Where things are at

I've been working downstairs on my laptop, the one my sister was nice enough to donate because she didn't want to spend the time and effort getting it fixed. I'm down here partly because it's comfortable to sit in a big padded chair, at least for a while, and write. More comfortable than my worn-out desk chair and the desk that's a door that is really too high up for me to type properly without sacrificing my wrists (one of which is bothering me even now, with the keyboard on my lap in what I guess is a not entirely neutral position), anyway. (Yes, I know that's a sentence fragment.) I'm really waiting for my wife to go to sleep so I don't need to worry about her walking in while I'm watching "Battlestar Galactica" on my desktop compouter, having downloaded it from iTunes for $1.99 just 24 hours after it aired on SciFi (which we don't get). It's not so much that I don't want her to know that I'm doing this; it's just that I don't want to see her disapproving of me while I'm doing it.

I had thought of imposing blogging on all of my students, but I've chickened out. There are a couple of reasons for this, one of the most important being that I'm uncertain about the legal implications of my asking them to join up with a commercial (though no-cost) service as a mandatory part of their university educations. The other reason is that while writing a journal would be a good habit for them to acquire, and although imposing it upon them for the semester might actually help them develop this habit, it's far enough removed from what I'm supposed to be teaching them in this class that I don't see a need to require it.

I wrote to one of the authors of the textbook I'm using for this course, asking whether they had prepared anything like a solutions manual for the exercises they provide at the end of each chapter. He said that no one else had asked, but they would think about it if they got more requests. One reason for this, I suppose, is that many of the people using their book are friends of theirs--fellow survey methodologists who they know and who share enough understanding about core principles that a solutions manual ought not be necessary. What I find is that although I almost always know what they're looking for, sometimes I'm not positive and at most times I'd like to see their solutions just to be sure that what I'm thinking is what they were thinking. Having asked and been told that nobody else has asked, I feel kind of foolish about having asked. I don't regret it, but I still feel foolish.

A few minutes ago I finished reading the book The Courage to Write : How Writers Transcend Fear by Ralph Keyes. I was steered toward this book by Holly Lisle, in her book Mugging the Muse: Writing Fiction for Love AND Money. Both books are primarily about writing fiction, but both ought to be useful to non-fiction writers as well, particularly Keyes' book. Keyes' story is essentially this: all writers are terrified by writing, all terrified of the blank page, all terrified of making fools of themselves. Successful writers are those who, in addition to telling a story that people want to hear, are persistent enough to write through their fears. More encouragingly, I think, he argues that successful writers use rather than overcome their fear and anxiety; they use it as a source of energy, and they use their specific fears as sources for ideas and as a way to locate the places they really need to go (that is, the places they most fear to go). He talks about some writers as counterphobics--people who have a need to work through their fears by living through the things they fear most--and that's not me!

I've been piling dread upon dread in working on a short article that I've mentioned already. I desperately want to finish it, but I'm not only afraid of screwing it up, not only afraid that I really need to do more data analysis--again and again, not only afraid of what my co-author is thinking and will think and so on, but ... oh, I already said that: I'm just afraid of screwing it up and looking foolish.

Anyway, the battery is almost dead so that's it for now ...

Course design

This is the fifth time I've taught this course. I never teach it the same way twice, for better or worse. Most of the time I've spent on revamping the course has been on locating new materials, reading the materials, trying to figure out how to squeeze everything in, and trying to figure out how to order everything. It's not easy.

The biggest course element I've added this time is a session on measurement theory. This isn't something I know a lot about; it's an area developed mainly by psychometricians—that is, psychologists who devote their time to measuring psychological states. We sociologists have had a tendency to "wing it" as far as measurement goes, ignoring what the psycholometricians have to tell us (although we're perfectly willing to use the scales they've developed, when they apply to our work). Anyway, that's why I've added a session on measurement theory.

The other thing I'm adding is a course component aimed at developing what C. Wright Mills calls "intellectual craftsmanship." This component is intended to inculcate positive, career-long intellectual habits that will (in theory!) serve students well regardless of what empirical methods they use in their research. The core assignment (for this component) will be to develop and refine a research agenda outline--what Mills calls a "master agenda." Our research agendas are the plans we formulate in order to translate our interests and desires into actions. I doubt that many graduate students not yet on the job market have taken the time to think about what their research agenda is and will be, but the second year of a graduate program is not too early to start thinking about this. (This is not the same thing as the "statement of research interests" sometimes required for academic job applications. The research agenda outline is not intended as a form of self-presentation; it is a repository for the thoughts and half-baked ideas that will eventually go into the more polished "statement.")

