Top of Form 1

Logged in as: echidne (Logout

Bottom of Form 1

##

Overtime, while something they compensated for, strikes me as something that bears investigation.

Companies often have a limited number of overtime hours available. In these cases, the overtime may be offered first tot he people who work well without supervision, are more experienced, or who are perceived to be these things, or the people the boss likes, best, feels sorry for, or expects to have evenings and weekends flexible.

Which is to say, the overtime hours may never even be offered to the woman trying to build her retirement account when the boss is giving Timmy the Fasttrack boy the chance to prove himself.
Samantha Vimes |   | Email | 01.23.06 - 7:00 am | #

#

A good point, Samantha Vimes. The study controlled for working hours so it should have controlled for overtime in that, though.
Echidne of the snakes |   | Email | Homepage | 01.23.06 - 7:24 am | #

#

Excellent posts. You have a clear, concise writing style and do a good job explaining complex techniques.

One thing I was wondering about was wage compression. I was under the impression that the longer you stayed with a particular employer, the less likely you were to get a big raise -- that you'd have to let's say go to another company to get that bump.

Are men more likely to switch employers than women? Or are they more likely to get promotions when they stay with the same employer.

The reason I wonder was all the talk in the previous post of "risk-taking".

Thanks,
MG
MG |   | 01.23.06 - 8:47 am | #

#

MG, I'm not sure about the effect of tenure with the same employer. I remember seeing one study which found out that men were rewarded more for staying with the same employer, while women were rewarded more for length of experience in general. What the "rewarded more" means here is that for each extra year with the same employer men got a bigger wage increase whereas women were rewarded more than men for each extra year of general experience.

It would still be true that men earned more, on average, and I don't remember seeing these results repeated in any other study. So they may have been a fluke. In general I want to see findings repeated at least once before I discuss them.

And I don't know if this is what you meant.

Thanks for the praise. Much appreciated.
Echidne of the snakes |   | Email | Homepage | 01.23.06 - 9:47 am | #

#

After thinking about your comment a little more, MG, I believe that what you are describing would work more in white-collar occupations than the rest, and the rest are more numerous. But I may be wrong about that.
Echidne of the snakes |   | Email | Homepage | 01.23.06 - 9:57 am | #

#

I have one question, and one request for a link or citation.

Question: You use the figures '80%' and '20%' to characterize wage differences and/or what is left unaccounted for by the variables analyzed in this study. I'm a bit confused about what these numbers represent at the different times you use them -- how much less women make than men or how much more men make than women, whether these numbers are different simpliciter or if this is what's left after analyzing away the variables controlled for, and so on. This is particularly opaque in light of your point from part I, that the gap can be measured in various ways, and the current standard is rather awkward. Perhaps you could clarify the results a little?

Request for a citation: I'm interested in these sorts of studies, looking at academia in particular, and especially those that consider different sectors of the academy ('hard' sciences, social sciences, humanities, etc.). Can you make any recommendations?

Excellent post. I look forward to reading part III. Thanks again for doing this!
Noumena |   | Email | Homepage | 01.23.06 - 10:57 am | #

#

Question: You use the figures '80%' and '20%' to characterize wage differences and/or what is left unaccounted for by the variables analyzed in this study. I'm a bit confused about what these numbers represent at the different times you use them -- how much less women make than men or how much more men make than women, whether these numbers are different simpliciter or if this is what's left after analyzing away the variables controlled for, and so on. This is particularly opaque in light of your point from part I, that the gap can be measured in various ways, and the current standard is rather awkward. Perhaps you could clarify the results a little?

The initial gender gap, with nothing controlled for, showed men earning 44% more than women. That is how the researchers formulated it. After they controlled for all the variables I discuss they were left with men still earning 20% more than women. The latter figure would be the unexplained residual, the part that was not explained by the variables the study used.

Request for a citation: I'm interested in these sorts of studies, looking at academia in particular, and especially those that consider different sectors of the academy ('hard' sciences, social sciences, humanities, etc.). Can you make any recommendations?

Let me think about this a little. Would the links have to be freely available on the net or would academic journals be ok?

I'm very glad that you liked the post. It's hard work, you know.
Echidne of the snakes |   | Email | Homepage | 01.23.06 - 11:03 am | #

#

Echidne, great series! Looking forward to part 3.

