This is an important issue, Lisa, and I am glad that you raised it. If young professors are trained in empirical methods, there is no reason they should refrain from using them. It is, of course, the norm for untenured professors in other departments to gather and analyze data. The problem lies in getting the training while the tenure clock is ticking. Not everyone can pull that off.

I agree with Gordon, especially about the issue's importance for younger (read: untenured) law faculty. I would only add that emerging law faculty hiring and legal scholarship trends suggests that, at some point, this issue's salience might fade over time.

As an untenured person doing (pretty much exclusively now) empirical work, I would like to offer a little push back.

In the fall of 2001, as I job-talked my first empirical article, I remember hearing the caveat on pre-tenure empirical work from the chair an appointments committee. But two years later, an IU colleague told me, "In this business, we spend a lot of time doing scholarship. You've got to follow your star."

So which advice is better? In my case, I could not follow both. So I picked empiricism because I loved the work.

Granted, I have spent a huge amount of time (and money) learning about statistical methodology, but always in the process of moving an empirical project forward. Further, I would caution against generalities on data collection and the amount of time it takes to write an empirical article. I have often been dumbfounded by the quality of data that can be collected by a little research and 60 to 120 hour of RA time. I have enough original data to write for years, and more and better data in the pipeline. Moreover, there are lots of datasets that are public and underutilized, with lots of still-unexplored research questions. People wholly committed to empirical research don't see nearly as many obstacles as non-empiricists.

Sure, two years from now I may not get tenure (I remain optimistic!), but intellectually, this has been the best five years of my life. Why give in to fear? Follow that damn star.

Some comments from someone who participates in both disciplines. First, you seem to distinguish "empirical" pieces and "theory" pieces. This strikes me as a horrible dichotomy. Quantitative work in the social sciences is not atheoretical (I know you didn't say that); theory motivates any good piece. What social science pieces tend to lack in comparison to law review pieces are (a) fluff and (b) the relative absence of passing off ideas as "original" when in fact they're clear derivations or one step removed copying from another piece or line of work.

On to your points:

"First, the research takes a long time, too long for people on a tenure clock."
--------------------------------------
This is an odd claim. Social science folks are on a tenure clock too, and many produce quite a lot of sound empirical work. Sure, it's relatively easy to fire off a good law review piece; if you mean that doing good social science work takes more effort than doing what passes for good (not all, certainly) law review type work, then you're probably right.
--------------------------------

"Second, data that is not public is extremely hard to get, feeding into the first problem and potentially undermining the saliency of the study."
--------------------------------------
Data is not that hard to collect, generally, if you have focused research questions. And sure there will be tough to collect data, but if the opportunity cost is too high, move on to another project -- there's plenty of them.
----------------------------------------

"Third, the finished written product is generally not that long and possibly too scientific for traditional scholarship, making it difficult to place the article in a traditional law review, and hence potentially undercutting the weight given to the article during tenure review."
------------------------------------------
This may be true, though, of course, the "not too long" is generally a substitute for the fluff that gets thrown in most law review articles. And with the move to shorter articles at the top reviews and competition from outlets like JLE, JLS, JELS, I suspect this either (a) no longer holds much, or (b) is on its way out as a norm.
-----------------------------------

"Fourth, for purposes of external reviews associated with tenure and promotion, it is difficult to find outside people who can evaluate the work. And apparently if you find someone with a social science background who understands how to conduct empirical research, there is the possibility that she will be overly critical if the law professor fails to appropriately defend her methodology for the study."
-----------------------------------
Law professors aren't often punished for shoddy work. If by "overly critical" you mean you will get called out on serious design flaws, model misspecification, or other serious errors, they you're probably right. But avoiding empirical work for this reason is hardly justified -- do good work, think about the methodological issues, and be prepared to justify specifications, assumptions, and the like.

This is such an interesting debate. It seems to me the claims on both sides are correct. If you have a passion for empirical work, you should do it. The challenges are surmountable, the rewards are substantial, and, in fact, almost any good legal research should have an empirical component to it. That said, Lisa’s advice is probably correct as well. Doing empirical work pre-tenure is probably a risky strategy, particularly if you are not already fairly adept at using the machinery of empirical research. Conclusion: the problem is ourselves and our tenure processes.

I would echo Bill H.'s and "Anon's" comments, especially disagreeing with at least the first 3 of the reasons against empirical research by junior people. The thinking may reflect a general - false - perception in the legal academy that "empirical" means "econometric" and "political science" research. A great deal of relevant empirical research is in other social science fields and is experimental (mock jury research as just one example). Taking that approach can give interesting results and, IMHO, belies these 3 reasons.

1. "takes too long" - I'm impressed by the activity in many psychology departments and by those doing "psych & law" research, running and submitting more experimental work in a year or two than many law profs publish before tenure. This ties into

2. "availability of data" - as "anon" suggests, collecting original data is common in law-related contexts. A number of examples at the recent Law & Society Association conference; a vast number at the annual American Psychology/Law Society conference and in primary empirical "law-and-" journals. Certainly the traditional undergraduate subject population is easily available, but many social scientists doing experimental work are using samples from the community, i.e., jury-eligible people for mock jury studies.