What do I (or we) lose by squeezing in these elements? For one, a week that I devoted last year to meetings with students. Two, time to work on data analysis methods. Three, unless I can figure out a way to work this in, time to spend on exemplars of good social science produced using survey research. I may still find time for this, but it will have to come at the expense of something else ...

Stuck on blog

I'm up past my bedtime, which is far too typical. I am unfortunately stuck on a piece of research that I've been working on for quite some time—really, far too long. The data comes from the same study that How the Other Half Works (my book co-authored with Roger Waldinger) is based on. What we did for this study was interview managers with hiring responsibilities in six industries in Los Angeles County in the early 1990s (a long time ago already!).

The paper I'm working on concerns the hotel industry specifically. I don't have time to go into all of the background on the piece right now. I really just want to get down some thoughts about it before I go to bed that I can reflect on later.

The problem I'm having is partly one of framing and partly one of organization. The evidence I have comes almost exclusively from relatively short conversations with managers about what they do when hiring, how they think about the jobs they're hiring for, and what they think about racially/ethnically distinctive groups of workers. (The other evidence I have is from U.S. government surveys, primarily the decennial Census of Population.) What I want to know about is how managers take race/ethnicity and citizenship/immigration status into account when making hiring decisions. They don't tell me that, of course—they (almost) universally say that they select the "best person" regardless of personal characteristics. It is evident, however, that personal characteristics are part of what makes someone the best person. For example, in area of the hotel dominated by Spanish-speaking workers, there is a strong bias against hiring English-language monolinguals even though speaking Spanish is not a formal qualification for the job (in most instances). Being able to get along with others is an important consideration to many managers, and managers readily admit that being "like" the others one has to get along with is an important part of being able to get along.

The framing problem has to do with what I can say relative to the literatures on labor market structure (dualism/segmentation), immigrant economic incorporation (ethnic niches, ethnic economies), native-immigrant and black-immigrant labor market competition (including both wage and job competition and the evidence at the national and local levels), labor market discrimination (including pure prejudice, statistical discrimination, cognitive models, and job/labor queues), and more! In the quantitative model, I'm supposed to formulate a research question, come up with some hypotheses that follow from the question given the literature, and then test the hypotheses with the given data. This is primarily a piece of qualitative research, however. Still, I should have a research question that drives everything. The problem is that I can't decide precisely what the research question is; it keeps shifting. It's clear, however, that the research question has to be one that can be answered (at least in part) given the data at hand, and as I said the data at hand is what managers say about themselves, what they do, and what they think about certain jobs and various kinds of workers. We know that what people tell us about themselves and what they do is not the same as an "objective" report on who they are, what they do, or what they think because (a) there are often parts of what people do and think that are invisible to them (typically because they think of these things as natural and given and therefore unworthy of notice), (b) there are parts of what people do and think that are embarrassing to them (and possibly dangerous to them, if articulated) and therefore either omitted or distorted, and (c) there is a well-known disconnection between what people think and what they do.

In any event, I had tried to refocus the piece around the "cognitive maps" of employers. Cognitive maps are mental tools that people use to help them "navigate" the world; in this instance, they are tools that help them make employment decisions. I conceive of these maps as linking personal attributes of individuals with the attributes that managers most desire in workers for specific positions. The idea was that an understanding of these cognitive maps is the primary insight that we need to gain in order to understand what employers actually do when they hire. That is, managers have told us how they think; what we've done is to organize what they've told us into a coherent map of the world. This is an appealing way to go, but I don't think I can sustain it.

Instead, I'm thinking that I need to focus on job queues/labor queues as an explanation for racial/ethnic/immigration status change in the hotel industry. Between 1970 and 2000, the racial/ethnic/immigration status composition of the hotel industry in Los Angeles County changed drastically. For those who are concerned about the impact of immigration on the labor market prospects of natives, the hotel industry is a case that clearly deserves attention. The key question is "what accounts for demographic change in the industry?" However, although many of our respondents addressed the past, the interviews were basically cross-sectional, focusing on the particular time (1992-93) in which they were conducted. These interviews are good at revealing certain aspects of the state of things at this time, but not so good at telling us what happened during the 1970s and the first part of the 1980s. We can, of course, make inferences about the past based on what we're told about the present, but the present is sufficiently different from the past, especially from the early 1970s, to make that exercise questionable.

