay for college life. . . .but no real college man or woman ever expects to live in the classroom" (1987:12). The outsiders are "studious, polite, and respectful of authority. . . .hardworking students [who] sought the approval of their teachers, not of their peers" (1987:14).
The findings are relevant to campus policy and programming because they could lead to some changes in campus policy and programming that are aimed at decreasing the amount of non-academic activities in the back of classrooms directly. However, while there are obviously a lot of non-academic activities occurring in lectures and seminars, I do not feel that there is really any reason to try and stop them, and really no way to stop them, because there is already an established culture of students sitting in the back of the room sleeping, eating, reading, etc. I fee! that no matter what rules and regulations are put in place to stop them, there will always be students in the back of the class doing other things besides taking notes. This is because there will always be students who are just showing up to class to appear as a "good student," but not to take notes.
5
THE ETHNOGRAPHIC FIELD RESEARCH PAPER
211
OUR COMMENTS
Instead, I would suggest that classes not be scheduled earlier than
11:00 a.m. to cut down on the amount of sleeping in class. Classes should also not be longer than two hours so students do not get bored and fidgety. Classes should not be scheduled around mealtimes so that people will not have to eat in class or fall asleep in class from being too tired from not eating. Also, I think online classes, video classes, or other types of distance learning should be considered for large lectures.
My findings could also be used to help foster a better understanding between students and faculty and therefore better relationships between the two, which would facilitate students' academic and personal development.
6
THE ETHNOGRAPHIC FIELD RESEARCH PAPER
213
OUR COMMENTS
Appropriately, Gloria started her References on a separate page.
Gloria should have included her interview and/or field notes in her list of references. Since the format was not specified in the assignment, she should have discussed the required information and preferred format with her instructor. Here is one possibility:
Fong, Gloria. May 11, 2004. Field notes, observations of seminar room, California.
REFERENCES
Astin, Alexander. 1993. "Effects of Involvement." Pp. 365-395 in What Matters in College? Four Critical Years Revisited, edited by A. Astin. San Francisco, CA: Jossey-Bass.
Horowitz, Helen Lefkowitz. 1987. "Introduction: The Worlds that
Undergraduates Make." Pp. 3-22 in Campus Life: Undergraduate Cultures from the End of the Eighteenth Century to the Present. New York: Alfred A. Knopf.
7