WEEK 14:

UNOBTRUSIVE MEASURES

Problems with obtrusive measures: (derived from the book Unobtrusive Measures: Nonreactive Research in the Social Sciences, by Eugene Webb, Donald T. Campbell, Richard D. Schwartz, and Lee Sechrest; Rand McNally Co., 1972)
1) Guinea Pig or Testing Effect: subjects may feel that they must leave a good impression, or the test may make them interested in the subject, so they learn more about it.
2) Role Selection: they pick a nonrepresentative role (unlike their usual self), especially if they are less educated and less familiar with the subject of the test.
3) Response Sets, such as acquiescence bias, especially occurring when there are response category having similar wordings; example of Mississippi Poll, state program spending items.
4) Interviewer Effect: the race, age, and sex of interviewer may affect the interviewee’s responses.

Unobtrusive Measures directly remove the researcher from the research setting.

Types of Unobtrusive Measures (Source: Research Methods in the Social Sciences, 5th edition, by Chava Frankfort-Nachmias and David Nachmias; St. Martin's, 1996, pages 315-324)

• 1) Physical traces- erosion
• 2) Physical traces- accretion
• 3) Simple observation- physical signs
• 4) Simple observation- body language
• 5) Simple observation- physical location (proximity to others).
  Validity problems with simple observation method: available only in public settings; observations may not represent a wider population; doesn't generate explanations.
• 6) Archival Records- actuarial records, political records, government documents, mass media archive.
• 7) Content Analysis- any technique for making inferences by systematically and objectively identifying specified characteristics of messages. Examples of M. Hermann's study of how a nation's leader's personality affects the likelihood of that country going to war (especially in closed political systems); also, Edith Efron's The News Twisters, and Richard Hofstetter's studies on whether television nightly news programs are ideologically biased towards one party's presidential candidates.

Unobtrusive Measures are cool. I think a lot of us use them, but we don’t really think about it.

Physical traces, erosion (taking away). You have limited money to spend on paving sidewalks. You look for erosion of the grass, where are many people walking? That’s how we got those diagonal sidewalks on the drill field.

Physical traces, accretion (adding to). You have limited funds to build playground equipment for a local park, and you want to figure out how many people use your parks, so you look at the amount of trash left behind. In Hawaii, you can tell that most people live on the island of Oahu than the other Hawaiian islands just by the litter along the highway that goes north along the shore out of Honolulu. Accretion, count the cars to determine population density; Oahu is so crowded, it actually has three interstates.

Simple observation. Physical Signs. Sociologists did studies of the cultural diversity of large cities like New York City by just counting the signs with different nationalities at all of the restaurants in the town. Same in San Francisco, Chicago. Washington D.C. is great in diversity, because of all of the embassies.

Simple observation, body language. As an employer, do candidates care about your organization, are they really interested in it? When one of our untenured faculty members interviewed for the job, she not only knew about each of the professors, but she even knew what type of election laws Mississippi had. In your own career pursuits, study the organizations’ websites, learn everything that you can about any place where you want a job, or an internship. Employers often ask: Do job candidates have the motivation or the energy to do the job? One untenured candidate came in and despite a full day of emotionally exhausting interviews she was walking at the end faster than me! We hired both of those candidates. On the other hand, one young male acted distracted, checked his cell phone while I was walking him to the Dean’s building and telling him about the Dean; another male made some snide remark about the university when talking to his wife in front of our secretary (she naturally told me); they did not get hired. Our current untenured faculty do a great job teaching (including using technology in their classes), and they also publish. But we had one faculty member who played on Facebook a lot, was listening to music as he walked in the halls. He is no longer here.

Simple observation- physical location. You can tell how important someone is by whether they have an office in the White House itself. And if so, how physically close to the Oval Office. In the old Soviet Union, we would look for who was standing next to the General Secretary of the Communist Party as likely to be next in line for leadership. A physical location application, do you ever see a student in some faculty offices? We used to have a few profs where the answer was no; they didn’t want to be bothered, they just wanted to publish. They left or were helped to leave; nobody missed them.

Archival Records. Age similarity may be related to marriage stability. You used to be able to read the ages of people getting married in the newspaper; in later papers, you could then read the divorce decrees. People of significant age differences often have different needs and life histories. Another example of archival records is roll call votes of state legislators and congress members. They do know that they are being studied, but they don’t fully know how they are being studied. Thus, I studied how southern state legislators voted differently based on their party and race. Republicans were most conservative, African American Democrats were most liberal, white Democrats were in the middle.

Content Analysis. You might begin by just listening objectively to what people say. It is clear to me that CNN and MSNBC are liberal, they despised Trump, they insinuated that he is a bigot, a criminal, and a traitor. FOX and One America News are conservative, love Trump, don’t like Obama or Biden. (In other words, watch what you say, since researchers can study you.) Trump’s Inaugural Address painted a bleak picture of American economic decline, a sad way to unite the country, especially when the former Presidents Obama and Bush whom he was implicitly criticizing were sitting in back of him (as was Hillary). Biden, on the other hand, had a great inaugural address that was vague, and that kept talking about unity. But early in his administration, he kept criticizing the Republican Party and supporters of Trump. An objective content analysis of both Presidents’ statements while in office might find that each President is really talking more to their base of supporters and appealing to them. Indeed, Trump used to brag about having “the highest approval rating among Republicans” of any President. Seems like both Trump and Biden (at least initially) acted as if they were President of only their own political party, instead of President of the entire nation. No wonder we are so divided as a nation!