Initially, I was going to argue that demographic change in the industry was an entirely benign process, wherein incoming Latino immigrants assumed positions that were being abandoned by natives, White, Black, and Latino. Examination of data on the experienced unemployed/non-employed seems to contradict this position, but I'm not sure if I want to include that data (it's iffy and somewhat of a digression). I want to trace the history of the demographic transition because that's the history of the present (or the present of the interviews)--the context within which the interviews make sense.

The article can be about what employers do, taking their word for it. The article can be about what employers think, taking their word for it, more or less. We can compare what they say they do and think against our understanding of discrimination and against our understanding of labor/job queues. But this doesn't answer our question about what accounts for demographic change.

OK, I'm completely running out of time ...

The main thought I had that I wanted to get down here was that we can talk about job queues and labor queues. Job queues are what workers construct when they evaluate jobs relative to each other (in the abstract), placing, e.g., a job as a supervisor in an ice cream factory above, e.g., scrubbing toilets in a hospital. Labor queues are what managers construct when they evaluate workers relative to each other (in the abstract), placing, e.g., women above men and Latinos above Blacks for hotel housekeeping positions. My contention would be that in 1970, hotel housekeeping jobs were higher in the job queues of native workers than they were in 2000. This is for a few reasons, including the fact that native workers were better-educated in 2000 than they were in 1970, leading to increased job prospects (and aspirations); that more occupations were open to women in 2000 than in 1970, so that women (who comprise almost all of the housekeeping workers) had more alternatives, and finally because, as Piore suggested, these jobs have become immigrant jobs and are therefore seen as undesirable. (Alternatively, they could be seen as undesirable immigrant jobs because immigrant dominance makes the jobs difficult to obtain as a result of linguistic shift, immigrant hostility, and other processes of social closure.)

Argh! This isn't getting me anywhere. Another place to start is with the bad attitudes/good soldiers dichotomy. In this view, working class workers are all sneered at by employers, but not for the same reasons. Black workers are thought of as too uppity and ambitious, whereas Latino workers are thought of as hard-working but unambitious. These are employer impressions of workers that (especially for the Latinos) do not correspond to notions of discriminatinon. They do suggest an ordering for the labor queue (they may disapprove of low ambition among housekeepers, but they really do count on it, so it's not truly a liability for them to hire people with low ambition except in the long run). OK, that's got to be it ...

Strongly Agree

My name is Michael Lichter, and I am an Assistant Professor of Sociology. Welcome to my blog, “Somewhat Agree.” This blog was created for my benefit and for the benefit of the students taking my Spring 2006 “Survey Research” course, otherwise known as SOC578 and SPM517. This course meets (much too early!) on Tuesday mornings in 474 Park Hall at the North Campus of the University at Buffalo, State University of New York.

My primary reason for starting this blog is to provide an example for my SOC578/SPM517 students. I don't expect this example to be particularly stellar, but I suspect it will still be good enough. My goal is to use the blog as a “file” in the sense of C. Wright Mills in his essay “On Intellectual Craftsmanship,” from his classic book The Sociological Imagination.

Over time, I will be adding the various elements that Mills suggests we keep in our files, but the point for the moment this blog is just a bookmark or place-keeper. This is where my stuff will go as I generate it.

What does the title of the blog signify? “Somewhat agree” is a common response option in surveys, usually found just to the right of "Strongly agree" and to the left of “Somewhat disagree” (or “Neither agree nor disagree” in surveys that are nice enough to provide a neutral option rather than forcing people to take a stand).

Like all ordinal response categories, “somewhat agree” is ambiguous. If I somewhat agree, how much less do I agree than if I “strongly agree”? Also, if I somewhat agree, doesn't that imply that I somewhat disagree? Yet, “somewhat disagree” is considered to be very distinctly different from “somewhat agree.” As a researcher, I’d prefer not to have the ambiguity, but as a blogger I welcome it. C’est la vie.