In January, I wrote a blog entry about job dissatisfaction , in which I discussed the results of an annual survey by the Conference Board  that found only 45 percent of respondents were satisfied with their jobs. In fact, the Conference Board has found job dissatisfaction rising ever since the survey was first fielded in 1987. Lots of dissatisfied workers, right?
That’s what I thought, until I came across another survey, this one by the National Opinion Research Center  at the University of Chicago (where I earned my master’s). This survey finds 86 percent of Americans satisfied with their work; and among these happy workers, 48 percent consider themselves “very satisfied.” Only 4 percent consider themselves “very dissatisfied.”
The report on the NORC findings notes that “job satisfaction is higher among those with more education, more prestigious occupations, and better-paying positions.” No surprise there, nor with their finding that “job satisfaction is greater among those who are still working after age 65 (71% very satisfied).” The workers who have not retired by then are understandably more likely to be those workers who enjoy their jobs. Satisfaction is lowest among workers under 29, with only 42 percent saying they’re very satisfied. The report comments, “In general, job satisfaction increases with age. This results from a combination of upward career mobility and from workers finding out what type of job best suits them.”
The study also found a lower level of satisfaction among African Americans than whites (40 percent, as opposed to 53 percent) and among Hispanic Americans than non-Hispanic Americans (46 percent, as opposed to 51 percent). Satisfaction for women averages out to almost exactly the same as for men; slightly more women are very satisfied, but slightly more women are very dissatisfied.
Like the Conference Board’s surveys, NORC’s goes back several decades (to 1972), and they find little change in level of satisfaction over that interval.
You may wonder how the Conference Board and NORC can find such different results from their research. But if you follow political polling at all, you know that the way a questionnaire is worded can have a great impact on the outcome. I don’t have access to the complete questionnaires for both studies, so I don’t know what differences may exist between their survey questions.
Both surveys were of households, rather than of workplaces, so one would expect the samples to be similar. I notice that the NORC study included homemakers, part-time, and even unemployed workers in its sample, but I don’t know whether the Conference Board considered these workers among the “jobs” it covered. On the other hand, if the samples do differ in this way, one would expect the NORC respondents to be less satisfied, because “very satisfied” responses were less frequent among part-time (46 percent) and unemployed (37 percent) workers than among full-time workers (51 percent), and only slight more frequent among homemakers (52 percent).
I’m sorry to say that I don’t have enough information to make a good hypothesis about why the two surveys differ so greatly in their findings.
In my next blog entry, I’ll share a couple of my favorite anecdotes about statistics. That may sound like a dry topic, but these anecdotes will surprise you.