3. Article length might be an issue, I suppose, despite the trend nowadays for journals to profess that they want shorter articles. But I know Cardozo, and some time ago the Penn Journal of Constitutional Law, have both stated an interest in empirical pieces, and of course more and more is appearing in law reviews. And I think distilling empirical work - "scientific" work - into comprehensible, legally relevant writing is a great opportunity for clear writing.

4. The external review issue I think is more salient, though increasingly, there are more people available and knowledgeable enough to review empirical work.

So I'm in favor! (Full disclosure, I do this kind of stuff.)

To amplify on the comments a little

The tenure standards at law schools are clearly less than in other more empirical fields, but those other areas provide more support, i.e. students, for the research. I would think the extent of law school support, such as through RAs, would be relevant to this decision.

There are many datasets publicly available for research on various topics, so the data collection problem can be exaggerated.

There's no intrinsic reason why it should take more time. The problem I see is the learning curve for law professors who are not yet empirically trained. Even in this csae, though, descriptive statistics are pretty easy and can represent a valuable empirical contribution.

I don't think there is any shortage of outside people to evaluate this research any more.

The "shorter article" issue is in part an artifact of the requirements of particular publication outlets. Empirical articles in law reviews are not generally so short. See Heise, Morriss & Sisk in NYU LR or Brudney & Ditslear in Vanderbilt LR as examples. Moreover, this problem should be dispelled by the recent law review preferences for shorter articles.

Good heavens, that's such an indictment of law schools! While I have my skepticism about "Emprical Legal Studies" as a movement, the notion that law faculties (and of course the students who edit the law reviews, godloveem) are unable to find anyone to review the work who understands it except Mean Evil Social Science Researchers who ask the poor beleagured law professor to -- gasp -- justify the methodology they use and thus ensure they will never get tenure is just horrrifying. If it really is that bad, the legal academy needs to undergo some serious reform.

I agree with most of the "push back" offered by Henderson, anon, Blumenthal, and Cross. I would like to add that perhaps "Should Young Scholars Engage in Empirical Legal Research?" is not the correct question to be asking. I often hear from fellow junior faculty that they "want" to engage in empirical work, but are concerned about the perceived potential costs (time, tenure, etc.). While I agree with Frank that start up costs to becoming empirically trained are potentially high, I also think that if a junior scholar wants to engage in empirical work or is already trained, the proper question might be: "If we think empirical scholarship adds value, what can institutions do to encourage young scholars to engage in empirical legal research?" The potential answers to this question (e.g., available RAs, on-site statistician/methodologist, funding for ISCPSR, ecourage co-authored work) would encourage empirical work from both junior and senior faculty.

Since I think it might be a more fruitful endeavor to discuss how junior scholars might reduce the roadblocks to empirical work, I thought I'd note that the Empirical Legal Studies Blog (www.elsblog.org) blog has offered suggestions before:

http://www.elsblog.org/the_empirical_legal_studi/2006/05/new_empirical_l.html

Thanks for all the comments. I think this is an important question and Jason is right that we should perhaps reframe it to ask how one best gets support for such research. I definitely think people should follow their passion and write the kind of scholarship that appeals to them. Also, the reasons given against such research pre-tenure may be generalizations, and hence people may have different experiences on this issue. However I would still think that a young scholar needs to be aware of some of the roadblocks so that she or he can at least find ways to navigate around them. Indeed, there were both professors and deans in the room who--though seeming supportive--also seemed to suggest that scholars needed to be cautious on this issue.

In regard to the timing, I do think part of the problem is a steep learning curve, but my perception of the discussion on this issue was that it just took a long time to gather data--particularly non-public data. (Maybe this is because we are not as clear about where the look). But even after people had found the source of data, it appeared to take much longer than people had anticipated, particularly when people were doing surveys or sifting through a lot of data. I think people found this particularly disconcerting when they were comparing themselves to faculty members producing quality scholarship at a higher rate.

My point about doing a theory piece reflected one response to the timing problem. Not that the theory and data were disconnected, but rather that young scholars could write a theory piece first outlining their ideas about why the data was important and what they thought the data would reveal. Some people even suggested that you could do this more than once--that is, if data was generated appearing to contradict a thesis, an author could write about that in a preliminary manner. This enables the author to get something on record--keeping in mind that people need to evaluate your writing and if you have little to nothing written, then that evaluation cannot occur.

Ultimately, the issue may be, how do law schools get comfortable with different forms of scholarship so that the people producing it do not feel less secure than their counterparts about their tenure prospects.

Jeremy Blumenthal's comment raises the following issue for me: In most (not all) prominent schools, the only people who do empirical work are law-and-econ types (or law-and-politics people who do the kind of political science that is like economics). People like that often think of experimental or sociological research as insufficiently rigorous. So an untentured person who does those kinds of empirical research might end up screwed, because the only empiricists on the faculty will dismiss the genre. What it takes is for senior faculty to stand up for that kind of research (which has to be part of a more general project of standing up for methodological pluralism).