One reference for a study in Academia: "Gender and Racial/Ethnic Differences in Salary and Other Characteristics of Postsecondary Faculty: Fall 1998 Statistical Analysis Report September 2002" from National Center for Education Statistics. Here is an excerpt from the study:

"After controlling for race, type of institution, teaching field, level of instruction, tenure status, rank, highest degree, years since highest degree, age, time spent teaching, number of classes taught, time spent engaged in research, and number of total publications or other permanent creative works in the previous 2 years, full-time female faculty members earned an average of $53,600 compared with $58,700 for men."

Note that several of the factors that were controlled for are themselves subject to discrimination.
Kali |   | Email | 01.23.06 - 1:47 pm | #

#

So if all variables were accounted for, and you found that women, on average were making more than men, the conclusion that you could draw was that there was systematic discrimination against men, right?
Tony |   | Email | Homepage | 01.23.06 - 2:38 pm | #

#

So if all variables were accounted for, and you found that women, on average were making more than men, the conclusion that you could draw was that there was systematic discrimination against men, right?

Right.
Echidne of the snakes |   | Email | Homepage | 01.23.06 - 3:44 pm | #

#

Great series, Echidne! Thanks for all your work.

And I'm afraid I have a seemingly simple explanation for the ongoing discrimination of women in the workforce and elsewhere: all too many women put up with it and see nothing wrong in being classified as second-class people. Comes from observations over the last four decades or so ..
Helga Fremlin |   | Email | 01.23.06 - 5:32 pm | #

#

Academic journals would be okay -- grad student with access to a nice library and all that -- but since I'm lazy and don't want to be bothered, webpages are preferable. :-D

Kali, thanks for the cite!
Noumena |   | Email | Homepage | 01.23.06 - 9:44 pm | #

#

Curses, not in my school's library. Oh well, there's an interesting one on the gender of grad students, and another one on the gender and racial breakdown of faculty. Perhaps Echidne will bless us with more sources soon. Ah well, I still have plenty of procrastination fodder for the time being.
Noumena |   | Email | Homepage | 01.23.06 - 9:55 pm | #

#

Noumena, there are loads of academic articles that you could look at, but the only one I found on the net within a short search time was this one on the UK academia. It still discusses a lot of shared issues with the US situation:

Gender Gap in UK Universities

Then there are lots of articles, some better than others. A few that you might be interested in looking at are:

Ginther, Donna K. 2004. "Why Women Earn Less: Economic Explanations for the Gender Salary Gap in Science" AWIS Magazine (Winter 2004) 33:1, 6-10.

Ginther, Donna K. and Kathy J. Hayes. 2003. "Gender Differences in Salary and Promotion for Faculty in the Humanities, 1977-1995." Journal of Human Resources.

But none of these are the very best ones.
Echidne of the snakes |   | Email | Homepage | 01.23.06 - 9:59 pm | #

#

Here are a few books on the general topics of my posts for those who wish to study the field more:

Joyce P. Jacobsen, The Economics of Gender

Jane Humphries, Gender and Economics

Francine Blau, Marianne Ferber and Ann Winkler: The Economics of Women, Men and Work.

I like the Jacobsen book, but it's a little outdated now. It's still good for theory and maybe there is a new edition out.

The Humphries reference is to a collection of classical articles in the field. It is pretty technical.

I haven't read the Blau et al.
book myself, but the writers are fairly balanced scholars of the field and the book has information on more recent events.

It's always a good thing to remember that people have their own biases, even researchers.
Echidne of the snakes |   | Email | Homepage | 01.23.06 - 10:19 pm | #

#

So if all variables were accounted for, and you found that women, on average were making more than men, the conclusion that you could draw was that there was systematic discrimination against men, right?

Right.


Well I do have to say you are being intellectually honest. Thank you.
Tony |   | Email | Homepage | 01.23.06 - 11:03 pm | #

#

What I meant was, they *did* specifically control for overtime.