I want to join the push-back club. There is no risk-free tenure strategy. Theory work can be boring. Or trivial. Or wrong. Or it can step on the tender toes of the senior person in the field who then feels compelled to squash the offending scholar. Or it can be too wishy-washy and satisfy no one. It isn't as if the alternative work comes with a put-in-a-nickle-get-out-a-successful-article outcome.

I think schools all over the country are waking up to the importance of empirical work. Even that soft social science sort of work can cut the legs out from underneatn old debates and introduce new insights that open debates.

Recently I've watched people get tenure based on what are basicly projects in mid-stride (good projects, to be sure). I've also listed to colleagues cut a LOT of slack for imperfect studies from junior faculty. And I've seen a lot senior empiricists step up to the plate to help the juniors.

Of course you should do whatever kind of work you want to do because it will be infused with your enthusiasm. But I'd argue that even a coolly rational actor might want to consider who is on the fence might want to tell a story backed up by data.

Prof. Warren said the very same things at a conference I attended on the HLS campus a short while ago. While I can't comment on the tenure process, I am in the middle of empirical training (Ph.D. in social science). Good empirical work, both quantitative and qualitative, takes time--but much of that time comes from learning the skills to be used. Trying to learn high quality social science as a junior faculty member sounds daunting to say the least. However, those with the training already in hand shouldn't face a disabling time crunch.

Ultimately, Warren gets it right in her last paragraph; do the work you are enthusiastic about doing (especially if you're an empiricist).

Lisa Fairfax started off this discussion by reporting on advice that junior scholars should not do empirical research. Whenever I want to utter the words "junior scholars should (or should not) do X," I try to remember to ask myself whether that is because it is good advice or whether that is because it helps me. For example, I strongly have the urge to recommend that junior scholars should wash the cars of more senior faculty because it promotes interaction with your colleague, but that runs afoul of my test.

Early in my career, a much more senior colleague suggested that one did not want to publish too much early in one's career because more articles could lead to more criticism. Was perhaps this comment motivated by the fact that highly productive junior faculty made less productive faculty not look so good? I certainly don't think conspiratorial intent or malevolence was present. Rather, we all tend to see what is the best for the world through the prism of self-interest.

I wonder how much of the advice against empiricism for junior faculty might stem from similar motivations. "Don't do empirical scholarship" can be shorthand for "do the sort of scholarship I do." Also, as someone who does empirical work, it helps me to hear that empirical work is so difficult and time-consuming that junior scholars need not attempt it! I am sure the advice against doing empirical work was sincere and in good faith. But we do work in a profession with its own special interests, and we should remember the motivations for any career advice we receive.

I think the advice at the top of this thread is outmoded, but the cautions are still worth keeping in mind. In the Eighties, Sullivan, Warren, and I often suggested that untenured faculty be careful about relying too heavily on empirical work before getting tenure, but that was at a time when empirical work was being done by very few in the legal academy. The wheel of fashion has turned and empirical articles are now welcome. In addition, there are now well-established approaches to locating existing databases and to gathering original data that did not then exist. Perhaps most important, there are now more senior faculty who feel comfortable in evaluating such work and are able to advise concerning the feasibility of specific projects. For all those reasons, I think untenured faculty should not consider empirical work out of bounds.

Having said that, there remain risks with empirical work that don’t exist in conventional scholarship. It is easier to have a project fail more or less completely. I recognize that the same fate befalls some conventional work. That generally happens when either an empirical or a conventional project is too ambitious or is undertaken without sufficient preliminary work to determine its feasibility. I think there is a somewhat greater chance of that happening with empirical work, in part because few law professors have been trained in it.

Empirical projects can take too long. It is especially important to be sure that an empirical project can be completed, or at least reach a stage of interesting first results, within the time appropriate for a tenure piece. It is also important to work the idea through with experienced people at the start to be sure that interesting results will not require statistical expertise or project resources that the investigator may not have.

Speaking frankly, there is everything to be said for doing some conventional work in parallel to your empirical work. The latter often has phases or cycles that leave periods of time when other work can be done. During our first empirical project, I wrote two conventional policy articles before our first empirical report appeared. Even on a faculty that is very open to different genres, that fact made life easier for everyone. On faculties with less flexibility, it may be a good idea to canvas a bit to be sure no one will reject empirical work out of hand.

Well, I feel too young to be "outmoded" and I am not sure why so many are using the word "pushback" on this thread. My point was cautionary and directed only at those who have no prior training in empirical work: "Not everyone can pull that off."

Geoff McGovern said it better: "Good empirical work, both quantitative and qualitative, takes time--but much of that time comes from learning the skills to be used. Trying to learn high quality social science as a junior faculty member sounds daunting to say the least."

If you expect someone without prior methods training to do sophisticated empirical work, you are basically asking them to complete PhD-level training while producing a package of tenurable publications. Of course it's possible, but don't downplay the risks or you are doing junior faculty a grave disservice.

I think Jay Westbrook offers wise counsel: "there is everything to be said for doing some conventional work in parallel to your empirical work."