But overtime has subjective elements in how it is given as well as how it is received, and in many jobs it makes for a large amount of extra income. Which means the gender gap is larger than what they measured, and some of it may be due to bias.
Samantha Vimes |   | Email | 01.24.06 - 7:21 am | #

#

Thank you for these posts on the wage gap. It's great to see economics taken out of the hands of conservatives. I look forward to the 3rd. The second will always be my favorite though because I did data collection for the PSID.
INWER |   | 01.24.06 - 12:19 pm | #

#

I've really enjoyed reading these posts and comments and look forward to the next.
Cynthia |   | 01.24.06 - 12:27 pm | #

#

I know I'm really late Echidne, but thank you for this wonderful series and I am looking forward to part three. 9For some reason you were blocked from my internet access for a while and I couldn't get to you from my bookmark. I could, however, get to you from a link from another site. Weird, huh?)

BTW, I love these posts that you think are long and dry. I know it's not your normal "voice" for the blog, but in my opinion, your "dry" posts read like fresh mountain spring water.
hanna joergel |   | 01.24.06 - 11:32 pm | #

#

Thanks, Cynthia and hanna joergel. That is so kind of you. I will add these comments to my bragfile for the rainy day.
Echidne of the snakes |   | Email | Homepage | 01.25.06 - 1:20 am | #

#

And thanks, INWER, too.

Let me know if you have any suggestions for improvement.
Echidne of the snakes |   | Email | Homepage | 01.25.06 - 1:21 am | #

#

I'm probably just being dense here, but I don't see how holding constant the variables of marital status and having children between genders can fully control for the differences caused by this if society views taking care of house and children as the woman's duty.


Minor tweaks that mothers might do to balance home and work might be intrepreted by the boss to mean they aren't as job-oriented.

For instance, my old work had very flexible hours, and my co-workers who were mothers worked full time, but one came to work at 6am and left at 3pm. The other worked slightly less during the week, but came in on Saturday. Since the boss worked from about 9 to 6, they weren't around for much of the time that he was, which could affect his opinion of their diligence, and therefor their pay.

My brother (who is a father), however, adjusts his schedule to coincide with his boss's. If his boss leaves at 5pm, he might leave at 5:15.

Other factors that may affect the opinion of the boss that don't seem to be controlled in this study are personal phone calls from work, more random "vacation" days used to take care of family business, and lessened ability to adjust personal schedule for work.

These don't seem to be controlled for in the study. Would judging work on these factors be considered discrimination since they don't necessarily affect work output, only the opinion of the boss?
Agent Bucky |   | Email | 01.25.06 - 9:09 pm | #

#

Agent Bucky, many of the factors you list would be captured by the fact that occupation is held constant in this study. IF certain occupations allow more flexibility and IF mothers choose them for that reason, then the occupation variables should pick up things like flexitime and a relaxed attitude about taking time off from work.

Other factors that may affect the opinion of the boss that don't seem to be controlled in this study are personal phone calls from work, more random "vacation" days used to take care of family business, and lessened ability to adjust personal schedule for work.

These don't seem to be controlled for in the study. Would judging work on these factors be considered discrimination since they don't necessarily affect work output, only the opinion of the boss?


Unless they affect output the behaviors would be discriminatory. As I mentioned in the first part of the series, it is very difficult to measure output in most white-collar work, and even in blue-collar work the output often depends on some team and machines used together, so it's hard to say how much one person does. In many places the boss's opinion is used as a measure of a worker's productivity, because he or she sees the worker at work more. But the problem with this is personality conflicts and also the possibility that the boss is a bigot.

I'm probably just being dense here, but I don't see how holding constant the variables of marital status and having children between genders can fully control for the differences caused by this if society views taking care of house and children as the woman's duty.

It doesn't control for that fully, but it should take out an enormous chunk of any household-chore related effects, especially as the study also controlled for working part-time and the occupation of the worker. Nothing controls for every single theoretical possibility completely, but it's probably safe to say that some proxies do wipe up an awful lot of the possible mess.

Using these measures does NOT tell us if the women who work part-time, say, do this from their own choice or because they are driven into it by family and societal pressure. So you could argue that I'm giving the wingnut explanations the maximum leeway in this series by showing how they fail even in the most generous interpretation of what is going on.
Echidne of the snakes |   | Email | Homepage | 01.25.06 - 10:14 pm | #

#

This isn't dry at all. It's impressively clear and very interesting. Moving on to the next section now. . .
Oscar |   | 01.26.06 - 9:47 am |